A Boost Circuit Control Method Based on Adaptive Weight Adjustment

Abstract

This invention discloses a Boost circuit control method based on adaptive weight adjustment. The method provided by this invention constructs a dynamically adjustable comprehensive performance index and utilizes an adaptive PID algorithm to optimize the weight coefficients of each controlled parameter online, thereby achieving adaptive adjustment of the Boost circuit control target according to operating conditions. In this invention, on the one hand, comprehensive monitoring and evaluation of multiple parameters provides a complete basis for circuit state judgment, improving monitoring accuracy. On the other hand, through dynamic adjustment of weights, the control system can adjust its performance priorities in real time (e.g., prioritizing current limiting or voltage regulation) under complex operating conditions such as sudden load changes and input fluctuations, thus significantly enhancing the system's stability, dynamic response capability, and adaptability. Ultimately, this method can guide the Boost circuit to autonomously approach the globally optimal comprehensive performance state under various operating conditions.

Description

Technical Field

[0001] This invention relates to the technical field of electronic circuits, and more specifically to a Boost circuit control method based on adaptive weight adjustment. Background Technology

[0002] In modern electronic circuit applications, Boost circuits are widely used in voltage management, power electronic conversion, and other fields. With the continuous enhancement of electronic device functions and the increasing demands on power supply performance, the precise control and efficient operation of Boost circuits have become increasingly important. However, obtaining a comprehensive understanding of the Boost circuit's operating status has always been a challenging problem. Traditional Boost circuit monitoring methods are limited to single or a few parameters, failing to fully reflect the circuit's true state. Because multiple parameters within the circuit are interconnected, a normal change in a single parameter does not necessarily indicate that the circuit is operating at its optimal point. Furthermore, traditional Boost circuit control methods lack dynamic adjustment mechanisms. Faced with sudden changes in circuit performance indicators, they cannot promptly optimize parameters to maintain stable output, thus affecting the Boost circuit's stability. Therefore, there is an urgent need for a circuit monitoring device and control method capable of acquiring multiple key parameters, comprehensively evaluating the Boost circuit's state, and dynamically adjusting relevant parameters.

[0003] Traditional Boost circuit monitoring methods suffer from two main drawbacks. First, they monitor only a single parameter, failing to comprehensively reflect the circuit's operating state. Since multiple parameters within the circuit are interconnected, a normal change in a single parameter does not necessarily indicate that the circuit is operating at its optimal point. Second, they lack a dynamic adjustment mechanism; when relevant circuit parameters change, traditional control methods cannot optimize them in a timely manner, thus affecting circuit stability.

[0004] In related technologies, when Boost circuits face complex operating conditions such as sudden load changes and component parameter drift, there is a technical problem that the overall system performance cannot be maintained at its optimal level due to the fixed weight coefficients of the multi-parameter control targets. Summary of the Invention

[0005] The technical problem this invention aims to solve is that, in related technologies, when Boost circuits face complex operating conditions such as sudden load changes and component parameter drift, the fixed weight coefficients of the multi-parameter control targets lead to a situation where the overall system performance cannot be consistently maintained at its optimal level. The purpose is to provide a Boost circuit control method based on adaptive weight adjustment, thereby solving the technical problem of the system's overall performance not being consistently optimal.

[0006] This invention is achieved through the following technical solution:

[0007] In a first aspect, the present invention provides a Boost circuit control method based on adaptive weight adjustment, characterized in that it includes:

[0008] Determine the expected values ​​of multiple circuit parameters, and calculate the weighted sum of the expected values ​​of multiple circuit parameters based on preset initial weighting coefficients to obtain the expected value of the overall system performance;

[0009] Real-time acquisition of actual values ​​of multiple circuit parameters;

[0010] Calculate the deviation between the actual values ​​and their corresponding expected values ​​for each circuit parameter;

[0011] Based on the deviation and the corresponding actual value, the adjustment amount of the current weight coefficient of each circuit parameter is calculated using an adaptive PID algorithm, and each weight coefficient is updated; wherein, the adjustment amount includes: a proportional term corresponding to the current weighted deviation, an integral term corresponding to the integral of the historical weighted deviation, and a differential term corresponding to the rate of change of the weighted deviation.

[0012] Calculate the actual value of the current system's overall performance based on the updated weighting coefficients and the actual values ​​of each circuit parameter;

[0013] Based on the deviation between the current actual value of the system's overall performance and the expected value of the system's overall performance, the control signal of the Boost circuit is adjusted, and when the deviation exceeds the preset range, the deviation is recalculated to update each weight coefficient.

[0014] Further, the step of determining the expected values ​​of multiple circuit parameters and calculating a weighted sum of the expected values ​​of the multiple circuit parameters based on preset initial weighting coefficients to obtain the expected value of the overall system performance includes:

[0015] Determine the expected values ​​of at least two circuit parameters of the energy storage element in the Boost circuit and the expected values ​​of at least two circuit parameters of the load loop in the Boost circuit.

[0016] Combine the desired values ​​of at least two circuit parameters of the energy storage element and at least two circuit parameters of the load loop into a desired parameter vector;

[0017] Combine the preset initial weight coefficients into an initial weight vector;

[0018] The dot product of the initial weight vector and the expected parameter vector is calculated to obtain the expected value of the overall system performance.

[0019] Furthermore, at least two circuit parameters of the energy storage element include inductor voltage and / or inductor current; at least two circuit parameters of the load circuit include load voltage and / or load current.

[0020] Furthermore, the step of acquiring the actual values ​​of multiple circuit parameters in real time includes:

[0021] Analog signals of multiple circuit parameters are sequentially acquired using a multi-channel time-division sampling method; for the acquisition of inductor voltage and / or inductor current, a first analog signal is obtained through a corresponding first sensor; for the acquisition of load voltage and / or load current, a second analog signal is obtained through a corresponding second sensor.

[0022] The first and second analog signals are converted into digital signals, which are used as the actual values ​​of the corresponding circuit parameters.

[0023] Furthermore, the step of calculating the adjustment amount of the current weight coefficient of each circuit parameter based on the deviation and the corresponding actual value using an adaptive PID algorithm, and updating each weight coefficient, includes:

[0024] For each circuit parameter, perform the following steps:

[0025] Multiply the deviation of the circuit parameter by its actual value to obtain the weighted deviation of the circuit parameter;

[0026] Multiply the weighted deviation by the proportional gain coefficient corresponding to the circuit parameter to obtain the proportional adjustment amount of the circuit parameter.

[0027] The weighted deviation is integrated, and the integral result is multiplied by the integral gain coefficient corresponding to the circuit parameter to obtain the integral adjustment amount of the circuit parameter.

[0028] Calculate the rate of change of the weighted deviation and multiply the rate of change by the differential gain coefficient corresponding to the circuit parameter to obtain the differential adjustment amount of the circuit parameter;

[0029] The total weighted adjustment of the circuit parameters is obtained by adding the proportional adjustment, integral adjustment, and derivative adjustment.

[0030] The total weight adjustment is added to the current weight coefficient of the circuit parameter to obtain the updated weight coefficient.

[0031] Furthermore, each circuit parameter corresponds to an independent proportional gain coefficient, integral gain coefficient, and differential gain coefficient;

[0032] The step of multiplying the weighted deviation by the proportional gain coefficient corresponding to the circuit parameter to obtain the proportional adjustment amount of the circuit parameter includes:

[0033] The weighted deviations of each circuit parameter are combined into a first vector. The proportional gain coefficients corresponding to each circuit parameter are used as diagonal elements to form a first diagonal matrix. The product of the first diagonal matrix and the first vector is calculated to obtain a proportional adjustment vector composed of the proportional adjustment amounts of each circuit parameter.

[0034] The step of integrating the weighted deviation and multiplying the integral result by the integral gain coefficient corresponding to the circuit parameter to obtain the integral adjustment amount of the circuit parameter includes:

[0035] The integral results of each circuit parameter are combined into a second vector. The integral gain coefficients corresponding to each circuit parameter are used as diagonal elements to form a second diagonal matrix. The product of the second diagonal matrix and the second vector is calculated to obtain the integral adjustment vector composed of the integral adjustment amount of each circuit parameter.

[0036] The steps of calculating the rate of change of the weighted deviation and multiplying the rate of change by the differential gain coefficient corresponding to the circuit parameter to obtain the differential adjustment amount of the circuit parameter include:

[0037] The rate of change of each circuit parameter is combined into a third vector, and the differential gain coefficients corresponding to each circuit parameter are used as diagonal elements to form a third diagonal matrix. The product of the third diagonal matrix and the third vector is calculated to obtain a differential adjustment vector composed of the differential adjustment amounts of each circuit parameter.

[0038] Furthermore, the step of adjusting the control signal of the Boost circuit based on the deviation between the current actual value of the system's overall performance and the expected value of the system's overall performance, and recalculating the deviation to update each weight coefficient when the deviation exceeds a preset range, includes:

[0039] The difference between the current actual value of the system's overall performance and the expected value of the system's overall performance is used to obtain the overall performance deviation.

[0040] If the absolute value of the overall performance deviation is less than or equal to the preset threshold, then based on the currently updated weight coefficients, the deviations of each circuit parameter and their actual values, a control signal for adjusting the duty cycle of the Boost circuit switching transistor is calculated and output.

[0041] If the absolute value of the overall performance deviation is greater than the preset threshold, then based on the latest collected actual values ​​of the circuit parameters, the deviation between each circuit parameter and its expected value is recalculated, and the process of calculating the adjustment amount of the current weight coefficient of each circuit parameter using the adaptive PID algorithm based on the deviation and the corresponding actual value is returned, and the weight coefficients are updated, until the absolute value of the recalculated overall performance deviation is less than or equal to the preset threshold.

[0042] Secondly, the present invention provides a Boost circuit based on adaptive weight adjustment, comprising:

[0043] The Boost circuit module is used to implement DC voltage boost conversion;

[0044] A multi-channel sampling module, the input of which is connected to the Boost circuit module, is used to acquire the actual values ​​of multiple circuit parameters of the Boost circuit module in real time;

[0045] A control module, which is connected to the output terminal of the multi-channel sampling module and the control terminal of the Boost circuit module, is configured to execute the control method as described in any one of claims 1 to 7 to generate a control signal to adjust the switching transistor of the Boost circuit module.

[0046] Thirdly, the present invention provides an electronic device, comprising: a memory, and one or more processors communicatively connected to the memory; the memory stores instructions executable by the one or more processors, the instructions being executed by the one or more processors to cause the one or more processors to implement the method described above.

[0047] Fourthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method.

[0048] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0049] 1. This invention collects multiple key parameters of the Boost circuit and performs a comprehensive evaluation, providing a reliable basis for fully determining whether the circuit is at its optimal operating point, and effectively improving the accuracy of monitoring the circuit status.

[0050] 2. An adaptive PID algorithm is employed to dynamically adjust control parameters, enabling the Boost circuit to optimize its performance in a timely manner when faced with sudden load changes or input voltage fluctuations, thus enhancing its stability. Furthermore, quantitative indicators are provided for the comprehensive evaluation of the Boost circuit.

[0051] 3. By using a reasonable weighting coefficient adjustment mechanism, the system performance has been further optimized, thereby ensuring that the Boost circuit can maintain its best performance under various operating conditions. Attached Figure Description

[0052] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings:

[0053] Figure 1This is one of the flowcharts for a Boost circuit control method based on adaptive weight adjustment provided in the embodiments of this specification;

[0054] Figure 2 A block diagram of a Boost circuit based on adaptive weight adjustment provided in the embodiments of this specification;

[0055] Figure 3 This is a schematic diagram illustrating the PID control principle provided in the embodiments of this specification;

[0056] Figure 4 This is a second flowchart of a Boost circuit control method based on adaptive weight adjustment provided in the embodiments of this specification. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0058] In related technologies, Boost circuits, as a typical DC-DC boost converter, are widely used in electronic devices requiring voltage boosting, new energy power generation systems, and energy storage devices. Their function is to store and release energy in the inductor by periodically controlling the switching of a transistor at a given input voltage, thereby obtaining a stable and higher DC voltage at the output. In practical applications, Boost circuits often face complex dynamic conditions such as input voltage fluctuations, sudden changes in load current, and drift of circuit component parameters with temperature and time. These changing conditions require the control strategy not only to maintain the stability and accuracy of the output voltage but also to consider multiple performance indicators such as inductor current ripple, system transient response speed, and overall conversion efficiency, i.e., to pursue the optimal overall system performance.

[0059] In related technologies, one possible approach to improve control performance is to simultaneously monitor multiple key parameters of the Boost circuit (e.g., inductor current, inductor voltage, load voltage, and load current) and attempt to construct a comprehensive performance evaluation index to guide control by assigning weights to these parameters. A typical approach is to pre-assign fixed weight coefficients to each parameter and use the weighted sum as the input target or feedback quantity of the controller. However, this fixed-weight multi-parameter control strategy has inherent limitations. The impact and priority of each parameter on the overall system performance (e.g., stability, response speed, efficiency) dynamically change under different operating conditions. For example, during a transient process with a sudden increase in load, limiting inductor current overshoot and preventing magnetic saturation and device overstress are primary tasks, requiring a significant increase in the weight of current-related parameters. However, after entering steady state, maintaining output voltage accuracy and improving conversion efficiency become more important objectives. Fixed weight coefficients cannot respond to these inherent changes in demand, leading to a mismatch between the control system's focus and the optimal objectives under the current operating conditions. The fundamental reason is that fixed weights essentially presuppose a single, rigid performance evaluation standard, while the optimal goal pursued by Boost circuits under complex dynamic operating conditions is a multi-faceted objective that adapts and evolves with the operating state. Therefore, even if multiple parameters are monitored, it is difficult to ensure that the circuit always operates at the point of optimal overall performance across the entire operating range by using fixed weight coefficients, thus limiting further improvement in the overall system performance.

[0060] The inventive concept of this invention is to propose a Boost circuit control method based on adaptive weight adjustment. By introducing an adaptive PID algorithm, the weight coefficients of each monitored circuit parameter in the comprehensive performance evaluation index are dynamically and online adjusted, so that the control target of the system can be autonomously optimized according to the real-time deviation and the trend of change. This solves the technical problem that the comprehensive performance of the system cannot be maintained at the optimal level across the entire operating range due to the fixed weight coefficients.

[0061] This embodiment provides a Boost circuit control method based on adaptive weight adjustment. The execution subject of the method can be the control module of the Boost circuit. The control module can be a microcontroller unit, a digital signal processor, a field-programmable gate array, or a system-on-a-chip. The system-on-a-chip integrates multiple functional modules such as the microprocessor core, digital signal processing unit, analog-to-digital converter, and PWM generator on a single chip to form a complete control system.

[0062] like Figure 1 As shown, the method may include:

[0063] Step S10: Determine the expected values ​​of multiple circuit parameters, and calculate the weighted sum of the expected values ​​of multiple circuit parameters based on the preset initial weight coefficients to obtain the expected value of the overall system performance.

[0064] In this embodiment, the circuit parameters can be represented as physical quantities characterizing the operating state of the Boost circuit, such as voltage, current, power, or temperature. Specifically, the multiple circuit parameters can be physical parameters related to the energy storage element (inductor) of the Boost circuit and physical parameters related to the load circuit, to comprehensively reflect the dynamic and output characteristics of the circuit.

[0065] In this embodiment, determining the desired value can be expressed as pre-calculating or setting the target values ​​of each selected circuit parameter under ideal steady state based on the rated operating conditions of the load, the circuit design specifications (e.g., input and output voltages, rated power), and safety margins. For example, the desired value of the load voltage can be set to a constant rated voltage value, and the desired value of the inductor current can be derived from the target output power and circuit efficiency.

[0066] In this embodiment, the initial weighting coefficient can be represented as a scaling factor pre-assigned to each circuit parameter, which characterizes the relative importance of each parameter to the overall system performance evaluation at the initial moment. The sum of all weighting coefficients can be a fixed value, such as 1. This coefficient can be a fixed value set based on engineering experience, or it can be a set of default values ​​loaded according to the initial operating conditions when the system starts up.

[0067] In this embodiment, the weighted sum can be a linear weighted sum or other weighted summation methods. Specifically, the expected values ​​of each circuit parameter can be arranged sequentially to form an expected parameter vector, and the corresponding initial weight coefficients can be arranged to form an initial weight vector. Then, by calculating the dot product of the two vectors (i.e., multiplying the corresponding elements and then summing them), the expected value of the overall system performance can be obtained.

[0068] Step S12: Real-time acquisition of actual values ​​of multiple circuit parameters.

[0069] In this embodiment, the executing entity can continuously measure and obtain the instantaneous or effective values ​​of the circuit parameters selected in step S10 during the actual operation of the Boost circuit at a certain sampling period.

[0070] In this embodiment, the actual values ​​of multiple circuit parameters can be acquired in real time using a pre-set multi-channel sampling module. Specifically, the execution unit (control module) can coordinate and drive a dedicated multi-channel sampling module to acquire the actual values ​​of circuit parameters in real time. This multi-channel sampling module can be a hardware functional unit used to safely and accurately convert key electrical quantities (high voltage, high current) in the Boost circuit into digital signals that the control module can process.

[0071] In one possible and specific implementation, the multi-channel sampling module may include: a sensor group, analog switches and sampling channels, signal conditioning circuitry, and an analog-to-digital converter.

[0072] Specifically, the sensor group may include sensors configured for different circuit parameter characteristics and measurement locations. The sensor group may include a first sensor and a second sensor.

[0073] More specifically, the first sensor is used to acquire the state parameters of the energy storage element (inductor). It is a device capable of converting the voltage across the inductor or the current flowing through the inductor into a low-voltage analog signal. For acquiring the inductor voltage, the first sensor can be a voltage divider network composed of high-precision resistors, proportionally attenuating the high voltage across the inductor to a suitable measurement range. For acquiring the inductor current, the first sensor can be a Hall effect-based current sensor or a current transformer, used to linearly convert the current signal into a voltage signal.

[0074] The second sensor is used to acquire the state parameters of the load circuit. It is a device that converts the output voltage or output load current across the load into a low-voltage analog signal. For acquiring the load voltage, the second sensor can also be a voltage divider resistor network. For acquiring the load current, the second sensor can be a sampling resistor connected in series in the load circuit (to measure its voltage drop with a differential amplifier) ​​or a Hall current sensor.

[0075] Specifically, the analog switch can be a multiplexer controlled by digital signals from the control module. It can have multiple input terminals, respectively connected to the output terminals of the first and second sensors and a common output terminal. The control module can send channel selection signals to the analog switch according to a preset timing sequence, causing it to connect the analog signal output from one of the sensors to the common output terminal within each extremely short sampling time window, thereby achieving cyclical and sequential access to each analog signal. Each path from the sensor to the analog switch can be a sampling channel.

[0076] Specifically, the signal conditioning circuit can be connected after the common output of the analog switch. It preprocesses the raw analog signal received during switching to improve signal quality and adaptability. The signal conditioning circuit may include amplifiers and filters. The amplifier (e.g., an operational amplifier) ​​amplifies the weak signal output from the sensor to a voltage amplitude suitable for processing by the analog-to-digital converter. The filter (e.g., a low-pass filter) filters out high-frequency switching noise and electromagnetic interference from the signal, retaining effective low-frequency components.

[0077] Specifically, the analog-to-digital converter (ADC) is located after the signal conditioning circuit. It can be a device that discretizes the conditioned continuous-time analog voltage signal and converts it into a proportionally proportional digital code. This ADC can operate at a certain sampling rate. Whenever the analog switch switches to a new channel, the ADC samples and converts the analog signal of that channel at that moment, outputting a digital code value. This digital code value represents the actual value of the corresponding circuit parameter at that sampling moment.

[0078] Step S14: Calculate the deviation between the actual value and the corresponding expected value of each circuit parameter.

[0079] In this embodiment, for each monitored circuit parameter, the expected value of the parameter determined in step S10 is subtracted from the actual value collected in step S12, and the result is the deviation value of the parameter. If the actual value is greater than the expected value, the deviation is positive; otherwise, it is negative.

[0080] Step S16: Based on the deviation and the corresponding actual value, use the adaptive PID algorithm to calculate the adjustment amount of the current weight coefficient of each circuit parameter, and update each weight coefficient; wherein, the adjustment amount includes: the proportional term corresponding to the current weighted deviation, the integral term corresponding to the integral of the historical weighted deviation, and the differential term corresponding to the rate of change of the weighted deviation.

[0081] In this embodiment, the adaptive PID algorithm can be expressed as an adjustment rule that takes the deviations of each parameter calculated in the aforementioned steps and their real-time actual values ​​as inputs, and the correction amount (i.e., adjustment amount) of the weight coefficients of each parameter as output.

[0082] It should be noted that in this embodiment, the adaptive PID algorithm draws on the idea of ​​proportional-integral-derivative control, but the object of action is not the direct control signal, but the weight coefficients in the evaluation system.

[0083] This refers to the adaptive PID algorithm in this implementation, which maps the circuit's operating state (manifested as deviations in various parameters) to a fine-grained adjustment of the evaluation criteria (weighting coefficients) in real time. When a sudden load change causes a sharp increase in current deviation, this algorithm allows the weight of the current parameter to be quickly increased, causing the control module to automatically shift its optimization focus to current limiting. When the system enters a steady state and voltage accuracy deviation becomes the primary concern, the weight of the voltage parameter can gradually take over. In this way, a leap from fixed-target control to target self-optimization control is achieved, fundamentally solving the technical problem that the overall system performance cannot be continuously optimized under dynamic operating conditions due to rigid control targets.

[0084] In one specific implementation, a weighted deviation is first calculated. That is, for each circuit parameter, the deviation calculated in step S14 is multiplied by the actual value of that parameter collected in step S12. This operation makes the influence of the deviation more clearly correlated with the actual operating level of the circuit (e.g., voltage level, current magnitude).

[0085] Then, calculate the three adjustment amounts separately:

[0086] The proportional term is proportional to the current weighted deviation. Specifically, the weighted deviation of each parameter can be multiplied by an independently adjustable proportional gain coefficient to obtain the proportional adjustment of that parameter's weight coefficient. This term is used for rapid response to current operational deviations.

[0087] The integral term is proportional to the integral of the historical weighted deviation. Specifically, the weighted deviation of each parameter can be integrated over time (or accumulated through discretization) to obtain the historical cumulative amount, which is then multiplied by an independent integral gain coefficient to obtain the integral adjustment amount for that parameter. This term is used to eliminate potential steady-state biases and ensure long-term accuracy.

[0088] The differential term is proportional to the rate of change of the weighted deviation. Specifically, the rate of change of the weighted deviation of each parameter over time (or the difference under discretization) can be calculated, and then multiplied by an independent differential gain coefficient to obtain the differential adjustment of that parameter. This term is used to predict the trend of deviation changes, suppress system oscillations, and improve stability.

[0089] Next, for each circuit parameter, its proportional adjustment, integral adjustment, and derivative adjustment are added together to obtain the total weighted adjustment of that parameter.

[0090] Finally, update the weighting coefficients. The current weighting coefficient for each circuit parameter is added to the calculated total weight adjustment for that parameter to obtain the updated weighting coefficient. This operation allows the weighting coefficients to be dynamically and automatically adjusted based on the circuit's real-time operating status (reflected by deviation) and trends (reflected by integration and differentiation).

[0091] Step S18: Calculate the actual value of the current system's overall performance based on the updated weighting coefficients and the actual values ​​of each circuit parameter.

[0092] In this embodiment, the updated weighting coefficients from step S16 can be used to weight and synthesize the actual values ​​of the circuit parameters collected in step S12, thereby calculating a scalar value reflecting the overall performance of the current system. Specifically, the actual values ​​of each circuit parameter can be constructed into an actual parameter vector, and the updated weighting coefficients from step S16 can be constructed into an updated weight vector. The dot product of this updated weight vector and the actual parameter vector is then calculated, and the result is the current overall system performance value. This current overall system performance value represents the comprehensive score of the circuit's current operating state based on the latest weighting evaluation system.

[0093] Step S110: Adjust the control signal of the Boost circuit according to the deviation between the current actual value of the system's overall performance and the expected value of the system's overall performance, and recalculate the deviation to update each weight coefficient when the deviation exceeds the preset range.

[0094] In this embodiment, this step is used to complete the control closed loop and achieve iterative optimization. First, the difference between the current actual value of the system's overall performance calculated in step S18 and the expected value of the system's overall performance determined in step S10 is used to obtain the overall performance deviation. Then, this overall performance deviation is used to adjust the control signal and trigger weight readjustment. Specifically, based on the magnitude and direction of this overall performance deviation, and combined with the current circuit parameter deviation (from step S14) and the updated weight coefficients (from step S16), a preset control law (e.g., another PID controller for the overall deviation, or a duty cycle lookup table) is used to calculate and finally generate and output a drive signal (which can be a PWM signal) for adjusting the on / off state of the switching transistor (e.g., MOSFET) in the Boost circuit. This signal directly controls the power conversion of the circuit so that the actual performance of the circuit approaches the overall expected value. It can also be determined whether the absolute value of the overall performance deviation exceeds a preset tolerance range (i.e., the preset range). If it does not exceed the preset range, it means that under the current weight coefficients, the overall performance of the circuit is close enough to the target, and the system maintains the current weights and controller parameters. If the deviation exceeds the limit, it indicates that the current adaptive weight adjustment result has not yet enabled the system to achieve satisfactory overall performance, and a new round of weight optimization needs to be initiated. At this point, the process returns to step S14, recalculates the deviation based on the latest acquired actual values ​​of the circuit parameters, and executes step S16 again, forming a closed-loop iteration of evaluation-adjustment-re-evaluation. This process continues until the overall performance deviation meets the requirements, thereby ensuring that the control system can dynamically track and maintain globally optimal performance.

[0095] In some implementations, the step of determining the expected values ​​of multiple circuit parameters and calculating a weighted sum of the expected values ​​of the multiple circuit parameters based on preset initial weighting coefficients to obtain the expected value of the overall system performance includes:

[0096] Step S102: Determine the expected values ​​of at least two circuit parameters of the energy storage element in the Boost circuit and the expected values ​​of at least two circuit parameters of the load loop in the Boost circuit.

[0097] In this embodiment, the energy storage element can be an inductor in a Boost topology, and its circuit parameters are used to reflect the energy storage and dynamic processes within the circuit. The load loop can be represented as a loop consisting of the output terminal of the Boost circuit and the load connected to it, and its circuit parameters can be used to reflect the final output performance of the circuit and external requirements.

[0098] In this embodiment, the executing entity can obtain pre-set target values ​​or values ​​calculated based on real-time load information from its internal memory or through an external interface. Specifically, the expected values ​​for the circuit parameters of the energy storage element may include the expected values ​​for inductor current and inductor voltage.

[0099] More specifically, the expected inductor current can be calculated using circuit theory formulas (e.g., formulas related to duty cycle in continuous conduction mode) based on the target output power, estimated circuit efficiency, and input voltage. Alternatively, the center value corresponding to the maximum allowable ripple current can be set as the expected inductor current. The expected inductor voltage characterizes the target voltage level that the inductor should achieve under ideal steady-state conditions. It can be calculated based on the fundamental principles of circuit topology (e.g., the volt-second balance principle) from the input voltage and the desired ratio of the on-time to off-time of the switching transistor (duty cycle). Expected values ​​for circuit parameters of the load loop can include expected load voltage and expected load current. The expected load voltage can be the DC voltage value that the Boost circuit needs to output stably. The expected load current can be expressed as the magnitude of the current that the circuit expects to deliver to the load. It can be determined based on the rated power of the load or the real-time power demand; it can be a fixed value or a dynamic value that updates with load changes.

[0100] Step S104: Combine the expected values ​​of at least two circuit parameters of the energy storage element and at least two circuit parameters of the load loop into an expected parameter vector.

[0101] In this implementation, the executing entity can create an ordered data list or array, i.e., a vector of expected parameters, in memory. Each element of this vector corresponds to the expected value of a circuit parameter. For example, if four parameters are selected: expected inductor current, expected inductor voltage, expected load voltage, and expected load current, a column vector with four elements can be constructed. The first element stores the expected inductor current, the second the expected inductor voltage, the third the expected load voltage, and the fourth the expected load current.

[0102] Step S106: Combine the preset initial weight coefficients into an initial weight vector.

[0103] In this embodiment, this step can assign a corresponding initial importance weight to each element in the expected parameter vector (i.e., each expected value of circuit parameter), and organize the weight values ​​into a vector as well.

[0104] In this embodiment, the preset initial weighting coefficients can be coefficients pre-set based on engineering experience, simulation analysis, or preliminary optimization strategies during the system startup or control process initialization phase. The magnitude of each coefficient represents the relative contribution or importance of the expected value of its corresponding circuit parameter to the overall system performance and the overall objective in the initial state. For example, if the stability of the output voltage is more important, the initial weighting coefficient corresponding to the expected load voltage value can be set larger. The control module can read the preset weighting coefficients and construct another data list or array in the exact same order as the parameters in the expected parameter vector.

[0105] Step S108: Calculate the dot product of the initial weight vector and the expected parameter vector to obtain the expected value of the overall system performance.

[0106] In this implementation, the first coefficient in the initial weight vector is multiplied by the first expected value in the expected parameter vector to obtain the first weighted term. The second coefficient in the initial weight vector is multiplied by the second expected value in the expected parameter vector to obtain the second weighted term, and so on. This multiplication operation can be performed on the weight coefficients and expected values ​​at all corresponding positions. Then, all the obtained weighted terms (products) are summed to obtain the final sum, which is the expected value of the system's overall performance.

[0107] In some embodiments, at least two circuit parameters of the energy storage element include inductor voltage and / or inductor current; at least two circuit parameters of the load circuit include load voltage and / or load current.

[0108] In this embodiment, the inductor current is a physical quantity that directly reflects the instantaneous energy stored in the inductor. Using it as a monitored parameter allows for real-time sensing of the circuit's input power, whether the inductor is approaching saturation, and the current stress in the switching transistor and freewheeling diode. In adaptive weight adjustment, if the inductor current deviation increases, the system can automatically increase its weight, prioritizing current control to ensure safety. Monitoring the inductor voltage helps indirectly determine the circuit's operating mode (e.g., continuous conduction mode, intermittent conduction mode) and whether the switching action is normal. The load voltage is the core output target of the Boost circuit and a crucial indicator of whether it meets load requirements. Monitoring the load current can sense load changes (e.g., sudden load increase or decrease), providing feedforward or predictive information for the control system.

[0109] In this embodiment, at least two circuit parameters of the energy storage element may further include the temperature of the inductor. Specifically, the temperature of the inductor can be the temperature of the inductor itself or its core and windings. The analog signal can be obtained by a temperature sensor (e.g., a thermistor, thermocouple) attached to the inductor and converted into a digital parameter.

[0110] Understandably, temperature is a key factor affecting inductor performance (e.g., inductance drift, increased losses) and long-term reliability. In high power density or high-temperature applications, inductor temperature can be included as one of the monitored parameters and weighted in adaptive weighting adjustments. This allows the system to automatically adjust its control strategy to prioritize reducing inductor losses or implementing overheat protection when the temperature approaches safe limits, thereby achieving a dynamic balance between electrical performance and thermal safety.

[0111] In this embodiment, at least two circuit parameters of the load loop may further include output power. It is understood that in applications requiring constant power control or maximum power point tracking, output power is a direct control objective. Using it as a monitoring parameter allows the adaptive weighting adjustment mechanism of this invention to serve power stability and optimization. For example, when load changes cause power to deviate from the target, the system can adaptively adjust the weights, making the control action more focused on quickly restoring the target power.

[0112] In this embodiment, at least two circuit parameters of the load circuit may further include output voltage ripple.

[0113] In some implementations, the step of acquiring the actual values ​​of multiple circuit parameters in real time includes:

[0114] Step S122: Simulated signals of multiple circuit parameters are sequentially acquired using a multi-channel time-division sampling method; wherein, for the acquisition of inductor voltage and / or inductor current, a first analog signal is obtained through a correspondingly set first sensor; for the acquisition of load voltage and / or load current, a second analog signal is obtained through a correspondingly set second sensor.

[0115] In this embodiment, the multi-channel time-division sampling method can be represented as using a time-multiplexing strategy to acquire multiple signal sources in turn using a single or limited signal processing link (e.g., amplification and filtering circuits and analog-to-digital converters) within a sampling period. That is, signals from different channels are turned on and measured in different time segments, rather than using completely independent parallel acquisition systems simultaneously.

[0116] In this embodiment, a preset multi-channel sampling module can be used to sequentially acquire analog signals of multiple circuit parameters. This multi-channel sampling module may include: a first sensor, a second sensor, an analog switch and sampling channel, a signal conditioning circuit, and an analog-to-digital converter. Specifically, the first sensor may further include: a high-precision resistor voltage divider network, a Hall effect current sensor, or a current transformer. The high-precision resistor voltage divider network can be used to acquire the inductor voltage, and can proportionally attenuate the potentially high voltage across the inductor (e.g., from tens of volts to a few volts), obtaining a low-voltage signal linearly proportional to the measured high voltage. The Hall effect current sensor or current transformer can acquire the inductor current. The second sensor may further include: a resistor voltage divider network, a Hall current sensor, a current transformer, or a sampling resistor plus a differential amplifier. The resistor voltage divider network is used to acquire the load voltage and to proportionally reduce the output voltage to the measurement range. The Hall current sensor, current transformer, or sampling resistor plus a differential amplifier is used to convert the output current into a voltage signal.

[0117] In this embodiment, an analog switch and sampling channel can be used to implement the multi-channel functionality. Specifically, the analog switch and sampling channel may include multiple input terminals, which can be connected to the output terminals of the aforementioned sensors respectively. It may also include a common output terminal and a digital control terminal. The control module (e.g., a microcontroller) can send a preset channel selection code to the digital control terminal of the analog switch. According to this code, the analog switch internally connects only one input terminal to the common output terminal at a given time, while the other input terminals are in a high-impedance, disconnected state. The control module cycles through the channel selection code in a preset, fixed time sequence, thereby sequentially connecting the inductor voltage signal, inductor current signal, load voltage signal, load current signal, etc., to the common output line. In this way, sequential access to multiple signals can be achieved.

[0118] Step S124: Convert the first analog signal and the second analog signal into digital signals, which are used as the actual values ​​of the corresponding circuit parameters.

[0119] In this embodiment, an analog-to-digital converter can be used to sample the conditioned continuous-time analog voltage signal that is alternately connected to the common output terminal in step S122 at specific time points, and quantize its amplitude into a digital code represented by binary numbers.

[0120] Specifically, the analog input of the analog-to-digital converter (ADC) can be connected to the common output of the analog switch (or the output of the signal conditioning circuit). The control module first sends a command to the analog switch to switch to the target channel (e.g., the inductor current channel). Then, after a short setup period, once the output signal of the analog switch and the conditioning circuit have stabilized, the control module can trigger the ADC to sample and convert the current analog voltage of that channel. After conversion, the ADC can generate an interrupt signal or the control module can read the status register to obtain the converted digital result. Next, the control module can read the digital result and, based on the current channel code, determine the circuit parameter represented by the digital value (e.g., representing inductor current). Finally, the digital value is stored in the memory variable or array element allocated to that parameter as the actual value of that parameter in the current sampling period. After completing the acquisition of one channel, the control module can update the channel selection code and switch to the next channel (e.g., switch to the inductor voltage channel), repeating the above process until all the circuit parameters to be monitored have completed data acquisition and conversion in this round of sampling.

[0121] In some implementations, the step of calculating the adjustment amount of the current weight coefficient of each circuit parameter based on the deviation and the corresponding actual value using an adaptive PID algorithm, and updating each weight coefficient, includes:

[0122] For each circuit parameter, perform the following steps:

[0123] Step S161: Multiply the deviation of the circuit parameter by its actual value to obtain the weighted deviation of the circuit parameter.

[0124] In this embodiment, for each monitored circuit parameter, the control module can read the deviation value calculated in step S14 from the memory and multiply it by the actual value acquired in real time in step S12. For example, for the inductor current parameter, its deviation value is multiplied by the actual current value to obtain the weighted deviation of the inductor current. The purpose of this operation is to convert the absolute deviation into a relative quantity related to the power or energy level under the current operating condition. For example, the same absolute current deviation has a more significant impact on the system when the system is running at a high current than when it is running at a low current. By introducing the actual value for weighting, the subsequent adjustment calculation can more reasonably reflect this difference.

[0125] Step S162: Multiply the weighted deviation by the proportional gain coefficient corresponding to the circuit parameter to obtain the proportional adjustment amount of the circuit parameter.

[0126] In this embodiment, the proportional adjustment can be represented as an adjustment component generated by scaling the current weighted deviation based on a pre-set proportional coefficient (proportional gain coefficient). That is, the larger the current weighted deviation, the further the current state of the parameter deviates from the desired state in a weighted sense. Therefore, a correspondingly larger immediate adjustment needs to be applied to the weight coefficient of this parameter to encourage the control system to pay more attention to the recovery of this parameter in the future. The proportional gain coefficient determines the strength of this response.

[0127] In one possible and specific implementation, the control module can multiply the weighted deviation of the parameter calculated in step S161 by a pre-set proportional gain coefficient for that parameter. This coefficient determines the system's response strength to the current weighted deviation; the larger the coefficient, the faster the response. The calculation result is the proportional adjustment amount of the parameter's weight coefficient, which drives the weight coefficient to change in the direction of reducing the current deviation. The proportional gain coefficients of all parameters can be independent of each other, can be stored in a gain coefficient table, and allow independent adjustment for the dynamic characteristics of different parameters.

[0128] Step S163: Perform an integral operation on the weighted deviation, and multiply the integral result by the integral gain coefficient corresponding to the circuit parameter to obtain the integral adjustment amount of the circuit parameter.

[0129] In this embodiment, the integral adjustment can be represented as an adjustment component obtained by accumulating (integrating) the weighted deviation of the parameter over time, and then scaling the accumulated value according to a preset integral gain coefficient. The purpose of the integral operation is to accumulate historical deviation information. Even if the weighted deviation at the current moment is very small, if the parameter has had a persistent small weighted deviation over a period of time, its integral value will gradually increase, thereby generating a continuous weight correction force through the integral adjustment until the historically accumulated deviation effect is eliminated.

[0130] In one possible and specific implementation, firstly, the control module performs an integral operation on the weighted deviation of the parameter obtained in step S161. This can be achieved by accumulating the historical values ​​of the weighted deviation (i.e., (Weighted bias × sampling period)). Next, the obtained integral value (cumulative sum) is multiplied by the integral gain coefficient set for this parameter. This coefficient determines the strength and speed at which the system eliminates steady-state errors. To prevent excessive accumulation of the integral term (integral saturation), an integral limiting mechanism can also be introduced. The final product is the integral adjustment amount of the parameter's weighting coefficient.

[0131] Step S164: Calculate the rate of change of the weighted deviation, and multiply the rate of change by the differential gain coefficient corresponding to the circuit parameter to obtain the differential adjustment amount of the circuit parameter.

[0132] In this embodiment, the differential adjustment can be expressed as an adjustment component scaled by a preset differential gain coefficient based on the rate of change of the weighted deviation (i.e., the difference between the current weighted deviation and the weighted deviation at the previous moment, divided by the time interval, approximating the rate of change). Its function is to sense the direction and speed of change of the weighted deviation. If the weighted deviation is increasing rapidly, the differential term will generate a positive adjustment (assuming a positive gain), thereby increasing the weight of this parameter in advance and more actively suppressing the expansion of the deviation. Conversely, if the weighted deviation is decreasing rapidly, the differential term can generate a negative adjustment to prevent overshooting due to excessive adjustment of the weights. It plays a role in damping and prediction.

[0133] In one possible and specific implementation, the control module can calculate the rate of change of the weighted deviation of the parameter obtained in step S161. This can be approximated by calculating the difference between the weighted deviation of the current sampling period and the weighted deviation of the previous sampling period, and then dividing by the sampling period (i.e., first-order difference). Then, this rate of change is multiplied by the differential gain coefficient set for the parameter. This coefficient determines the system's sensitivity to the trend of deviation change. Since the differential action is sensitive to noise, the differential signal can also be smoothed and filtered. The calculation result is the differential adjustment amount of the parameter's weight coefficient, which can exert a reverse suppression effect when the deviation has an increasing trend and a maintenance effect when it has a decreasing trend.

[0134] Step S165: Add the proportional adjustment, integral adjustment and derivative adjustment to obtain the total weight adjustment of the circuit parameters.

[0135] In this embodiment, the control module can algebraically add the proportional adjustment, integral adjustment, and derivative adjustment calculated in steps S162, S163, and S164 respectively for the same circuit parameter. The result of this summation is the total weight adjustment of the circuit parameter, which comprehensively indicates the adjustment direction and magnitude of the parameter's weight coefficient at the current moment.

[0136] Step S166: Add the total weight adjustment amount to the current weight coefficient of the circuit parameter to obtain the updated weight coefficient.

[0137] In this embodiment, the control module can read the current weight coefficient of the parameter from the memory, and then add it to the total weight adjustment amount of the parameter calculated in step S165. The result of the addition is the updated weight coefficient.

[0138] In some implementations, each circuit parameter corresponds to an independent proportional gain coefficient, integral gain coefficient, and differential gain coefficient.

[0139] Step S162: The step of multiplying the weighted deviation by the proportional gain coefficient corresponding to the circuit parameter to obtain the proportional adjustment amount of the circuit parameter includes:

[0140] Step S1620: Combine the weighted deviations of each circuit parameter into a first vector, use the proportional gain coefficients corresponding to each circuit parameter as diagonal elements to form a first diagonal matrix, calculate the product of the first diagonal matrix and the first vector, and obtain a proportional adjustment vector composed of the proportional adjustment amounts of each circuit parameter.

[0141] In this embodiment, combining the weighted deviations of each circuit parameter into a first vector can be represented as creating a column vector. Each element of this vector corresponds to a weighted deviation value of a monitored circuit parameter calculated in step S161. That is, the current weighted deviation states of all parameters are organized into an ordered set of numbers.

[0142] In this embodiment, constructing a first diagonal matrix using the proportional gain coefficients corresponding to each circuit parameter as diagonal elements can be represented as creating a diagonal matrix. This matrix has non-zero elements only on its main diagonal, and these diagonal elements are arranged in the same order as the parameters in the first vector, with each parameter's independent proportional gain coefficient placed sequentially. The off-diagonal elements are zero, ensuring that the adjustment calculation for each parameter depends only on its own gain coefficient and weighted deviation, and is independent of other parameters, thus maintaining decoupling between parameters.

[0143] In this embodiment, calculating the product of the first diagonal matrix and the first vector can be represented as performing a matrix-vector multiplication operation. According to the rules of linear algebra, multiplying a diagonal matrix by a vector is equivalent to multiplying each element on the diagonal of the matrix (i.e., the gain coefficient of each parameter) by the element at the corresponding position in the vector (i.e., the weighted deviation of that parameter).

[0144] Step S163: The step of integrating the weighted deviation and multiplying the integration result by the integral gain coefficient corresponding to the circuit parameter to obtain the integral adjustment amount of the circuit parameter includes:

[0145] Step S1630: Combine the integration results of each circuit parameter into a second vector, use the integral gain coefficients corresponding to each circuit parameter as diagonal elements to form a second diagonal matrix, calculate the product of the second diagonal matrix and the second vector, and obtain the integral adjustment vector composed of the integral adjustment amount of each circuit parameter.

[0146] In this embodiment, combining the integral results of each circuit parameter into a second vector can be represented as creating a column vector whose elements are the cumulative value of the historical weighted deviation of each parameter (i.e., the integral result). This integral result can be updated in each control cycle (by adding the weighted deviation of the current cycle). Similarly, constructing a second diagonal matrix using the integral gain coefficients corresponding to each circuit parameter as diagonal elements can be represented as constructing a diagonal matrix whose main diagonal is composed of the independent integral gain coefficients of each parameter. That is, the control module can construct a second diagonal matrix based on the integral gain coefficients independently set for each parameter. Its form is similar to the first diagonal matrix, with the main diagonal elements being the integral gain coefficients of each parameter and the off-diagonal elements being zero. The control module performs a multiplication operation between the second diagonal matrix and the second vector. This operation simultaneously calculates the product of the integral results of all parameters and their respective integral gain coefficients, and the output result is an integral adjustment vector. Each element of this vector is the integral adjustment amount of the corresponding circuit parameter.

[0147] Step S164: Calculating the rate of change of the weighted deviation and multiplying the rate of change by the differential gain coefficient corresponding to the circuit parameter to obtain the differential adjustment amount of the circuit parameter, includes:

[0148] Step S1640: Combine the rate of change of each circuit parameter into a third vector, use the differential gain coefficients corresponding to each circuit parameter as diagonal elements to form a third diagonal matrix, calculate the product of the third diagonal matrix and the third vector, and obtain a differential adjustment vector composed of the differential adjustment amounts of each circuit parameter.

[0149] In this implementation, a column vector can also be created, whose elements are the rate of change (i.e., differential approximation) of the weighted deviation of each parameter at the current time relative to the previous time. It can be understood that the rate of change can be calculated using a first-order backward difference, i.e., subtracting the weighted deviation of the previous sampling period from the weighted deviation of the current sampling period, and then dividing by the sampling period. The rates of change of the weighted deviations of all parameters are arranged in order, forming the third vector. Similarly, this third diagonal matrix is ​​a diagonal matrix composed of the independent differential gain coefficients of each parameter.

[0150] In some implementations, the step of adjusting the control signal of the Boost circuit based on the deviation between the current actual value of the system's overall performance and the expected value of the system's overall performance, and recalculating the deviation to update each weight coefficient when the deviation exceeds a preset range, includes:

[0151] Step S1102: Subtract the current actual value of the system's overall performance from the expected value of the system's overall performance to obtain the overall performance deviation.

[0152] In this embodiment, the control module can read the actual value of the current system overall performance calculated in step S18 from its memory, and perform an algebraic subtraction operation with the expected value of the system overall performance determined and stored in step S10 to obtain the overall performance deviation. This value can be positive, negative, or zero, and its absolute value reflects the degree of deviation.

[0153] Step S1104: If the absolute value of the overall performance deviation is less than or equal to a preset threshold, then based on the currently updated weight coefficients, the deviations of each circuit parameter and their actual values, calculate and output a control signal for adjusting the duty cycle of the Boost circuit switching transistor.

[0154] In this embodiment, this step is the execution phase of the control process. When the absolute value of the overall performance deviation does not exceed the preset, acceptable error range (i.e., the preset threshold), the control module determines that the current set of weighting coefficients and the circuit state have reached a relatively optimal balance point, and no further weighting adjustment is needed. At this time, the control module can use all the latest available information (including the newly updated weighting coefficients, the actual deviations of each circuit parameter and their measured values) to calculate the final control command through a preset control algorithm. This command directly acts on the power switch of the Boost circuit, maintaining or fine-tuning the circuit operation by changing the ratio of its on and off times (duty cycle).

[0155] Specifically, the preset threshold can be a positive number pre-set according to system performance requirements (e.g., steady-state accuracy). This preset threshold defines the acceptable performance range boundary of the expected value, for example, it can be set to ±5% of the expected value.

[0156] The control module can use the currently updated weighting coefficients (from the latest result in step S16, representing the system's current trade-off strategy for various performance indicators), the deviations of each circuit parameter (from step S14, reflecting the instantaneous difference between each individual parameter and the target), and the actual values ​​of each circuit parameter as inputs to the algorithm. These inputs are fed into a preset algorithm to obtain the algorithm's output. This preset algorithm can be a digital PID controller, fuzzy control, or sliding mode control, etc. The output of the algorithm can be a control quantity used to adjust the duty cycle of the switching transistor. Specifically, it can be the duty cycle setting value of the pulse width modulation signal. The control module can write this setting value into the corresponding register of its internal PWM generator module. The PWM generator can then generate a high-frequency pulse signal with the corresponding duty cycle based on this value and output it to the drive circuit of the switching transistor (e.g., MOSFET), thereby completing the direct control of the Boost circuit power stage.

[0157] Step S1106: If the absolute value of the overall performance deviation is greater than the preset threshold, then based on the latest collected actual values ​​of the circuit parameters, recalculate the deviation between each circuit parameter and its expected value, and return to execute the step of calculating the adjustment amount of the current weight coefficient of each circuit parameter using the adaptive PID algorithm based on the deviation and the corresponding actual value, and updating each weight coefficient, until the absolute value of the recalculated overall performance deviation is less than or equal to the preset threshold.

[0158] In this embodiment, this step is the learning and adjustment phase of the control process. When it is determined that the overall performance deviation exceeds the acceptable range, it indicates that the currently used set of weighting coefficients has failed to guide the circuit to achieve the expected overall performance. The system will not immediately generate control signals based on this unsatisfactory weighting, but will trigger a new perception-calculation-adjustment cycle. The goal of this process is to optimize the system's performance evaluation criteria (i.e., weight allocation) itself by readjusting the weighting coefficients, thereby ultimately finding a weighting configuration that enables the overall performance to meet the target.

[0159] Specifically, the control module first re-executes step S14 based on the actual values ​​of the circuit parameters acquired in step S12 from the latest sampling period, that is, recalculates the deviation between each circuit parameter and its expected value. Then, the program flow returns to step S16, that is, based on the deviation and the corresponding actual value, the adaptive PID algorithm is used to calculate the adjustment amount of the current weight coefficient of each circuit parameter, and the weight coefficients are updated. After steps S16 and S18 are completed, a new actual value of the overall system performance calculated based on the new weights can be obtained, and a new overall performance deviation is calculated accordingly. Finally, the control module can again determine whether the new overall performance deviation is less than or equal to a preset threshold. If yes, the loop is exited and step S1104 is executed to output the control signal. If no, the iterative process of step S1106 above is repeated.

[0160] This embodiment provides a Boost circuit based on adaptive weight adjustment, including:

[0161] The Boost circuit module is used to implement DC voltage boost conversion;

[0162] A multi-channel sampling module, the input of which is connected to the Boost circuit module, is used to acquire the actual values ​​of multiple circuit parameters of the Boost circuit module in real time;

[0163] A control module, which is connected to the output terminal of the multi-channel sampling module and the control terminal of the Boost circuit module, is configured to execute the control method described above to generate a control signal to adjust the switching transistor of the Boost circuit module.

[0164] In one possible and specific implementation, a Boost circuit monitoring device based on an adaptive PID algorithm is provided, comprising: a Boost circuit module, a multi-channel sampling module, and an algorithm processing module. This device system is visible. Figure 2 .

[0165] The Boost circuit module consists of a Boost circuit and a load circuit. The main function of the Boost circuit is to boost the input voltage to a higher output voltage. It is typically composed of basic components such as inductors, switching transistors (usually MOSFETs), diodes, and capacitors. The periodic switching on and off of the switch stores and releases energy, thereby achieving the purpose of boosting the voltage. The electrical characteristics of the inductor, especially the inductor current and voltage, play a decisive role in the stability of the circuit. The performance of the load circuit has a significant impact on the stability and efficiency of the Boost circuit; fluctuations in load current and voltage directly affect the instability of the Boost circuit.

[0166] The multi-channel sampling module includes an analog switch, sampling channels, and an analog-to-digital converter (ADC). The analog switch input connects to the load terminal of the Boost circuit module and the inductor. The analog switch switches between different sampling channels to achieve time-division sampling of parameters such as inductor voltage, inductor current, load voltage, and load current in the Boost circuit. The sampling channels are connected to the analog switch output and include sensors and signal conditioning circuitry. Sensors can be selected based on the type of parameter being measured; for voltage measurements, a combination of high-precision voltage divider resistors and differential amplifiers is used, while for current measurements, a Hall effect sensor or current transformer is used. The signal conditioning circuit is connected to the sensor output to amplify and filter the weak signals acquired by the sensors to improve signal quality. The ADC is connected to the multi-channel sampling module to convert the analog signals into digital signals.

[0167] The algorithm processing module's input is connected to the output of the multi-channel sampling module, and its output is connected to the MOSFET in the Boost circuit module. This module receives the digital signals output from the multi-channel sampling module, i.e., the values ​​of each sampling parameter. During actual system operation, this module can optimize performance indicators based on the error between the actual and expected values, thereby generating control signals to adjust the duty cycle of the MOSFET, ultimately achieving dynamic adjustment of system performance indicators. For example, if monitoring reveals a significant difference between the inductor current-related terms and the expected value, and this has a substantial impact on overall performance, the module's internal adaptive mechanism will quickly activate, adjusting the inductor current weighting coefficient to approach the expected value, while simultaneously reducing other weighting coefficients according to a predetermined ratio, and always strictly ensuring that the weighting coefficients remain within a reasonable range. This reasonable range is typically set so that all weighting coefficients are non-negative and their sum falls within a specific interval, which can be determined based on the circuit's characteristics. If the control signal changes suddenly, it may cause circuit oscillation; to ensure the stability of the Boost circuit, the control signal will be smoothed. Especially when the load changes suddenly, the control signal will not increase to the maximum value instantly, but will be adjusted gradually according to the dynamic response characteristics of the system to avoid impacting the circuit components.

[0168] The algorithm processing module uses an adaptive PID (Proportional Integral Derivative) algorithm that is derived from the classic PID control principle (the principle is as follows). Figure 3 (As shown). The PID algorithm calculates three control signals—proportional, integral, and derivative—based on the error between the actual and desired system values. The proportional control is adjusted according to the magnitude of the error, allowing the system to quickly approach the desired value; the integral control eliminates the system's static error; and the derivative control is adjusted according to the rate of change of the error to prevent system oscillation.

[0169] Specifically, the output of the PID controller It can be calculated using the following formula:

[0170] (1);

[0171] In the formula, For the controller's output; The input to the controller is the difference between the given value and the output value of the controlled object, called the deviation signal. The proportional gain of the controller; The integral coefficient of the controller; The differential coefficients of the controller; The sampling period.

[0172] The adaptive PID algorithm not only improves the stability and efficiency of the Boost circuit, but also ensures that it maintains optimal performance under various operating conditions, providing a comprehensive performance indicator for system evaluation.

[0173] like Figure 4 As shown, a Boost circuit control method based on an adaptive PID algorithm is implemented as follows:

[0174] S100: Calculate the expected value of the system.

[0175] Calculate the inductor voltage (based on the specific load circuit) as the expected performance parameter of the system. ), inductor current ( ), load voltage ( ), load current ( Based on the obtained expected indicators, substitute them into the following formula:

[0176] (2);

[0177] In the formula, It can be used as the expected value of the overall system performance, representing the global and comprehensive performance quantification benchmark that the entire Boost circuit is expected to achieve under ideal and stable operating conditions. , , , The weighting coefficients are the initial importance coefficients pre-assigned to the four desired parameters. This is the desired value of the inductor voltage, which is the target value that the average voltage across the inductor in the Boost circuit is expected to reach. This is the desired value of the inductor current, which is the target value that the average current flowing through the inductor of the Boost circuit is expected to reach. This represents the expected load voltage. This represents the expected value of the load current.

[0178] The weighting coefficients and their constraints can be determined based on the specific load circuit. All weighting coefficients are non-negative and their sum is within a specific range, such as: .

[0179] S200: Obtain actual system metrics.

[0180] The inductor voltage under actual conditions is collected according to the sampling period. ), inductor current ( ), load voltage ( ), load current ( ).

[0181] S300: Calculates the deviation of the system indicators.

[0182] Substitute the actual indicators collected by S200 into the following formula to obtain the deviation from the expected value.

[0183] (3);

[0184] In the formula, , , as well as These four deviation vectors represent the instantaneous algebraic differences between the current actual operating value and the pre-set ideal expected value.

[0185] S400: Calculate the proportional adjustment amount.

[0186] Let the proportional gain matrix be:

[0187] (4);

[0188] In the formula, - These are the diagonal elements in the proportional gain matrix, each corresponding to a monitored circuit parameter (e.g., inductor voltage, inductor current, load voltage, and load current in sequence).

[0189] The deviation calculated in step S300 is substituted into the formula to obtain the proportional adjustment amount.

[0190] (5);

[0191] In the formula, For the proportional adjustment vector, - The four elements are the weighting coefficients of the four monitored circuit parameters (which can be inductor voltage, inductor current, load voltage, and load current in that order) and are adjusted proportionally. This is the proportional gain matrix. This is the inductor current weighted deviation. This is the weighted deviation of the inductor voltage. This is the load current weighted deviation. This is the load voltage weighted deviation.

[0192] S500: Calculate the integral adjustment.

[0193] Define the integral error vector function:

[0194] (6)

[0195] In the formula, This is the integral error vector. For a moment The actual value of the inductor current. For a moment The actual value of the inductor voltage. For a moment The actual value of the load current. For a moment The actual value of the load voltage.

[0196] Let the integral gain matrix be:

[0197] (7)

[0198] In the formula, Here is the integral gain matrix. - It consists of four diagonal elements, each of which is an independent integral gain coefficient.

[0199] Substitute the deviation calculated in step S300 into the following formula to obtain the proportional adjustment amount.

[0200] (8)

[0201] In the formula, Adjust the vector for integration. - The four elements of the integral adjustment vector correspond to the integral adjustment of the weighting coefficients of the four monitored circuit parameters (which can be in the order of inductor current, inductor voltage, load current, and load voltage). This is the integral gain matrix. This is the historical integral of the inductor current weighted deviation. This is the historical integral of the weighted deviation of the inductor voltage. It is the historical integral of the load current weighted deviation. It is the historical integral of the load voltage weighted deviation.

[0202] S600: Calculate the differential adjustment amount.

[0203] Define the error rate of change vector function:

[0204] (9)

[0205] In the formula, Let the error change rate vector be... The rate of change of the weighted deviation of the inductor current. The rate of change of the weighted deviation of the inductor voltage. The rate of change of the load current weighted deviation. This represents the rate of change of the weighted deviation of the load voltage.

[0206] Let the differential gain matrix be:

[0207] (10);

[0208] In the formula, The differential gain matrix is... - There are four non-zero elements, each corresponding to one of the four rate of change terms.

[0209] Calculate the differential adjustment amount based on the deviation obtained in step S300:

[0210] (11);

[0211] In the formula, Let be the differential adjustment vector. - The differential adjustment of the weighting coefficients corresponds to the four monitored circuit parameters (inductor current, inductor voltage, load current, and load voltage in that order). - These are the four independent differential gain coefficients in the differential gain matrix.

[0212] S700: Update weight coefficients

[0213] Based on the proportional adjustment, integral adjustment, and differential adjustment obtained from S300 to S600, substitute them into formula (12) to obtain the updated weight coefficients.

[0214] (12);

[0215] In the formula, The updated weight coefficients for the nth parameter. This indicates that the weight coefficient of the nth circuit parameter at the current time (t) is updated from the previous time. This represents the proportional adjustment amount obtained by calculating the proportional term for the nth parameter. This represents the integral adjustment amount obtained by calculating the integral term for the nth parameter. This represents the differential adjustment amount obtained by calculating the differential term for the nth parameter.

[0216] S800: Calculates the actual values ​​of system indicators.

[0217] Substitute the weighting coefficients calculated in S700 and the actual parameters collected in S200 into the formula:

[0218] (13);

[0219] In the formula, This is the global comprehensive performance score of the Boost circuit at the current moment, based on the latest weighted coefficient evaluation system. - These are the new weight coefficients obtained after calculation and updating using the adaptive PID algorithm.

[0220] S900: Calculates the deviation of the actual value of the system.

[0221] Calculate the actual value Compared with expected value The difference Its calculation formula is This difference will be used to accurately evaluate the effect achieved by this weighting coefficient adjustment. It will then determine whether the updated weighting coefficients can effectively bring the actual operating state of the circuit closer to the desired state. Given that different load circuits have different characteristics and requirements, appropriate weighting coefficients need to be set in advance. The range, for example, can be set to exist Within this range. If the calculation result exceeds the preset range, then it is necessary to return to step S300 and re-perform the relevant calculation and adjustment operations; conversely, if... If the values ​​are within the set range, it indicates that the updated weighting coefficients have successfully brought the actual state of the circuit close to the desired state, and no further adjustments are needed.

[0222] According to an embodiment of the present invention, an electronic device is provided. The electronic device in this embodiment may include one or more of the following components: a processor, a network interface, memory, non-volatile memory, and one or more application programs, wherein the one or more application programs may be stored in the non-volatile memory and configured to be executed by one or more processors, and the one or more programs are configured to perform the methods described in the foregoing method embodiments.

[0223] According to embodiments of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a computer, causes the computer to perform the method described in any of the above embodiments.

[0224] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A Boost circuit control method based on adaptive weight adjustment, characterized in that, include: Determine the expected values ​​of multiple circuit parameters, and calculate the weighted sum of the expected values ​​of multiple circuit parameters based on preset initial weighting coefficients to obtain the expected value of the overall system performance; Real-time acquisition of actual values ​​of multiple circuit parameters; Calculate the deviation between the actual values ​​and their corresponding expected values ​​for each circuit parameter; Based on the deviation and the corresponding actual value, the adjustment amount of the current weight coefficient of each circuit parameter is calculated using an adaptive PID algorithm, and each weight coefficient is updated; wherein, the adjustment amount includes: a proportional term corresponding to the current weighted deviation, an integral term corresponding to the integral of the historical weighted deviation, and a differential term corresponding to the rate of change of the weighted deviation. Calculate the actual value of the current system's overall performance based on the updated weighting coefficients and the actual values ​​of each circuit parameter; Based on the deviation between the current actual value of the system's overall performance and the expected value of the system's overall performance, the control signal of the Boost circuit is adjusted, and when the deviation exceeds the preset range, the deviation is recalculated to update each weight coefficient.

2. The method according to claim 1, characterized in that, The step of determining the expected values ​​of multiple circuit parameters and calculating the weighted sum of the expected values ​​of the multiple circuit parameters based on preset initial weighting coefficients to obtain the expected value of the overall system performance includes: Determine the expected values ​​of at least two circuit parameters of the energy storage element in the Boost circuit and the expected values ​​of at least two circuit parameters of the load loop in the Boost circuit. Combine the desired values ​​of at least two circuit parameters of the energy storage element and at least two circuit parameters of the load loop into a desired parameter vector; Combine the preset initial weight coefficients into an initial weight vector; The dot product of the initial weight vector and the expected parameter vector is calculated to obtain the expected value of the overall system performance.

3. The method according to claim 2, characterized in that, The energy storage element has at least two circuit parameters including inductor voltage and / or inductor current; the load circuit has at least two circuit parameters including load voltage and / or load current.

4. The method according to claim 3, characterized in that, The step of acquiring the actual values ​​of multiple circuit parameters in real time includes: Analog signals of multiple circuit parameters are sequentially acquired using a multi-channel time-division sampling method; for the acquisition of inductor voltage and / or inductor current, a first analog signal is obtained through a corresponding first sensor; for the acquisition of load voltage and / or load current, a second analog signal is obtained through a corresponding second sensor. The first and second analog signals are converted into digital signals, which are used as the actual values ​​of the corresponding circuit parameters.

5. The method according to claim 1, characterized in that, The step of calculating the adjustment amount of the current weight coefficient of each circuit parameter based on the deviation and the corresponding actual value, and updating each weight coefficient, includes: For each circuit parameter, perform the following steps: Multiply the deviation of the circuit parameter by its actual value to obtain the weighted deviation of the circuit parameter; Multiply the weighted deviation by the proportional gain coefficient corresponding to the circuit parameter to obtain the proportional adjustment amount of the circuit parameter. The weighted deviation is integrated, and the integral result is multiplied by the integral gain coefficient corresponding to the circuit parameter to obtain the integral adjustment amount of the circuit parameter. Calculate the rate of change of the weighted deviation and multiply the rate of change by the differential gain coefficient corresponding to the circuit parameter to obtain the differential adjustment amount of the circuit parameter; The total weighted adjustment of the circuit parameters is obtained by adding the proportional adjustment, integral adjustment, and derivative adjustment. The total weight adjustment is added to the current weight coefficient of the circuit parameter to obtain the updated weight coefficient.

6. The method according to claim 5, characterized in that, Each circuit parameter corresponds to an independent proportional gain coefficient, integral gain coefficient, and differential gain coefficient. The step of multiplying the weighted deviation by the proportional gain coefficient corresponding to the circuit parameter to obtain the proportional adjustment amount of the circuit parameter includes: The weighted deviations of each circuit parameter are combined into a first vector. The proportional gain coefficients corresponding to each circuit parameter are used as diagonal elements to form a first diagonal matrix. The product of the first diagonal matrix and the first vector is calculated to obtain a proportional adjustment vector composed of the proportional adjustment amounts of each circuit parameter. The step of integrating the weighted deviation and multiplying the integral result by the integral gain coefficient corresponding to the circuit parameter to obtain the integral adjustment amount of the circuit parameter includes: The integral results of each circuit parameter are combined into a second vector. The integral gain coefficients corresponding to each circuit parameter are used as diagonal elements to form a second diagonal matrix. The product of the second diagonal matrix and the second vector is calculated to obtain the integral adjustment vector composed of the integral adjustment amount of each circuit parameter. The steps of calculating the rate of change of the weighted deviation and multiplying the rate of change by the differential gain coefficient corresponding to the circuit parameter to obtain the differential adjustment amount of the circuit parameter include: The rate of change of each circuit parameter is combined into a third vector, and the differential gain coefficients corresponding to each circuit parameter are used as diagonal elements to form a third diagonal matrix. The product of the third diagonal matrix and the third vector is calculated to obtain a differential adjustment vector composed of the differential adjustment amounts of each circuit parameter.

7. The method according to claim 5, characterized in that, The step of adjusting the control signal of the Boost circuit based on the deviation between the current actual value of the system's overall performance and the expected value of the system's overall performance, and recalculating the deviation to update each weight coefficient when the deviation exceeds a preset range, includes: The difference between the current actual value of the system's overall performance and the expected value of the system's overall performance is used to obtain the overall performance deviation. If the absolute value of the overall performance deviation is less than or equal to the preset threshold, then based on the currently updated weight coefficients, the deviations of each circuit parameter and their actual values, a control signal for adjusting the duty cycle of the Boost circuit switching transistor is calculated and output. If the absolute value of the overall performance deviation is greater than the preset threshold, then based on the latest collected actual values ​​of the circuit parameters, the deviation between each circuit parameter and its expected value is recalculated, and the process of calculating the adjustment amount of the current weight coefficient of each circuit parameter using the adaptive PID algorithm based on the deviation and the corresponding actual value is returned, and the weight coefficients are updated, until the absolute value of the recalculated overall performance deviation is less than or equal to the preset threshold.

8. A Boost circuit based on adaptive weight adjustment, characterized in that, include: The Boost circuit module is used to implement DC voltage boost conversion; A multi-channel sampling module, the input of which is connected to the Boost circuit module, is used to acquire the actual values ​​of multiple circuit parameters of the Boost circuit module in real time; A control module, which is connected to the output terminal of the multi-channel sampling module and the control terminal of the Boost circuit module, is configured to execute the control method as described in any one of claims 1 to 7 to generate a control signal to adjust the switching transistor of the Boost circuit module.

9. An electronic device, characterized in that, include: A memory, and one or more processors communicatively connected to the memory; The memory stores instructions that can be executed by the one or more processors to cause the one or more processors to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1 to 7.

Images