Embedded multi-mode power management system based on intelligent switching strategy

Abstract

The invention discloses an embedded multi-mode power supply management system based on an intelligent switching strategy, and relates to the technical field of embedded equipment, and the system comprises a power supply input sampling module which is used for receiving a DC power supply from a battery end and carrying out the continuous sampling of an input voltage and an input current, so as to represent a current power supply state and a load working condition; the power supply conversion output module is used for selecting a target path in a DCDC power supply conversion path and an LDO power supply conversion path to execute voltage conversion under the constraint of sampling data output by the power supply input sampling module and a control instruction given by the intelligent control switching module, S1, a power supply input sampling step; the power input sampling module synchronously samples the voltage and the current of the battery input end in a preset sampling period, and the embedded equipment management circuit can adaptively select DCDC and LDO working modes in a full-load range and perform time sequence fine management on the switching process, so that the energy efficiency, the noise and the switching stability are considered.

Description

Technical Field

[0001] This invention relates to the field of embedded device technology, specifically to an embedded multi-mode power management system based on an intelligent switching strategy. Background Technology

[0002] DC-DC (Direct-to-DC converter) and LDO (Low Dropout Linear Regulator) are widely used in wearable devices, industrial sensors, medical electronics and other fields.

[0003] Existing embedded power management circuits face the following main problems in practical applications: First, when using a single power conversion structure, the conversion efficiency is significantly low under both light and heavy load conditions, making it difficult to improve the overall energy efficiency across the entire load range and limiting the battery life of embedded devices. Second, when the system uses a combination of DC-DC converters and LDOs to cover different load conditions, multiple power modules are prone to significant voltage oscillations and transient offsets during switching, which may lead to device reset or output signal distortion in extreme cases, affecting system stability. Third, common fixed threshold switching strategies mainly rely on a single current or voltage threshold for judgment, lacking comprehensive analysis of load change trends, resulting in delayed switching response, frequent jitter, and difficulty in balancing speed and stability in the switching window. Fourth, existing solutions generally lack the ability of embedded intelligent algorithms to predict and adaptively optimize load conditions, relying on manual static configuration for parameter configuration and switching logic, making it difficult to dynamically optimize the DC-DC operating frequency, LDO voltage regulation accuracy, and switching criteria based on actual operating data, thus limiting the overall intelligence level and energy efficiency potential of the system. Summary of the Invention

[0004] The purpose of this invention is to provide an embedded multi-mode power management system based on an intelligent switching strategy to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: an embedded multi-mode power management system based on intelligent switching strategy, including a power input sampling module, used to receive DC power from the battery and continuously sample the input voltage and input current to characterize the current power state and load conditions. The power conversion output module is used to select the target path in the DC-DC power conversion path and the LDO power conversion path to perform voltage conversion under the constraints of the sampling data output by the power input sampling module and the control command given by the intelligent control switching module, and output a stable DC power supply to the embedded device. The intelligent control switching module receives sampling data from the power input sampling module, runs a preset intelligent algorithm to identify and predict the load conditions, and generates mode selection commands and switching timing control signals for the power conversion output module, thereby realizing the coordinated management of the DC-DC power conversion path and the LDO power conversion path.

[0006] According to the above technical solution, the power input sampling module includes a battery input unit, a current sampling circuit, a voltage sampling circuit, and an analog-to-digital conversion and data buffering unit. The battery input unit is used to connect the DC power output from the battery to the input node of the management circuit. The current sampling circuit is used to convert the input current into a sampling signal corresponding to the magnitude of the input current. The voltage sampling circuit is used to convert the input voltage into a sampling signal corresponding to the amplitude of the input voltage. The analog-to-digital conversion and data buffering unit is used to perform analog-to-digital conversion on the current sampling signal and the voltage sampling signal and buffer the output within a preset time window.

[0007] The power conversion output module includes a DC-DC power conversion unit, an LDO power conversion unit, a DC-DC input control circuit, an LDO input control circuit, an output switching circuit, and an output filtering circuit. The DC-DC power conversion unit is used to achieve high-efficiency voltage conversion under heavy load conditions using a switching conversion method. The LDO power conversion unit is used to achieve low-noise voltage conversion under light load conditions using a linear regulation method. The DC-DC input control circuit controls the on / off state of the input terminal of the DC-DC power conversion unit based on the control signal output by the intelligent control switching module. The LDO input control circuit controls the on / off state of the input terminal of the LDO power conversion unit based on the control signal output by the intelligent control switching module. The output switching circuit selects a power supply path between the output terminals of the DC-DC power conversion unit and the LDO power conversion unit to connect to the load output terminal. The output filtering circuit filters and suppresses ripple on the voltage output from the selected power supply path to obtain a stable output voltage.

[0008] The intelligent control switching module includes an MCU microcontroller unit, a memory unit, a load status identification unit, a switching logic control unit, and a timing coordination unit. The MCU microcontroller unit is used as the core computing and control core to execute the instruction set. The memory unit is used to store sampled data, historical load records, and intelligent algorithm parameters. The load status identification unit is used to identify the current load size and load change trend based on voltage and current sampled data. The switching logic control unit is used to generate power path selection decisions based on the load status identification results and output them to the DC-DC input control circuit and the LDO input control circuit. The timing coordination unit is used to coordinate the switching timing of the DC-DC power conversion unit, the LDO power conversion unit, and the output switching circuit during power path switching to reduce voltage oscillations during the switching process.

[0009] According to the above technical solution, the working method of the system includes: S1, Power input sampling step: The power input sampling module synchronously samples the voltage and current at the battery input terminal within a preset sampling period and sends the sampling results to the intelligent control switching module to obtain the basic measurement data of the current load condition. S2, Efficiency Assessment and Mode Determination Step: The intelligent control switching module assesses the expected conversion efficiency of the DC-DC power conversion path and the LDO power conversion path under the current operating conditions based on the sampling data of step S1, and determines the preferred power conversion path and generates the power mode determination result by combining the load range and preset strategy. S3, power conversion module control steps: the intelligent control switching module controls the conduction state of the DC-DC input control circuit and the LDO input control circuit according to the mode determination result of step S2, so that the target power conversion unit in the power conversion output module enters the working state, and the unselected power conversion unit is in the off or standby state. S4, Output switching and filtering steps: Under the timing control of the intelligent control switching module, the output switching circuit connects the output terminal of the target power conversion unit to the load output terminal. The output filtering circuit filters the output voltage after connection to reduce ripple and noise, thereby providing a stable DC power supply to the embedded device. S5, Adaptive Optimization Step: During the repeated execution of the above steps, the intelligent control switching module continuously records the sampled data, mode determination results, and switching effects. Based on historical records, it updates the load prediction model and switching threshold, thereby achieving adaptive optimization of the DC-DC and LDO switching strategies and related parameters in subsequent operations.

[0010] According to the above technical solution, S1 specifically includes: S1-1, within the sampling period Ts, the power input sampling module obtains multiple sample values ​​Uin(k) of the voltage at the battery input terminal and multiple sample values ​​Iin(k) of the current at the battery input terminal at fixed time intervals, where k is the sampling sequence number, Ts is the sampling period, Uin(k) represents the input voltage sampled in the kth time, and Iin(k) represents the input current sampled in the kth time. S1-2, the power input sampling module processes the sequence Uin(k) and the sequence Iin(k) using a preset smoothing algorithm to obtain the average input voltage Uinavg and the average input current Iinavg for efficiency estimation, thereby reducing the impact of sampling noise on subsequent judgment results; S1-3, the intelligent control switching module estimates the expected conversion efficiency of the DC-DC power conversion path and the LDO power conversion path based on Uinavg, Iinavg, the target output voltage Uout, and the estimated load current Iload. The conversion efficiency of the DC-DC path is denoted as ηDD, and the conversion efficiency of the LDO path is denoted as ηLD. Based on the typical efficiency range constraint that ηDD is within 70% to 95% and ηLD is within 50% to 90%, the higher value of ηDD and ηLD is selected as the preferred power conversion path under the current operating condition. Here, Uout is the target output voltage that the power conversion output module needs to provide, Iload is the load current generated by the embedded device, ηDD is the conversion efficiency of the DC-DC power conversion unit, and ηLD is the conversion efficiency of the LDO power conversion unit.

[0011] According to the above technical solution, S2 specifically includes: S2-1, The intelligent control switching module compares the priority power conversion path obtained in step S1 with the power conversion path currently in operation. When the two are consistent, the current operating mode remains unchanged. When the two are inconsistent, a power path switching request flag is generated. S2-2, the intelligent control switching module judges the switching request flag based on the preset upper limit switching threshold Ihigh and lower limit switching threshold Ilow of the load current and the time window Thys for suppressing frequent switching. When the load current is continuously higher than Ihigh within the time window Thys, the switching request flag is confirmed as a heavy load switching request and the target power conversion path is set to the DC-DC power conversion path. When the load current is continuously lower than Ilow within the time window Thys, the switching request flag is confirmed as a light load switching request and the target power conversion path is set to the LDO power conversion path. In other cases, the switching request flag is kept in a suppressed state. Here, Ihigh is the load current threshold that triggers the switching to the DC-DC path, Ilow is the load current threshold that triggers the switching to the LDO path, and Thys is the length of the time window used to avoid frequent back-and-forth switching. S2-3 After receiving the confirmed switching request, the intelligent control switching module updates the enable signals of the DC-DC input control circuit and the LDO input control circuit within a control cycle on the order of a single microsecond. This causes the power conversion unit corresponding to the target power conversion path to enter an effective working state, while the non-target power conversion unit enters a shutdown or low-power standby state. Moreover, this control cycle is shortened to approximately one-fifth to one-tenth of the original control cycle compared to the control cycle of the traditional fixed threshold switching strategy.

[0012] According to the above technical solution, S3 specifically includes: S3-1 After confirming the target power conversion path, the intelligent control switching module sends a switching warning signal to the output switching circuit and sets the overlap conduction time Δtoverlap between the old and new power conversion paths in the output switching circuit, so that the DC-DC power conversion unit and the LDO power conversion unit remain in the conducting state for a short time during the switching moment, thereby reducing the transient voltage change at the load end. S3-2, within the overlap conduction time Δtoverlap, the output switching circuit controls the new and old power supply conversion paths to be connected to the load output terminal through preset current limiting elements or slope limiting switches respectively, and the output filtering circuit filters the superimposed voltage, so that the voltage change rate during the switching process is constrained. S3-3, at the end of the overlap conduction time Δtoverlap, the output switching circuit completely disconnects the original power conversion path and only retains the connection of the target power conversion path. The transient offset amplitude of the load output voltage during the entire switching process is denoted as ΔUswitch. Through the above overlap conduction and filtering control, ΔUswitch is controlled within a range of approximately ±2 millivolts, which is significantly reduced compared to the switching voltage offset of approximately ±50 millivolts in the traditional fixed threshold switching scheme. Here, Δtoverlap is the time interval between the simultaneous conduction of the old and new power conversion paths, and ΔUswitch is the transient offset amplitude of the output voltage during the switching process.

[0013] According to the above technical solution, S4 specifically includes: S4-1, the output filter circuit consists of a filter inductor unit and a filter capacitor unit. The filter inductor unit is connected in series between the output switching circuit and the load output terminal, and the filter capacitor unit is connected in parallel across the load output terminal. The combination of inductor and capacitor achieves the attenuation of high-frequency switching components. S4-2, the inductance and capacitance values ​​of the output filter circuit are configured according to the target output voltage, maximum load current and allowable output ripple index, so that the switching ripple when the DC-DC power conversion unit is working can be fully attenuated by the filter network, and the linear regulation ripple when the LDO power conversion unit is working is further suppressed. S4-3, the design goal of the output filter circuit is to ensure that the peak-to-peak value of the output voltage ripple measured on the load side is denoted as ΔUripple, and the value of ΔUripple is no greater than five millivolts, thereby ensuring that the analog circuits and high-precision digital modules of the embedded device obtain a stable power reference when working. Here, ΔUripple is the peak-to-peak value of the output voltage ripple at the load output terminal after filtering.

[0014] According to the above technical solution, S5 specifically includes: S5-1, the intelligent control switching module will compile the corresponding input voltage, input current, output voltage, load current and the currently used power conversion path into a historical record after each significant change in load current and each power path switching, and store it in the memory unit to continuously form a historical load dataset containing multiple records. S5-2, the intelligent control switching module selects the most recent N historical records from the historical load dataset, and performs statistical analysis on the energy consumption and switching times when using the DC-DC power conversion path and the LDO power conversion path according to the load change trajectory, and constructs a load prediction model to characterize the load change trend, where N is the number of historical records participating in the modeling. In subsequent operation, the intelligent control switching module matches the real-time sampled load change sequence with the load prediction model. When the predicted load is about to enter a long-term heavy load condition, it pre-selects the DC-DC power conversion path and appropriately increases the target efficiency operating point of the DC-DC power conversion unit. When the predicted load is about to enter a long-term light load condition, it pre-selects the LDO power conversion path and appropriately reduces the static loss of the LDO power conversion unit. As a result, the average energy efficiency across the entire load range is improved by about 15% to 30% compared to the conventional solution without a load prediction model. At the same time, the number of switching operations under the same load conditions is significantly reduced, and the adaptive optimization capability is improved by no less than 40%.

[0015] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: The present invention utilizes the high efficiency of DC-DC power converters under heavy load conditions and the low noise and low static power consumption characteristics of LDO power conversion modules under light load conditions. Through the power input sampling module, voltage and current data are collected in real time, and the intelligent control switching module runs load prediction algorithms and adaptive switching strategies to optimize the working path in different load ranges. This makes the overall energy efficiency of the system in the full load range about 15% to 30% higher than that of using only a single power supply solution, significantly extending the battery life of battery-powered embedded devices.

[0016] Under the timing control of the intelligent control switching module, the power conversion output module smoothly switches between the DC-DC and LDO paths. By introducing brief overlapping conduction and output filtering before and after switching, the voltage oscillation amplitude during multi-power module switching can be compressed from approximately ±50 millivolts in traditional solutions to approximately ±2 millivolts, reducing the risk of device reset caused by switching and improving output signal quality. The intelligent control switching module avoids using a simple fixed threshold strategy. With the help of the MCU's internal load prediction and adaptive optimization algorithms, it completes load status identification and power path switching within microseconds. The switching response speed is about five to ten times faster than traditional solutions. At the same time, during long-term operation, it automatically adjusts the switching threshold, time window, and target parameters based on historical load records, ensuring that the operating parameters of the DC-DC and LDO continuously match the actual load requirements. The adaptive optimization capability is improved by no less than 40%, thereby comprehensively improving the intelligence and stability of embedded device power management. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is a schematic diagram of the overall modular structure of the present invention; Figure 2 This is a flowchart illustrating the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Please see Figure 1 and Figure 2 The present invention provides a technical solution: an embedded multi-mode power management system based on intelligent switching strategy, including a power input sampling module for receiving DC power from the battery and continuously sampling the input voltage and input current to characterize the current power state and load conditions. The power conversion output module is used to select the target path in the DC-DC power conversion path and the LDO power conversion path to perform voltage conversion under the constraints of the sampling data output by the power input sampling module and the control command given by the intelligent control switching module, and output a stable DC power supply to the embedded device. The intelligent control switching module is used to receive the sampling data from the power input sampling module, run a preset intelligent algorithm to identify and predict the load conditions, and generate mode selection instructions and switching timing control signals for the power conversion output module, thereby realizing the coordinated management of the DC-DC power conversion path and the LDO power conversion path. The power input sampling module includes a battery input unit, a current sampling circuit, a voltage sampling circuit, and an analog-to-digital converter and data buffer unit. The battery input unit is used to connect the DC power output from the battery to the input node of the management circuit. The current sampling circuit is used to convert the input current into a sampling signal corresponding to the magnitude of the input current. The voltage sampling circuit is used to convert the input voltage into a sampling signal corresponding to the amplitude of the input voltage. The analog-to-digital converter and data buffer unit is used to perform analog-to-digital conversion on the current sampling signal and the voltage sampling signal and buffer the output within a preset time window.

[0020] The power conversion output module includes a DC-DC power conversion unit, an LDO power conversion unit, a DC-DC input control circuit, an LDO input control circuit, an output switching circuit, and an output filtering circuit. The DC-DC power conversion unit achieves high-efficiency voltage conversion under heavy load conditions using a switching conversion method. The LDO power conversion unit achieves low-noise voltage conversion under light load conditions using a linear regulation method. The DC-DC input control circuit controls the on / off state of the DC-DC power conversion unit's input terminal based on the control signal output from the intelligent control switching module. The LDO input control circuit controls the on / off state of the LDO power conversion unit's input terminal based on the control signal output from the intelligent control switching module. The output switching circuit selects a power supply path between the output terminals of the DC-DC and LDO power conversion units and connects it to the load output terminal. The output filtering circuit filters and suppresses ripple on the voltage output from the selected power supply path to obtain a stable output voltage.

[0021] The intelligent control switching module includes an MCU microcontroller unit, a memory unit, a load status identification unit, a switching logic control unit, and a timing coordination unit. The MCU microcontroller unit is used as the core computing and control core to execute the instruction set. The memory unit is used to store sampled data, historical load records, and intelligent algorithm parameters. The load status identification unit is used to identify the current load size and load change trend based on voltage and current sampled data. The switching logic control unit is used to generate power path selection decisions based on the load status identification results and output them to the DC-DC input control circuit and the LDO input control circuit. The timing coordination unit is used to coordinate the switching timing of the DC-DC power conversion unit, the LDO power conversion unit, and the output switching circuit during power path switching to reduce voltage oscillations during the switching process. The working method of this system includes: S1, Power input sampling step: The power input sampling module synchronously samples the voltage and current at the battery input terminal within a preset sampling period and sends the sampling results to the intelligent control switching module to obtain the basic measurement data of the current load condition. S2, Efficiency Assessment and Mode Determination Step: The intelligent control switching module assesses the expected conversion efficiency of the DC-DC power conversion path and the LDO power conversion path under the current operating conditions based on the sampling data of step S1, and determines the preferred power conversion path and generates the power mode determination result by combining the load range and preset strategy. S3, power conversion module control steps: the intelligent control switching module controls the conduction state of the DC-DC input control circuit and the LDO input control circuit according to the mode determination result of step S2, so that the target power conversion unit in the power conversion output module enters the working state, and the unselected power conversion unit is in the off or standby state. S4, Output switching and filtering steps: Under the timing control of the intelligent control switching module, the output switching circuit connects the output terminal of the target power conversion unit to the load output terminal. The output filtering circuit filters the output voltage after connection to reduce ripple and noise, thereby providing a stable DC power supply to the embedded device. S5, Adaptive optimization step: During the repeated operation of the above steps, the intelligent control switching module continuously records the sampled data, mode determination results and switching effects, and updates the load prediction model and switching threshold based on historical records, thereby achieving adaptive optimization of the DC-DC and LDO switching strategies and related parameters in subsequent operations. S1 specifically includes: S1-1, within the sampling period Ts, the power input sampling module obtains multiple sample values ​​Uin(k) of the voltage at the battery input terminal and multiple sample values ​​Iin(k) of the current at the battery input terminal at fixed time intervals, where k is the sampling sequence number, Ts is the sampling period, Uin(k) represents the input voltage sampled in the kth time, and Iin(k) represents the input current sampled in the kth time. S1-2, the power input sampling module processes the sequence Uin(k) and the sequence Iin(k) using a preset smoothing algorithm to obtain the average input voltage Uinavg and the average input current Iinavg for efficiency estimation, thereby reducing the impact of sampling noise on subsequent judgment results; S1-3, the intelligent control switching module estimates the expected conversion efficiency of the DC-DC power conversion path and the LDO power conversion path based on Uinavg, Iinavg, the target output voltage Uout, and the estimated load current Iload. The conversion efficiency of the DC-DC path is denoted as ηDD, and the conversion efficiency of the LDO path is denoted as ηLD. Based on the typical efficiency range constraint that ηDD is in the range of 70% to 95% and ηLD is in the range of 50% to 90%, the higher value of ηDD and ηLD is selected as the preferred power conversion path under the current operating condition. Uout is the target output voltage that the power conversion output module needs to provide, Iload is the load current generated by the embedded device, ηDD is the conversion efficiency of the DC-DC power conversion unit, and ηLD is the conversion efficiency of the LDO power conversion unit. S2 specifically includes: S2-1, The intelligent control switching module compares the priority power conversion path obtained in step S1 with the power conversion path currently in operation. When the two are consistent, the current operating mode remains unchanged. When the two are inconsistent, a power path switching request flag is generated. S2-2, the intelligent control switching module judges the switching request flag based on the preset upper limit switching threshold Ihigh and lower limit switching threshold Ilow of the load current and the time window Thys for suppressing frequent switching. When the load current is continuously higher than Ihigh within the time window Thys, the switching request flag is confirmed as a heavy load switching request and the target power conversion path is set to the DC-DC power conversion path. When the load current is continuously lower than Ilow within the time window Thys, the switching request flag is confirmed as a light load switching request and the target power conversion path is set to the LDO power conversion path. In other cases, the switching request flag is kept in a suppressed state. Here, Ihigh is the load current threshold that triggers the switching to the DC-DC path, Ilow is the load current threshold that triggers the switching to the LDO path, and Thys is the length of the time window used to avoid frequent back-and-forth switching. S2-3, after receiving the confirmed switching request, the intelligent control switching module updates the enable signals of the DC-DC input control circuit and the LDO input control circuit within a control cycle on the order of a single microsecond, so that the power conversion unit corresponding to the target power conversion path enters the effective working state, and the non-target power conversion unit enters the off or low-power standby state. Moreover, this control cycle is shortened to about one-fifth to one-tenth of the original control cycle compared with the control cycle of the traditional fixed threshold switching strategy. S3 specifically includes: S3-1 After confirming the target power conversion path, the intelligent control switching module sends a switching warning signal to the output switching circuit and sets the overlap conduction time Δtoverlap between the old and new power conversion paths in the output switching circuit, so that the DC-DC power conversion unit and the LDO power conversion unit remain in the conducting state for a short time during the switching moment, thereby reducing the transient voltage change at the load end. S3-2, within the overlap conduction time Δtoverlap, the output switching circuit controls the new and old power supply conversion paths to be connected to the load output terminal through preset current limiting elements or slope limiting switches respectively, and the output filtering circuit filters the superimposed voltage, so that the voltage change rate during the switching process is constrained. S3-3, at the end of the overlap conduction time Δtoverlap, the output switching circuit completely disconnects the original power conversion path and only retains the connection of the target power conversion path. The transient offset amplitude of the load output voltage during the entire switching process is denoted as ΔUswitch. Through the above overlap conduction and filtering control, ΔUswitch is controlled within a range of approximately ±2 millivolts, which is significantly reduced compared to the approximately ±50 millivolts switching voltage offset in the traditional fixed threshold switching scheme. Here, Δtoverlap is the time interval between the simultaneous conduction of the old and new power conversion paths, and ΔUswitch is the transient offset amplitude of the output voltage during the switching process. S4 specifically includes: S4-1, the output filter circuit consists of a filter inductor unit and a filter capacitor unit. The filter inductor unit is connected in series between the output switching circuit and the load output terminal, and the filter capacitor unit is connected in parallel across the load output terminal. The combination of inductor and capacitor achieves the attenuation of high-frequency switching components. S4-2, the inductance and capacitance values ​​of the output filter circuit are configured according to the target output voltage, maximum load current and allowable output ripple index, so that the switching ripple when the DC-DC power conversion unit is working can be fully attenuated by the filter network, and the linear regulation ripple when the LDO power conversion unit is working is further suppressed. S4-3, the design goal of the output filter circuit is to ensure that the peak-to-peak value of the output voltage ripple measured on the load side is denoted as ΔUripple, and the value of ΔUripple is no greater than five millivolts, thereby ensuring that the analog circuits and high-precision digital modules of the embedded device obtain a stable power reference when working. Here, ΔUripple is the peak-to-peak value of the output voltage ripple at the load output terminal after filtering. S5 specifically includes: S5-1, the intelligent control switching module will compile the corresponding input voltage, input current, output voltage, load current and the currently used power conversion path into a historical record after each significant change in load current and each power path switching, and store it in the memory unit to continuously form a historical load dataset containing multiple records. S5-2, the intelligent control switching module selects the most recent N historical records from the historical load dataset, and performs statistical analysis on the energy consumption and switching times when using the DC-DC power conversion path and the LDO power conversion path according to the load change trajectory, and constructs a load prediction model to characterize the load change trend, where N is the number of historical records participating in the modeling. In subsequent operation, the intelligent control switching module matches the real-time sampled load change sequence with the load prediction model. When the predicted load is about to enter a long-term heavy load condition, it pre-selects the DC-DC power conversion path and appropriately increases the target efficiency operating point of the DC-DC power conversion unit. When the predicted load is about to enter a long-term light load condition, it pre-selects the LDO power conversion path and appropriately reduces the static loss of the LDO power conversion unit. As a result, the average energy efficiency across the entire load range is improved by about 15% to 30% compared to the conventional solution without a load prediction model. At the same time, the number of switching operations under the same load conditions is significantly reduced, and the adaptive optimization capability is improved by no less than 40%.

[0022] A DC-DC power converter module handles high-load power conversion, leveraging its high-efficiency switching mode (typical efficiency 70%-95%). Combined with an LDO power converter module for low-load scenarios, it utilizes the advantages of linear regulation and low power consumption, solving the problem of low efficiency of a single power supply under light or full load. This achieves a 15%-30% improvement in overall system energy efficiency across the entire load range, effectively extending the battery life of embedded devices. The power switching control circuit precisely manages the input source on / off and power module switching. Through hardware logic and timing control, it significantly reduces voltage oscillation amplitude during multi-power module switching (e.g., reducing the traditional switching oscillation voltage from ±50mV to within ±2mV), avoiding problems such as device restarts and signal distortion caused by oscillation, and improving system power supply stability.

[0023] Abandoning fixed threshold switching strategies, this solution leverages the MCU's built-in intelligent algorithms to dynamically analyze load conditions and adjust switching logic in real time, resolving response latency issues. Compared to traditional solutions, switching response speed is improved by 5-10 times, achieving load change identification and power switching within microseconds (μs), meeting the high-speed dynamic load requirements of embedded devices. The MCU integrates intelligent algorithms that learn and analyze historical load data to achieve load prediction. For example, by predicting load change trends, it can switch power modules in advance while adaptively optimizing the operating parameters of DC-DC converters and LDOs (such as switching frequency and voltage regulation accuracy), enabling the circuit to automatically match load demands. Compared to conventional circuits, adaptive optimization capabilities are improved by over 40%, comprehensively enhancing the intelligence level of power management in embedded devices.

[0024] This patented circuit integrates multiple modules working collaboratively: using a battery as the power input, the sampling circuit collects the battery's voltage and current data in real time and transmits it to the MCU; the MCU analyzes the data to determine the load type. If it is a large load, the control circuit activates the DC-DC module; if it is a small load, the control circuit activates the LDO module; subsequently, the output switching circuit selects the output path of the DC-DC module or the LDO module under the MCU's command. After the filtering circuit filters out ripple and noise, it provides a stable power supply to the electrical equipment. Throughout the process, the MCU continuously and dynamically adjusts the power supply based on the sampled data, achieving efficient and intelligent power management for the electrical equipment.

[0025] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0026] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An embedded multi-mode power management system based on an intelligent switching strategy, characterized in that: include: The power input sampling module is used to receive DC power from the battery and continuously sample the input voltage and input current to characterize the current power state and load conditions. The power conversion output module is used to select the target path in the DC-DC power conversion path and the LDO power conversion path to perform voltage conversion under the constraints of the sampling data output by the power input sampling module and the control command given by the intelligent control switching module, and output a stable DC power supply to the embedded device. The intelligent control switching module receives sampling data from the power input sampling module, runs a preset intelligent algorithm to identify and predict the load conditions, and generates mode selection commands and switching timing control signals for the power conversion output module, thereby realizing the coordinated management of the DC-DC power conversion path and the LDO power conversion path.

2. The embedded multi-mode power management system based on intelligent switching strategy according to claim 1, characterized in that: The power input sampling module includes a battery input unit, a current sampling circuit, a voltage sampling circuit, and an analog-to-digital conversion and data buffering unit. The battery input unit is used to connect the DC power output from the battery to the input node of the management circuit. The current sampling circuit is used to convert the input current into a sampling signal corresponding to the magnitude of the input current. The voltage sampling circuit is used to convert the input voltage into a sampling signal corresponding to the amplitude of the input voltage. The analog-to-digital conversion and data buffering unit is used to perform analog-to-digital conversion on the current sampling signal and the voltage sampling signal and buffer the output within a preset time window. The power conversion output module includes a DC-DC power conversion unit, an LDO power conversion unit, a DC-DC input control circuit, an LDO input control circuit, an output switching circuit, and an output filtering circuit. The DC-DC power conversion unit is used to achieve high-efficiency voltage conversion based on switching conversion under heavy load conditions. The LDO power conversion unit is used to achieve low-noise voltage conversion based on linear regulation under light load conditions. The DC-DC input control circuit controls the on / off state of the input terminal of the DC-DC power conversion unit according to the control signal output by the intelligent control switching module. The LDO input control circuit controls the on / off state of the input terminal of the LDO power conversion unit according to the control signal output by the intelligent control switching module. The output switching circuit selects a power supply path between the output terminals of the DC-DC power conversion unit and the LDO power conversion unit to connect to the load output terminal. The output filtering circuit filters and suppresses ripple on the voltage output from the selected power supply path to obtain a stable output voltage. The intelligent control switching module includes an MCU microcontroller unit, a memory unit, a load status identification unit, a switching logic control unit, and a timing coordination unit. The MCU microcontroller unit is used as the core computing and control core to execute the instruction set. The memory unit is used to store sampled data, historical load records, and intelligent algorithm parameters. The load status identification unit is used to identify the current load size and load change trend based on voltage and current sampled data. The switching logic control unit is used to generate power path selection decisions based on the load status identification results and output them to the DC-DC input control circuit and the LDO input control circuit. The timing coordination unit is used to coordinate the switching timing of the DC-DC power conversion unit, the LDO power conversion unit, and the output switching circuit during power path switching to reduce voltage oscillations during the switching process.

3. An embedded multi-mode power management system based on an intelligent switching strategy according to claim 2, characterized in that: The working method of this system includes: S1, Power input sampling step: The power input sampling module synchronously samples the voltage and current at the battery input terminal within a preset sampling period and sends the sampling results to the intelligent control switching module to obtain the basic measurement data of the current load condition. S2, Efficiency Assessment and Mode Determination Step: The intelligent control switching module assesses the expected conversion efficiency of the DC-DC power conversion path and the LDO power conversion path under the current operating conditions based on the sampling data of step S1, and determines the preferred power conversion path and generates the power mode determination result by combining the load range and preset strategy. S3, power conversion module control steps: the intelligent control switching module controls the conduction state of the DC-DC input control circuit and the LDO input control circuit according to the mode determination result of step S2, so that the target power conversion unit in the power conversion output module enters the working state, and the unselected power conversion unit is in the off or standby state. S4, Output switching and filtering steps: Under the timing control of the intelligent control switching module, the output switching circuit connects the output terminal of the target power conversion unit to the load output terminal. The output filtering circuit filters the output voltage after connection to reduce ripple and noise, thereby providing a stable DC power supply to the embedded device. S5, Adaptive Optimization Step: During the repeated execution of the above steps, the intelligent control switching module continuously records the sampled data, mode determination results, and switching effects. Based on historical records, it updates the load prediction model and switching threshold, thereby achieving adaptive optimization of the DC-DC and LDO switching strategies and related parameters in subsequent operations.

4. An embedded multi-mode power management system based on an intelligent switching strategy according to claim 3, characterized in that: S1 specifically includes: S1-1, within the sampling period Ts, the power input sampling module obtains multiple sample values ​​Uin(k) of the voltage at the battery input terminal and multiple sample values ​​Iin(k) of the current at the battery input terminal at fixed time intervals, where k is the sampling sequence number, Ts is the sampling period, Uin(k) represents the input voltage sampled in the kth time, and Iin(k) represents the input current sampled in the kth time. S1-2, the power input sampling module processes the sequence Uin(k) and the sequence Iin(k) using a preset smoothing algorithm to obtain the average input voltage Uinavg and the average input current Iinavg for efficiency estimation, thereby reducing the impact of sampling noise on subsequent judgment results; S1-3, the intelligent control switching module estimates the expected conversion efficiency of the DC-DC power conversion path and the LDO power conversion path based on Uinavg, Iinavg, the target output voltage Uout, and the estimated load current Iload. The conversion efficiency of the DC-DC path is denoted as ηDD, and the conversion efficiency of the LDO path is denoted as ηLD. Based on the typical efficiency range constraint that ηDD is within 70% to 95% and ηLD is within 50% to 90%, the higher value of ηDD and ηLD is selected as the preferred power conversion path under the current operating condition. Here, Uout is the target output voltage that the power conversion output module needs to provide, Iload is the load current generated by the embedded device, ηDD is the conversion efficiency of the DC-DC power conversion unit, and ηLD is the conversion efficiency of the LDO power conversion unit.

5. An embedded multi-mode power management system based on an intelligent switching strategy according to claim 4, characterized in that: S2 specifically includes: S2-1, The intelligent control switching module compares the priority power conversion path obtained in step S1 with the power conversion path currently in operation. When the two are consistent, the current operating mode remains unchanged. When the two are inconsistent, a power path switching request flag is generated. S2-2, the intelligent control switching module judges the switching request flag based on the preset upper limit switching threshold Ihigh and lower limit switching threshold Ilow of the load current and the time window Thys for suppressing frequent switching. When the load current is continuously higher than Ihigh within the time window Thys, the switching request flag is confirmed as a heavy load switching request and the target power conversion path is set to the DC-DC power conversion path. When the load current is continuously lower than Ilow within the time window Thys, the switching request flag is confirmed as a light load switching request and the target power conversion path is set to the LDO power conversion path. In other cases, the switching request flag is kept in a suppressed state. Here, Ihigh is the load current threshold that triggers the switching to the DC-DC path, Ilow is the load current threshold that triggers the switching to the LDO path, and Thys is the length of the time window used to avoid frequent back-and-forth switching. S2-3 After receiving the confirmed switching request, the intelligent control switching module updates the enable signals of the DC-DC input control circuit and the LDO input control circuit within a control cycle on the order of a single microsecond. This causes the power conversion unit corresponding to the target power conversion path to enter an effective working state, while the non-target power conversion unit enters a shutdown or low-power standby state. Moreover, this control cycle is shortened to approximately one-fifth to one-tenth of the original control cycle compared to the control cycle of the traditional fixed threshold switching strategy.

6. An embedded multi-mode power management system based on an intelligent switching strategy according to claim 5, characterized in that: S3 specifically includes: S3-1 After confirming the target power conversion path, the intelligent control switching module sends a switching warning signal to the output switching circuit and sets the overlap conduction time Δtoverlap between the old and new power conversion paths in the output switching circuit, so that the DC-DC power conversion unit and the LDO power conversion unit remain in the conducting state for a short time during the switching moment, thereby reducing the transient voltage change at the load end. S3-2, within the overlap conduction time Δtoverlap, the output switching circuit controls the new and old power supply conversion paths to be connected to the load output terminal through preset current limiting elements or slope limiting switches respectively, and the output filtering circuit filters the superimposed voltage, so that the voltage change rate during the switching process is constrained. S3-3, at the end of the overlap conduction time Δtoverlap, the output switching circuit completely disconnects the original power conversion path and only retains the connection of the target power conversion path. The transient offset amplitude of the load output voltage during the entire switching process is denoted as ΔUswitch. Through the above overlap conduction and filtering control, ΔUswitch is controlled within a range of approximately ±2 millivolts, which is significantly reduced compared to the switching voltage offset of approximately ±50 millivolts in the traditional fixed threshold switching scheme. Here, Δtoverlap is the time interval between the simultaneous conduction of the old and new power conversion paths, and ΔUswitch is the transient offset amplitude of the output voltage during the switching process.

7. An embedded multi-mode power management system based on an intelligent switching strategy according to claim 6, characterized in that: S4 specifically includes: S4-1, the output filter circuit consists of a filter inductor unit and a filter capacitor unit. The filter inductor unit is connected in series between the output switching circuit and the load output terminal, and the filter capacitor unit is connected in parallel across the load output terminal. The combination of inductor and capacitor achieves the attenuation of high-frequency switching components. S4-2, the inductance and capacitance values ​​of the output filter circuit are configured according to the target output voltage, maximum load current and allowable output ripple index, so that the switching ripple when the DC-DC power conversion unit is working can be fully attenuated by the filter network, and the linear regulation ripple when the LDO power conversion unit is working is further suppressed. S4-3, the design goal of the output filter circuit is to ensure that the peak-to-peak value of the output voltage ripple measured on the load side is denoted as ΔUripple, and the value of ΔUripple is no greater than five millivolts, thereby ensuring that the analog circuits and high-precision digital modules of the embedded device obtain a stable power reference when working. Here, ΔUripple is the peak-to-peak value of the output voltage ripple at the load output terminal after filtering.

8. An embedded multi-mode power management system based on an intelligent switching strategy according to claim 7, characterized in that: S5 specifically includes: S5-1, the intelligent control switching module will compile the corresponding input voltage, input current, output voltage, load current and the currently used power conversion path into a historical record after each significant change in load current and each power path switching, and store it in the memory unit to continuously form a historical load dataset containing multiple records. S5-2, the intelligent control switching module selects the most recent N historical records from the historical load dataset, and performs statistical analysis on the energy consumption and switching times when using the DC-DC power conversion path and the LDO power conversion path according to the load change trajectory, and constructs a load prediction model to characterize the load change trend, where N is the number of historical records participating in the modeling. In subsequent operation, the intelligent control switching module matches the real-time sampled load change sequence with the load prediction model. When the predicted load is about to enter a long-term heavy load condition, it pre-selects the DC-DC power conversion path and appropriately increases the target efficiency operating point of the DC-DC power conversion unit. When the predicted load is about to enter a long-term light load condition, it pre-selects the LDO power conversion path and appropriately reduces the static loss of the LDO power conversion unit. As a result, the average energy efficiency across the entire load range is improved by about 15% to 30% compared to the conventional solution without a load prediction model. At the same time, the number of switching operations under the same load conditions is significantly reduced, and the adaptive optimization capability is improved by no less than 40%.

Images