Power supply system and method for controlling multiple power converters using composite efficiency

Abstract

A power supply system and a method for controlling a plurality of power converters using a composite efficiency are provided. The power supply system includes: a plurality of power converters configured to deliver a system load current to a load, the system load current being a combination of individual load currents provided by the plurality of power converters; and a system performance controller configured to detect a value of the system load current. The system performance controller is configured to determine the value of the individual load current using power loss information so that the composite efficiency reaches a threshold condition. The system performance controller is configured to generate a control signal to operate the plurality of power converters at the determined value.

Description

[0001] Cross-references to related patent applications

[0002] This application claims priority to U.S. non-provisional application No. 16 / 794,820 filed on February 19, 2020, the entirety of which is incorporated herein by reference. In addition, this application claims priority to U.S. non-provisional application No. 16 / 547,803 filed on August 22, 2019, the entirety of which is incorporated herein by reference.

[0003] In addition, this application claims priority to U.S. Provisional Application No. 62 / 823,441, filed on March 25, 2019, and U.S. Provisional Application No. 62 / 854,527, filed on May 30, 2019, each of which is incorporated herein by reference in its entirety. Technical Field

[0004] The present disclosure relates to the field of power supply technology, and in particular to a power supply system and a method of controlling a plurality of power converters using composite efficiency. Background Art

[0005] A power system may be designed to achieve one or more performance goals (e.g., high efficiency, high speed, low noise, low emissions, etc.) However, manufacturing variations in system components and / or effects on the system during system use (e.g., thermal effects, component wear) may affect achievement of those goals. Summary of the invention

[0006] According to one aspect, a power system includes: a power stage; a power controller configured to control the operation of the power stage; a metering circuit configured to sense a measurement condition of the power stage; and a system performance controller configured to be coupled to the power controller and the metering circuit. The system performance controller is configured to set or adjust a control parameter of the power stage based on an energy conversion efficiency of the power stage. The system performance controller includes: an efficiency calculation circuit configured to calculate the energy conversion efficiency of the power stage based on the measurement condition; and a control manipulation module configured to modify the control parameter until the energy conversion efficiency reaches a threshold condition.

[0007] According to various aspects, the power supply system may include one or more of the following features (or any combination thereof). The system performance controller may select a combination of control parameters that causes the energy conversion efficiency to reach a threshold condition. The system performance controller may determine a first value of the control parameter that causes the energy conversion efficiency to reach the threshold condition for a first environmental condition, and determine a second value of the control parameter that causes the energy conversion efficiency to reach the threshold condition for a second environmental condition. The first environmental condition may be a first input voltage, and the second environmental condition may be a second input voltage, wherein the second input voltage is different from the first input voltage. The control manipulation module may iteratively modify the control parameter by selecting different values ​​for the control parameter based on an artificial intelligence (AI) algorithm. The control manipulation module may apply historical data to a neural network of the AI ​​algorithm to predict the value of the control parameter that causes the energy conversion efficiency to reach the threshold condition. The control manipulation module may: select a first value for the control parameter; send the first value to the power supply controller; based on the measurement condition, calculate the energy conversion efficiency of the power level operated according to the first value; modify the first value to obtain a second value in response to the energy conversion efficiency not reaching the threshold condition; send the second value to the power supply controller; and based on the measurement condition, calculate the energy conversion efficiency of the power level operated according to the second value. The system performance controller may update the control parameters in response to a triggering event. The triggering event may include detection of a change in one or more environmental conditions of the power stage. The triggering event may include detection of an energy conversion efficiency not meeting a threshold condition. The triggering event may include activation of the power stage.

[0008] According to one aspect, a power system includes: a power stage; a power controller storing control parameters to control the operation of the power stage; a metering circuit configured to sense a measurement condition of the power stage; and a system performance controller connected to the power controller and the metering circuit. The system performance controller is configured to periodically monitor the energy conversion efficiency of the power stage during operation of the power stage and periodically update the control parameters. The system performance controller includes: an efficiency calculation circuit configured to calculate the energy conversion efficiency of the power stage based on the measurement condition; and a control manipulation module configured to modify the control parameters until the energy conversion efficiency reaches a threshold condition. The system performance controller is configured to provide updated control parameters to the power controller.

[0009] According to various aspects, the power supply system may include one or more of the above / below features (or any combination thereof). The measurement conditions may include an input voltage, an input current, an output voltage, and an output current. The metering circuit may include one or more analog-to-digital converters configured to convert the measurement conditions from an analog format to a digital format. The control parameters may include zero voltage switching (ZVS) control parameters, switching frequency, pulse width, duty cycle, or dead time. The power stage may be a first power stage, and the power supply may include a second power stage, wherein the control parameters include phase change parameters related to phase reduction or phase addition.

[0010] According to one aspect, a method for controlling a control parameter of a power stage using energy conversion efficiency includes: detecting a trigger event; selecting a value for the control parameter based on the energy conversion efficiency, including: receiving a measurement condition of the power stage; calculating the energy conversion efficiency of the power stage based on the measurement condition; and modifying the control parameter until the energy conversion efficiency reaches a threshold condition. The method includes providing the selected value to a power supply controller to control the power stage using the selected value.

[0011] According to various aspects, the method may include one or more of the above / below features (or any combination thereof). The modification may include applying historical data to a neural network of an artificial intelligence (AI) algorithm to predict a combination of control parameters that results in the energy conversion efficiency reaching a threshold condition. The control parameters may include at least one of a zero voltage switching (ZVS) control parameter, a switching frequency, a pulse width, a duty cycle, or a dead time. The triggering event may include activation of a power stage, detection of a change in one or more environmental conditions of the power stage, expiration of a timer, or detection that the energy conversion efficiency of the power stage does not reach a threshold condition.

[0012] According to one aspect, a system performance controller for controlling a power stage of a power system includes: a computing circuit configured to monitor a performance metric of a power stage by receiving a measurement condition of the power stage and calculating the performance metric based on the measurement condition; and a control manipulation module configured to execute an artificial intelligence (AI) algorithm in response to a triggering event to identify a value of a control parameter of the power stage that causes the performance metric to reach a threshold condition. In some examples, the triggering event includes activation of the power stage, detection of a change in one or more environmental conditions of the power stage, detection that the performance metric does not reach a threshold condition, or detection of the expiration of a timer. In some examples, the AI ​​algorithm is a target search algorithm. In some examples, the AI ​​algorithm is an applied learning algorithm.

[0013] According to one aspect, a power system includes: a power stage including a power switch and an inductor; a power controller connected to the power stage; a metering circuit configured to sense a measurement condition of the power stage; and a system performance controller configured to be coupled to the power controller and the metering circuit. The system performance controller is configured to calculate an energy conversion efficiency based on the measurement condition and select a value for a zero voltage switching (ZVS) control parameter that causes the energy conversion efficiency to reach a threshold condition. The ZVS control parameter indicates the magnitude of a reverse current flowing through the inductor to discharge a parasitic capacitance of the power switch.

[0014] According to various aspects, the power supply system may include one or more of the above / below features (or any combination thereof). The ZVS control parameter may be a current sensing voltage threshold indicating the magnitude of the reverse current. The ZVS control parameter may include a time value for delaying the activation of the power switch so that the magnitude of the reverse current causes the parasitic capacitance of the power switch to discharge. The ZVS control parameter may be a first ZVS control parameter, and the system performance is configured to select a combination of values ​​of the first ZVS control parameter and the second ZVS control parameter that result in the energy conversion efficiency reaching a threshold condition. The second ZVS control parameter may include a time delay. The power switch may be a pulse switching modulation (PWM) field effect transistor (FET). The system performance controller may include: an efficiency calculation circuit configured to calculate the energy conversion efficiency of the power stage based on the measurement conditions; and a control manipulation module configured to execute an artificial intelligence (AI) algorithm to iteratively modify the ZVS control parameter until the energy conversion efficiency reaches the threshold condition. The metering circuit may include: one or more voltage sensors configured to sense the input voltage and output voltage of the power stage; and one or more current sensors configured to sense the input current and output current of the power stage. The metering circuit may include one or more analog-to-digital converters (ADCs) configured to convert the measured conditions into a digital format. During operation of the power stage, the system performance controller may update the ZVS control parameters in response to a triggering event, wherein the triggering event may include at least one of: detection of a change in one or more environmental conditions of the power stage; or detection that the energy conversion efficiency does not meet a threshold condition. The system performance controller may update the ZVS control parameters in response to activation of the power stage.

[0015] According to one aspect, a power system includes a power stage including a first power switch, a second power switch, and an inductor. The power system includes a power controller storing a zero voltage switching (ZVS) control parameter, wherein the ZVS control parameter indicates a magnitude of a reverse current flowing through the inductor to discharge a parasitic capacitance of the second power switch. The power system includes: a metering circuit configured to sense a measurement condition of the power stage; and a system performance controller configured to update the ZVS control parameter in response to a triggering event. The system performance controller includes: an efficiency calculation circuit configured to calculate an energy conversion efficiency of the power stage based on the measurement condition; and a control manipulation module configured to modify the ZVS control parameter until the energy conversion efficiency reaches a threshold condition.

[0016] According to various aspects, the power supply system may include one or more of the above / below features (or any combination thereof). The triggering event may include detection of a change in an environmental condition of the power stage, wherein the environmental condition includes an input voltage of the power stage. The triggering event may include detection of an energy conversion efficiency not reaching a threshold condition. The metering circuit may include: a multiplexer configured to receive a plurality of analog signals representing at least a portion of the measurement condition and generate a multiplexed analog signal; and an analog-to-digital converter (ADC) configured to convert the multiplexed analog signal into a digital signal. The ZVS control parameter may include a time value for delaying activation of the second power switch such that the magnitude of the reverse current discharges the parasitic capacitance of the second power switch. During switching operation of the first power switch and the second power switch, the ZVS control parameter may cause the first power switch to be activated for a longer time than the second power switch to discharge the parasitic capacitance of the second switch.

[0017] According to one aspect, a method for controlling a zero voltage switching (ZVS) control parameter of a power stage using energy conversion efficiency includes: detecting a trigger event of a power stage, wherein the power stage includes a power switch and an inductor; and updating a value of a ZVS control parameter based on the energy conversion efficiency, wherein the ZVS control parameter indicates a magnitude of a reverse current flowing through the inductor to discharge a parasitic capacitance of the power switch. The updating step includes: receiving a measurement condition of the power stage; calculating an energy conversion efficiency of the power stage based on the measurement condition; and modifying the ZVS control parameter until the energy conversion efficiency reaches a threshold condition.

[0018] According to various aspects, the method may include one or more of the above / below features (or any combination thereof). The triggering event may include: activation of the power stage; detection of a change in one or more environmental conditions of the power stage; or detection that the energy conversion efficiency of the power stage does not meet a threshold condition. The ZVS control parameter may be a first ZVS control parameter, and the updating further includes selecting a combination of values ​​of the first ZVS control parameter and a second ZVS control parameter that results in the energy conversion efficiency meeting the threshold condition, wherein the second ZVS control parameter includes a time delay.

[0019] According to one aspect, a power supply system includes: a plurality of power stages configured to receive an input voltage and generate an output voltage; and a system host device connected to the plurality of power stages. The system host device includes a system performance controller configured to use the overall energy conversion efficiency to set or update one or more control parameters for controlling the plurality of power stages. The system host device includes: an efficiency calculation circuit configured to calculate the overall energy conversion efficiency based on a measurement condition; and a control manipulation module configured to modify one or more control parameters until the overall energy conversion efficiency reaches a threshold condition.

[0020] According to various aspects, the power supply system may include one or more of the above / below features (or any combination thereof). One or more control parameters may include a phase change parameter defining one or more efficiency transition points, which indicate the timing of when to deactivate or activate one or more of the multiple stages. One or more control parameters may include an intermediate bus voltage, which indicates the voltage between two adjacent power stages on the power bus. One or more control parameters may include at least one of an input condition or an output condition. Multiple power stages may be connected in series. Multiple power states may be connected in parallel. The efficiency calculation circuit may calculate the overall energy conversion efficiency based on the input voltage, the input current, the output voltage, and the output current. The system performance controller may update one or more control parameters in response to a triggering event. The triggering event may include: activation of multiple power stages; detection of changes in measurement conditions of multiple power stages; or detection that the overall energy conversion efficiency does not reach a threshold condition. Multiple power stages may include a first power stage and a second power stage. The system host device may be connected to the first power stage via a first power communication bus, and the system host device may be connected to the second power stage via a second power communication bus. The control manipulation module may execute an artificial intelligence (AI) algorithm defining a neural network, wherein the control manipulation module may apply training data to the neural network to predict one or more values ​​of one or more control parameters that result in the overall energy conversion efficiency reaching a threshold condition.

[0021] According to one aspect, a power system includes a plurality of power stages configured to receive an input voltage and generate an output voltage, wherein the plurality of power stages include a first power stage and a second power stage. The power system may include a system host device connected to the plurality of power stages, wherein the system host device stores control parameters for controlling the plurality of power stages. The system host device may include a system performance controller configured to receive measurement conditions from the plurality of power stages and, in response to a triggering event, update the control parameters based on the measurement conditions. The system performance controller includes: an efficiency calculation circuit configured to calculate an overall energy conversion efficiency based on the measurement conditions; and a control manipulation module configured to modify the control parameters until the overall energy conversion efficiency reaches a threshold condition.

[0022] According to various aspects, the power supply system may include one or more of the above / below features (or any combination thereof). The control parameter may include a phase change point for deactivating the first power stage. The control parameter may include a phase change point for activating the first power state. The control parameter may include an intermediate bus voltage, wherein the intermediate bus voltage is a voltage between the first power stage and the second power stage. The control parameter may include an input voltage or an output voltage of the first power stage. The triggering event may include: activation of multiple power stages; detection of a change in a measurement condition; or detection that the overall energy conversion efficiency does not meet a threshold condition.

[0023] According to one aspect, a method for controlling multiple power stages using overall energy conversion efficiency includes: detecting a trigger event of multiple power stages, wherein the multiple power stages include a first power stage and a second power stage connected in series or in parallel with the first power stage; updating the value of a control parameter for controlling the multiple power stages based on the overall energy conversion efficiency, wherein the control parameter includes a phase change parameter, an intermediate bus voltage, or an input condition or an output condition of each power stage, and the updating includes receiving measurement conditions of the multiple power stages, calculating the overall energy conversion efficiency based on the measurement conditions, and modifying the control parameter until the overall energy conversion efficiency reaches a threshold condition. In some examples, the trigger event may include: activation of multiple power stages, detection of a change in the measurement conditions, or detection that the overall energy conversion efficiency does not reach a threshold condition. The overall energy conversion efficiency may be calculated based on input voltage, input current, output voltage, and output current.

[0024] According to one aspect, a power system includes: a power stage; a power controller configured to control the operation of the power stage; a metering circuit configured to sense a measurement condition of the power stage; and a system performance controller configured to be coupled to the power controller and the metering circuit. The system performance controller is configured to set or adjust a control parameter of the power stage based on a standby power of the power stage. The system performance controller includes: a standby power calculation circuit configured to calculate the standby power of the power stage based on the measurement condition; and a control manipulation module configured to modify the control parameter until the standby power reaches a threshold condition.

[0025] According to various aspects, the power supply system may include one or more of the above / below features (or any combination thereof). The standby power calculation circuit may calculate the standby power based on the input voltage and input current of the power stage. The control parameter may include a skip cycle parameter. During the operation of the power stage, the system performance controller may update the control parameter in response to a triggering event. The control parameter may be a first control parameter, and the system performance controller may select a combination of values ​​of the first control parameter and the second control parameter that causes the standby power to reach a threshold condition. The control manipulation module may iteratively modify the control parameter by selecting different values ​​for the control parameter based on an artificial intelligence (AI) algorithm. The control manipulation module may apply training data to a neural network of the artificial intelligence (AI) algorithm to predict the value of the control parameter that causes the standby power to reach a threshold condition. The control manipulation module may: select a first value for the control parameter; send the first value to the power supply controller; calculate the standby power of the power stage based on a measurement condition using the first value; modify the first value to obtain a second value in response to the standby power not reaching the threshold condition; send the second value to the power supply controller; and calculate the standby power based on the measurement condition using the second value. The metering circuit may include: a voltage sensor configured to sense an input voltage; and a current sensor configured to sense an input current. The metering circuit may include an analog-to-digital converter (ADC) configured to convert the monitored condition into a digital format.

[0026] According to one aspect, a power system includes: a power stage; a power controller storing control parameters for controlling operation of the power stage; and a metering circuit configured to sense a measurement condition of the power stage, wherein the measurement condition includes an input voltage of the power stage and an input current of the power stage. The power system may include a system performance controller configured to be coupled to the power controller and the metering circuit, wherein the system performance controller is configured to update the control parameters to minimize the standby power of the power stage. The system performance controller includes: a standby power calculation circuit configured to calculate the standby power of the power stage based on the input voltage and the input current; and a control manipulation module configured to modify the control parameters until the standby power reaches a threshold condition.

[0027] According to various aspects, the power supply system may include one or more of the above / below features (or any combination thereof). The control parameter may include a skip cycle parameter. The system performance controller may update the control parameter in response to a trigger event. The control parameter may be a first control parameter, and the system performance controller may select a combination of values ​​of the first control parameter and the second control parameter that causes the standby power to reach a threshold condition. The control manipulation module may iteratively modify the control parameter by selecting different values ​​for the control parameter based on an artificial intelligence (AI) algorithm. The control manipulation module may apply training data to a neural network of the artificial intelligence (AI) algorithm to predict the value of the control parameter that causes the standby power to reach a threshold condition. The metering circuit may include: a voltage sensor configured to sense an input voltage; and a current sensor configured to sense an input current; and an analog-to-digital converter (ADC) configured to convert the input voltage and the input current into a digital format.

[0028] According to one aspect, a method of controlling a control parameter of a power stage to minimize standby power includes: detecting a trigger event of a power stage; and updating a value of a control parameter controlling the power stage to minimize the standby power of the power stage, wherein the control parameter includes a skip cycle parameter. The updating may include: receiving a measurement condition of the power stage; calculating the standby power based on the measurement condition; and modifying the control parameter until the standby power reaches a threshold condition. In some examples, the standby power is calculated based on an input voltage and an input current. In some examples, the method includes: measuring an input voltage and an input current of the power stage; and converting the input voltage and the input current into a digital format.

[0029] According to one aspect, a power system includes a plurality of power converters, each of which is configured to deliver a system load current to a load, the system load current being a combination of individual load currents provided by the plurality of power converters, and the power system further includes a system performance controller configured to detect a value of the system load current. The system performance controller is configured to determine the value of the individual load current using power loss information so that the composite efficiency reaches a threshold condition. The system performance controller is also configured to generate a control signal to operate the plurality of power converters at the determined value.

[0030] According to various aspects, the power supply system may include one or more of the following features (or any combination thereof). A system performance controller is configured to periodically update power loss information by monitoring input conditions and output conditions of a plurality of power converters and by calculating a composite efficiency based on the input conditions and output conditions. The power loss information includes a plurality of load current combinations within a range of values ​​of the system load current, each load current combination being associated with a corresponding composite efficiency, and each load current combination providing a separate combination of values ​​of a single load current. The system performance controller is configured to select a load current combination that provides the highest composite efficiency from the plurality of load current combinations. In response to a triggering event, the system performance controller is configured to use the power loss information to determine a new value of the single load current so that the composite efficiency reaches a threshold condition. The system performance controller is configured to determine the most efficient power converter among the plurality of power converters based on the power loss information. The system performance controller is configured to set the single load current of the most efficient power converter to a value up to the maximum single value provided by the most efficient power converter and to allocate any remaining amount to the next most efficient converter. The system performance controller is configured to generate or update power loss information in response to a triggering event. The system performance controller includes a control manipulation module configured to iteratively modify at least one of the values ​​of the single load currents, and an efficiency calculation circuit configured to calculate a composite efficiency based on measured input conditions and output conditions in each iteration. The control manipulation module is configured to store the value of the single load current and the corresponding composite efficiency in each iteration. The control manipulation module is configured to execute an artificial intelligence (AI) algorithm using a neural network, and the control manipulation module is configured to apply power loss information to the neural network to predict at least one of the values ​​of the single load current for a next iteration.

[0031] According to one aspect, a power system includes a plurality of power converters including at least a first power converter and a second power converter, the plurality of power converters being collectively configured to deliver a system load current to a load, the system load current being at least a combination of a first load current provided by the first power converter and a second load current provided by the second power converter. The power system also includes a system performance controller configured to detect a value of the system load current, and to determine at least a first value of the first load current and a second value of the second load current using power loss information so that a composite efficiency reaches a threshold condition. The system performance controller is configured to generate a first control signal to operate the first power converter at least at a first value, and to generate a second control signal to operate the second power converter at a second value. The system performance controller is configured to set a first value at the value of the detected system load current, and to control the second power converter to operate in a standby mode. The power loss information includes a plurality of load current combinations within a range of values ​​of the system load current, each load current combination being associated with a corresponding composite efficiency, and each load current combination providing a separate combination of at least a value of the first load current and a value of the second load current. The system performance controller is configured to select a combination of load currents that provides the highest composite efficiency from the plurality of load current combinations. In response to a trigger event, the system performance controller is configured to use the power loss information to determine a new first value of at least the first load current and a new second value of the second load current so that the composite efficiency is maximized. The system performance controller is configured to determine the first value and the second value according to the most efficient converter (MEC) method in response to the system load current being detected to be equal to or less than the maximum load current of the first power converter or the second power converter. The system performance controller is configured to determine the first value and the second value according to the equal sharing method in response to the detected system load current being greater than the maximum load current of the first power converter or the second power converter. The system performance controller is configured to recalculate or update the power loss information in response to the trigger event, wherein the trigger event includes: activation of one or more power converters among the plurality of power converters, detection of a change in one or more measured conditions, detection that the composite efficiency is not maximized and / or expiration of a timer. The system performance controller includes a control manipulation module configured to iteratively modify at least one of the values ​​of the first load current and the second load current in response to the trigger event, and the system performance controller also includes an efficiency calculation circuit configured to calculate the composite efficiency based on the measured input conditions and output conditions at each iteration. The control manipulation module is configured to store the value of the first load current, the value of the second load current and the corresponding composite efficiency at each iteration. The power system includes a power device including a plurality of power converters, a system performance controller, a power controller and a metering circuit.

[0032] According to one aspect, a method for controlling multiple power converters using a composite efficiency includes: delivering a system load current to a load through a power system, wherein the system load current is a combination of individual load currents provided by multiple power converters; detecting a value of the system load current through a system performance controller; using power loss information to determine the value of the individual load currents so that the composite efficiency reaches a threshold condition; and generating a control signal to operate the multiple power converters at the determined value. The power loss information includes multiple load current combinations within a range of values ​​of the system load current, each load current combination is associated with a corresponding composite efficiency, and each load current combination provides a separate combination of values ​​of the individual load currents. The method includes selecting a load current combination that provides the highest composite efficiency from the multiple load current combinations. The method includes detecting a change in the value of the system load current, and using the power loss information to determine a new value of the individual load current so that the composite efficiency reaches the threshold condition. The method includes updating the power loss information in response to a triggering event, wherein the triggering event includes activation (e.g., switching on) of one or more power converters among the multiple power converters, detecting a change in one or more measured conditions, detecting that the composite efficiency is not maximized, and / or a timer expires. The updating includes iteratively modifying at least one of the values ​​of the individual load currents, an efficiency calculation circuit that calculates a composite efficiency based on measured input and output conditions at each iteration, and storing the values ​​of the individual load currents and the corresponding composite efficiency at each iteration.

[0033] According to one aspect, a method for controlling multiple power converters using composite efficiency includes delivering a system load current to a load through a power system, wherein the system load current is a combination of at least a first load current provided by a first power converter and a second load current provided by a second power converter; the method also includes calculating a first integral value of a composite efficiency curve within a load range of the first power converter; calculating a second integral value of the composite efficiency curve within the load range of the second power converter; determining a value of the first load current and a value of the second load current based on the first integral value and the second integral value; and, generating a control signal to operate the first power converter and the second power converter at the determined values.

[0034] In some aspects, the method includes: determining that a first integrated value provides a greater composite efficiency than the second integrated value; and selecting a first power converter to provide power until a maximum current threshold provided by the first power converter is reached. The method may include determining that a system load current is greater than a maximum current threshold provided by the first power converter; and selecting a second power converter to provide power until a maximum current threshold provided by the second power converter is reached. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1A A power supply system having a system performance controller according to one aspect is shown.

[0036] Figure 1B According to one aspect Figure 1B System performance controller.

[0037] Figure 2 A flow chart depicting exemplary operation of a system performance controller according to one aspect is shown.

[0038] Figure 3 A flow chart depicting exemplary operation of a power supply system according to one aspect is shown.

[0039] Figure 4 A power supply system having multiple system performance controllers and multiple power stages connected in series is shown according to one aspect.

[0040] Figure 5 A power supply system having multiple system performance controllers and multiple power stages connected in parallel is shown according to one aspect.

[0041] Figure 6 A power supply system having a system performance controller and multiple power stages according to one aspect is shown.

[0042] Figure 7 A flow chart depicting exemplary operation of a system performance controller using a goal seeking algorithm is shown according to one aspect.

[0043] Figure 8 A flow chart depicting exemplary operation of a system performance controller using an applied learning algorithm is shown according to one aspect.

[0044] Fig.9A A flow chart depicting exemplary operation of a system performance controller using a goal seeking algorithm according to another aspect is shown.

[0045] Fig. 9B A flow chart depicting exemplary operation of a system performance controller using an applied learning algorithm according to another aspect is shown.

[0046] Fig. 10A A power supply system having a system performance controller for zero voltage switching control according to one aspect is shown.

[0047] Fig. 10B Power levels of a power supply system according to one aspect are shown.

[0048] Fig. 10C Switching losses of a power stage according to one aspect are shown.

[0049] Fig. 10D A current sensing voltage threshold as a zero voltage switching control parameter is shown according to one aspect.

[0050] Fig. 10E Efficiency curves showing values ​​of current sensing voltage threshold according to one aspect.

[0051] Fig.11 A power supply system having a system performance controller for zero voltage switching control according to one aspect is shown.

[0052] Fig.12 A power supply controller of a power supply system according to one aspect is shown.

[0053] Fig.13 A flow chart depicting exemplary operation of a system performance controller for zero voltage switching control is shown according to one aspect.

[0054] Fig.14 A flow chart depicting exemplary operation of a system performance controller for zero voltage switching control using a goal seeking algorithm is shown according to one aspect.

[0055] Fig.15 A flow chart depicting exemplary operation of a system performance controller for zero voltage switching control using an applied learning algorithm is shown according to one aspect.

[0056] Fig.16 A flow chart depicting exemplary operation of a system performance controller for zero voltage switching control using a goal seeking algorithm according to another aspect is shown.

[0057] Fig.17 A flow chart depicting exemplary operation of a system performance controller for zero voltage switching control using an applied learning algorithm according to another aspect is shown.

[0058] Fig.18A and Fig.18B A power supply system having a system performance controller for controlling multiple series power stages using overall efficiency according to one aspect is shown.

[0059] Fig.19 Phase change points along an overall efficiency power curve as a control parameter of an electrical power system are shown according to one aspect.

[0060] Fig. 20A and Fig. 20B A power supply system having a system performance controller for controlling multiple parallel power stages using overall efficiency according to one aspect is shown.

[0061] Fig.21A flow chart depicting exemplary operation of a system performance controller for controlling multiple power stages using overall efficiency is shown according to one aspect.

[0062] Fig. 22 A power supply system having a system performance controller for setting or adjusting control parameters based on standby power according to one aspect is shown.

[0063] Fig.23 Control signals for controlling one or more power stages during skip cycle mode are shown according to one aspect.

[0064] Fig.24 A flow chart depicting exemplary operation of a system performance controller for controlling power levels based on standby power is shown according to one aspect.

[0065] Fig.25 A power supply system for controlling load sharing across multiple converters to optimize overall efficiency according to one aspect of the present application is shown.

[0066] Fig.26 A power supply system for controlling load sharing across a first converter and a second converter to optimize overall efficiency according to one aspect of the present application is shown.

[0067] Fig. 27 A power supply system for controlling load sharing across multiple converters to optimize overall efficiency according to one aspect of the present application is shown.

[0068] Fig.28 A flow chart depicting an exemplary operation of a power supply system according to one aspect of the present application is shown.

[0069] Fig.29 A graph depicting the efficiency of a single converter expressed as a percentage as a function of load current according to one aspect of the present application is shown.

[0070] Fig.30 A graph depicting the efficiency of a single converter expressed as power loss as a function of load current according to one aspect of the present application is shown.

[0071] Fig.31 A graph depicting composite power loss as a function of load current according to one aspect of the present application is shown.

[0072] Fig.32 A graph depicting composite power savings and efficiency as a function of load current according to one aspect of the present application is shown.

[0073] Fig.33A graph depicting the efficiency of a single converter expressed as a percentage as a function of load current according to one aspect of the present application is shown.

[0074] Fig.34 A graph depicting the efficiency of a single converter expressed as power loss as a function of load current according to one aspect of the present application is shown.

[0075] Fig.35 A graph depicting composite power loss as a function of load current according to one aspect of the present application is shown.

[0076] Fig.36 A graph depicting composite power savings and efficiency as a function of load current according to one aspect of the present application is shown.

[0077] Fig.37 A graph depicting composite power loss as a function of load current according to one aspect of the present application is shown.

[0078] Fig.38 A graph depicting composite power loss as a function of load current according to one aspect of the present application is shown. DETAILED DESCRIPTION

[0079] The present invention relates to a power supply system having or using a system performance controller that calculates the energy conversion efficiency (e.g., η = (V out *I out ) / (V in *I in )) and sets or adjusts one or more parameters of the power stage by adjusting the value of the control parameter until the energy conversion efficiency reaches a threshold condition (e.g., is maximized and / or reaches a target threshold). The system performance controller may execute an artificial intelligence (AI) algorithm to step through (or cycle through) the values ​​of one or more control parameters until the efficiency is maximized and / or equal to or greater than a threshold level. The system performance controller may set or update the control parameters based on the energy conversion efficiency during production or during the life of the system, which may cause the power stage to continue to meet the target or optimal energy conversion efficiency regardless of changes in environmental conditions, aging of circuit components of the power stage, manufacturing variations of circuit components of the power stage, and / or thermal effects of the power stage.

[0080] For example, when the power switch is activated, the power stage may cause switching losses in response to the drain-source voltage of the power switch becoming non-zero, thereby reducing energy conversion efficiency. In some examples, the control parameters may include zero voltage switching (ZVS) control parameters that enable the power switch (or multiple power switches) to switch at ZVS (or substantially ZVS). The energy conversion efficiency may be negatively affected by the environment of the power system (e.g., external stimuli) and manufacturing variations of components of the power system.

[0081] However, the system performance controller calculates the energy conversion efficiency and then sets or adjusts one or more control parameters that will cause the energy conversion efficiency to reach a threshold condition. In addition, in some examples, the system performance controller monitors and calculates the energy conversion efficiency during the life of the power stage and can programmatically re-evaluate and potentially select new values ​​for the ZVS control parameters that optimize the energy conversion efficiency or maintain it at a threshold level, regardless of changes in the environment, aging of components, and / or thermal effects on the power stage.

[0082] In some examples, the power system includes a plurality of power stages (e.g., connected in series or in parallel), and the system performance controller is configured to control the power stages during production or during the life of the power stages in a manner that maximizes the overall energy conversion efficiency or causes the overall energy conversion efficiency to be greater than a threshold level. The overall energy conversion efficiency may be an end-to-end energy conversion efficiency. In some examples, the system performance controller may set or adjust phase shedding / phase adding parameters that control when the power stages may be deactivated (or activated) to maximize the overall energy conversion efficiency. In some examples, the system performance controller may set or adjust one or more intermediate bus voltages to cause the overall energy conversion efficiency to be maximized or greater than a threshold level.

[0083] In some examples, the system performance controller directly calculates the standby power (e.g., (V IN *I IN )) and sets or adjusts one or more control parameters of the power stage by adjusting the values ​​of the control parameters until the standby power is minimized or less than a threshold level. The standby power may be the electrical energy used by an inactive device (e.g., the device is coupled to the power stage but is turned off). For example, the system performance controller directly calculates the standby power of the power stage based on the measured conditions (e.g., (V IN *I IN)) and selects a value for the control parameter so that the standby power is minimized and / or less than a standby power threshold level. In some examples, the skip cycle parameter may affect the standby power. For example, when a deactivated device is coupled to the power stage, the power controller may control the power stage in a skip cycle mode. In the skip cycle mode, the power controller generates a pulse set to switch the power switch during a burst period (followed by a skip cycle period) and then generates another pulse set during the burst period (followed by a skip cycle period). The skip cycle parameters may include a pulse frequency during a burst period, a duty cycle during a burst period, a time between burst periods, and / or a length of a skip cycle period.

[0084] In some examples, the system performance controller monitors and calculates other performance metrics such as noise (e.g., total harmonic distortion), emissions, ripple, and / or transient response, and sets or adjusts one or more control parameters of the power stage by adjusting the value of the control parameter until the monitored performance metric is maximized (or minimized) and / or reaches a target threshold.

[0085] FIG. 1 shows a power supply system 100 having one or more power stages 102, a power supply controller 104, a metering circuit 106, and a system performance controller 108 configured to calculate a performance metric 101 (or a set of performance metrics 101) based on a measurement condition 103 and set or adjust one or more control parameters 105 so that the performance metric 101 (or the set of performance metrics 101) reaches a threshold condition (e.g., below / above a target level, maximized or minimized, or falls within an acceptable range). In some examples, the system performance controller 108 is configured to set or adjust one or more control parameters so that the performance metric 101 is maximized (or minimized). The control parameters 105 determined by the system performance controller 108 may be stored at the power supply controller 104. Figure 1B A schematic diagram illustrating an example of a system performance controller 108 according to one aspect.

[0086] The performance metric 101 may be negatively affected by the environment of the power stage 102 (e.g., external stimuli) and manufacturing variations of components of the power stage 102. However, the system performance controller 108 may set or update the control parameters 105 based on the performance metric 101 during production or during the life of the system, which may cause the power stage 102 to continue to meet the performance metric 101 despite changes in environmental conditions, aging of the circuit components 112 of the power stage 102, manufacturing variations of the circuit components 112 of the power stage 102, and / or thermal effects on the power stage 102.

[0087] The control parameters 105 may be any type of control parameters of the power stage 102 that may affect one or more performance metrics 101. In some examples, the control parameters 105 may include zero voltage switching (ZVS) parameters 142, dead time 144, pulse width 146, duty cycle 148, switching frequency 150, skip cycle parameters 152, phase change parameters 158 (e.g., phase cut / phase add), intermediate bus voltage parameters 160, and / or input / output (I / O) conditions 161. In some examples, the control parameters 105 may include zero current switching (ZCS). In some examples, the control parameters 105 may include synchronous / asynchronous rectification, line voltage multiplication, energy recovery, buffering, and / or safe operating area (SOA). In some examples, the control parameters 105 may include modes and transitions, such as continuous, discontinuous, critical, fixed frequency, variable frequency, on time, off time, resonant transition, switching frequency center / modulation, current limit, and / or power stage drive (e.g., intensity, waveform).

[0088] In some examples, the performance metric 101 is an energy conversion efficiency 123. The energy conversion efficiency 123 is the ratio of output power to input power. In some examples, the energy conversion efficiency 123 is a single efficiency 123A. The single efficiency 123A refers to the energy conversion efficiency of a single power stage 102. In some examples (such as, in the case of multiple power stages), the energy conversion efficiency 123 may refer to an overall efficiency 123B. The overall efficiency 123B refers to the end-to-end energy conversion efficiency of the power stage 102. In some examples, instead of using an indirect parameter as a function of the energy conversion efficiency, the system performance controller 108 directly calculates the energy conversion efficiency 123 of the power stage 102 based on the measurement conditions 103 (e.g., input voltage, input current, output voltage, output current) (e.g., η=(V OUT *I OUT ) / (V IN *I IN )) and selects a value for the control parameter 105 so that the energy conversion efficiency 123 is maximized and / or equals or exceeds the efficiency threshold level. For example, the system performance controller 108 may adjust one or more control parameters 105 until the energy conversion efficiency 123 is maximized and / or equals or exceeds the efficiency threshold level.

[0089] In some examples, the system performance controller 108 may adjust the ZVS control parameter 142 (or multiple ZVS control parameters 142) until the energy conversion efficiency 123 is maximized and / or equals or exceeds the efficiency threshold level. For example, if the power switch 141 is activated when its drain-source voltage is non-zero, switching losses may occur. The ZVS control parameter 142 can be a control parameter that causes the power switch to be activated at zero voltage (or substantially zero voltage) to reduce or eliminate switching losses, thereby improving the energy conversion efficiency 123. In some examples, the ZVS control parameter 142 indicates the magnitude of the reverse current flowing through the inductor to discharge the parasitic capacitance of the power switch. In some examples, the ZVS control parameter 142 is a current sensing voltage threshold indicating the magnitude of the reverse current. In some examples, the ZVS control parameter 142 includes a time value for delaying the activation of the power switch so that the magnitude of the reverse current causes the parasitic capacitance of the power switch to discharge. In some examples, the ZVS control parameter 142 includes a dead time 144.

[0090] Other control parameters 105 may also affect the energy conversion efficiency 123 of the power stage 102. For example, the system performance controller 108 may set or adjust the value of the pulse width 146 by modifying the pulse width 146 until the energy conversion efficiency 123 reaches a threshold condition. In some examples, the threshold condition is that the energy conversion efficiency 123 is equal to or greater than a threshold level. In some examples, the threshold condition is that the energy conversion efficiency 123 is maximized. In some examples, the system performance controller 108 may set or adjust the value of the duty cycle 148 by modifying the duty cycle 148 until the energy conversion efficiency 123 reaches the threshold condition. In some examples, the system performance controller 108 may set or adjust the value of the switching frequency 150 by modifying the switching frequency 150 until the energy conversion efficiency 123 reaches the threshold condition.

[0091] In some examples, the system performance controller 108 may set or adjust the value of the dead time 144 by modifying the dead time 144 until the energy conversion efficiency 123 reaches a threshold condition. The dead time 144 may relate to a time period during which the first power switch and the second power switch are deactivated. When the dead time is too long, the power stage 102 may cause losses due to the conduction and reverse recovery of the body diodes of the first power switch and the second power switch, thereby reducing the energy conversion efficiency 123 of the power stage 102. When the dead time 144 is too short, the power stage 102 may cause losses due to the simultaneous conduction of the first power switch and the second power switch, resulting in a relatively large current spike, thereby reducing the energy conversion efficiency 123 of the power stage 102.

[0092] In some examples, the performance metric 101 is standby power 125. Standby power 125 can be the power used by a disabled device (e.g., the device is coupled to the power stage 102 but is turned off). For example, the system performance controller 108 can directly calculate the standby power 125 (e.g., V ) of the power stage 102 based on the measured conditions 103 (e.g., input voltage and input current). IN *I IN )) and selects a value for the control parameter 105 so that the standby power 125 reaches a threshold condition (e.g., minimized and / or less than a standby power threshold level). For example, the system performance controller 108 may adjust one or more control parameters 105 until the standby power 125 is minimized and / or less than the standby power threshold level. In some examples, the skip cycle parameter 152 may affect the standby power 125. For example, when a deactivated device is coupled to the power stage 102, the power controller 104 may control the power stage 102 in a skip cycle mode (e.g., skip mode, skip cycle modulation, etc.). In the skip cycle mode, the power controller 104 generates a pulse set to switch the power switch 141 during a burst period (followed by a skip cycle period) and then generates another pulse set during the burst period (followed by a skip cycle period). The skip cycle parameter 152 may include a pulse frequency during a burst period, a duty cycle during a burst period, a frequency between burst periods, and / or a length of a skip cycle period.

[0093] In some examples, performance metric 101 is a transient response. For example, system performance controller 108 directly calculates a transient response metric for power stage 102 based on measurement conditions 103 and selects a value for control parameter 105 such that the transient response metric reaches a threshold condition (e.g., is minimized and / or is less than a transient response threshold level). For example, system performance controller 108 may adjust one or more control parameters 105 until the transient response metric is minimized and / or is less than a transient response threshold level.

[0094] In some examples, the performance metric 101 is total harmonic distortion (THD). For example, the system performance controller 108 may directly calculate the THD metric (e.g., indicating the level of THD within the power stage 102) based on the measurement conditions 103 and select values ​​for the control parameters 105 so that the THD metric reaches a threshold condition (e.g., minimized and / or less than a THD threshold level). For example, the system performance controller 108 may adjust one or more control parameters 105 until the THD metric is minimized and / or less than the THD threshold level.

[0095] In some examples, performance metric 101 is an emission (e.g., electromagnetic radiation). For example, system performance controller 108 may directly calculate an emission metric (e.g., indicating a level of emissions within power stage 102) based on measurement conditions 103 and select a value for control parameter 105 so that the emission metric reaches a threshold condition (e.g., minimized and / or less than an emission threshold level). For example, system performance controller 108 may adjust one or more control parameters 105 until the emission metric is minimized and / or less than an emission threshold level.

[0096] In some examples, the performance metric 101 is ripple. For example, the system performance controller 108 may directly calculate the ripple metric (e.g., indicating the level of ripple within the power stage 102) based on the measurement conditions 103 and select a value for the control parameter 105 so that the ripple metric reaches a threshold condition (e.g., minimized and / or less than a ripple threshold level). For example, the system performance controller 108 may adjust one or more control parameters 105 until the ripple metric is minimized and / or less than a ripple threshold level. However, the performance metric 101 may include other FOMs, such as output voltage, power factor, etc.

[0097] System performance controller 108 is connected to power controller 104. In some examples, system performance controller 108 is configured to provide control parameters 105 to power controller 104 so that power controller 104 can control power stage 102 according to the set or updated value determined by system performance controller 108. In some examples, system performance controller 108 is configured to receive environmental conditions and / or control conditions from power controller 104. The environmental and / or control conditions may include input voltage, input current, current sensing, switching frequency, duty cycle, temperature, operating mode parameters, etc.

[0098] In some examples, the system performance controller 108 is wired to the power controller 104. In some examples, the system performance controller 108 is wirelessly connected (e.g., a network-based connection, such as a dedicated network or the Internet) to the power controller 104. In some examples, the power controller 104 and the power stage 102 are included in an integrated product (e.g., a semiconductor package). In some examples, the system performance controller 108 is located outside the integrated product. In some examples, the system performance controller 108 is associated with a display device that provides an external input / output interface to view results, set environmental conditions, and / or modify controls. In some examples, the integrated product includes the system performance controller 108 and the metering circuit 106. In some examples, the system performance controller 108 is located on a server remote from the power controller 104 and the power stage 102. In some examples, the system performance controller 108 is removably connected to the power controller 104. For example, system performance controller 108 may be connected to power controller 104 during testing and / or production of power stage 102 to determine control parameters 105 (which are then stored at power controller 104 ), but system performance controller 108 is not included in an integrated product.

[0099] System performance controller 108 is connected to metering circuit 106. In some examples, system performance controller 108 is wired to metering circuit 106. In some examples, system performance controller 108 is wirelessly connected to metering circuit 106. System performance controller 108 may receive measurement conditions 103 regarding power level 102 to calculate one or more performance metrics 101.

[0100] The power controller 104 is connected to the power stage 102. The power controller 104 can be any type of controller configured to control the operation of the power stage 102. Generally speaking, the type of the power controller 104 can depend on the type of the power stage 102. In some examples, the power controller 104 is a PWM controller. The power controller 104 is configured to generate control signals to control the power stage 102 according to the control parameters 105, and the control signals are provided to the power stage 102. The power controller 104 can include analog drivers 110 that are configured to generate control signals for one or more of the circuit components 112 of the power stage 102. In some examples, the power controller 104 receives environmental conditions (e.g., input voltage, input current, current sensing, switching frequency, duty cycle, operating mode parameters, temperature, etc.) from the power stage 102.

[0101] In some examples, the power controller 104 may include a combination of digital components and analog components. In some examples, the power controller 104 may include logic circuits, comparators, memory devices, etc. Control parameters 105 are stored at a program memory of the power controller 104, and these parameters may be determined by the system performance controller 108 so that one or more performance metrics 101 are optimized. In some examples, the control parameters 105 stored at the program memory of the power controller 104 are updated (e.g., automatically updated) by the system performance controller 108 to account for changing environmental conditions, aging of components, and / or thermal effects of the system.

[0102] although Figure 1A 102, but the power supply system discussed herein may include multiple power stages 102, as discussed further later in this disclosure. For example, the power supply system 100 may include a single power stage 102 or a combination of parallel and / or series power stages 102. The power stage 102 may include power conversion circuitry that converts electrical power, such as converting between alternating current (AC) and direct current (DC), converting DC to DC, converting DC to AC, or changing voltage or frequency for any combination thereof.

[0103] In general, the power stage 102 may include one of a variety of topologies, such as a buck converter, a boost converter, an inverting buck-boost converter, a flyback converter, an active clamp forward converter, a single switch forward converter, a dual switch forward converter, a push-pull converter, a half-bridge converter, a full-bridge converter, a phase-axis full-bridge converter, etc. In some examples, the power stage 102 includes a switching resonant converter.

[0104] Power stage 102 includes circuit components 112. Circuit components 112 may include diodes, switches (e.g., transistors, field effect transistors (FETs), metal oxide semiconductor field effect transistors (MOSFETs)), capacitors, inductors, and / or transformers, among others. In some examples, circuit components 112 include inductors 143. In some examples, inductor 143 is a boost inductor. In some examples, circuit components 112 include one or more power switches 141. In some examples, power switch 141 includes a FET. In some examples, power switch 141 includes a first power switch and a second power switch. In some examples, the first power switch is a pulse width modulation (PWM) switch (e.g., a FET). In some examples, the first power switch is a high-side PWM FET. In some examples, the second power switch is a PWM switch (e.g., a FET). In some examples, the second power switch is a low-side PWM FET.

[0105] The metering circuit 106 is connected to the power stage 102. In some examples, the power stage 102 includes the metering circuit 106. The metering circuit 106 is configured to sense the measurement condition 103 of the power stage 102. For example, the metering circuit 106 can monitor and obtain the measurement condition 103 of the power stage 102. In some examples, the metering circuit 106 is configured to obtain information about the measurement condition 103 from the power stage 102 in an analog format and convert the information about the measurement condition 103 into a digital format.

[0106] In some examples, the measurement conditions 103 include input voltage, input current, output voltage, and output current. In some examples, the measurement conditions 103 include the temperature of the power stage 102. In some examples, the measurement conditions 103 include the measured noise of the power stage 102, such as the detected total harmonic distortion (THD) of the power stage 102. In some examples, the measurement conditions 103 include the detected emissions (e.g., electromagnetic interference (EMI), electromagnetic compatibility (EMC)) of the power stage 102. In some examples, the measurement conditions 103 include intermediate conditions, such as branch currents, node voltages (e.g., zero current conditions, zero voltage conditions), slew rates, and waveforms of the power stage 102. In some examples, the measurement conditions 103 include parameter changes or aging levels of one or more components of the power stage 102. In some examples, the measurement conditions 103 include inferred conditions, such as boost inductance, which may change over time due to aging. The inductance (e.g., L=(V*dt) / dI) can be inferred by measuring the differential voltage and differential current during the switching period. Similarly, capacitance can be inferred by measuring the differential voltage and differential current during the switching period (eg, C = (I*dt) / dV, etc.).

[0107] In some examples, the metering circuit 106 includes one or more sensors 114 that are coupled to the power stage 102 and configured to measure and obtain one or more of the above-mentioned measurement conditions 103. In some examples, the sensor 114 includes one or more current sensors, such as: a first current sensor that obtains the input current of the power stage 102; and a second current sensor that obtains the output current of the power stage 102 (or any other current sensor that obtains one or more intermediate currents within the power stage 102). In some examples, the sensor 114 includes one or more voltage sensors, such as: a first voltage sensor that obtains the input voltage of the power stage 102; and a second voltage sensor that obtains the output voltage of the power stage 102 (or any other voltage sensor that obtains one or more intermediate voltages within the power stage 102). In some examples, the sensor 114 includes a temperature sensor that is configured to obtain the temperature of the power stage 102 (or various components of the power stage 102). In some examples, the sensor 114 includes one or more noise detection sensors that are configured to obtain the noise level of the power stage 102. In some examples, sensors 114 include emissions detection sensors configured to detect one or more types of emissions.

[0108] In some examples, the metering circuit 106 includes one or more analog-to-digital converters (ADCs) 116 configured to convert sensed information from the sensors 114 into a digital format. In some examples, the metering circuit 106 includes one or more multiplexers 118. For example, a particular multiplexer 118 is configured to receive signals (e.g., each corresponding to a different measurement condition 103) and then multiplex the signals into a single signal, which is then converted into a digital signal by the ADC 116. In some examples, the multiplexers 118 include: a first multiplexer configured to receive an input voltage and an input current (from a corresponding sensor 114) and generate a single signal representing the input voltage and current; and a second multiplexer configured to receive an output voltage and an output current (from a corresponding sensor 114) and generate a single signal representing the output voltage and the output current. In some examples, ADC 116 includes a first ADC configured to receive a signal from a first multiplexer and generate a digital signal, and a second ADC configured to receive a signal from a second multiplexer and generate a digital signal.

[0109] The system performance controller 108 may include: one or more computation circuits 120 configured to compute the corresponding performance metric 101; and a control manipulation module 130 configured to modify one or more control parameters 105 until the performance metric 101 reaches a threshold condition (e.g., maximized, minimized, or above / below a threshold level). In some examples, the system performance controller 108 includes one computation circuit 120. In some examples, the system performance controller 108 includes more than one computation circuit 120.

[0110] In some examples, the system performance controller 108 is a computing device having one or more memory devices 162 and one or more digital processors 164. The memory device 162 includes executable instructions that cause the digital processor 164 to perform operations of the system performance controller 108. In some examples, the system performance controller 108 includes one or more integrated circuits 166. In some examples, the system performance controller 108 includes one or more field programmable gate arrays (FPGAs) 168. In some examples, the system performance controller 108 includes one or more state machines 170. In some examples, the computing circuit 120 (or multiple computing circuits 120) and the control manipulation module 130 are included on the same device (which may be included as part of an integrated product including the power controller 104 and the power stage 102 or located outside the integrated product including the power controller 104 and the power stage 102). In some examples, the computing circuit 120 (or multiple computing circuits 120) is included on a first computing device, and the control manipulation module 130 is included in a second computing device. In some examples, the first computing device may be included in an integrated product (along with power controller 104 and power stage 102 ), and the second computing device may be external to the integrated product.

[0111] The system performance controller 108 may include an efficiency calculation circuit 122 configured to calculate an energy conversion efficiency 123 of the power stage 102 based on the measurement conditions 103. In some examples, the efficiency calculation circuit 122 continuously (e.g., periodically) calculates the energy conversion efficiency 123 of the power stage 102 based on the measurement conditions 103. For example, the system performance controller 108 receives the measurement conditions 103 (e.g., in a digital format) from the metering circuit 106. The measurement conditions 103 may include an input voltage, an input current, an output voltage, and an output current (e.g., transient or horizontal). The efficiency calculation circuit 122 calculates the energy conversion efficiency 123 of the power stage 102 (e.g., η=(V out *I out ) / (V in *I in)). In some examples, as explained later in this disclosure with respect to a power supply system having multiple power stages, the energy conversion efficiency 123 may be an overall efficiency 123B.

[0112] The control manipulation module 130 is configured to repeatedly modify the control parameter 105 (or a set of control parameters 105) until the energy conversion efficiency 123 reaches a threshold condition (e.g., maximized or equal to or greater than a threshold level). For example, in a first iteration, the control manipulation module 130 is configured to select a first value for the control parameter 105, and the control manipulation module 130 is configured to provide the first value to the power supply controller 104 so that the power stage 102 operates at the first value. The metering circuit 106 senses the measurement condition 103 (e.g., input voltage, input current, output voltage, output current) of the power stage 102. The efficiency calculation circuit 122 calculates the energy conversion efficiency 123 of the power stage 102 operating according to the first value.

[0113] If the energy conversion efficiency 123 is less than the threshold level, the control manipulation module 130 selects a second value for the control parameter 105, and the control manipulation module 130 provides the second value to the power controller 104 so that the power stage 102 operates according to the second value. Then, the metering circuit 106 senses the measurement conditions 103 (e.g., input current, input voltage, output current, output voltage) of the power stage 102. The efficiency calculation circuit 122 calculates the energy conversion efficiency 123 of the power stage 102 operating at the second value. If the energy conversion efficiency 123 does not reach the threshold condition, the control manipulation module 130 selects another value, and the process repeats until the energy conversion efficiency 123 reaches the threshold condition.

[0114] The system performance controller 108 may include a standby power calculation circuit 124 configured to calculate a standby power 125 based on the measurement condition 103. In some examples, the efficiency calculation circuit 122 continuously (e.g., periodically) calculates the standby power 125 of the power stage 102 based on the measurement condition 103. For example, the system performance controller 108 receives the measurement condition 103 (e.g., in a digital format) from the metering circuit 106. The measurement condition 103 may include an input voltage and an input current. The standby power calculation circuit 124 calculates the standby power 125 (e.g., V in *I in ). The control manipulation module 130 is then configured to repeatedly modify the control parameter 105 (or a set of control parameters 105) until the standby power 125 is minimized or less than a threshold level.

[0115] In some examples, the calculation circuit 120 may include one or more other calculation circuits 120, such as: a transient response calculation circuit configured to calculate a transient response metric; a THD calculation circuit configured to calculate a THD metric; an emissions calculation circuit configured to calculate an emissions metric; and / or a ripple communication circuit configured to calculate a ripple metric. These other calculation circuits 120 may be executed in conjunction with the control manipulation module 130 to adjust one or more control parameters 105 to minimize or maximize the corresponding performance metric 101 in the same manner as described above.

[0116] In some examples, the control manipulation module 130 iteratively modifies the control parameters 105 by selecting different values ​​for the control parameters 105 based on an artificial intelligence (AI) algorithm 132. For example, the control manipulation module 130 may execute the AI ​​algorithm 132 (with the historical data 140 input) to identify which value of one or more control parameters 105 to select next. The historical data 140 may include data derived from the power system 100. For example, the historical data 140 may include information previously calculated and / or derived by the system performance controller 108. In some examples, the historical data 140 may include the measured conditions 103, the control parameters 105, and the calculated performance metrics 101 of the previous iterations performed by the system performance controller 108. In some examples, the historical data 140 may include data derived from other power systems (e.g., power systems similar to the power system 100).

[0117] The AI ​​algorithm 132 may obtain a maximum weighted combination of the calculated metrics (e.g., efficiency) for the combination of environmental conditions and control parameters 105, and then select a value corresponding to the maximum weighted combination for the control parameter 105. In some examples, the AI ​​algorithm 132 obtains the maximum weighted combination based on a predetermined minimization, objective, or maximization of each of the calculated metrics.

[0118] In some examples, the AI ​​algorithm 132 includes a target search algorithm 134. The target search algorithm 134 may be a search algorithm that selects values ​​for the control parameters 105 in a manner that results in a target performance metric. In some examples, the AI ​​algorithm 132 includes an applied learning algorithm 136 that defines a neural network 138 (input along with historical data 140) to predict values ​​for one or more control parameters 105. For example, according to the applied learning algorithm 136, the control manipulation module 130 is configured to apply the historical data 140 (e.g., also referred to as training data) to the neural network 138 to predict values ​​for one or more control parameters 105 that result in the performance metric 101 meeting a threshold condition.

[0119] The neural network 138 can be a group of interconnected nodes, each node representing an artificial neuron. These nodes are connected to each other in the form of layers, where the output of one layer becomes the input of the next layer. The neural network 138 transforms the input received by the input layer, transforms the input through a series of hidden layers, and produces an output via the output layer. Each layer consists of a subset of the node set. The nodes in the hidden layer are fully connected to all nodes in the previous layer, and their outputs are provided to all nodes in the next layer. The nodes in a single layer act independently of each other (i.e., do not share connections). The nodes in the output provide the transformed input to the requesting process. In some examples, the neural network 138 is a convolutional neural network, which is a neural network that is not fully connected. Therefore, the complexity of the convolutional neural network is lower than that of a fully connected neural network. The convolutional neural network can also use pooling or maximum pooling to reduce the dimensionality of the data flowing through the neural network (thereby reducing the complexity), so the required level of calculation can be reduced. This makes the calculation of the output faster in the convolutional neural network than in the neural network.

[0120] In some examples, system performance controller 108 is used during testing (or production) of power stage 102 to select one or more control parameters 105 for a range of environmental conditions that achieve a target performance metric 101 (or a set of performance metrics 101). For example, system performance controller 108 may select values ​​for control parameter 105 (or multiple control parameters 105) for an operating range of power stage 102 (e.g., input voltage, supply voltage, temperature, output current, etc.) such that a monitored performance metric 101 (or multiple monitored performance metrics 101) may meet a threshold condition.

[0121] Relative to a first environmental condition (e.g., an input voltage having a value A), system performance controller 108 may determine a first value of control parameter 105 that causes performance metric 101 to reach a threshold condition. Relative to a second environmental condition (e.g., an input voltage having a value B), system performance controller 108 may determine a second value of control parameter 105 that causes performance metric 101 to reach a threshold condition. In some examples, system performance controller 108 determines the values ​​of control parameter 105 that cause performance metric 101 to reach a threshold condition for the entire operating range of power stage 102 (e.g., a range of input voltages at which power stage 102 may operate).

[0122] In some examples, the control parameter 105, the corresponding environmental condition, and the calculated performance metric 101 are stored at the power supply controller 104. For example, the power supply controller 104 may store a first value of the control parameter 105 in conjunction with its corresponding environmental condition (e.g., input voltage A) and the corresponding calculated performance metric 101. The power supply controller 104 may store a second value of the control parameter 105 in conjunction with its corresponding environmental condition (e.g., input voltage B) and the corresponding calculated performance metric 101.

[0123] In some examples, in response to power stage 102 being activated (e.g., turned on) or while power stage 102 is operating, power controller 104 can detect environmental conditions (e.g., input voltage) and apply appropriate values ​​for control parameters 105. For example, power controller 104 can receive one or more environmental conditions such as input voltage, current sensing, temperature, etc. If power controller 104 detects a change in the environment, power controller 104 can control power stage 102 using appropriate values ​​for control parameters 105.

[0124] In some examples, when power stage 102 is activated (e.g., power stage 102 is turned on), power controller 104 can determine that the environmental condition is input voltage B. In this example, power controller 104 can use the second value of control parameter 105. On a subsequent activation, power controller 104 can determine that the environmental condition is input voltage A. In this example, power controller 104 can select the first value of control parameter 105.

[0125] In some examples, the system performance controller 108 is configured to continuously (e.g., periodically) monitor the performance metric 101 during the life of the system and update the one or more control parameters 105 so that the performance metric 101 continues to meet their threshold conditions. In some examples, the system performance controller 108 is configured to execute the control manipulation module 130 to update the one or more control parameters 105 in response to a triggering event. In some examples, the triggering event includes activation of the power stage 102. In some examples, the triggering event includes detection of the expiration of a timer. In some examples, the triggering event includes detection of a change in one or more environmental conditions. In some examples, the triggering event includes detection that the performance metric 101 does not meet the threshold condition.

[0126] For example, the system performance controller 108 is configured to execute periodically at fixed time intervals, or based on predefined events such as changes in environmental conditions, activations, low performance metrics, etc. For example, during operation of the power stage 102 , the system performance controller 108 monitors environmental conditions, and if one or more environmental conditions change by more than a threshold amount, the system performance controller 108 is configured to execute to re-evaluate the control parameters 105 based on the monitored performance metrics 101 and then provide updated values ​​to the power controller 104 .

[0127] Figure 2 A flowchart 200 depicting exemplary operation of the system performance controller 108 is shown according to one aspect. Figure 2 In an example of , the system performance controller 108 includes: an efficiency calculation circuit 122, which is configured to monitor and calculate the energy conversion efficiency 123; and a control manipulation module 130, which is configured to manipulate one or more control parameters 105 according to the AI ​​algorithm 132. In some examples, the system performance controller 108 is configured to monitor the energy conversion efficiency 123 and update the one or more control parameters 105 in response to a triggering event so that the energy conversion efficiency 123 is optimized, maintained, or improved over time.

[0128] In operation 202, the system performance controller 108 detects a triggering event. In some examples, the triggering event includes activation of the power stage 102. In some examples, the triggering event includes detection of expiration of a timer. In some examples, the triggering event includes detection of a change in one or more environmental conditions (e.g., a change in input voltage). In some examples, the triggering event includes detection that the energy conversion efficiency 123 does not meet a threshold condition.

[0129] In operation 204, the system performance controller 108 selects a value for the control parameter 105 based on the energy conversion efficiency 123 (e.g., to maximize the energy conversion efficiency 123). In some examples, the control parameter 105 is a ZVS control parameter 142 that controls ZVS. In some examples, the control parameter 105 is a dead time 144 that maximizes the energy conversion efficiency 123 instead of cross conduction. In some examples, the control parameter 105 is a pulse width 146, a duty cycle 148, a switching frequency 150, a skip cycle parameter 152, or an I / O condition 161.

[0130] For example, in operation 201, system performance controller 108 receives measurement conditions 103 of power stage 102. Measurement conditions 103 include input voltage, input current, output voltage, and output current. In some examples, system performance controller 108 receives measurement conditions 103 from metering circuit 106. In some examples, metering circuit 106 converts the input voltage, input current, output voltage, and output current into a DC format. In some examples, metering circuit 106 includes one or more ADCs 116. In some examples, metering circuit 106 includes one or more multiplexers 118.

[0131] In operation 203, the efficiency calculation circuit 122 calculates the energy conversion efficiency 123 (eg, (V OUT *I OUT ) / (V IN *I IN )). In operation 205, the control manipulation module 130 modifies the control parameter 105 until the energy conversion efficiency 123 reaches a threshold condition (e.g., is maximized or equal to or greater than a threshold level). In some examples, the control manipulation module 130 selects a first value for the control parameter 105 and sends the first value to the power controller 104. Then, the efficiency calculation circuit 122 calculates the energy conversion efficiency 123 of the power stage 102 operating according to the first value based on the measurement condition 103. Then, the control manipulation module 130 modifies the first value to obtain a second value in response to the energy conversion efficiency 123 not reaching the threshold condition, and sends the second value to the power controller 104. Then, the efficiency calculation circuit 122 calculates the energy conversion efficiency 123 of the power stage 102 operating according to the second value based on the measurement condition 103.

[0132] In some examples, the control manipulation module 130 uses an AI algorithm 132 to modify the control parameter 105. In some examples, the AI ​​algorithm 132 is a target search algorithm 134. In some examples, the AI ​​algorithm is an applied learning algorithm 136. In some examples, the control manipulation module 130 applies historical data 140 to a neural network 138 to predict a value of the control parameter 105 that causes the energy conversion efficiency 123 to reach a threshold condition. In some examples, the system performance controller 108 determines a first value of the control parameter 105 that causes the energy conversion efficiency 123 to reach a threshold condition for a first environmental condition, and determines a second value of the control parameter 105 that causes the energy conversion efficiency 123 to reach the threshold condition for a second environmental condition. In some examples, the first environmental condition and the second environmental condition are different voltages.

[0133] In operation 206, system performance controller 108 provides the selected value to power controller 104 to utilize the selected value to control power stage 102. In some examples, the selected value is stored in a program memory at power controller 104. In some examples, system performance controller 108 updates the program memory with the selected value.

[0134] Figure 3 1 shows a flowchart 300 depicting an exemplary operation of the system performance controller 108 according to one aspect. Although the flowchart 300 shows the operation of the system performance controller 108 using the energy conversion efficiency 123 as the monitored performance metric 101, Figure 3 The operation can be applied to any performance metric 101. Figure 3 As shown, the system performance controller 108 can continuously monitor the energy conversion efficiency 123 of the power stage 102, and if the energy conversion efficiency 123 of the power stage 102 does not meet the threshold condition (for example, is not maximized and does not drop below the threshold condition), the system performance controller 108 can perturb the power supply system 100 to set or adjust the value of the control parameter 105 in a single step through the value of the control parameter 105, which causes the energy conversion efficiency 123 to reach the threshold condition.

[0135] In operation 302, system performance controller 108 may detect a triggering event. In some examples, the triggering event includes detection of activation of power stage 102. In some examples, the triggering event includes detection of a change in one or more environmental conditions. In some examples, the triggering event includes detection that energy conversion efficiency 123 does not meet a threshold condition.

[0136] In operation 304, the system performance controller 108 may determine whether the system performance controller 108 includes historical data 140. In some examples, the historical data 140 includes values ​​of the control parameters 105. In some examples, during previous executions of the system performance controller 108, the system performance controller 108 may have used the efficiency calculation circuit 122 and the control manipulation module 130 to determine values ​​that caused the energy conversion efficiency 123 to reach the threshold condition, and these optimized control parameters 105 are stored in the historical data 140. Additionally, it is noted that the historical data 140 may include control parameters 105 that have been optimized from AI learning.

[0137] If the system performance controller 108 includes the historical data 140 (yes), then in operation 308, the system performance controller 108 loads the control parameter 105 from the historical data 140. For example, the system performance controller 108 may provide the control parameter 105 to the power supply controller 104. If the system performance controller 108 does not include the historical data 140 (no), then in operation 306, the control manipulation module 130 is configured to perturb the system, for example, by stepping through the value of the control parameter 105. Alternatively, if the system performance controller 108 includes the historical data 140, and after the system performance controller 108 loads the control parameter 105 from the historical data 140, the process proceeds to operation 306. In operation 306, the control manipulation module 130 selects a first value (e.g., using the AI ​​algorithm 132) and provides the first value to the power supply controller 104 to operate the power stage 102 according to the first value. The system performance controller 108 obtains the measurement condition 103 from the metering circuit 106. For example, the measurement conditions 103 include input voltage, input current, output voltage, and output current.

[0138] In operation 312, the efficiency calculation circuit 122 calculates the energy conversion efficiency 123. For example, the efficiency calculation circuit 122 calculates the energy conversion efficiency 123 (eg, (V OUT *I OUT ) / (V IN *I IN )). In operation 314, the system performance controller 108 determines whether the energy conversion efficiency 123 reaches a threshold condition. In some examples, the system performance controller 108 uses the AI ​​algorithm 132 to determine whether the energy conversion efficiency 123 is maximized. In some examples, the system performance controller 108 determines whether the energy conversion efficiency 123 is greater than or equal to a target level. If not, the process returns to operation 306 to use the AI ​​algorithm 132 to select another value for the control parameter 105.

[0139] Figure 4 A power system 400 is shown having a plurality of power stages 402 connected in series with each other. For example, the power stages 402 include a first power stage 402-1, a second power stage 402-2, and a third power stage 402-3 coupled to a load 409. Figure 4 Three power stages 402 are shown, but the power system 400 can include any number of power stages 402, such as two power stages 402 or more than three power stages 402. In some examples, the power system 400 includes a power controller 404 configured to control each of the power stages 402. In some examples, the power system 400 includes a separate power controller 404 for each of the power stages 402.

[0140] Each of the power stages 402 is associated with a separate system performance controller. For example, the power system 400 includes: a first system performance controller 408-1, which is configured to optimize the first power stage 402-1; a second system performance controller 408-2, which is configured to optimize the second power stage 402-2; and a third system performance controller 408-3, which is configured to optimize the third power stage 402-3. A separate metering circuit can be included in each power stage 402 (or coupled to each power stage).

[0141] Figure 5 A power supply system 500 is shown having a plurality of power stages 502 connected in parallel to each other. For example, the power stages 502 include a first power stage 502-1, a second power stage 502-2, and a third power stage 502-3 coupled to a load 509. Figure 5 Three power stages 502 are shown, but the power system 500 can include any number of power stages 502, such as two power stages 502 or more than three power stages 502. In some examples, the power system 500 includes a power controller 504 configured to control each of the power stages 502. In some examples, the power system 500 includes a separate power controller 504 for each of the power stages 502.

[0142] Each of the power stages 502 is associated with a separate system performance controller. For example, the power system 500 includes: a first system performance controller 508-1, which is configured to optimize the first power stage 502-1; a second system performance controller 508-2, which is configured to optimize the second power stage 502-2; and a third system performance controller 508-3, which is configured to optimize the third power stage 502-3. A separate metering circuit can be included in each power stage 502 (or coupled to each power stage).

[0143] Figure 6 A power system 600 with multiple power stages 602 and a single system performance controller 608 is shown according to one aspect. For example, the power stages 602 include a first power stage 602-1, a second power stage 602-2, and a third power stage 602-3. In some examples, the power stages 602 are coupled in series with respect to each other. In some examples, the power stages 602 are coupled in parallel with respect to each other. Although Figure 6 Three power stages 602 are shown, but the power system 600 can include any number of power stages 602, such as two power stages 602 or more than three power stages 602. In some examples, the power system 600 includes a power controller 604 configured to control each of the power stages 602. In some examples, the power system 600 includes a separate power controller 604 for each of the power stages 602.

[0144] In some examples, the system performance controller 608 is configured to optimize performance for each power stage 602 individually. Figure 1A , Figure 1B and Figure 6 , the system performance controller 608 is configured to set or adjust the control parameter 105 (or multiple control parameters 105) for each power stage 602 so that each power stage 502 achieves the target performance metric 101 (or multiple performance metrics 101). If the monitored performance metric 101 is the energy conversion efficiency 123, the system performance controller 108 is configured to calculate the energy conversion efficiency 123 of the first power stage 602-1 and adjust one or more control parameters 105 for the first power stage 602-1 until the energy conversion efficiency 123 reaches a threshold condition. The system performance controller 608 is configured to monitor the energy conversion efficiency 123 of the second power stage 602-2 (and make any adjustments to one or more of its control parameters 105), and monitor the energy conversion efficiency 123 of the third power stage 602-3 (and make any adjustments to one or more of its control parameters 105), respectively.

[0145] In some examples, the system performance controller 608 is configured to monitor the overall efficiency 123B of the power stages 602 and then adjust one or more control parameters 105 for one or more of the power stages 602 until the overall efficiency 123B reaches a threshold condition. For example, the system performance controller 608 may receive an input voltage and an input current, which are inputs to the power stages 602. Additionally, the system performance controller 608 may receive an output voltage and an output current, which are outputs of the power stages 602. The system performance controller 108 is configured to adjust one or more control parameters 105 for one or more of the power stages 602 until the overall efficiency 123B reaches a threshold condition.

[0146] Figure 7 A flowchart 700 depicting an exemplary operation of the power supply system 100 is shown according to one aspect. Figure 7 Flowchart 700, but Figure 7 The operation of may be applicable to any power system discussed herein. In some examples, the control parameters 105 are set using the system performance controller 108 and then stored at the power controller 104. The system performance controller 108 may use the target search algorithm 134 to perform Figure 7 The power controller 104 periodically measures environmental conditions and applies appropriate control parameters 105 to the power stage 102. In some examples, the system performance controller 108 is configured to initiate control during development, testing, and / or production of the power system. Figure 7In some examples, system performance controller 108 is configured to initiate a Figure 7 operation.

[0147] In some examples, the performance metric 101 is an energy conversion efficiency 123. In some examples, the energy conversion efficiency 123 is a single efficiency 123A. In some examples, the energy conversion efficiency 123 is an overall efficiency 123B. In some examples, the performance metric 101 is a standby power 125. In some examples, the performance metric 101 is a THD metric, a ripple metric, an emissions metric, or a transient response metric. The control parameter 105 can be any of the above parameters for controlling the power stage 102.

[0148] In some examples, the system performance controller 108 is configured to provide the first value to the power supply controller 104 so that the power stage 102 operates according to the first value. In operation 708, the calculation circuit 120 is configured to calculate the performance metric 101. For example, the system performance controller 108 receives the measurement condition 103 from the metering circuit 106 and calculates the performance metric 101 (e.g., the energy conversion efficiency 123). In operation 710, the system performance controller 108 stores the environmental condition, the first value of the control parameter 105, and the calculated performance metric 101.

[0149] In operation 712, the system performance controller 108 determines whether the desired Z response is achieved, e.g., whether the performance metric 101 reaches a threshold condition (e.g., maximized, minimized, or greater than / less than a threshold level). If not, the operation returns to operation 706, and the control manipulation module 130 selects a second value for the control parameter 105 based on the target search algorithm 134. If yes, then in operation 714, the system performance controller 108 determines whether the environmental range is complete, e.g., whether the range of environmental conditions has been tested. If no, the system performance controller 108 directs the process back to operation 704 to set another value for the environmental condition. For example, the system performance controller 108 may select input voltage B as the environmental condition and then modify the value of the control parameter 105 until the performance metric 101 reaches the threshold condition. If yes, then the process proceeds to operation 716.

[0150] In operation 716, a program control may be selected that results in a desired performance metric 101 (e.g., selecting a value of a control parameter 105 that causes the performance metric 101 to reach a threshold condition for a range of environmental conditions). For example, for a corresponding environmental condition, the program control may include a value of a control parameter 105 that causes the performance metric 101 to reach a threshold condition.

[0151] In operation 718, the program control may be stored in a product memory of the power controller 104. In operation 720, the power controller 104 applies the program control to the power stage 102. In some examples, in operation 722, the power controller 104 measures the environmental conditions and applies the program control at periodic intervals. In some examples, the periodic interval may refer to only one time or more than one time according to more than one event (such as a power cycle) to ensure effective coverage.

[0152] For example, the power controller 104 may receive one or more environmental conditions such as input voltage, current sensing, temperature, etc. If the power controller 104 detects a change in the environment, the power controller 104 may use an appropriate value of the control parameter 105 to control the power stage 102. For example, when the power stage 102 is activated (e.g., the power stage 102 is turned on), the power controller 104 may determine that the environmental condition is input voltage B. In this example, the power controller 104 may use an appropriate value of the control parameter 105. Upon subsequent activation, the power controller 104 may determine that the environmental condition is input voltage A, and the power controller 104 may select an appropriate value of the control parameter 105.

[0153] Figure 8 A flowchart 800 depicting an exemplary operation of the power supply system 100 is shown according to one aspect. Figure 8 Flowchart 800, but Figure 8 The operation can be applied to any power system discussed in this article. Figure 8 The flowchart 800 may be similar to Figure 7 700, except that Figure 8 Flowchart 800 uses the applied learning algorithm 136 . Figure 8 The operation may include reference Figure 7 In some examples, system performance controller 108 is configured to initiate a power system 100 during development, testing, and / or production. Figure 8 In some examples, system performance controller 108 is configured to initiate a power stage 102 in response to being activated (eg, turned on). Figure 8 operation.

[0154] Relative to Figure 8In some examples, the performance metric 101 is an energy conversion efficiency 123. In some examples, the energy conversion efficiency 123 is a single efficiency 123A. In some examples, the energy conversion efficiency 123 is an overall efficiency 123B. In some examples, the performance metric 101 is a standby power 125. In some examples, the performance metric 101 is a THD metric, a ripple metric, an emissions metric, or a transient response metric. The control parameter 105 can be any of the above parameters for controlling the power stage 102.

[0155] In operation 802, the power stage 102 is activated (e.g., initially powered on). In operation 804, the system performance controller 108 sets one or more environmental conditions. In operation 806, the control manipulation module 130 uses the applied learning algorithm 136 to modify the value of the control parameter 105. Although the flowchart 800 depicts the modification of a single control parameter 105, the control manipulation module 130 may set and modify a plurality of different control parameters 105. The control parameter 105 may have a control range that defines a range of values. The control manipulation module 130 may select a first value within the control range (e.g., a certain value of the ZVS control parameter 142). In some examples, the system performance controller 108 is configured to provide the first value to the power supply controller 104 so that the power stage 102 operates according to the first value. In operation 808, the calculation circuit 120 calculates the performance metric 101. For example, the calculation circuit 120 receives the measurement condition 103 from the metering circuit 106 and calculates the performance metric 101 (e.g., the energy conversion efficiency 123). In operation 810 , the system performance controller 108 stores the environmental conditions, the values ​​of the control parameters 105 , and the calculated performance metrics 101 .

[0156] In operation 812, the system performance controller 108 determines whether the control range of the control parameter 105 is complete. If not, the process returns to operation 806, and the control manipulation module 130 uses the application learning algorithm 136 to select another value for the control parameter 105. If yes, then in operation 814, the system performance controller 108 determines whether the environmental range is complete, for example, whether the range of environmental conditions has been tested. If not, the system performance controller 108 directs the process back to operation 804 to set another value for the environmental condition. For example, the control manipulation module 130 may select another value for the environmental condition and then modify the value of the control parameter 105 for its control range. If yes, then the process proceeds to operation 816.

[0157] In operation 816, a program control may be selected that results in a desired performance metric (e.g., a value of control parameter 105 that results in the best performance metric for a range of environments). For example, the program control may include a value of control parameter 105 that optimizes performance metric 101 for a corresponding environmental condition. In operation 818, the program control may be stored in a product memory of power controller 104. In operation 820, power controller 104 applies the program control to power stage 102. In operation 822, in some examples, such as with respect to Figure 7 As discussed in flowchart 700 , power controller 104 measures environmental conditions and applies application controls at periodic intervals.

[0158] Fig.9A A flowchart 900 depicting an exemplary operation of the power supply system 100 is shown according to another aspect. Fig.9A Flowchart 900, but Fig.9A The operation may be applicable to any of the power supply systems discussed herein. Fig.9A The flowchart 900 may be similar to Figure 7 700 (eg, both use the goal search algorithm 134), except that flowchart 900 depicts a continuous learning example.

[0159] Relative to Fig.9A In some examples, the performance metric 101 is an energy conversion efficiency 123. In some examples, the energy conversion efficiency 123 is a single efficiency 123A. In some examples, the energy conversion efficiency 123 is an overall efficiency 123B. In some examples, the performance metric 101 is a standby power 125. In some examples, the performance metric 101 is a THD metric, a ripple metric, an emissions metric, or a transient response metric. The control parameter 105 can be any of the above parameters for controlling the power stage 102.

[0160] exist Fig.9AIn operation 902, the initiation of the operation may be event-driven and / or periodically driven. In operation 902, the system performance controller 108 may detect a triggering event. In some examples, the triggering event includes the power stage 102 being activated. In some examples, the triggering event occurs during the operation of the power stage 102. For example, while the power stage 102 is executing, the system performance controller 108 may detect an environmental change. In some examples, while the power stage 102 is executing, the system performance controller 108 may continuously (e.g., periodically) calculate the performance metric 101, and in response to the performance metric 101 not reaching a threshold condition, the system performance controller 108 may detect a triggering event. In some examples, the system performance controller 108 monitors a timer, and in operation 901, upon detecting that the timer has expired, the system performance controller 108 is configured to trigger Fig.9A operation.

[0161] Upon detecting a trigger event or expiration of a timer, in operation 904, the control manipulation module 130 uses the target search algorithm 134 to modify the value of the control parameter 105. Although the flowchart 900 depicts the modification of a single control parameter 105, the control manipulation module 130 may set and modify a plurality of different control parameters 105. In some examples, the control parameter 105 may have a control range that defines a range of values. The control manipulation module 130 may select a value within the control range (e.g., a certain value of the ZVS control parameter 142). In some examples, the system performance controller 108 provides the value to the power supply controller 104 so that the power stage 102 operates according to the first value. In operation 906, the calculation circuit 120 calculates the performance metric 101. For example, the calculation circuit 120 receives the measurement condition 103 from the metering circuit 106 and calculates the performance metric 101 (e.g., the energy conversion efficiency 123). In operation 908, the system performance controller 108 stores or modifies the environmental conditions, the value of the control parameter 105, and the calculated performance metric 101. In some examples, system performance controller 108 updates historical data 140 using environmental conditions, values ​​of control parameters 105 , and calculated performance metrics 101 .

[0162] In operation 910, the system performance controller 108 determines whether the desired Z response is achieved, e.g., whether the performance metric 101 reaches a threshold condition (e.g., maximized or equal to or greater than a threshold level). If not, the operation returns to operation 904, and the control manipulation module 130 selects another value for the control parameter 105 based on the target search algorithm 134. If yes, then in operation 912, the system performance controller 108 determines whether the environment is the same. For example, the system performance controller 108 may determine whether one or more environmental conditions have changed (e.g., indicating that the power stage 102 has not stabilized). If one or more environmental conditions have changed, the system performance controller 108 directs the process back to operation 902 to wait until stability is detected.

[0163] In response to the environmental conditions being detected as being the same or similar, in operation 914, the system performance controller 108 selects a program control that results in a desired performance metric (e.g., selects a value of the control parameter 105 that causes the performance metric 101 to reach a threshold condition). For example, for the corresponding environmental condition, the program control may include a value of the control parameter 105 that causes the performance metric 101 to reach the threshold condition. In operation 916, the system performance controller 108 may update the product memory of the power controller 104 with the selected program control. In some examples, in operation 918, the power controller 104 measures the environmental conditions and applies the program control at periodic intervals. In some examples, in the learning example, in addition to the example Figure 7 and Figure 8 Outside of the intervals noted in flowcharts 700 and 800, periodic events may be triggered by environmental changes and measured after the system has stabilized.

[0164] Fig. 9B A flowchart 950 depicting exemplary operation of the system performance controller 108 is shown according to another aspect. Although described with respect to the power supply system 100 Fig. 9B Flowchart 950, but Fig. 9B The operation may be applicable to any of the power supply systems discussed herein. Fig. 9B The flowchart 950 may be similar to Fig.9A 900 (eg, both are continuous learning examples), except that Fig. 9B Flowchart 950 uses the applied learning algorithm 136.

[0165] Relative to Fig. 9BIn some examples, the performance metric 101 is an energy conversion efficiency 123. In some examples, the energy conversion efficiency 123 is a single efficiency 123A. In some examples, the energy conversion efficiency 123 is an overall efficiency 123B. In some examples, the performance metric 101 is a standby power 125. In some examples, the performance metric 101 is a THD metric, a ripple metric, an emissions metric, or a transient response metric. The control parameter 105 can be any of the above parameters for controlling the power stage 102.

[0166] exist Fig. 9B In operation 952, the initiation of the operation may be event-driven and / or periodically driven. In operation 952, the system performance controller 108 may detect a triggering event. In some examples, the triggering event includes the power stage 102 being activated. In some examples, the triggering event occurs during operation of the power stage 102. For example, while the power stage 102 is executing, the system performance controller 108 may detect an environmental change. In some examples, while the power stage 102 is executing, the system performance controller 108 may continuously (e.g., periodically) calculate the performance metric 101, and in response to the performance metric 101 not reaching a threshold condition, the system performance controller 108 may detect a triggering event. In some examples, the system performance controller 108 monitors a timer, and in operation 951, upon detecting that the timer has expired, the system performance controller 108 is configured to trigger Fig. 9B operation.

[0167] Upon detection of a triggering event or expiration of a timer, the control manipulation module 130 modifies the value of the control parameter 105 using the applied learning algorithm 136 in operation 954. Fig. 9B Flowchart 950 of FIG. 1 depicts modification of a single control parameter 105, but the system performance controller 108 may set and modify multiple different control parameters 105. In some examples, the control parameter 105 may have a control range that defines a range of values. The system performance controller 108 may select a value within the control range (e.g., a certain value of the ZVS control parameter 142). In some examples, the system performance controller 108 provides the value to the power supply controller 104 so that the power stage 102 operates according to the first value. In operation 956, the calculation circuit 120 calculates the performance metric 101. For example, the calculation circuit 120 receives the measurement condition 103 from the metering circuit 106 and calculates the performance metric 101 (e.g., the energy conversion efficiency 123). In operation 958, the system performance controller 108 stores or modifies the environmental conditions, the value of the control parameter 105, and the calculated performance metric 101. In some examples, the system performance controller 108 updates the historical data 140 using the environmental conditions, the value of the control parameter 105, and the calculated performance metric 101.

[0168] In operation 960, the system performance controller 108 determines whether the control range of the control parameter 105 is complete. If not, the process returns to operation 954, and the control manipulation module 130 uses the application learning algorithm 136 to select another value of the control parameter 105. If yes, then in operation 962, the system performance controller 108 determines whether the environment is the same. For example, the system performance controller 108 may determine whether one or more environmental conditions have changed (e.g., indicating that the power stage 102 has not stabilized). If one or more environmental conditions have changed, the system performance controller 108 directs the process back to operation 952. In response to the environmental conditions being detected as being the same or similar, in operation 964, the system performance controller 108 selects a program control that results in a desired performance metric (e.g., selects a value for the control parameter 105 that results in the best performance metric for the environmental range). For example, for the corresponding environmental conditions, the program control may include a value of the control parameter 105 that optimizes the performance metric 101.

[0169] In operation 966, system performance controller 108 may update the product memory of power controller 104 with the selected program control. In some examples, in operation 968, power controller 104 measures environmental conditions and applies program controls at periodic intervals.

[0170] Fig. 10A An example of a power supply system 1000 according to one aspect is shown. The power supply system 1000 includes a power stage 1002, a metering circuit 1006, a system performance controller 1008, and a power supply controller 1004. The power stage 1002, the metering circuit 1006, and the system performance controller 1008 may include any of the features described with reference to the previous figures. Fig. 10A As shown, the system performance controller 1008 includes an efficiency calculation circuit 122 and a control manipulation module 130. Figure 1A , Figure 1B and Fig. 10A , the system performance controller 1008 may set or adjust the value of the ZVS control parameter 142 (or multiple ZVS control parameters 1042 ) by monitoring and calculating the energy conversion efficiency 123 and adjusting the value of the ZVS control parameter 142 (or multiple ZVS control parameters 1042 ) to maximize the energy conversion efficiency 123 .

[0171] Fig. 10B An example of a portion of a power stage 1002 is shown, the power stage including an AC power source 1041, an inductor 1043, a first power switch 1045, a second power switch 1047, a power switch 1051, a capacitor 1057, and a resistor 1059. In some examples, the first power switch 1045 is a high-side PWM FET. In some examples, the second power switch 1047 is a low-side PWM FET. Fig. 10C A graph 1050 is shown, which depicts the drain-source voltage (V DS ) is a function of the level of switching losses. Fig. 10D A graph 1052 is shown which depicts the D The current sensing voltage threshold (V TH,CS ). Fig. 10E An efficiency curve is shown which depicts the efficiency for a given dead time t D , power stage 1002 relative to the current sensing voltage threshold (V TH,CS )’s energy conversion efficiency is 123.

[0172] refer to Figures 10B to 10E , when the drain-source voltage (V- DS ) is non-zero, activation (eg, switching on) of the second power switch 1047 may generate switching losses (eg, considerable switching losses). The switching losses are related to the parasitic capacitance 1046 (eg, C OSS ), the drain-source voltage of the second power switch 1047 (V DS ) squared and the switching frequency (F SW ). In some examples, the current sensing voltage threshold V TH,CS The appropriate value of V DS =0, which may indicate that for a given time delay t between deactivation (eg, opening) of the first power switch 1045 and activation (eg, switching on) of the second power switch 1047 D , so that the parasitic capacitance 1046 (C OSS ) required to discharge the reverse inductor current (e.g., I L,R If the magnitude of the current allowed to flow through the inductor 1043 in the reverse direction is too large (for example, I L,R If the conduction loss is above a certain level, the power stage 1002 may suffer efficiency losses as energy is extracted from the load. In other words, the conduction losses will offset any benefits of lower switching losses.

[0173] In some examples, the power supply system 1000 discussed herein includes a forced negative (or reverse) inductor current (eg, I L,R ) and uses forced commutation to achieve ZVS. However, due to the large negative inductor current (I L,R ) may reduce ZVS (which reduces switching losses) but may increase conduction losses, so an appropriate value for the reverse inductor current needs to be determined (e.g., I L,RHowever, the system performance controller 1008 may adjust the reverse inductor current (I ) based on historical or observed energy conversion efficiency 123 that may adapt to changes in environmental conditions. L,R ).

[0174] In some examples, the ZVS control parameters 142 include the inductor reverse current (I L,R ) for the time delay t between the deactivation of the first power switch 1045 and the activation of the second power switch 1047 D , which is sufficient to make the parasitic capacitance 1046 (C OSS ) discharge. In some examples, the ZVS control parameters 1042 include a current sensing voltage threshold V TH,CS , the current sensing voltage threshold indicates that for a given time delay the parasitic capacitance 1046 (C OSS )The reverse inductor current (I L,R In some examples, the ZVS control parameter 1042 includes a time value for activating the second power switch 1047 for a longer time than the first power switch 1045 so that the reverse inductor current (I L,R ) causes the parasitic capacitance 1046 (C OSS ) discharge.

[0175] In some examples, the system performance controller 1008 sets or adjusts the ZVS control parameter 142 until the energy conversion efficiency 123 reaches a threshold condition. For example, the system performance controller 1008 can determine a value of the ZVS control parameter 1042 that optimizes the energy conversion efficiency 123. TH,CS As an example of the ZVS control parameter 1042, the control manipulation module 1030 may be a current sensing voltage threshold V TH,CS The first value is set and provided to the power controller 1004 so that the power stage 1002 operates according to the first value. The metering circuit 1006 can obtain the measurement conditions 103 (e.g., input voltage, input current, output voltage, and output current) and provide the measurement conditions 103 to the system performance controller 1008 in a digital format.

[0176] The efficiency calculation circuit 122 can calculate the energy conversion efficiency 123 based on the measurement condition 103. If the energy conversion efficiency 123 does not reach the threshold condition, the control manipulation module 130 can set the current sensing voltage threshold V TH,CSThe second value is selected and provided to the power controller 1004 so that the power stage 1002 operates according to the second value. The metering circuit 1006 may obtain the measurement condition 103, and the efficiency calculation circuit 122 may calculate the energy conversion efficiency 123 based on the measurement condition 103. If the energy conversion efficiency 123 reaches the threshold condition, the control manipulation module 130 may select the second value and instruct the power controller 104 to set or update the ZVS control parameter 1042 with the second value.

[0177] In some conventional methods, developers can measure indirect parameters as a measure of energy conversion efficiency, and then adjust the ZVS control parameters to optimize the indirect parameters. For example, the sudden time discharge rate can be measured as an indirect way to measure the energy conversion efficiency. For example, the sudden time discharge rate can indirectly correspond to a certain negative current flowing through the inductor, which can indirectly correspond to certain switching losses, which can indirectly correspond to the energy conversion efficiency. In this way, developers can adjust parameters to optimize the sudden time discharge rate, which can indirectly improve the energy conversion efficiency of the power stage. However, using indirect parameters as a function of energy conversion efficiency 123 may result in a relatively long time to market. In addition, the energy conversion efficiency 123 may be negatively affected by the environment of the power system 1000 (e.g., external stimuli) and manufacturing variations of components of the power system 1000.

[0178] However, the system performance controller 1008 directly calculates the energy conversion efficiency 123 (e.g., it does not use an indirect parameter as a measure of the energy conversion efficiency), and then sets or adjusts the ZVS control parameters 142 that will cause the energy conversion efficiency 123 to reach a threshold condition. Furthermore, in some examples, the system performance controller 1008 monitors and calculates the energy conversion efficiency 123 during the life of the power stage, and can programmatically re-evaluate and potentially select new values ​​for the ZVS control parameters 142 that maximize the energy conversion efficiency 123 or maintain it at a threshold level, regardless of changes in the environment, aging of components, and / or thermal effects of the power stage 1002.

[0179] Fig.11 An example of a power system 1100 according to one aspect is shown. In some examples, the power system 1100 is a totem pole PFC power system. The power system 1000 may include any of the features discussed herein with respect to other figures. The power system 1100 includes a power stage 1102, a metering circuit 1106, a system performance controller 1108, and a power controller 1104. In some examples, the system performance controller 1108 is configured to obtain one or more ZVS control parameters 142 in a manner that optimizes the energy conversion efficiency 123 by monitoring and calculating the energy conversion efficiency 123 and adjusting the one or more ZVS control parameters 142 based on the energy conversion efficiency 123. Fig.11As shown, the system performance controller 1108 includes the efficiency calculation circuit 122 and the control manipulation module 130 .

[0180] The power stage 102 may include an AC power source 1141, an inductor 1143 (eg, L BOOST ), a first power switch 1145, a second power switch 1147, a third power switch 1149, and a fourth power switch 1151. In some examples, the first power switch 1145 is a high-side PWM FET. In some examples, the second power switch 1147 is a low-side PWM FET. In some examples, the first power switch 1145 and the second power switch 1147 are switched at a PWM frequency. In some examples, the first power switch 1145 and the second power switch 1147 are operated according to a critical conduction mode (CrM), wherein the on-time of the first power switch 1145 is constant, and the switching cycle is terminated when the inductor current drops to zero, resulting in a variable PWM frequency or a continuous conduction mode (CCM) at a fixed frequency. In some examples, the third power switch 1149 and the fourth power switch 1151 are switched at an AC frequency. In some examples, the power stage 102 may include a variable resistor 1153, a resistor 1155, and a capacitor 1157, wherein the power stage 1102 is coupled to a load 1159.

[0181] The metering circuit 1106 is configured to monitor the measurement conditions 103 of the power stage 1102. The measurement conditions 103 include input voltage, input current, output voltage, and output current. The metering circuit 1106 includes: a first metering circuit 1106-1, which is configured to detect input conditions (e.g., input voltage and input current); and a second metering circuit 1106-2, which is configured to detect input conditions (e.g., output voltage and output current).

[0182] The first metering circuit 1106-1 includes a first current sensor 1114-1 configured to detect an input current (eg, I- IN). In some examples, the first current sensor 114-1 is connected between the AC power source 1141 and the inductor 1143. The first metering circuit 1106-1 includes a plurality of resistors coupled to the AC line at a location close to the AC power source 1141. The plurality of resistors include a resistor 1163, a resistor 1164, a resistor 1165, and a resistor 1166. The first metering circuit 1106-1 includes a first multiplexer 1118-1 and a first ADC 1116-1. The first ADC 1116-1 is connected to an output of the first multiplexer 1118-1, and the output of the first ADC 1116-1 is connected to an input of the system performance controller 1108. The first multiplexer 1118-1 is configured to receive a signal representing an input voltage and a signal representing an input current, and multiplex the signals to generate a single signal, which is provided to the first ADC 1116-1. The first ADC 1116 - 1 is configured to convert the signal into a digital format detectable by the system performance controller 1108 .

[0183] The second metering circuit 1106-2 includes a second current sensor 1114-2 configured to detect an output current (eg, I OUT ) (e.g., also referred to as load current). The second metering circuit 1106-2 includes a plurality of resistors coupled to the output of the power stage 1102. The plurality of resistors include resistor 1161 and resistor 1162. The second metering circuit 1106-2 includes a second multiplexer 1118-2 and a second ADC 1116-2. The second ADC 1116-2 is connected to the output of the second multiplexer 1118-2, and the output of the second ADC 1116-2 is connected to the input of the system performance controller 1108. The second multiplexer 1118-2 is configured to receive a signal representing the output voltage and a signal representing the output current, and multiplex the signals to generate a single signal, which is provided to the second ADC 1116-2. The second ADC 1116-2 is configured to convert the signal into a digital format detectable by the system performance controller 1108.

[0184] The efficiency calculation circuit 122 is configured to monitor and calculate the energy conversion efficiency 123 of the power stage 1102 based on the input voltage, input current, output voltage, and output current received from the metering circuit 1106. The control manipulation module 130 manipulates one or more ZVS parameters 142 based on the energy conversion efficiency 123 according to any of the techniques described herein. The system performance controller 1108 is configured to provide the ZVS parameters 142 to the power supply controller 1104.

[0185] Fig.12An example of a power supply controller 1104 is shown according to an aspect. The power supply controller 1104 is configured to receive the ZVS parameters 142 from the system performance controller 1108 .

[0186] The power controller 1104 includes a plurality of analog drivers configured to drive control of power switches of the power stage 1102. For example, the power controller 1104 may include: a first analog driver 1181 configured to generate a control signal (e.g., SRH) for the third power switch 1149; a second analog driver 1182 configured to generate a control signal (e.g., SRL) for the fourth power switch 1151; a third analog driver 1183 configured to generate a control signal (e.g., PWML) for the second power switch 1147; and a fourth analog driver 1184 configured to generate a control signal (e.g., PWMH) for the first power switch 1145.

[0187] The power controller 1104 may include: a differential sensor 1191 configured to receive a signal (BR1) and a signal (BR2) and generate a difference signal reflecting the difference between BR1 and BR2; and a polarity logic 1168 configured to determine the polarity of the difference signal. The output of the polarity logic 1168 is connected to the first analog driver 1181 and the second analog driver 1182.

[0188] The power supply controller includes a digital compensator 1170 configured as a digital compensation signal (FB), a multi-slope ramp generator 1171, a PWM logic circuit 1169, and a gate drive logic 1175. The PWM logic circuit 1169 is configured to receive the ZVS parameter 142 from the system performance controller 1108. The PWM logic circuit 1169 includes a PWM comparator 1172, a logic component 1173, and a trigger circuit 1174. The PWM comparator 1172 can compare the output of the digital compensator 1170 and the output of the multi-slope ramp generator 1171. The logic component 1173 is connected to the PWM comparator 1172. The logic component 1173 is connected to an input terminal (e.g., a reset terminal) of the trigger circuit 1174. The trigger circuit 1174 is configured to receive a zero crossing detector (ZCD) signal and / or a watchdog timer (WDT) signal. In some examples, a delay is added to the ZCD signal, where the amount of the delay is determined by the ZVS control parameters 142 received from the system performance controller 1108. The output of the trigger circuit 1174 is connected to the input of the gate drive logic 1175. The output of the gate drive logic 1175 is connected to the third analog driver 1183 and the fourth analog driver 1184.

[0189] like Fig.13As shown, the power controller 1104 may include comparators 1176, 1177, 1178, 1179, and a logic component 1180. The comparator 1176 converts the voltage of the current sensing signal (eg, V CS,TH ) is compared with a first voltage threshold, comparator 1177 compares the voltage of the current sensing signal with a second voltage threshold, comparator 1178 compares the voltage of the current sensing signal with a third voltage threshold, and comparator 1179 compares the voltage of the current sensing signal with a negative threshold voltage (e.g., V CS,TH ) is compared. In some examples, the negative threshold voltage V CS,TH Indicates the amount of delay added to the ZCD signal.

[0190] Fig.13 A flowchart 1300 depicting exemplary operation of the power supply system 1100 is shown according to one aspect. Fig.11 The power supply system 1100 is described in the flowchart 1300, but Fig.13 The operation can be applied to any power supply system described in this article.

[0191] In operation 1302, the system performance controller 1108 detects a triggering event. In some examples, the triggering event includes activation of the power stage 1102. In some examples, the triggering event includes detection of a change in one or more environmental conditions. In some examples, the triggering event includes expiration of a timer. In some examples, the triggering event includes detection that the energy conversion efficiency 123 does not meet a threshold condition.

[0192] In operation 1304, the system performance controller 1108 selects a value for the ZVS control parameter 142 based on the energy conversion efficiency 123. In some examples, the ZVS control parameter 142 indicates a reverse current (I ) flowing through the inductor 1143 to discharge the parasitic capacitance of the second power switch 1147. L,R ). In some examples, the ZVS control parameter 142 is an indication of the reverse inductor current (I L,R ) of the current sensing voltage threshold (V TH,CS In some examples, the ZVS parameters include a time value for delaying activation of the second power switch 1147 so that the reverse inductor current (I L,R ) causes the parasitic capacitance of the second power switch 1147 to discharge. In some examples, the system performance controller 1108 selects a combination of ZVS control parameters 142 (eg, V CS,TH , time delay t D ).

[0193] For example, in operation 1301, system performance controller 1108 receives measurement conditions 103 of power stage 1102. In some examples, system performance controller 1108 receives measurement conditions 103 from metering circuit 1106. Measurement conditions 103 may include input voltage, input current, output voltage, and output current. In some examples, metering circuit 1106 includes one or more ADCs (e.g., ADC 1116-1, ADC 1116-2) and one or more multiplexers (e.g., multiplexer 1118-1, multiplexer 1118-2).

[0194] In operation 1303, the efficiency calculation circuit 122 calculates the energy conversion efficiency 123 (eg, (V OUT *I OUT ) / (V IN *I IN )). In operation 1305, the control manipulation module 130 modifies the ZVS control parameter 142 until the energy conversion efficiency 123 reaches the threshold condition. In some examples, the control manipulation module 130 selects a first value for the ZVS parameter 142 and sends the first value to the power controller 1104. Then, the efficiency calculation circuit 122 calculates the energy conversion efficiency 123 of the power stage 1102 operating according to the first value based on the measurement condition 103. Then, the control manipulation module 130 modifies the first value to obtain a second value in response to the energy conversion efficiency 123 not reaching the threshold condition, and sends the second value to the power controller 1104. Then, the efficiency calculation circuit 122 calculates the energy conversion efficiency 123 of the power stage 1102 operating according to the second value based on the measurement condition 103.

[0195] In some examples, the control manipulation module 130 uses the AI ​​algorithm 132 to modify the ZVS parameter 142. In some examples, the AI ​​algorithm 132 is a target search algorithm 134. In some examples, the AI ​​algorithm is an applied learning algorithm 136. In some examples, the control manipulation module 130 applies the historical data 140 to the neural network 138 to predict a value of the ZVS parameter 142 that causes the energy conversion efficiency 123 to reach a threshold condition. In some examples, the system performance controller 1108 determines a first value of the ZVS parameter 142 that causes the energy conversion efficiency 123 to reach a threshold condition for a first environmental condition, and determines a second value of the ZVS parameter 142 that causes the energy conversion efficiency 123 to reach a threshold condition for a second environmental condition. In some examples, the first environmental condition and the second environmental condition are different voltages.

[0196] In operation 1306, the system performance controller 1108 provides the selected value to the power controller 1104 to utilize the selected value to control the power stage 1102. In some examples, the selected value is stored in a program memory at the power controller 1104. In some examples, the system performance controller 1108 updates the program memory with the selected value.

[0197] Fig.14 1400 is shown depicting an exemplary operation of the power system 1100 according to one aspect. The flowchart 1400 involves determining a ZVS control parameter 142 that optimizes or provides an energy conversion efficiency 123 that reaches a threshold condition using a target search algorithm 134. In some examples, the ZVS control parameter 142 is a current sensing voltage threshold (V TH,CS ). TH,CS Indicator is used to indicate the time delay (t D ) period makes the parasitic capacitance (C OSS ) discharges to achieve the reverse current (I L,R ) value. Although the power supply system 1100 is described Fig.14 Flowchart 1400, but Fig.14 The operation can be applied to any power system discussed in this article.

[0198] Similar to Figure 7 In some examples, the operations of flowchart 1400 may involve a fixed product solution, where the system performance controller 1108 is used to determine V TH,CS The system performance controller 1108 is configured to initiate the ZVS parameter 142 during development, testing, and / or production of the power system 1100. Fig.14 In some examples, system performance controller 1108 is configured to initiate a power stage 1102 in response to being activated (eg, turned on). Fig.14 operation.

[0199] In operation 1402, the power stage 1102 is activated (eg, powered on). In operation 1404, the system performance controller 108 sets one or more environmental conditions. For example, the system performance controller 108 sets the input voltage (V IN ) and output current (I O ). In operation 1406, the control manipulation module 130 modifies V TH,CS In some examples, the system performance controller 108 modifies V TH,CS , while other ZVS control parameters 142 are fixed (e.g., time delay t D). Note that with time delay t D becomes smaller, the required reverse inductor current (I L,R , such as V TH,CS As shown in the figure, the value of the parasitic capacitance C OSS Discharge and obtain ZVS. In some examples, the system performance controller 1108 can identify V TH,CS and t D A combination of the two and storing the preferred result.

[0200] In some examples, V TH,CS The control range is 0mV to -300mV. The system performance controller 1108 sets V TH,CS The value of V is provided to the power controller 1104 so that the power stage 1102 operates according to the value. In operation 1408, the efficiency calculation circuit 122 is configured to calculate the energy conversion efficiency 123. For example, the system performance controller 1108 receives the measurement condition 103 from the metering circuit 1106, and the efficiency calculation circuit 122 calculates the energy conversion efficiency 123 based on the measurement condition 103. In operation 1410, the system performance controller 1108 stores V TH,CS The value of and the calculated energy conversion efficiency 123. In some examples, the system performance controller 108 stores the environmental conditions (e.g., V IN and I O Setting value), V TH,CS The value of and the calculated energy conversion efficiency 123.

[0201] In operation 1412, the system performance controller 1108 determines whether the desired energy conversion efficiency is achieved, for example, whether the energy conversion efficiency 123 reaches a threshold condition (e.g., maximized or equal to or greater than a threshold level). If not, the operation returns to operation 1406, and the control manipulation module 130 sets a target value for V based on the target search algorithm 134. TH,CS Select another value. If yes, then in operation 1414, the system performance controller 1108 determines whether the environmental range is complete, for example, whether the range of environmental conditions has been tested. If no, the system performance controller 1108 directs the process back to operation 1404 to set V IN and I O At least one of sets another value.

[0202] In operation 1416, a program control that achieves a desired energy conversion efficiency 123 may be selected (e.g., V is selected for a range of environments). TH,CSThe program control may be stored in a product memory of the power controller 1104 in operation 1418. In operation 1420, the power controller 1104 applies the program control to the power stage 1102. In some examples, in operation 1422, the power controller 1104 measures the environmental conditions and applies the program control at periodic intervals.

[0203] Fig.15 1500 is shown depicting an exemplary operation of the power system 1100 according to one aspect. The flowchart 1500 involves determining a ZVS control parameter 142 that optimizes or provides energy conversion efficiency 123 that reaches a threshold condition using an applied learning algorithm 136. In some examples, the ZVS control parameter 142 is a current sensing voltage threshold (V TH,CS ). TH,CS Indicator is used to indicate the time delay (t D ) period makes the parasitic capacitance (C OSS ) discharges to achieve the reverse current (I L,R ) value. Although the power supply system 1100 is described Fig.15 Flowchart 1500, but Fig.15 The operation can be applied to any power system discussed in this article. Fig.15 The flowchart 1500 may be similar to Fig.14 1400, except that Fig.15 Flowchart 1500 uses the applied learning algorithm 136. In addition, Fig.15 The flowchart 1500 may be similar to Figure 7 The flowchart 700, in a sense, Fig.15 Flowchart 1500 of FIG. 1500 relates to a fixed product solution, where the system performance controller 1108 is used to determine V TH,CS The optimal value of the control parameter 105 is determined and stored in the product memory of the power controller 1104.

[0204] In some examples, system performance controller 1108 is configured to initiate a power system during development, testing, and / or production. Fig.15 In some examples, system performance controller 1108 is configured to initiate a power stage 1102 in response to being activated (eg, turned on). Fig.15 operation.

[0205] In operation 1502, the power stage 1102 is activated (eg, powered on). In operation 1504, the system performance controller 1108 sets one or more environmental conditions. For example, the system performance controller 1108 may set the input voltage (V IN )、output current (IO ) and one or more properties of the second power switch 1147 (in some examples, a FET). In some examples, the properties of the second power switch 1147 include an on-resistance (R ON ) and capacitance (C OSS In operation 1506, the control manipulation module 130 modifies V according to the application learning algorithm 136. TH,CS ,t D (time delay) and V- O (output voltage). In some examples, the control manipulation module 130 is configured to train and adjust the neural network 138 based on the historical data 140 to predict the energy conversion efficiency 123. The system performance controller 1108 stores the results in a memory (e.g., the environmental conditions (e.g., V IN ,I O and FET), ZVS control parameters 142 (e.g., V TH,CS 、T D and V O ) and the predicted energy conversion efficiency). Then, the control manipulation module 130 is configured to select V that maximizes the energy conversion efficiency 123. TH,CS 、T D and V O .

[0206] The system performance controller 1108 sets the selected V TH,CS 、T D and V- O The power controller 1104 is provided so that the power stage 1102 operates according to these values. In operation 1508, the efficiency calculation circuit 122 calculates the energy conversion efficiency 123. For example, the system performance controller 1108 receives the measurement conditions 103 from the metering circuit 1106, and the efficiency calculation circuit 122 calculates the energy conversion efficiency 123 based on the measurement conditions 103. In operation 1510, the system performance controller 1108 stores the environmental conditions (e.g., V IN ,I O and FET), ZVS control parameters 142 (e.g., V TH,CS 、T D and V O ) and the calculated energy conversion efficiency 123.

[0207] In operation 1512, the system performance controller 1108 determines whether the control range of the control parameter 105 is complete. If not, the process returns to operation 1506 to cycle through V TH,CS ,t D and V O other values, and the control manipulation module 130 modifies V according to the application learning algorithm 136 TH,CS 、TD and V O One or more of .

[0208] Table 1 below shows an example of the history data 140 .

[0209]

[0210]

[0211] In some examples, the control manipulation module 130 is configured to first step through the inputs that are likely to be easiest to control while holding the more difficult inputs fixed. In some examples, the following inputs are listed from most difficult to least difficult: GaN FET, V O ,t D 、V TH,CS 、V IN ,I O In some examples, in operation 1504, the control manipulation module 130 cycles through I O , then change V IN , and again loops through I O , and single-step through V IN , until all steps are completed. Then, the control module 130 changes V TH,CS .

[0212] If yes, then in operation 1514, the system performance controller 1108 determines whether the environmental range is complete, for example, whether the range of environmental conditions (e.g., V IN ,I O and FET range). If not, the system performance controller 1108 directs the process back to operation 1504 to modify V IN ,I O and one or more of the FETs. If so, the process proceeds to operation 1516.

[0213] In operation 1516, the control manipulation module 130 is configured to respond to one or more environmental conditions (e.g., V IN ,I O The V that achieves the target energy conversion efficiency is selected by using different combinations of TH,CS 、T D and V O In operation 1518, V TH,CS and t D The selected value of is stored in the product memory of the power controller 1104. In operation 1520, the power controller 1104 sets the program control (e.g., V TH,CS and t D) is applied to the power stage 1102. In operation 1522, in some examples, the power controller 1104 measures the environmental conditions and applies application control at periodic intervals (e.g., based on V IN , IO At each subsequent power-up, V TH,CS ,t D ).

[0214] Fig.16 16 is a flowchart 1600 depicting an exemplary operation of the power system 1100 according to another aspect. For example, when the power stage 1102 is operating, the system performance controller 1108 is configured to perform Fig.16 1600 (including activation of power stage 1102). Flowchart 1600 involves determining, using target search algorithm 134, a ZVS control parameter 142 that optimizes or provides energy conversion efficiency 123 that reaches a threshold condition. In some examples, ZVS control parameter 142 is a current sensing voltage threshold (V TH,CS ). TH,CS Indicator is used to indicate the time delay (t D ) period makes the parasitic capacitance (C OSS ) discharges to achieve the reverse current (I L,R ) value. Although the power supply system 1100 is described Fig.16 Flowchart 1600, but Fig.16 The operation can be applied to any power system discussed in this article.

[0215] In operation 1602, the system performance controller 1108 may detect a triggering event. In some examples, the triggering event includes the power stage 102 being activated. In some examples, the triggering event occurs during operation of the power stage 1102. For example, while the power stage 1102 is executing, the system performance controller 1108 may detect a change in environmental conditions, thereby detecting the triggering event. In some examples, while the power stage 1102 is executing, the efficiency calculation circuit 122 may continuously (e.g., periodically) calculate the energy conversion efficiency 123, and in response to the energy conversion efficiency 123 not reaching a threshold condition, the system performance controller 1108 may detect the triggering event. In some examples, the system performance controller 1108 monitors a timer, and in operation 1601, upon detecting that the timer has expired, the system performance controller 1108 is configured to trigger Fig.16 operation.

[0216] Upon detecting a trigger event or timer expiration, in operation 904, the control manipulation module 130 uses the target search algorithm 134 to modify V TH,CS In some examples, the system performance controller 1108 modifies VTH,CS , while other ZVS control parameters 142 are fixed (e.g., time delay t D ). The system performance controller 1108 converts V TH,CS The value of is provided to the power controller 1104 so that the power stage 1102 operates according to the value.

[0217] In operation 1606, the efficiency calculation circuit 122 is configured to calculate the energy conversion efficiency 123. For example, the system performance controller 1108 receives the measurement condition 103 from the metering circuit 1106, and the efficiency calculation circuit 122 calculates the energy conversion efficiency 123 based on the measurement condition 103. In operation 1608, the system performance controller 1108 stores V TH,CS The value of and the calculated energy conversion efficiency 123. In some examples, the system performance controller 1108 stores the environmental conditions (e.g., V IN and I O The measured value of V TH,CS The value of and the calculated energy conversion efficiency 123.

[0218] In operation 1610, the system performance controller 1108 determines whether the desired energy conversion efficiency is achieved, for example, whether the energy conversion efficiency 123 reaches a threshold condition (e.g., maximized or equal to or greater than a threshold level). If not, the operation returns to operation 1604, and the control manipulation module 130 sets a target value for V based on the target search algorithm 134. TH,CS Another value is selected. If yes, then in operation 1612, the system performance controller 1108 determines whether the environment is substantially the same. If no, the process loops back to operation 1602 until system stability is detected. If yes, then the process proceeds to operation 1614.

[0219] In operation 1614, the system performance controller 1108 selects a V that yields a describable energy conversion efficiency. TH,CS , for example, V TH,CS The value that causes the energy conversion efficiency 123 to reach the threshold condition is selected. In operation 1616, the system performance controller 1108 sets or modifies the selected V in the product memory of the power supply controller 104. TH,CS In some examples, at operation 1618 , power controller 1104 measures environmental conditions and applies application controls at periodic intervals.

[0220] Fig.17 1700 is shown depicting an exemplary operation of the power system 1100 according to another aspect. For example, when the power stage 1102 is operating, the system performance controller 1108 is configured to perform Fig.171700 (including periodic activation of power stage 1102). Flowchart 1700 involves determining, using applied learning algorithm 136, a ZVS control parameter 142 that optimizes or provides energy conversion efficiency 123 that reaches a threshold condition. In some examples, ZVS control parameter 142 is a current sensing voltage threshold (V TH,CS ). TH,CS Indicator is used to indicate the time delay (t D ) period makes the parasitic capacitance (C OSS ) discharges to achieve the reverse current (I L,R ) value. Although the power supply system 1100 is described Fig.17 Flowchart 1700, but Fig.17 The operation can be applied to any power system discussed in this article.

[0221] In operation 1702, the system performance controller 1108 may detect a triggering event. In some examples, the triggering event includes the power stage 1102 being activated. In some examples, the triggering event occurs during operation of the power stage 1102. For example, while the power stage 1102 is executing, the system performance controller 1108 may detect a change in environmental conditions, thereby detecting the triggering event. In some examples, while the power stage 1102 is executing, the system performance controller 1108 may continuously (e.g., periodically) calculate the energy conversion efficiency 123, and in response to the energy conversion efficiency 123 not reaching a threshold condition, the system performance controller 1108 may detect the triggering event. In some examples, the system performance controller 1108 monitors a timer, and in operation 1701, upon detecting that the timer has expired, the system performance controller 1108 is configured to trigger Fig.17 operation.

[0222] When a trigger event is detected or the timer expires, in operation 1704, the control manipulation module 130 modifies V according to the application learning algorithm 136. TH,CS ,t D (time delay) and V- O (output voltage). In some examples, the control manipulation module 130 is configured to train and adjust the neural network 138 based on the historical data 140 to predict the energy conversion efficiency 123. The system performance controller 1108 stores the results in a memory (e.g., the environmental conditions (e.g., V IN ,I O and FET), ZVS control parameters 142 (e.g., V TH,CS 、T D and V O ) and the predicted energy conversion efficiency). Then, the control manipulation module 130 is configured to select V that maximizes the energy conversion efficiency 123.TH,CS , T D and V O .

[0223] The system performance controller 1108 sets the selected V TH,CS ,t D and V- O The power controller 1104 is provided so that the power stage 1102 operates according to these values. In operation 1706, the efficiency calculation circuit 122 calculates the energy conversion efficiency 123. For example, the system performance controller 1108 receives the measurement conditions 103 from the metering circuit 1106 and calculates the energy conversion efficiency 123 based on the measurement conditions 103. In operation 1708, the system performance controller 1108 stores the environmental conditions (e.g., V IN ,I O and FET), ZVS control parameters 142 (e.g., V TH,CS , T D and V O ) and the calculated energy conversion efficiency 123.

[0224] In operation 1710, the system performance controller 1108 determines whether the control range of the ZVS control parameter 142 is complete. If not, the process returns to operation 1704 to cycle through V TH,CS ,t D and V O other values, and the control manipulation module 130 modifies V according to the application learning algorithm 136 TH,CS ,t D and V O If yes, then in operation 1712, the system performance controller 1108 determines whether the environment is substantially the same. If no, the system performance controller 108 directs the process back to operation 1702 until stability is detected. If yes, then the process proceeds to operation 1714.

[0225] In operation 1714, the control and manipulation module 130 is configured to select a V that obtains the target energy conversion efficiency 123 for the current environmental conditions. TH,CS , T D and V O In operation 1716, the system performance stores or modifies V in the product memory of the power controller 1104. TH,CS and t D In operation 1718, the power controller 1104 detects conditions based on the periodic events and applies the application.

[0226] Fig.18A and Fig.18BA power system 1800 is shown with a system performance controller 1808, which is configured to optimize the performance of the power system 1800 based on the overall efficiency 123B of the power system 1800. In some examples, the power system 1800 includes a cascaded power system. In some examples, the power system 1800 is configured to power one or more server computers (e.g., cloud-based computers). In some examples, the power system 1800 is configured to power one or more telecommunication components (e.g., base stations). The power system 1800 may include any of the features discussed with reference to the previous figures.

[0227] like Fig.18A and Fig.18B As shown, the power supply system 1800 may include: a front-end module 1804, which is configured to receive an AC voltage and generate an input voltage (eg, a DC voltage) (V IN ); a plurality of power stages 1802, which are configured to receive an input voltage (VIN) and generate an output voltage (V OUT ), the output voltage is provided to a load 1809. In some examples, the front end module 1804 includes a PFC front end AC-DC module.

[0228] The plurality of power stages 1802 include a first power stage 1802-1 and a second power stage 1802-2. Fig.18A and Fig.18B Only two power stages are shown in FIG. 1 , but the power supply system 1800 may include any number of power stages, such as three power stages, four power stages, five power stages, or more than five power stages. In some examples, the first power stage 1802-1 is a DC power stage (e.g., a DC-DC converter), and the second power stage 1802-2 is a DC power stage. Fig.18A and Fig.18B As shown, power stage 1802 is connected in series with a second power stage 1802 - 2 coupled to a load 1809 .

[0229] The power system 1800 may include a system host device 1810 that is communicatively coupled to each of the power stages 1802. In some examples, the system host device 1810 includes a multiphase controller. In some examples, the system host device 1810 may include any of the features described with reference to the power controller of the previous figures. In some examples, the system host device 1810 is located at a location separate from the power stage 1802. In some examples, the system host device 1810 is included in a server computer. The system host device 1810 is communicatively coupled to the first power stage 1802-1 via a first communication link. In some examples, the first communication link is a network communication link (e.g., a network-based communication link, such as the Internet). In some examples, the communication link is a power management bus (e.g., PMBus) (e.g., a two-wire communication protocol). The system host device 1810 is communicatively coupled to the second power stage 1802-2 via a second communication link. In some examples, the second communication link is a network communication link. In some examples, the second communication link is a power management bus. Additionally, a system host device 1810 may be communicatively coupled to the front end module 1804 .

[0230] The system host device 1810 may include a system performance controller 108. The system performance controller 108 may include any of the features described with respect to the previous figures. In general, the system host device 1810 (which executes the system performance controller 108) is configured to monitor environmental conditions, measurement conditions, and control parameters 105, and apply any operations discussed herein to obtain a target overall efficiency 123B. In some examples, as described above, the system performance controller 1808 may determine the optimized control parameters 105 that cause the overall efficiency 123B to reach a threshold condition, and then the control parameters 105 may be stored as historical data 140, which is then used during operation of the power system 1800. In some examples, as discussed above, the system performance controller 108 may be executed during the life of the system so that the control parameters 105 are readjusted as the environment changes (including thermal effects on the power system 1800) and the aging of the components of the power system 1800 to help ensure that the power system 1800 achieves the target overall efficiency threshold.

[0231] In some examples, system performance controller 1808 is configured to control power stage 1802 in a manner that optimizes or causes overall energy conversion efficiency 123B to reach a threshold condition. Fig.18A As shown, both the first power stage 1802-1 and the second power stage 1802-2 can be activated to provide power to the load 1809. However, if both power stages 1802 are activated, the overall efficiency 123B will decrease when operating at a relative load. Fig.18BAs shown, it may be more efficient to disable (e.g., isolate) one of the power stages 1802. Fig.18B As shown, the first power stage 1802 - 1 is disabled and the load 1809 is powered by the second power stage 1802 - 2 .

[0232] The system performance controller 1808 can set or adjust one or more control parameters 105 of the power stage 1802 based on the overall efficiency 123B. In some examples, the control parameters 105 include phase change parameters 158 (e.g., phase cut / add transition points). The phase change parameters 158 may include one or more efficiency transition points that indicate when to deactivate (or activate) one or more power stages 1802, such that the efficiency transition points cause the overall efficiency 123B to reach a threshold condition. In some examples, the efficiency transition points are at a particular I O Activation / deactivation of power stage 102 at the value.

[0233] For example, the system performance controller 108 is configured to determine when to enable and / or disable phases (e.g., power stages 1802) in a manner that optimizes overall efficiency 123B. In some examples, the system performance controller 108 may optimize phase shedding transitions to maximize overall efficiency 123B over a power curve (e.g., an entire power curve). For example, one way to reduce switching losses (e.g., at light loads) is to reduce the number of power stages 1802 in operation, e.g., to reduce the number of phases in operation (e.g., referred to as phase shedding). Thus, phase shedding may be defined as when the load current (e.g., I O ) is a process of being able to cut an increasing number of phases when a threshold is reduced. Or conversely, the process can be defined as phase addition. In some examples, under light load, phase cutting can increase the overall efficiency 123B of the power system 1800.

[0234] For example, the system performance controller 108 may include any of the features previously described, such as the efficiency calculation circuit 122 and the control manipulation module 130. The system performance controller 108 is configured to receive the measurement conditions 103 from the first power stage 1802-1 via a first communication link and to receive the measurement conditions 103 from the second power stage 1802-2 via a second communication link. For each of the power stages 1802, the measurement conditions 103 may include any of the previously described information, such as input voltage, input current, output voltage, output current, etc. The system performance controller 108 is configured to calculate the overall efficiency 123B of the power stage 1802 as a whole, and then determine or adjust the phase change parameters 158 so that the overall efficiency 123B reaches the threshold condition. In some examples, the system performance controller 108 calculates a single efficiency 123A for each of the power stages 1802. The system host device 1810 is configured to control the power stage 1802 according to the set or adjusted phase change parameters 158.

[0235] Fig.19 A graph 1900 is shown which illustrates the energy conversion efficiency of the power system 1800 versus increasing load conditions (eg, I O ). For example, graph 1900 depicts an efficiency curve 1951 for a first power stage, an efficiency curve 1953 for a second power stage, an efficiency curve 1955 for a third power stage, and an efficiency curve 1957 from a fourth power stage. Under light load conditions, the first power stage may be activated, and the second to fourth power stages may be deactivated (or isolated). As load conditions increase (and relative to phase addition), the system performance controller 108 is configured to determine a transition point 1952 for activating the second power stage, a transition point 1954 for activating the third power stage, and a transition point 1956 for activating the fourth power stage in a manner that optimizes the overall efficiency curve 1950. These transition points may also be used for phase reduction. The system host device 1810 may be configured to enable / disable phases based on efficiency transition points, such as Fig.19 As indicated by the arrow in .

[0236] In some examples, the control parameter 105 includes the intermediate bus voltage 1895. For example, referring to Fig.18A and Fig.18B , the voltage between the first power stage 1802-1 and the second power stage 1802-2 is denoted as V IB In some examples, system performance controller 1808 may determine or adjust V in a manner that causes overall efficiency 123B to reach a threshold condition. IBIn some examples, control parameters 105 include I / O conditions 161. I / O conditions 161 may include input voltages and / or output voltages of one or more of power stages 1802. For example, the level of input / output voltages may be control parameters 105 optimized by system performance controller 108 to achieve a target overall efficiency condition.

[0237] Fig. 20A and Fig. 20B A power system 2002 is shown having a plurality of power stages 2000 connected in parallel. For example, the power system 2000 is similar to Fig.18A and Fig.18B 1800, except that the power stages 2002 are connected in parallel (e.g., as opposed to being connected in series). Thus, the power system 2000 may include any of the features of the power system 1800. For example, the power system 2000 may include a system host device 2010 having a system performance controller 108. The power system 2000 may include: a front end module 2004 configured to receive an AC voltage and generate an input voltage (e.g., a DC voltage) (V IN ); a plurality of power stages 2002, which are configured to receive an input voltage (VIN) and generate an output voltage (V OUT ), the output voltage is provided to a load 2009. The plurality of power stages 2002 include a first power stage 2002-1 and a second power stage 2002-2.

[0238] Fig.21 According to one aspect, a FIG. 18A to FIG. 18B and FIG. 20A to FIG. 20B 2100 of an exemplary operation of a power supply system. FIG. 18A to FIG. 18B The power supply system 1800 is described in the flowchart 2100, but Fig.21 The operation may be applicable to any of the power systems described herein.

[0239] In operation 2102, the system performance controller 108 detects a triggering event. In some examples, the triggering event includes activation of the power stage 1802. In some examples, the triggering event includes detection of a change in one or more environmental conditions. In some examples, the triggering event includes expiration of a timer. In some examples, the triggering event includes detection that the overall efficiency 123B does not meet a threshold condition.

[0240] In operation 2104, the system performance controller 108 updates the value of the control parameter 105 used to control the plurality of power stages 1802 based on the overall efficiency 123B. In some examples, the control parameter 105 is a phase change parameter 158. In some examples, the phase change parameter 158 defines one or more efficiency transition points that indicate when to deactivate or activate the timing of one or more power stages 1802. In some examples, the power stages 1802 are connected in series. In some examples, they are connected in parallel (e.g., as Fig. 20A and Fig. 20B In some examples, the control parameter 105 is an intermediate bus voltage 160. In some examples, the control parameter 105 is an I / O condition 161.

[0241] For example, in operation 2101, the system performance controller 108 may receive measurement conditions 103 of the power stage 1802. The measurement conditions 103 include a DC input voltage, an input current, a DC output voltage, and an output current. In operation 2103, the system performance controller 108 calculates an overall efficiency 123B based on the measurement conditions 103. In operation 2105, the system performance controller 108 modifies the control parameters 105 until the overall efficiency 123B reaches a threshold condition.

[0242] In some examples, the control manipulation module 130 uses an AI algorithm 132 to modify the control parameter 105. In some examples, the AI ​​algorithm 132 is a target search algorithm 134. In some examples, the AI ​​algorithm is an applied learning algorithm 136. In some examples, the control manipulation module 130 applies historical data 140 to a neural network 138 to predict a value of the control parameter 105 that causes the energy conversion efficiency 123 to reach a threshold condition. In some examples, the system performance controller 108 determines a first value of the control parameter 105 that causes the energy conversion efficiency 123 to reach a threshold condition for a first environmental condition, and determines a second value of the control parameter 105 that causes the energy conversion efficiency 123 to reach the threshold condition for a second environmental condition. In some examples, the first environmental condition and the second environmental condition are different voltages.

[0243] Fig. 22 A power supply system 2200 is shown, the power supply system having: a power supply controller 2204 configured to control a power level 2202; a metering circuit 2206 configured to sense a measurement condition 103; and a system performance controller 2208, the system performance controller including: a standby power calculation circuit 124 configured to calculate a standby power 125 based on the measurement condition 103; and a control manipulation module 130 configured to adjust one or more control parameters 105 until the standby power 125 is minimized and / or reaches a threshold level. In some examples, the control parameters 105 include a skip cycle parameter 152.

[0244] For example, the system performance controller 2208 receives the measurement conditions 103 (eg, in digital format) from the metering circuit 2206. The measurement conditions 103 may include an input voltage and an input current. The standby power calculation circuit 124 calculates the standby power 125 (eg, V in *I in ). The control manipulation module 130 is then configured to repeatedly modify the control parameter 105 (or a set of control parameters 105) until the standby power 125 reaches a threshold condition.

[0245] Standby power 125 can be power used by a disabled device (e.g., the device is coupled to power stage 2202 but is turned off). For example, when a disabled device is coupled to power stage 102, power controller 2204 can control power stage 102 in a skip cycle mode (e.g., skip mode, skip cycle modulation, etc.). In some examples, skip cycle parameters 152 include one or more parameters that define control signals for switching operations during skip cycle mode.

[0246] Fig.23 An example of a control signal 2238 for a skip cycle mode is shown. The power controller 2204 may generate pulses 2240 during a burst period 2242 and then wait until after the skip cycle period 2244 to generate another set of pulses 2240. For example, after the burst period 2242, the power controller 2204 does not generate pulses 2240 during the skip cycle period 2244. After the skip cycle period 2244, the power controller 2204 generates another set of pulses 2240 during a subsequent burst period 2242. In some examples, the skip cycle parameters 152 may include the frequency of the pulses 2240 during the burst period 2242, the duty cycle during the burst period 2242, the frequency between burst periods 2242, and / or the length of the skip cycle period 2244.

[0247] Fig.24 According to one aspect, a Fig. 22 2400 of an exemplary operation of the power supply system 2200. Although reference is made to Fig.24 The power supply system 2200 is described in the flowchart 2400, but Fig.24 The operation may be applicable to any of the power systems described herein.

[0248] In operation 2402, the system performance controller 2208 detects a triggering event. In some examples, the triggering event includes activation of the power stage 2202. In some examples, the triggering event includes detection of a change in one or more environmental conditions. In some examples, the triggering event includes expiration of a timer. In some examples, the triggering event includes detection that the standby power 125 does not meet a threshold condition.

[0249] In operation 2404, system performance controller 2208 updates the value of control parameter 105 controlling power stage 2202 to minimize standby power 125. In some examples, control parameter 105 includes one or more skip cycle parameters 152. For example, in operation 2401, system performance controller 2208 may receive measurement conditions 103 of power stage 2202. In some examples, measurement conditions 103 include input voltage and input current. In operation 2403, system performance controller 1108 calculates standby power 125 (e.g., V IN *I IN ). In operation 2405, the system performance controller 2208 modifies the control parameter 105 until the standby power 125 reaches a threshold condition.

[0250] In some examples, the control manipulation module 130 selects a first value for the control parameter 105 and sends the first value to the power controller 2204. Then, the standby power calculation circuit 124 calculates the standby power 125 of the power stage 2202 operating according to the first value based on the measurement condition 103. Then, in response to the standby power 125 not reaching the threshold condition, the control manipulation module 130 modifies the first value to obtain a second value and sends the second value to the power controller 104. Then, the standby power calculation circuit 124 calculates the standby power 125 of the power stage 2202 operating according to the second value based on the measurement condition 103.

[0251] In some examples, the control manipulation module 130 uses the AI ​​algorithm 132 to modify the control parameter 105. In some examples, the AI ​​algorithm 132 is a target search algorithm 134. In some examples, the AI ​​algorithm 132 is an applied learning algorithm 136. In some examples, the control manipulation module 130 applies the historical data 140 to the neural network 138 to predict a value of the control parameter 105 that causes the standby power 125 to reach the threshold condition. In some examples, the system performance controller 2208 determines a first value of the control parameter 105 that causes the standby power 125 to reach the threshold condition for a first environmental condition, and determines a second value of the control parameter 105 that causes the standby power 125 to reach the threshold condition for a second environmental condition.

[0252] Fig.25A power system 2500 for controlling load distribution across multiple converters 2502 to optimize a composite efficiency 2523 is shown in accordance with one aspect of the present application. In some examples, the power system 2500 can be similar to the phase shearing embodiments (described above) and include evaluating the efficiency of each power converter 2502 and applying them to obtain an improved or optimal composite efficiency 2523, which can be beneficial when the power converters 2502 have different efficiencies over a load range. In general, the power system 2500 can include any of the details discussed with reference to the figures above. In some examples, the power system 2500 is Figure 1A The power supply system 100, Figure 5 Power supply system 500, Figure 6 Power supply system 600, Fig.11 The power supply system 1100 (including Fig.12 Power controller 1104 in) and / or Figure 20A-20B The example of power system 2000 in FIG. 25 is shown, and power system 2500 may include any details discussed in the present invention with respect to these figures, and / or any flow charts associated with these figures.

[0253] Multiple power converters 2502 are connected to load 2509. Multiple power converters 2502 are connected to load 2509 so that the output currents of power converters 2502 are combined to provide a system load current (I SL In some examples, multiple power converters 2502 are connected in parallel with respect to each other. In some examples, one or more power converters 2502 are not connected in parallel with one or more other power converters 2502, but are configured such that their current outputs are combined to generate a system load current (I SL ).

[0254] The power system 2500 includes a power source 2541 coupled to a plurality of power converters 2502. In some examples, the power source 2541 is an AC power source. In some examples, the power source 2541 is a DC power source. In some examples, the power system 2500 includes a plurality of power sources 2541 (e.g., independent power sources that may be DC power sources and / or AC power sources). For example, two or more power converters 2502 may draw power from independent power sources, but their outputs may be combined to provide a system load current (I SL The plurality of power converters 2502 are collectively configured to convert the system load current (I SL ) is delivered to the load 2509. The system load current (I SL ) is a single load current (eg, I L1 and I L2). In some examples, the plurality of power converters 2502 are power converters that convert from AC to DC. In some examples, the plurality of power converters 2502 are power converters that convert from DC to DC.

[0255] The plurality of power converters 2502 include a first power converter 2502-1 and a second power converter 2502-2. Fig.25 Only two power converters 2502 are shown, but power system 2500 may include any number of power converters 2502, such as three power converters, four power converters, or more than four power converters. In some examples, a single power converter 2502 may be referred to as a power stage (e.g., Figure 1A The plurality of power converters 2502 may include one of a variety of topologies, such as a buck converter, a boost converter, an inverting buck-boost converter, a flyback converter, an active clamp forward converter, a single switch forward converter, a dual switch forward converter, a push-pull converter, a half-bridge converter, a full-bridge converter, a phase-shifted full-bridge converter, etc. In some examples, the plurality of power converters 2502 include a switching resonant converter.

[0256] In some examples, the first power converter 2502-1 includes a metering circuit (eg, Figure 1A The metering circuit 106 in the first power converter 2502-1 can sense input conditions and output conditions (e.g., input current, input voltage, output voltage, and output current). In some examples, the metering circuit of the first power converter 2502-1 includes the following: Fig.11 In some examples, the first power converter 2502-1 may include a power supply controller (e.g., Figure 1A The power controller 104 in the embodiment of the present invention is configured to control the operation of the first power converter 2502-1. In some examples, the first power converter 2502-1 and / or the second power converter 2502-2 include: Fig.12 1 and 2. The power controller 1104 in FIG. 1 may include various details described in detail (e.g., PWM logic circuits, differential sensors, digital compensators, multi-slope ramp generators, polarity logic and gate drive logic, etc.). In some examples, the second power converter 2502-2 may include the same / similar components as the first power converter 2502-1, such as metering circuits and a power controller.

[0257] The power system 2500 includes a system performance controller 2508 that is communicatively coupled to the power converter 2502. The system performance controller 2508 may include any one or more details discussed with reference to the system performance controller in the previous figures of the present invention. For example, the system performance controller 2508 is a computing device having one or more memory devices and one or more digital processors. The memory device includes executable instructions that cause the digital processor to perform operations of the system performance controller 2508. In some examples, the system performance controller 2508 includes one or more integrated circuits. In some examples, the system performance controller 2508 includes one or more field programmable gate arrays (FPGAs). In some examples, the system performance controller 2508 includes one or more state machines.

[0258] System performance controller 2508 may include a control manipulation module 2530 configured to adjust control parameters 2505, and an efficiency calculation circuit 2522 configured to calculate a composite efficiency 2523 based on measured input conditions and output conditions of multiple power converters 2502. In some examples, control parameters 2505 include a single load current of multiple power converters 2502, such as a first load current (I ) provided by first power converter 2502-1. L1 ) and the second load current (I L2 System performance controller 2508 may use control manipulation module 2530 and efficiency calculation circuit 2522 to generate power loss information 2540, which system performance controller 2508 uses to select values ​​of individual load currents to increase or optimize composite efficiency 2523 (in some examples, provide a maximum or highest composite efficiency 2523).

[0259] System performance controller 2508 may activate first power converter 2502-1 (and deactivate second power converter 2502-2) and control manipulation module 2530 to step through (or cycle through) a first load current (I L1 ), and the system performance controller 2508 may calculate a single efficiency of the first power converter 2502-1 at each iteration.

[0260] If the current load range of the first power converter 2502-1 is 100A, in some examples, the control manipulation module 2530 may set the first load current (I L1 ) is initially set to 0A, and the efficiency calculation circuit 2522 may receive the measured input conditions and output conditions (e.g., V IN 、V OUT ,IIN ,I OUT ), and calculate a single efficiency (e.g., (V OUT *I OUT ) / (V IN *I IN )). In some examples, the efficiency calculation circuit 2522 calculates the power loss caused by the first power converter 2502-1 (eg, (V IN *I IN )-(V OUT *I OUT )). For example, the power loss caused by the first power converter 2502-1 may indicate the efficiency level of the first power converter 2502-1. In some examples, the efficiency calculation circuit 2522 converts the first load current (I L1 ) values ​​(e.g. 0 A) together with the corresponding individual efficiencies and / or power losses are stored as efficiency data.

[0261] Then, the control manipulation module 2530 selects another value of the first load current (e.g., 10A), and the efficiency calculation circuit 2522 receives the measured input condition and output condition of the first power converter 2502-1, calculates the single efficiency and / or power loss with the new value, and stores the first load current (I L1 ) value (e.g., 10A) together with the corresponding individual efficiencies and / or power losses. This process will continue until the efficiencies and / or power losses are calculated for the entire load current range of the first power converter 2502-1. In some examples, if the load current range is 100A, the control manipulation module 2530 can step through the first load current (I L1 ) (e.g., 0A, 10A, 20A, 30A, up to 100A). Although the above example uses an increment of 10A, the control manipulation module 2530 can step through or cycle through the first load current (I L1 ) value.

[0262] The system performance controller 2508 may activate the second power converter 2502-2 (and deactivate the first power converter 2502-1) while the control manipulation module 2530 steps through (or cycles through) values ​​of the second load current within the current load range of the second power converter 2502-2, and the system performance controller 2508 may calculate a single efficiency and / or power loss for the first power converter 2502-1 at each iteration (in the same manner as described above with reference to the first power converter 2502-1).

[0263] In some examples, the system performance controller 2508 can be configured to step through (or cycle through) the first load current (I L1 ) and the efficiency data generated by stepping through (or cycling through) the values ​​of the second load current to generate power loss information 2540. In some examples, for each combination of load currents, system performance controller 2508 can combine the efficiencies (e.g., power losses) of multiple power converters 2502 to generate a total power loss (also referred to as a composite efficiency 2523).

[0264] Table 2 below provides an example of a portion of the power loss information 2540 .

[0265]

[0266] In some examples, as shown in Table 2, the composite efficiency 2523 may be expressed as a total power loss (e.g., the power loss caused by the first power converter 2502-1 plus the power loss caused by the second power converter 2502-2). For example, the lowest total power loss may indicate the highest composite efficiency 2523, or the highest total power loss may indicate the lowest composite efficiency 2523. In some examples, the power loss information 2540 may express the composite efficiency 2523 as a composite percentage (e.g., 93%) reflecting the overall (or end-to-end) efficiency. The composite efficiency 2523 may refer to the end-to-end power conversion efficiency. In some examples, the composite efficiency 2523 is Figure 1B Examples of overall efficiency 123 and Figure 1B In some examples, composite efficiency 2523 can be expressed as a percentage and / or as a total loss caused by multiple power converters 2502.

[0267] If the total system maximum current is 200A, the power loss information 2540 will specify the combination of load currents and the composite efficiency 2523 for the entire 0-200A range, not just the combination of load currents at load currents of 10A, 20A, and 30A as shown in Table 2 (e.g., the total system current in Table 2 will be up to 200A). The power loss information 2540 includes the first load current (I L1 ) and the second load current (I L2) values. In some examples, the power loss information 2540 includes combinations of load currents within a range of system load current values, and each load current combination provides a separate combination of values ​​of a single load current and its calculated composite efficiency 2523. As shown in Table 2, a particular load current combination is (10A, 0A), which can provide a composite efficiency 2523 based on a power loss of 79.6W, while the next load current combination is (0A, 10A), which can provide a composite efficiency 2523 based on a power loss of 89W.

[0268] In some examples, the control module 2530 can be implemented by using an artificial intelligence (AI) algorithm (e.g. Figure 1A The AI ​​algorithm 132 in the embodiment selects different values ​​for the control parameter 2505 to iteratively modify the first load current (I L1 ) and the second load current (I L2 ) value. For example, the control and manipulation module 2530 may execute an AI algorithm (input together with the power loss information 2540 or a portion of the power loss information 2540) to identify that the first load current (I L1 ) and the second load current (I L2 )

[0269] In some examples, for a combination of load currents, the AI ​​algorithm 132 may obtain a maximum weighted combination of the calculated metrics (e.g., efficiency) and then select the value of the load current corresponding to the maximum weighted combination. In some examples, the AI ​​algorithm obtains the maximum weighted combination based on a predetermined minimization, objective, or maximization of each of the calculated metrics.

[0270] In some examples, the AI ​​algorithms include goal-seeking algorithms (e.g., Figure 1A The target search algorithm 134 in FIG. 25 may be a target search algorithm that selects the first load current (I ) in a manner that maximizes the composite efficiency 2523. L1 ) and / or the second load current (I L2 ). In some examples, the AI ​​algorithm includes applying a learning algorithm (e.g., Figure 1A 2540) to predict the value of the one or more control parameters 2505. For example, according to the applied learning algorithm, the control manipulation module 2530 is configured to apply the power loss information 2540 (e.g., also referred to as training data) to the neural network to predict the first load current (I L1 ) and the second load current (I L2) so that the composite efficiency 2523 reaches the threshold condition.

[0271] The system performance controller 2508 is configured to detect the system load current (I SL ) value. In some examples, upon activation, the plurality of power converters 2502 are initialized according to a startup operation, which may involve the first power converter 2502-1 and the second power converter 2502-2 providing equal load currents to the load 2509. For example, if the load capacity of the first power converter 2502-1 is 100A and the load capacity of the second power converter 2502-2 is 100A, the power system 250 may operate each of these converters at a ratio of 50% so that the first power converter 2502-1 provides a first load current (I L1 ), and the second power converter 2502-2 provides a second load current (I L2 ), and the total system current (I SL ) is 100A. However, for a total system current of 100A (I SL ), operating each of the first power converter 2502-1 and the second power converter 2502-2 at a ratio of 50% may not be the most efficient in terms of the composite efficiency 2523.

[0272] After the plurality of power converters 2502 have stabilized, the system performance controller 2508 may detect the system load current (I SL ) and uses the power loss information 2540 to determine a single load current (e.g., I L1 and I L2 ) such that the compound efficiency 2523 at the detected value reaches a threshold condition. In some examples, the threshold condition is a maximum or highest compound efficiency 2523. In some examples, the threshold condition is a compound efficiency 2523 that exceeds a threshold level. In some examples, the threshold condition is not the highest compound efficiency, but an efficiency that is relatively close to the highest compound efficiency.

[0273] The system performance controller 2508 may detect that the system load current is 20A, and, in some examples, the system performance controller 2508 may use the power loss information in Table 2 to select the first load current (I L1 ) is 20A, the second load current (I L2 ) is 0A, because the first load current (I L1 ) is 20A and the second load current (I L2) provides the highest composite efficiency 2523 (e.g., results in the lowest total power loss). This method can be referred to as an optimal sharing (OS) method. For example, according to the OS method, the system performance controller 2508 can determine which load current combinations provide the highest composite efficiency 2523 (e.g., results in the lowest power loss). In this way, the system performance controller 2508 can select the first load current (I ) that maximizes the composite efficiency 2523. L1 ) and the second load current (I L2 ) value, which may not require equal current to be provided by power converter 2502. Then, system performance controller 2508 generates a control signal (e.g., CTR 1 and CTR 2 ) to operate the plurality of power converters 2502 at the determined values. However, in addition to the OS method (which selects the combination of load currents that provides the highest composite efficiency 2523), in order to increase or optimize the composite efficiency 2523, the system performance controller 2508 may also use other algorithms or methods, such as the most efficient converter (MEC) method, a combination of the MEC method and the equal sharing (ES) method, and / or an integrated efficiency method, to determine the first load current (I L1 ) and the value of the second load current (I L2 ), which will be further explained below in the present invention.

[0274] The system performance controller 2508 can detect the system load current (I SL ) and uses the power loss information 2540 to determine the first load current (I L1 ) and the second load current (I L2 ) to make the composite efficiency 2523 at the system load current (I SL ) reaches the threshold condition at the change value of . For example, if the system load current (I SL ) changes from 50A to 160A, the system performance controller 2508 uses the power loss information 2540 to determine the first load current (I L1 ) and the second load current (I L2 ) at which the composite efficiency 2523 is optimized. For example, the power loss information 2540 may indicate that for a system load current of 160 A, the first load current (I L1 ) is 90A and the second load current (I L2 ) of 70A will provide the highest composite efficiency 2523. According to the OS method, the system performance controller 2508 can select these values ​​and generate a control signal (CTR 1 and CTR 2) to operate the first power converter 2502-1 and the second power converter 2502-2.

[0275] In some examples, the system performance controller 2508 is configured to continuously (e.g., periodically) monitor the composite efficiency 2523 during the life cycle of the system, and update one or more control parameters 2505 so that the composite efficiency 2523 continues to reach the threshold condition. In some examples, the system performance controller 2508 is configured to execute the control manipulation module 2530 in response to a triggering event to update the one or more control parameters 2505. In some examples, the triggering event includes the activation of multiple power converters 2502. In some examples, the triggering event includes the detection of a timer expiration. In some examples, the triggering event includes the detection of a change in one or more environmental conditions. In some examples, the triggering event includes the detection that the composite efficiency 2523 does not reach the threshold condition. In addition to continuously (e.g., periodically) monitoring the composite efficiency 2523, the system performance controller 2508 can also be configured to continuously (e.g., periodically) re-characterize (e.g., recalculate) the efficiency within the load current range for each power converter 2502 independently so that the composite efficiency 2523 can be calculated and the control can be accurately set.

[0276] In some examples, using power loss information 2540, system performance controller 2508 is configured to determine a single load current (eg, I L1 and I L2 ) to optimize the composite efficiency 2523. In the MEC approach, the system performance controller 2508 selects the most efficient converter (e.g., lowest power loss) to deliver all the power for each required load, and then uses the next most efficient converter (e.g., the converter with the next lowest power loss) to deliver any remaining power (if required).

[0277] Table 3 shows another portion of the power loss information 2540 of the two power converters 2502 when the total system current is 110A.

[0278]

[0279] The first power converter 2502-1 can deliver a current of a maximum of 100A, and the second power converter 2502-2 can also deliver a current of a maximum of 100A, that is, the maximum system current that can be provided is 200A. Similar to Table 2, Table 3 only shows a portion of the power loss information 2540. For example, if the total system maximum current is 200A (as shown in this example), the power loss information 2540 will specify the combination of load currents and the composite efficiency 2523 within the entire current range of 0-200A, rather than only specifying the combination of load currents when the current is 110A as shown in Table 3 (for example, Table 3 will start from 10A and rise to 200A). In addition, it should be noted that the composite efficiency 2523 is provided in increments of 10A for explanation purposes only.

[0280] When the system performance controller 2508 detects the system load current (I SL ), the system performance controller 2508 uses the power loss information 2540 (eg, as shown in Table 3 above) to determine the first load current (I L1 ) and the second load current (I L2 ) value so that the composite efficiency 2523 is optimized when the current is 110 A. If the system performance controller 2508 implements the MEC method, the system performance controller 2508 selects the first power converter 2502-1 to deliver a current of 100 A, and then selects the second power converter 2502-2 to deliver a current of 10 A, thereby providing a system load current (I SL ). For example, in the MEC method, a power converter is selected to provide the maximum current provided by the power converter, and any remaining power is then allocated to the next most efficient converter. In this example, each of the first power converter 2502-1 and the second power converter 2502-2 has a maximum current of 100A (thus, one of these converters is selected to deliver 100A current).

[0281] According to Table 3, the single power loss of the first power converter 2502-1 (operating at 100A) is 132W. In contrast, according to Table 3, the single power loss of the second power converter 2502-2 (operating at 100A) is 216W. Thus, according to the MEC method, the system performance controller 2508 may select the first power converter 2502-1 to deliver 100A current (for example, because the first power converter 2502-1 is a more efficient power converter), and then select the second power converter 2502-1 to deliver the remaining current, for example, 10A. Thus, the system performance controller 2508 may select the first load current (I L1 ) is determined to be 100A, and the second load current (I L2) is determined to be 10 A. However, if the system performance controller 2508 uses the best sharing (OS) method (as described above), the system performance controller 2508 may select a current of 50 A for the first power converter 2502-1 and a current of 60 A for the second power converter 2502-2 because the 50 / 60 combination provides the lowest total power loss of 134.4 W (while the 100 / 10 combination provides a total power loss of 216 W).

[0282] In some examples, if the system performance controller 2508 uses the MEC method or the OS method, and when the plurality of power converters 2502 are not used to provide power to the load 2509, the system performance controller 2508 can instruct the power converters 2502 that are not used to deliver power to be placed in a standby mode, for example, to be connected to the power source 2541 and consume the minimum power that is still sufficient to be ready to respond to a command to deliver power. For the example of Table 2, if the detected system load current (I SL ) is 20A, then according to the OS method or the MEC method, the system performance controller 2508 can L1 ) is determined to be 20A, and the second load current (I L2 ) is determined to be 0A. In this example, system performance controller 2508 may send a control signal to second power converter 2502-2 to place it in standby mode, so that second power converter 2502-2 is connected to power source 2541 and consumes a minimum amount of power that is still sufficient to be ready to respond to a command to deliver additional power. Then, if system performance controller 2508 detects that system load current (I SL ) has been changed to 110A, according to the OS method, the system performance controller 2508 can generate control signals (such as CTR 1 and CTR 2 ) so that the first power converter 2502-1 operates at a current of 50 A and the second power converter 2502-2 operates at a current of 60 A. If the MEC method is used, the system performance controller 2508 may generate a control signal (eg, CTR 1 and CTR 2 ) so that the first power converter 2502-1 operates at a current of 100 A and the second power converter 2502-2 operates at a current of 10 A. Regardless of whether the OS method or the MEC method is used, the second power converter 2502-2 is first in the standby mode and can respond relatively quickly to provide the requested power.

[0283] In some examples, system performance controller 2508 may use a combination of the MEC method and the equal share (ES) method to select the values ​​of the individual power converters 2502 to optimize the composite efficiency 2523. The MEC / ES combination may indicate that the MEC method is used until the first power converter reaches its maximum power, and then power is delivered in an equal share manner between the most efficient power converters. For example, if the detected system load current (I SL ) is 20A, the system performance controller 2508 can use the MEC method to select the value of the single current (for example, because the detected current is less than the maximum value of the first power converter 2502-1 or the second power converter 2502-2), according to Table 2, the value of the single current of the first power converter 2502-1 will be 20A, and the value of the single current of the second power converter 2502-2 will be 0A. Then, if the detected current is 110A, since 110A is greater than the maximum value of the first power converter 2502-1 or the second power converter 2502-2, the system performance controller 2508 can be converted to use the ES method. According to the ES method, the system performance controller 2508 can control each of the first power converter 2502-1 and the second power converter 2502-2 to operate at 55A (for example, the equal sharing device provides the same current to each power converter 2502). In some examples, at lighter loads, the MEC approach may be preferred (e.g., using one power converter to deliver power), while at heavier loads, the ES approach may be preferred (e.g., using two or more or all power converters to deliver power).

[0284] In some examples, the system performance controller 2508 using the combination of MEC / ES can perform a controlled transition process between lighter loads and heavier loads. Upon sensing the need to deliver more output power to the load 2509, the system performance controller 2508 can first control the second power converter 2502-2 to deliver all the required additional power to the load 2509. After considering or sensing the load demand and the stabilization of power delivery, the system performance controller 2508 can control the first power converter 2502-1 and the second power converter 2502-2 to transition to the above-mentioned ES method.

[0285] In some examples, system performance controller 2508 uses an integrated efficiency method to determine a single current (eg, I L1 and I L2) value. In some examples, the integrated efficiency method is similar to the MEC method, but does not require that a single power converter 2502 be switched in and out of standby mode. For example, the system performance controller 2508 can calculate the integral of the efficiency over the entire load range (e.g., the integrated efficiency) for each individual power converter 2502, and enable the power converters 2502 in a single step as the load demand increases, starting with the most efficient converter and ending with the least efficient converter. For example, the system performance controller 2508 can calculate the integral of the composite efficiency curve (e.g., the first integrated efficiency) over the load range of the first power converter 2502-1, and the system performance controller 2508 can calculate the integral of the composite efficiency curve (e.g., the second integrated efficiency) over the load range of the second power converter 2502-2. If the power loss provided by the first integrated efficiency is lower (e.g., more efficient) than the power loss provided by the second integrated efficiency, the system performance controller 2508 selects the first power converter 2502-1 to deliver power to the load 2509 (until it is a maximum), and then uses the second power converter 2502-2 to deliver any remaining amount (if necessary). Use of the least integrally efficient power stage may be enabled only if its use does not cause a standby transition of the next least integrally efficient power stage during the last two full range load steps.

[0286] It should be noted that all of the control methods described above anticipate the need to respond promptly to dynamically changing load conditions. Therefore, independent of the control method used, the system performance controller 2508 continuously monitors the load demand and controls one or more parallel power stages (e.g., power converters 2502) to respond quickly (e.g., immediately) to changes in the load 2509. After sensing the stabilization of the load demand based on one or more load ranges within the total power delivery range or based on other factors (e.g., voltage regulation within the load 2509 limit range), the system performance controller 2508 can control multiple power converters 2502 according to the control methods described above. In some examples, one or more load ranges can be based on a load current at which the composite efficiency 2523 (or power loss) is at a minimum (or maximum) value for the control method used, or they can be based on other factors, such as minimizing the number of transitions of a single power stage into and out of standby mode.

[0287] Fig.26 26 shows a power system 2600 for controlling load sharing on a plurality of converters 2602 to increase or optimize a composite efficiency 2623 according to one aspect of the present application. The power system 2600 may be Fig.25 An example of a power supply system 2500, and a power supply system 2600 may include reference Fig.25In general, the power supply system 2600 can include any of the details discussed with reference to the previous figures. In some examples, the power supply system 2600 is Figure 1A The power supply system 100, Figure 5 Power supply system 500, Figure 6 Power supply system 600, Fig.11 The power supply system 1100 (including Fig.12 Power controller 1104 in) and / or Figures 20A-20B 2000, and may include any details discussed herein with respect to these figures, and / or any flow charts associated with these figures.

[0288] A plurality of power converters 2602 are connected to a load 2609. A plurality of power converters 2602 are connected to the load 2609 in such a way that their output currents are combined to provide a system load current to the load 2609. In some examples, a plurality of power converters 2602 are connected in parallel relative to each other. In some examples, one or more power converters 2602 are not connected in parallel with one or more other power converters 2602, but are configured in such a way that their current outputs are combined to generate a system load current. The power supply system 2600 includes a power supply 2641 coupled to a plurality of power converters 2602. In some examples, the power supply 2641 is an AC power supply. In some examples, the power supply 2641 is a DC power supply. In some examples, the power supply system 2600 includes a plurality of power supplies 2641 (e.g., independent power supplies). For example, two or more power converters 2602 can draw power from independent power sources, but their outputs will be combined to provide a system load current. A plurality of power converters 2602 are collectively configured to deliver a system load current to the load 2609. The system load current is a combination of individual load currents provided by the plurality of power converters 2602. In some examples, the plurality of power converters 2602 are power converters that convert from AC to DC. In some examples, the plurality of power converters 2602 are power converters that convert from DC to DC.

[0289] The plurality of power converters 2602 include a first power converter 2602-1 and a second power converter 2602-2. Fig.26 Only two power converters 2602 are shown, but power system 2600 may include any number of power converters 2602, such as three power converters, four power converters, or more than four power converters. In some examples, a single power converter 2602 may be referred to as a power stage (e.g., Figure 1A power stage 102).

[0290] The power system 2600 includes a current sensor 2614 configured to sense input currents and output currents (e.g., Isns1, Isns2, CS1, and CS2) of the plurality of power converters 2602, and a voltage sensor 2616 configured to sense input voltages and output voltages (e.g., Vin, Vout1, and Vout2) of the plurality of power converters 2602. In some examples, the first power converter 2602-1 and the second power converter 2602-2 may include a metering circuit (e.g., Figure 1A In some examples, the metering circuit of the first power converter 2502-1 includes a metering circuit 106 in the power converter 2502-1, which can sense input conditions and output conditions (e.g., input current, input voltage, output voltage, and output current). Fig.11 In some examples, the first power converter 2602-1 and the second power converter 2602-2 may include a power controller (e.g., Figure 1A The power controller 104 in the embodiment is configured to control the operation of the first power converter 2502-1. In some examples, the first power converter 2602-1 and / or the second power converter 2602-2 include: Fig.12 The various details described in the power supply controller 1104 (e.g., PWM logic circuit, differential sensor, digital compensator, multi-slope ramp generator, polarity logic and gate drive logic, etc.).

[0291] The power system 2600 includes a system performance controller 2608 that is communicatively coupled to the second power converter 2602-2, the current sensor 2614, and the voltage sensor 2616. The system performance controller 2608 may include a control manipulation module 2630 configured to adjust the value of the single load current, and an efficiency calculation circuit 2622 configured to calculate a composite efficiency 2623 based on measured input conditions and output conditions of the plurality of power converters 2602. The system performance controller 2608 may use the control manipulation module 2630 and the efficiency calculation circuit 2622 to generate power loss information 2640, which the system performance controller 2608 uses to select the value of the single load current, the value of the single load current being compared to the reference Fig.25 The same approach discussed to provide maximum or highest compound efficiency 2623.

[0292] The system performance controller 2608 is configured to detect the value of the system load current in response to the plurality of power converters 2602 being activated (e.g., turned on) in an operating environment, and to determine the value of a single load current using the power loss information 2640 so that the composite efficiency 2623 at the detected value reaches a threshold condition (a maximum or highest composite efficiency 2623, a composite efficiency 2623 exceeding a threshold level, or a relatively high composite efficiency 2623, or an integrated efficiency that is maximized or above a threshold amount). In some examples, the system performance controller 2608 is configured to perform an OS method that may select a combination of load currents that provides the maximum composite efficiency 2623. In some examples, the system performance controller 2608 is configured to perform an MEC method that may select the most efficient converter to deliver power (until it reaches its maximum value), and then use the next most efficient converter to deliver the remaining power (if any). In some examples, the system performance controller 2608 is configured to perform a combination of the MEC method and the ES method, wherein the MEC method is applied at a system load level equal to or lower than one of the plurality of power converters, and then switches to the ES method when the system load current is greater than the maximum current provided by one of the plurality of power converters. In some examples, the system performance controller 2608 uses an integrated efficiency (IE) method. The OS method, the MEC method, the combination of the MEC / ES method, and the IE method can be described with reference to Fig.25 The method described is the same.

[0293] System performance controller 2608 is configured to generate control signals (e.g., Ctrl1 and Ctrl2) to operate multiple power converters 2602 at determined values. In some examples, system performance controller 2608 can detect changes in the value of the system load current in the operating environment and use power loss information 2640 to determine the first load current (I L1 ) and the second load current (I L2 ) such that the composite efficiency 2623 reaches a threshold condition at the changed value of the system load current. The system performance controller 2608 can then generate control signals (e.g., Ctrl1 and Ctrl2) to control the plurality of power converters 2602 at the new value. In some examples, the system performance controller 2608 is configured to continuously (e.g., periodically) monitor the composite efficiency 2623 during the life cycle of the system. In addition to continuously (e.g., periodically) monitoring the composite efficiency 2623, the system performance controller 2608 can also be configured to continuously (e.g., periodically) re-characterize (e.g., recalculate) the efficiency over the load current range for each power converter 2602 independently so that the composite efficiency 2623 can be calculated and the control can be accurately set.

[0294] Fig. 27 27 shows a power system 2700 for controlling load sharing on multiple converters 2702 to increase or optimize composite efficiency 2723 according to one aspect of the present application. In general, the power system 2700 can include any of the details discussed with reference to the previous figures. The power system 2700 can be Fig.25 Power system 2500 or Fig.26 An example of a power supply system 2600, and a power supply system 2700 may include reference Figure 25-26 The power system 2700 may include three or more power converters 2702. In some examples, the power system 2700 is Figure 1A The power supply system 100, Figure 5 Power supply system 500, Figure 6 Power supply system 600, Fig.11 The power supply system 1100 (including Fig.12 Power controller 1104 in) and / or Figures 20A-20B 2000, and may include any details discussed herein with respect to these figures, and / or any flow charts associated with these figures.

[0295] A plurality of power converters 2702 are connected to a load 2709. A plurality of power converters 2702 are connected to a load 2709 in such a way that their output currents are combined to provide a system load current to the load 2709. In some examples, a plurality of power converters 2702 are connected in parallel relative to each other. In some examples, one or more power converters 2702 are not connected in parallel with one or more other power converters 2702, but their current outputs are combined to form a system load current. The power supply system 2700 includes a power supply 2741 coupled to a plurality of power converters 2702. In some examples, the power supply 2741 is an AC power supply. In some examples, the power supply 2741 is a DC power supply. In some examples, the power supply system 2700 includes a plurality of power supplies 2741 (e.g., independent power supplies). For example, two or more power converters 2702 can draw power from independent power sources, but their outputs will be combined to provide a system load current. A plurality of power converters 2702 are collectively configured to deliver a system load current to a load 2709. The system load current is a combination of individual load currents provided by the plurality of power converters 2702. In some examples, the plurality of power converters 2702 are power converters that convert from AC to DC. In some examples, the plurality of power converters 2702 are power converters that convert from DC to DC.

[0296] Multiple power converters 2702 include a first power converter 2702-1, a second power converter 2702-2, and up to an Nth power converter 2702-2, where N may be an integer greater than or equal to 3. Multiple power converters 2702 include a current sensor 2714 configured to sense input currents and output currents (e.g., Isns1, Isns2, CS1, and CS2) of multiple power converters 2702, and a voltage sensor 2716 configured to sense input voltages and output voltages (e.g., Vin, Vout1, and Vout2) of multiple power converters 2702. In some examples, each power converter 2702 may include a metering circuit (e.g., Figure 1A In some examples, each power converter 2702 may include a power controller (e.g., Figure 1A The power controller 104 in FIG. 1 , and Fig.12 The power controller 1104 in FIG.

[0297] The power system 2700 includes a system performance controller 2708 that is communicatively coupled to the second power converter 2702-2, the current sensor 2714, and the voltage sensor 2716. The system performance controller 2708 may include a control manipulation module 2730 configured to adjust the value of the single load current, and an efficiency calculation circuit 2722 configured to calculate a composite efficiency 2723 based on measured input conditions and output conditions of the plurality of power converters 2702. The system performance controller 2708 may use the control manipulation module 2730 and the efficiency calculation circuit 2722 to generate power loss information 2740, wherein the system performance controller 2708 uses the power loss information 2740 to select the value of the single load current, and the value of the single load current is compared with the reference Fig.25 The same approach discussed to provide maximum or highest compounding efficiency 2723.

[0298] The system performance controller 2708 is configured to detect the value of the system load current in response to the plurality of power converters 2702 being activated (e.g., turned on) in an operating environment, and to determine the value of a single load current using the power loss information 2740 so that the composite efficiency 2723 at the detected value reaches a threshold condition (a maximum or highest composite efficiency 2723, a composite efficiency 2723 exceeding a threshold level, or a relatively high composite efficiency 2723). In some examples, the system performance controller 2708 is configured to perform an OS method that may select a combination of load currents that provides the maximum composite efficiency 2723. In some examples, the system performance controller 2708 is configured to perform an MEC method that may select the most efficient converter to deliver power (until it reaches its maximum value) and then use the next most efficient converter to deliver the remaining power (if any). In some examples, the system performance controller 2708 is configured to perform a combination of the MEC method and the ES method, wherein the MEC method is applied at a system load level equal to or lower than one of the plurality of power converters, and then switches to the ES method when the system load current is greater than the maximum current provided by one of the plurality of power converters. In some examples, the system performance controller 2708 uses an integrated efficiency (IE) method. The OS method, the MEC method, the combination of the MEC / ES method, and the IE method can be described with reference to Fig.25 The method described is the same.

[0299] The system performance controller 2708 is configured to generate control signals (eg, Ctrl 1 , Ctrl 2 ...Ctrl n ) to operate the plurality of power converters 2702 at the determined values. In some examples, the system performance controller 2708 can detect a change in the value of the system load current in the operating environment and use the power loss information 2740 to determine new values ​​of the first load current and the second load current so that the composite efficiency 2723 reaches a threshold condition at the changed value of the system load current. Then, the system performance controller 2708 can generate a control signal (e.g., Ctrl 1 , Ctrl 2 ...Ctrl n) to control the plurality of power converters 2702 at the new value. In some examples, the system performance controller 2708 is configured to continuously (e.g., periodically) monitor the composite efficiency 2723 during the life cycle of the system. In addition to continuously (e.g., periodically) monitoring the composite efficiency 2723, the system performance controller 2708 can also be configured to continuously (e.g., periodically) re-characterize (e.g., recalculate) the efficiency over the load current range for each power converter 2702 independently so that the composite efficiency 2723 can be calculated and the control can be accurately set.

[0300] Fig.28 A diagram depicting an embodiment of the present invention according to one aspect of the present invention is shown. Fig.25 2800 of an exemplary operation of the power system 2500 is shown. Fig.25 The flowchart 2800 is described with reference to the power supply system 2500, but Fig.25 The operation can be applied to any power supply system described in the present invention.

[0301] Operation 2802 includes delivering a system load current to a load 2509 through a power system 2500 having a plurality of power converters 2502, wherein the system load current is a combination of individual load currents provided by the plurality of power converters 2502. In some examples, the plurality of power converters 2502 are connected in parallel.

[0302] Operation 2804 includes detecting the value of the system load current by the system performance controller 2508, and operation 2806 includes using the power loss information 2540 to determine the value of the single load current so that the composite efficiency reaches a threshold condition. In some examples, the threshold condition is the maximum composite efficiency. In some examples, the value of the single load current is determined based on the OS method. In some examples, the value of the single load current is determined based on the MEC method. In some examples, the value of the single load current is determined based on a combination of the OS method and the MEC method. In some examples, the value of the single load current is determined based on the integral efficiency method. The power loss information 2540 includes a plurality of load current combinations within a range of values ​​of the system load current, each load current combination is associated with a corresponding composite efficiency 2523, and each load current combination provides a separate combination of values ​​of the single load current. In some examples, the determination operation includes selecting a combination of load currents that provides the highest composite efficiency 2523 from the plurality of load current combinations.

[0303] Operation 2808 includes generating a control signal to operate the plurality of power converters 2502 at determined values. In some examples, the operation includes detecting a change in a value of a system load current under an operating environment, and using the power loss information 2540 to determine a new value of a single load current so that the composite efficiency 2523 reaches a threshold condition. In some examples, the operation includes updating the power loss information 2540 in response to a trigger event, wherein the trigger event includes activation (e.g., switching on) of one or more power converters of the plurality of power converters, detecting a change in one or more measured input or output conditions, detecting that the composite efficiency is not maximized, and / or detecting that a timer has expired. In some examples, the updating operation includes iteratively modifying at least one of the values ​​of the single load currents, calculating the composite efficiency 2523 based on the measured input conditions and output conditions at each iteration, and storing the value of the single load current and the corresponding composite efficiency 2523 as the power loss information 2540 at each iteration.

[0304] Fig.29 A graph 2900 is shown depicting the efficiency of a single converter expressed as a percentage as a function of load current according to one aspect of the present application. Fig.25 Power supply system 2500 pairs Fig.29 , but the description may be applicable to any power supply system discussed in the present invention. Graph 2900 depicts the individual efficiencies of two power converters (e.g., a first power converter 2502-1 and a second power converter 2502-2). Graph 2900 includes a curve 2901 depicting a single power efficiency curve (e.g., expressed in efficiency %) of the first power converter 2502-1 over a load current range (e.g., from 0A to 100A) of the first power converter 2502-1, and a curve 2903 depicting a single power efficiency curve (e.g., expressed in efficiency %) of the second power converter 2502-2 over a load current range (e.g., 0A to 100A) of the second power converter 2502-2.

[0305] In some examples, system performance controller 2508 may activate first power converter 2502-1, and control manipulation module 2530 may step through (or cycle through) values ​​of load current within a current load range of first power converter 2502-1, and system performance controller 2508 may calculate a single efficiency (V OUT *I OUT ) / (V IN *I IN)). These results can generate curve 2901. Then, in some examples, system performance controller 2508 can activate second power converter 2502-2, and control manipulation module 2530 can step through (or cycle through) the value of the load current within the current load range of second power converter 2502-2, and system performance controller 2508 can calculate a single efficiency ((V OUT *I OUT ) / (V IN *I IN )). These results can produce curve 2903.

[0306] Fig.30 A graph 3000 is shown depicting the efficiency of a single converter expressed as power loss as a function of load current according to one aspect of the present application. Fig.25 Power supply system 2500 pairs Fig.30 The description is for illustration only, but the description can be applied to any power supply system discussed in the present invention. Fig.29 The curve graph of Fig.30 The single efficiency of two power converters (e.g., first power converter 2502-1 and second power converter 2502-2) is depicted. The graph 3000 includes a line 3001 depicting a single power efficiency curve (e.g., expressed in power loss (W)) of the first power converter 2502-1 over a load current range (e.g., from 0A to 100A) of the first power converter 2502-1, and a line 3003 depicting a single power efficiency curve (e.g., expressed in power loss (W)) of the second power converter 2502-2 over a load current range (e.g., 0A to 100A) of the second power converter 2502-2.

[0307] In some examples, system performance controller 2508 may activate first power converter 2502-1, and control manipulation module 2530 may step through (or cycle through) values ​​of load current within a current load range of first power converter 2502-1, and system performance controller 2508 may calculate a single efficiency (V OUT *I OUT ) / (V IN *I IN)). These results can generate line 3001. Then, in some examples, system performance controller 2508 can activate second power converter 2502-2, and control manipulation module 2530 can step through (or cycle through) the value of the load current within the current load range of second power converter 2502-2, and system performance controller 2508 can calculate a single efficiency ((V OUT *I OUT ) / (V IN *I IN )). These results can produce line 3003.

[0308] Fig.31 A graph 3100 depicting composite power loss as a function of load current according to one aspect of the present application is shown. Fig.25 Power supply system 2500 pairs Fig.31 , but the description can be applied to any power system discussed in the present invention. Graph 3100 includes composite power efficiency curves for the MEC method, the OS method, and the ES method. As described above, the MEC method is used to select the most efficient (e.g., the lowest loss) converter to deliver all power specifically for each required load, and use the next most efficient (e.g., the next lowest loss) converter to deliver the remaining power (if necessary). The OS method is used to select (or apply AI to select) a combination of power shares that produces the best composite efficiency (e.g., the lowest loss). The ES method is used to select an equal share of power between all power converters.

[0309] Graph 3100 includes line 3101 depicting a composite efficiency curve for the MEC method as a function of load current. Line 3101 depicts a composite efficiency curve expressed in power losses (W) as the value of the load current increases when system performance controller 2508 performs the MEC method (e.g., a combination of power losses from first power converter 2502-1 and second power converter 2502-2).

[0310] Graph 3100 includes line 3103 depicting a composite efficiency curve as a function of load current for the OS method. Line 3103 depicts a composite efficiency curve expressed in power loss (W) as the value of the load current increases when the system performance controller 2508 performs the OS method. Graph 3100 also includes line 3105 depicting a composite efficiency curve as a function of load current for the ES method. Line 3105 depicts a composite efficiency curve expressed in power loss (W) as the value of the load current increases when the system performance controller 2508 performs the OS method.

[0311] In some examples, such as Fig.31 As shown, at lighter loads, the MEC approach may be ideal (e.g., power is delivered by one power converter), while at heavier loads, the ES approach may be ideal (e.g., power is delivered by all power converters). In some examples, the OS approach is ideal because it produces the best composite efficiency for both lighter and heavier loads. In addition, Fig.31 Graph 3100 in can show that a combination of MEC / ES approaches is ideal where the power system can deliver all power starting from the most efficient converter, particularly for each required load, until the first power converter reaches maximum power and then deliver power in an equal share among the most efficient converters, particularly for each required load (e.g., line 3101 for light loads and subsequent line 3105 for higher loads).

[0312] Fig.32 A graph 3200 is shown depicting composite power savings and efficiency as a function of load current according to one aspect of the present application. Graph 3200 includes line 3201, which represents the power savings (in watts) of the MEC method as the value of the load current increases. For example, line 3201 may represent the amount of power saved when the OS method is used compared to when the MEC method is used. As shown in line 3201, the power savings are relatively high at higher loads and relatively low at lower loads. Graph 3200 includes line 3203, which represents the power savings of the ES method as the value of the load current increases. For example, line 3203 may represent the amount of power saved when the OS method is used compared to when the ES method is used. As shown in line 3203, the power savings are relatively high at lower loads and relatively low at higher loads. Graph 3200 includes line 3205 depicting the compound efficiency (expressed as a percentage) of the OS method as the value of the load current increases, line 3207 depicting the compound efficiency (expressed as a percentage) of the MEC method as the value of the load current increases, and line 3209 depicting the compound efficiency (expressed as a percentage) of the ES method as the value of the load current increases.

[0313] Fig.33 A graph 3300 is shown that depicts the efficiency of a single converter expressed as a percentage as a function of load current according to one aspect of the present application. For example, graph 3300 depicts the efficiency of a single converter of more than two power converters (e.g., a first power converter, a second power converter, and a third power converter). Although reference is made to Fig.25 Power supply system 2500 pairs Fig.33Some parts of the description of are described in detail, but the description can be applicable to any power supply system discussed in the present invention. The curve graph 3300 includes a line 3301 depicting a single power efficiency curve (e.g., expressed in efficiency %) of a first power converter over a load current range (e.g., from 0A to 100A), a curve 2903 depicting a single power efficiency curve (e.g., expressed in efficiency %) of a second power converter over a load current range (e.g., from 0A to 100A), and a line 3305 depicting a single power efficiency curve (e.g., expressed in efficiency %) of a third power converter over a load current range (e.g., from 0A to 100A).

[0314] Similar to Fig.29 In some examples, the system performance controller 2508 may activate the first power converter, and the control manipulation module 2530 may step through (or cycle through) the values ​​of the load current within the current load range of the first power converter, and the system performance controller 2508 may calculate a single efficiency (V OUT *I OUT ) / (V IN *I IN )). These results can generate line 3301. Then, in some examples, system performance controller 2508 can activate the second power converter, and control manipulation module 2530 can step through (or cycle through) values ​​of the load current within the current load range of the second power converter, and system performance controller 2508 can calculate a single efficiency ((V OUT *I OUT ) / (V IN *I IN )). These results can generate line 3303. Then, in some examples, system performance controller 2508 can activate the third power converter, and control manipulation module 2530 can step through (or cycle through) values ​​of the load current within the current load range of the third power converter, and system performance controller 2508 can calculate a single efficiency ((V OUT *I OUT ) / (V IN *I IN )). These results can produce line 3305.

[0315] Fig.34 A graph 3400 is shown that depicts the efficiency of a single converter expressed as power loss as a function of load current according to one aspect of the present application. For example, graph 3400 depicts the efficiency of a single converter of more than two power converters (e.g., a first power converter, a second power converter, and a third power converter). Although reference is made to Fig.25 Power supply system 2500 pairs Fig.34 3400 is not intended to illustrate some parts of the description of the present invention, but the description may be applicable to any power supply system discussed in the present invention. Graph 3400 includes a line 3401 depicting a single power efficiency curve (e.g., expressed in power loss (W)) of a first power converter over a load current range (e.g., from 0A to 100A), a line 3403 depicting a single power efficiency curve (e.g., expressed in power loss (W)) of a second power converter over a load current range (e.g., 0A to 100A), and a line 3405 depicting a single power efficiency curve (e.g., expressed in power loss (W)) of a third power converter over a load current range (e.g., 0A to 100A).

[0316] In some examples, system performance controller 2508 may activate the first power converter, and control manipulation module 2530 may step through (or cycle through) values ​​of the load current within a current load range of the first power converter, and system performance controller 2508 may calculate a single power loss (e.g., V IN *I IN )-(V OUT *I OUT )). These results can generate line 3401. Then, in some examples, system performance controller 2508 can activate the second power converter, and control manipulation module 2530 can step through (or cycle through) the values ​​of the load current within the current load range of the second power converter, and system performance controller 2508 can calculate a single power loss (e.g., (V IN *I IN )-(V OUT *I OUT )). These results can generate line 3403. Then, in some examples, system performance controller 2508 can activate the third power converter, and control manipulation module 2530 can step through (or cycle through) values ​​of the load current within the current load range of the third power converter, and system performance controller 2508 can calculate a single power loss (e.g., (V IN *I IN )-(V OUT *I OUT )). These results can produce line 3405.

[0317] Fig.35 A graph 3500 is shown depicting composite power loss as a function of load current according to one aspect of the present application. Fig.25 Power supply system 2500 pairs Fig.34The graph 3500 is illustrated in some parts of the description of the present invention, but the description can be applied to any power system discussed in the present invention. The graph 3500 includes composite power efficiency curves for the MEC method, the OS method, and the ES method.

[0318] Graph 3500 includes line 3501 depicting a composite efficiency curve as a function of load current for the MEC method. Line 3501 depicts a composite efficiency curve (e.g., a combination of power losses from three power converters) expressed in terms of power losses (W) as the value of the load current increases when the system performance controller 2508 performs the MEC method. Graph 3500 includes line 3503 depicting a composite efficiency curve as a function of load current for the OS method. Line 3503 depicts a composite efficiency curve as a function of load current expressed in terms of power losses (W) as the value of the load current increases when the system performance controller 2508 performs the OS method. Graph 3500 also includes line 3505 depicting a composite efficiency curve as a function of load current for the ES method. Line 3505 depicts a composite efficiency curve as a function of load current expressed in terms of power losses (W) as the value of the load current increases when the system performance controller 2508 performs the OS method.

[0319] In some examples, such as Fig.35 As shown, at lighter loads, the MEC approach may be ideal (e.g., power is delivered by one power converter), while at heavier loads, the ES approach may be ideal (e.g., power is delivered by all power converters). In some examples, the OS approach is ideal because it produces the best composite efficiency for both lighter and heavier loads. In addition, Fig.35 Graph 3500 in can show that a combination of MEC / ES approaches is also ideal, wherein the power system can deliver all power starting from the most efficient converter, particularly for each required load, until the first power converter reaches maximum power, and then deliver power in an equal share among the most efficient converters, particularly for each required load (e.g., line 3501 for light loads and subsequent line 3505 for higher loads).

[0320] Fig.36A graph 3600 is shown that depicts composite power savings and efficiency as a function of load current according to one aspect of the present application. For example, graph 3600 depicts composite power savings and efficiency of three power converters. Graph 3600 includes line 3601, which represents the power savings (in watts) of the MEC method as the value of the load current increases. For example, line 3601 may represent the amount of power saved when the OS method is used compared to when the MEC method is used. As shown in line 3601, the power savings are relatively high at higher loads and relatively low at lower loads. Graph 3600 includes line 3603, which represents the power savings of the ES method as the value of the load current increases. For example, line 3603 may represent the amount of power saved when the OS method is used compared to when the ES method is used. As shown in line 3603, the power savings are relatively high at lower loads and relatively low at higher loads. Graph 3600 includes line 3605 depicting the compound efficiency (expressed as a percentage) of the OS method as the value of the load current increases, line 3607 depicting the compound efficiency (expressed as a percentage) of the MEC method as the value of the load current increases, and line 3609 depicting the compound efficiency (expressed as a percentage) of the ES method as the value of the load current increases.

[0321] Fig.37 A graph 3700 depicting the composite power loss as a function of load current according to one aspect of the present application is shown. Graph 3700 depicts a more detailed graph of the composite power loss expressed as a percentage of efficiency as the value of the load current increases. Although referring to Fig.25 Power supply system 2500 pairs Fig.37 3700 is illustrated in some parts of the description of the present invention, but the description can be applied to any power system discussed in the present invention. Graph 3700 includes composite power efficiency curves for the MEC method, the OS method, and the ES method. Graph 3700 includes line 3701 depicting a composite efficiency curve for the MEC method as a function of the load current. Line 3701 depicts a composite efficiency curve expressed as a percentage of efficiency as the value of the load current increases when the system performance controller 2508 performs the MEC method. Graph 3700 includes line 3703 depicting a composite efficiency curve for the OS method as a function of the load current. Line 3703 depicts a composite efficiency curve expressed as a percentage of efficiency as the value of the load current increases when the system performance controller 2508 performs the OS method. Graph 3700 includes line 3705 depicting a composite efficiency curve for the ES method as a function of the load current. Line 3705 depicts a composite efficiency curve expressed as a percentage of efficiency as the value of the load current increases when the system performance controller 2508 performs the ES method.

[0322] Fig.38A graph 3800 is shown that depicts the composite power loss as a function of load current according to one aspect of the present application. For example, the graph 3800 depicts the composite power loss as a percentage for the OS method, the MEC method, and the integrated efficiency (IE) method as the value of the load current increases. Although referring to Fig.25 Power supply system 2500 pairs Fig.38 Graph 3800 is illustrated in some parts of the description of , but the description may apply to any power supply system discussed in the present invention. Graph 3800 includes line 3801 depicting a composite efficiency curve for the OS method as a function of load current. Line 3801 depicts a composite efficiency curve expressed as a percentage of efficiency as the value of the load current increases when the system performance controller 2508 performs the OS method. Graph 3800 includes line 3803 depicting a composite efficiency curve for the MEC method as a function of load current. Line 3803 depicts a composite efficiency curve expressed as a percentage of efficiency as the value of the load current increases when the system performance controller 2508 performs the MEC method. Graph 3800 includes line 3805 depicting a composite efficiency curve for the IE method as a function of load current. Line 3805 depicts a composite efficiency curve expressed as a percentage of efficiency as the value of the load current increases when the system performance controller 2508 performs the IE method.

[0323] In some examples, the IE method is similar to (e.g., substantially the same as) the MEC method, but the amount of transitions into and out of standby mode can be reduced (or eliminated). For example, the MEC method can indicate that a power converter that is not providing any power to the load is to be placed in standby mode (e.g., connected to the input and consuming only enough power to be ready to respond to a command to deliver power). In some examples, the load demand is dynamic (e.g., not static), and the power converter needs to respond accordingly.

[0324] In some examples, according to the IE method, the system performance controller 2508 can calculate the total efficiency (e.g., integrated efficiency) over the entire load range for each individual power converter 2502, and use the power level with the lowest integrated efficiency "last" (i.e., deliver power "last", or only when use does not cause a standby transition to the next lowest integrated efficiency power level in the last two full range load steps). In some examples, the dynamic characteristics of the load need to be taken into account, and the system performance controller 2508 can allocate power among the power converters within discrete ranges (e.g., as opposed to continuous load ranges) in combination with reallocation after the loads within these ranges stabilize.

[0325] Various embodiments of the systems and techniques described herein may be implemented in digital electronic circuits, integrated circuits, specially designed ASICs (application specific integrated circuits), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include implementations in one or more computer programs that may be executed and / or interpreted on a programmable system including at least one programmable processor, which may be dedicated or general purpose, coupled to receive data and instructions from and transmit data and instructions to a storage system, at least one input device, and at least one output device. Various embodiments of the systems and techniques described herein may be implemented as and / or may be generally referred to herein as circuits, modules, blocks, or systems that may combine software and hardware aspects. For example, a module may include functions / actions / computer program instructions executed on a processor (e.g., a processor formed on a silicon substrate, a GaAs substrate, etc.) or some other programmable data processing device.

[0326] Some of the exemplary embodiments described above are described as processes or methods, which are depicted as flow charts. Although the flow charts describe the operations as sequential processes, many of these operations can be performed in parallel, concurrently, or simultaneously. In addition, the order of operations can be rearranged. These processes can terminate when their operations are completed, but can also have additional steps not included in the accompanying drawings. These processes can correspond to methods, functions, processes, subroutines, subprograms, etc.

[0327] The methods discussed above (some of which are represented by flow charts) may be implemented by hardware, software, firmware, middleware, microcode, hardware description language, or any combination thereof. When implemented in software, firmware, middleware, or microcode, the program code or code segments for performing the necessary tasks may be stored in a machine or computer readable medium, such as a storage medium. A processor may perform the necessary tasks.

[0328] The specific structural and functional details disclosed herein are merely representative and are intended only for purposes of describing example embodiments. Example embodiments may, however, be embodied in many alternate forms and should not be construed as limited to only the embodiments set forth herein.

[0329] It should be understood that although the terms first, second, etc. can be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, the first element can be named as the second element, and similarly, the second element can be named as the first element without departing from the scope of the exemplary embodiment. As used herein, the term "and / or" includes any combination and all combinations of one or more associated items listed.

[0330] The terms used herein are only used for the purpose of describing specific embodiments and are not intended to limit the exemplary embodiments. As used herein, the singular forms one, a, and the are intended to also include the plural forms, unless the context clearly indicates otherwise. It should also be understood that when the terms include and / or include are used herein, the presence of the features, integers, steps, operations, elements, and / or parts is specified, but the presence or addition of one or more other features, integers, steps, operations, elements, parts, and / or groups thereof is not excluded.

[0331] It should also be noted that, in some alternative embodiments, the functions / actions indicated may not occur in the order indicated in the drawings. For example, two figures shown in succession may actually be executed simultaneously or may sometimes be executed in the reverse order, depending on the functions / actions involved.

[0332] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the exemplary embodiments belong. It should also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art, and will not be interpreted as an idealized or overly formal meaning unless explicitly defined herein.

[0333] The various parts and corresponding detailed descriptions of the above exemplary embodiments are presented in terms of software or algorithms and symbolic representations of the operations of the data bits in the computer memory. These descriptions and representations are descriptions and representations that are used by those of ordinary skill in the art to effectively convey the essence of their work to other persons of ordinary skill in the art. The terms used herein and commonly used, i.e., algorithms, are considered to be self-consistent sequences of steps that obtain desired results. These steps are steps that require physical manipulation of physical quantities. Usually, although not necessarily, these quantities take the form of optical signals, electrical signals, or magnetic signals that can be stored, transferred, combined, compared, and otherwise manipulated. It has been shown that sometimes, mainly for general purposes, it is convenient to refer to these signals as positions, values, elements, symbols, characters, items, numbers, etc.

[0334] In the above exemplary embodiments, references to actions and symbolic representations (e.g., in the form of flowcharts) of operations that can be implemented as program modules or functional processes include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types and can be described and / or implemented using existing hardware at existing structural elements. Such existing hardware may include one or more central processing units (CPUs), digital signal processors (DSPs), application specific integrated circuits, field programmable gate arrays (FPGAs), computers, etc.

[0335] It should be remembered, however, that all of these and similar terms are to be associated with the appropriate physical quantities and are merely convenient labels applied to these quantities. Unless specifically stated otherwise or apparent from the discussion, terms such as processing or computing or calculating or determining display refer to the actions and processes of a computer system or similar electronic computing device that manipulates and transforms data represented as physical quantities, electronic quantities within the computer system's registers or memories and converts it into other data similarly represented as physical quantities within the computer system's memories or registers or other such information storage devices, transmission devices or display devices.

[0336] It is also noted that the software-implemented aspects of the exemplary embodiments are typically encoded on some form of non-transitory program storage medium or implemented on some type of transmission medium. The program storage medium may be magnetic (e.g., a floppy disk or hard drive) or optical (e.g., a compact disk read-only memory or CD ROM), and may be read-only or random access. Similarly, the transmission medium may be twisted pair, coaxial cable, optical fiber, or some other suitable transmission medium known in the art. The exemplary embodiments are not limited by these aspects of any given implementation.

[0337] Finally, it should also be noted that although the attached claims list specific combinations of features described herein, the scope of the present disclosure is not limited to the specific combinations claimed thereafter, but extends to cover any combination of features or embodiments disclosed herein, regardless of whether the specific combination is specifically listed in the attached claims at this time.

Claims

1. A power supply system, characterized in that: The power supply system comprises: a plurality of power converters configured to deliver a system load current to a load, the system load current being a combination of the individual load currents provided by the plurality of power converters; and a system performance controller configured to detect values ​​of the system load currents, the system performance controller configured to determine values ​​of the respective load currents using power loss information so that a composite efficiency reaches a threshold condition, the system performance controller configured to generate control signals to operate the plurality of power converters at the determined values; The power loss information includes a plurality of load current combinations within a range of system load current values, each load current combination is associated with a corresponding composite efficiency, and each load current combination provides a separate combination of values ​​of the respective load currents.

2. The power supply system according to claim 1, characterized in that: The system performance controller is configured to periodically update the power loss information by monitoring input conditions and output conditions of the plurality of power converters and by calculating the composite efficiency based on the input conditions and the output conditions, The system performance controller is configured to select a load current combination that provides the highest composite efficiency from the multiple load current combinations.

3. The power supply system according to claim 1, characterized in that: In response to a triggering event, the system performance controller is configured to use the power loss information to determine new values ​​for the respective load currents such that the composite efficiency reaches the threshold condition.

4. The power supply system according to claim 1, characterized in that: The system performance controller is configured to determine a most efficient power converter among the multiple power converters based on the power loss information, and the system performance controller is configured to set the value of the single load current of the most efficient power converter to a maximum single value provided by the most efficient power converter and allocate any remaining amount to the next most efficient converter.

5. The power supply system according to claim 1, characterized in that: The system performance controller is configured to generate or update the power loss information in response to a trigger event, and the system performance controller includes: a control manipulation module configured to iteratively modify at least one of the values ​​of the respective load currents; and an efficiency calculation circuit configured to calculate the composite efficiency based on measured input conditions and output conditions in each iteration, The control manipulation module is configured to store the values ​​of the respective load currents and the corresponding composite efficiencies in each iteration, The control and manipulation module is configured to execute an artificial intelligence (AI) algorithm using a neural network, and the control and manipulation module is configured to apply the power loss information to the neural network to predict at least one of the values ​​of the respective load currents for the next iteration.

6. A power supply system, characterized in that: The power supply system comprises: a plurality of power converters, the plurality of power converters comprising at least a first power converter and a second power converter, the plurality of power converters being collectively configured to deliver a system load current to a load, the system load current being a combination of at least a first load current provided by the first power converter and a second load current provided by the second power converter; and a system performance controller configured to detect a value of the system load current and to determine at least a first value of the first load current and a second value of the second load current using power loss information so that a composite efficiency reaches a threshold condition, The system performance controller is configured to generate a first control signal to operate the first power converter at least at the first value, and to generate a second control signal to operate the second power converter at the second value.

7. The power supply system according to claim 6, characterized in that: the power loss information comprises a plurality of load current combinations within a range of values ​​of the system load current, each load current combination being associated with a respective composite efficiency, each load current combination providing a separate combination of a value of at least the first load current and a value of the second load current, The system performance controller is configured to select a load current combination that provides the highest composite efficiency from the multiple load current combinations.

8. The power supply system according to claim 6, characterized in that: The system performance controller is configured to: determine the first value and the second value according to a most efficient converter (MEC) method in response to the system load current being detected to be equal to or less than a maximum load current of the first power converter or the second power converter, The system performance controller is configured to determine the first value and the second value according to an equal sharing method in response to the system load current being detected as being greater than a maximum load current of the first power converter or the second power converter.

9. A method for controlling a plurality of power converters using composite efficiency, characterized in that: The method comprises: delivering a system load current to a load through the power system, the system load current being a combination of at least a first load current provided by the first power converter and a second load current provided by the second power converter; calculating a first integral value of a composite efficiency curve within a load range of the first power converter; calculating a second integral value of the composite efficiency curve within the load range of the second power converter; determining a value of the first load current and a value of the second load current based on the first integrated value and the second integrated value; and Control signals are generated to operate the first power converter and the second power converter at determined values.

10. The method according to claim 9, characterized in that The method further comprises: determining that the first integral value provides a greater compound efficiency than the second integral value; selecting the first power converter to provide power until a maximum current threshold provided by the first power converter is reached; determining that the system load current is greater than a maximum current threshold provided by the first power converter; and The second power converter is selected to provide power until a maximum current threshold provided by the second power converter is reached.

Images