A power management adaptive feedback regulation method and system with wide voltage input range

Abstract

The application provides a power management adaptive feedback regulation method and system with a wide voltage input range. In the application, the output current ripple waveform of the power supply is collected, wavelet packet decomposition is performed to extract the energy distribution of the characteristic frequency band, and the dynamic offset characteristics of the energy distribution with the wide voltage input fluctuation are used to identify the load change mode. The electromagnetic vibration caused by the current ripple is captured and converted into recovered electrical energy, which is directly supplied to the feedback circuit to realize energy self-supply. The recovered electrical energy is introduced into the ripple injection inverting proportional circuit, the gain coefficient and phase offset are dynamically set according to the load mode, and the inverting ripple signal is generated. The inverting ripple signal is superimposed on the feedback voltage to offset the ripple fluctuation by dynamically tracking the ripple amplitude change caused by the wide voltage input. Finally, the response speed of the rectifier tube driving timing is adjusted according to the input voltage fluctuation amplitude, and the optimized feedback control signal is generated. The application realizes the load fluctuation compensation and energy recovery closed loop.

Description

Technical Field

[0001] This application relates to the fields of power management and power electronic control technology, and in particular to a power management adaptive feedback regulation method and system with a wide voltage input range. Background Technology

[0002] Portable electronic devices need to be compatible with wide-voltage input environments such as solar panels, vehicle power supplies, and unstable power grids. These devices are often under light load conditions, but traditional power management systems experience a significant drop in conversion efficiency under light loads, and wide fluctuations in input voltage can easily cause dynamic changes in ripple amplitude. The core requirement is to achieve wide-voltage adaptive regulation while efficiently recovering load ripple energy to supply the feedback circuit, thereby reducing static power consumption and improving light-load energy efficiency.

[0003] Currently, mainstream solutions adopt a cloud-based collaborative static threshold control architecture: Local voltage monitoring devices collect operating voltage data from the equipment, upload it to a cloud server to generate a standard reference voltage, and preset a fixed threshold as a safety range. The local control module adjusts the switching frequency of the Buck-Boost circuit based on this threshold. When the input voltage exceeds the threshold, the synchronous rectifier is triggered to turn off to avoid overvoltage damage; under light load, it switches to a low-power mode to reduce static losses.

[0004] However, while existing solutions can achieve basic compatibility with wide voltage input, their core flaw lies in their failure to effectively resolve the contradiction between energy efficiency and dynamic response under light load scenarios. Specifically, the ripple energy recovery efficiency is extremely low. Due to the mismatch between the frequency band characteristics of the RC absorption circuit and the low-frequency energy of the load ripple, most of the electromagnetic vibration energy is converted into heat energy and dissipated, failing to supply the feedback circuit to reduce static power consumption. Summary of the Invention

[0005] This application provides a power management adaptive feedback regulation method and system with a wide voltage input range to solve the problem of feedback loop response delay in the prior art.

[0006] In a first aspect, this application provides a power management adaptive feedback regulation method with a wide voltage input range, comprising:

[0007] Acquire the current ripple waveform of the power supply output, perform wavelet packet decomposition on the current ripple waveform, and extract the energy distribution of the characteristic frequency band. Based on the dynamic offset characteristics of the energy distribution of the characteristic frequency band fluctuating with a wide voltage input range, identify the dynamic change pattern of the load.

[0008] By using a piezoelectric cantilever beam array positioned at the edge of the printed circuit board, the electromagnetic vibrations caused by the current ripple waveform are captured, the electromagnetic vibrations are converted into recovered electrical energy, and the recovered electrical energy is directly supplied to the feedback circuit to provide adaptive power supply energy for the feedback circuit.

[0009] The ripple of the recovered electrical energy is injected into the inverting proportional circuit. The gain coefficient and phase offset are dynamically set according to the load dynamic change mode to generate an inverting ripple signal.

[0010] The system dynamically tracks the ripple amplitude changes caused by a wide voltage input range and superimposes the inverted ripple signal onto the feedback voltage to dynamically cancel the ripple amplitude changes, generating a feedback control signal after ripple cancellation.

[0011] The feedback control signal is input into the adaptive ripple lockout loop. Based on the ripple amplitude change caused by the wide voltage input range, the response speed of the rectifier drive timing is adjusted so that the rectifier drive timing matches the trough of the output current ripple. The response speed is positively correlated with the input voltage fluctuation amplitude. At high voltage, the matching is accelerated to suppress losses, and at low voltage, the speed is reduced to avoid false triggering.

[0012] Optionally, the current ripple waveform of the power supply output is acquired, wavelet packet decomposition is performed on the current ripple waveform, and the energy distribution of the characteristic frequency band is extracted. Based on the dynamic offset characteristics of the energy distribution of the characteristic frequency band fluctuating with a wide voltage input range, the dynamic change pattern of the load is identified, including:

[0013] Acquire the current ripple waveform at the power supply output terminal, perform wavelet packet decomposition on the current ripple waveform, and generate multiple sub-frequency band signals;

[0014] Extract at least one preset target sub-frequency band, calculate the average energy value of the target sub-frequency band within a continuous time window, and arrange the average energy values ​​of all time windows in a time series to form a characteristic frequency band energy distribution;

[0015] Under wide voltage input range fluctuations, based on the energy distribution of the characteristic frequency band, the dynamic offset characteristics of the target sub-frequency band are detected, and the dynamic load change pattern is identified according to the dynamic offset characteristics.

[0016] If the average energy value of the high-frequency sub-band increases more than that of the low-frequency sub-band, the energy distribution shifts to higher frequencies, which is determined to be a dynamic load increase mode. If the average energy value of the low-frequency sub-band decreases less than that of the high-frequency sub-band, the energy distribution shifts to lower frequencies, which is determined to be a dynamic load decrease mode.

[0017] Optionally, dynamically tracking the ripple amplitude change caused by a wide voltage input range and superimposing the inverted ripple signal onto the feedback voltage to dynamically cancel the ripple amplitude change, generating a ripple-canceled feedback control signal, including:

[0018] The system dynamically tracks the ripple amplitude changes caused by a wide voltage input range at the power input terminal and calculates the instantaneous increment of the current ripple amplitude based on the ripple amplitude changes.

[0019] The inverted ripple signal is input to the first input terminal of the adder, and the feedback voltage signal is input to the second input terminal of the adder. When the instantaneous increment of the ripple amplitude exceeds a preset threshold, the adder is immediately triggered to perform a signal superposition operation, so that the inverted ripple signal and the feedback voltage signal are dynamically synthesized to cancel the change in ripple amplitude and generate a feedback control signal after ripple cancellation.

[0020] Optionally, when the instantaneous increment of the ripple amplitude exceeds a preset threshold, the adder is immediately triggered to perform a signal superposition operation, so that the inverted ripple signal and the feedback voltage signal are dynamically synthesized to cancel the change in ripple amplitude and generate a feedback control signal after ripple cancellation, including:

[0021] When the instantaneous increment of the ripple amplitude exceeds a preset threshold, the triggering unit immediately sends a superposition enable signal to the signal synthesis unit. The signal synthesis unit responds to the superposition enable signal and superimposes the inverted ripple signal of the first input port and the feedback voltage signal of the second input port.

[0022] According to the load dynamic change mode, the resistance ratio on the inverted ripple signal path is adjusted by a digitally controlled potentiometer so that the amplitude of the inverted ripple signal matches the ripple amplitude in the feedback voltage signal, thereby canceling the ripple amplitude change and generating a feedback control signal after ripple cancellation.

[0023] Optionally, the feedback control signal is input to an adaptive ripple lockout loop. Based on the ripple amplitude variation caused by a wide voltage input range, the response speed of the rectifier drive timing is adjusted so that the rectifier drive timing matches the trough of the output current ripple. The response speed is positively correlated with the input voltage fluctuation amplitude; it accelerates matching and suppresses losses at high voltages and slows down to avoid false triggering at low voltages. This includes:

[0024] When the input voltage fluctuation amplitude continuously exceeds the preset range for three ripple cycles, the feedback control signal is input into the adaptive ripple lockout loop.

[0025] Based on the ripple amplitude variation caused by the wide voltage input range, the response speed of the rectifier drive timing is dynamically adjusted. If the input voltage is higher than the rated value, the hysteresis voltage of the comparator is reduced to accelerate the rectifier drive timing matching and suppress switching losses. If the input voltage is lower than the rated value, the hysteresis voltage of the comparator is increased to reduce the response speed to avoid noise-induced false triggering.

[0026] The comparator detects the falling edge slope of the output current ripple. When the falling edge slope approaches zero, the rectifier drive signal is triggered so that the rectifier drive timing matches the trough of the output current ripple.

[0027] Optionally, by using a piezoelectric cantilever beam array disposed at the edge of the printed circuit board, the electromagnetic vibration caused by the current ripple waveform is captured, the electromagnetic vibration is converted into recovered electrical energy, and the recovered electrical energy is directly supplied to the feedback circuit to provide adaptive power supply energy for the feedback circuit, including:

[0028] A ring-shaped array of piezoelectric cantilever beams is set at the edge of the printed circuit board, and each cantilever beam contains symmetrically distributed displacement sensing piezoelectric elements and energy conversion piezoelectric elements.

[0029] The displacement state of the cantilever beam is detected by the displacement sensing piezoelectric element, a displacement detection signal is generated and input to the synchronous switch control circuit, and the synchronous switch control circuit processes the displacement detection signal to generate a trigger pulse that is synchronized with the maximum displacement of the cantilever beam.

[0030] The energy conversion piezoelectric element directly captures the electromagnetic vibration caused by the induced current ripple and outputs alternating charge. The alternating charge is input into the synchronous switching inductor circuit. The digital switch is closed when the displacement of the cantilever beam reaches its maximum value through the trigger pulse, so that the inductor and the energy conversion piezoelectric element form an oscillation circuit to obtain a high-frequency pulse current.

[0031] The high-frequency pulse current is converted into DC power through a full-bridge rectifier circuit, and the DC power is directly connected to the power input terminal of the feedback circuit as the recovered electrical energy, providing adaptive power supply energy for the feedback circuit.

[0032] Optionally, the ripple from the recovered electrical energy is injected into the inverting proportional circuit. Based on the load dynamic change mode, the gain coefficient and phase offset are dynamically set to generate the inverting ripple signal, including:

[0033] The input ripple of the DC power is injected into the inverting amplifier circuit, in which the inverting input terminal of the operational amplifier is connected to the positive terminal of the DC power, and the non-inverting input terminal is grounded.

[0034] According to the load dynamic change mode switching resistor network, if the load dynamic change mode is the load dynamic increase mode, a low resistance feedback resistor is connected to set a high gain coefficient; if the load dynamic change mode is the load dynamic decrease mode, a high resistance feedback resistor is connected to set a low gain coefficient.

[0035] An adjustable delay line is connected in series at the output of the operational amplifier, and the delay time is proportional to the rate of change of the input voltage, thereby generating an inverted ripple signal with a phase offset.

[0036] Secondly, this application provides a power management adaptive feedback regulation system with a wide voltage input range, comprising:

[0037] The identification module is used to acquire the current ripple waveform of the power supply output, perform wavelet packet decomposition on the current ripple waveform, extract the energy distribution of the characteristic frequency band, and identify the dynamic change pattern of the load based on the dynamic offset characteristics of the energy distribution of the characteristic frequency band fluctuating with a wide voltage input range.

[0038] The recycling module is used to capture the electromagnetic vibration caused by the current ripple waveform through a piezoelectric cantilever beam array set on the edge of the printed circuit board, convert the electromagnetic vibration into recycled electrical energy, and directly supply the recycled electrical energy to the feedback circuit to provide adaptive power supply energy for the feedback circuit.

[0039] The generation module is used to inject the ripple of the recovered electrical energy into the inverting proportional circuit, and dynamically set the gain coefficient and phase offset according to the load dynamic change mode to generate an inverting ripple signal.

[0040] The cancellation module is used to dynamically track the ripple amplitude change caused by a wide voltage input range, and superimpose the inverted ripple signal onto the feedback voltage to dynamically cancel the ripple amplitude change and generate a feedback control signal after ripple cancellation.

[0041] The adjustment module is used to input the feedback control signal into the adaptive ripple lockout loop, and adjust the response speed of the rectifier drive timing according to the ripple amplitude change caused by the wide voltage input range, so that the rectifier drive timing matches the trough of the output current ripple. The response speed is positively correlated with the input voltage fluctuation amplitude. At high voltage, the matching is accelerated to suppress loss, and at low voltage, the speed is reduced to avoid false triggering.

[0042] Thirdly, this application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement a power management adaptive feedback regulation method with a wide voltage input range as described in the first aspect above.

[0043] Fourthly, this application provides a computer storage medium storing a computer program, which, when executed by a computer, implements a power management adaptive feedback regulation method with a wide voltage input range as described in the first aspect.

[0044] This application acquires the current ripple waveform of the power supply output and performs wavelet packet decomposition. By extracting the dynamic offset characteristics of the energy distribution in the characteristic frequency band with wide voltage fluctuations, it achieves accurate identification of the dynamic change pattern of the load, providing accurate load status information for subsequent adaptive control. A piezoelectric cantilever beam array captures the electromagnetic vibrations caused by the current ripple and converts them into recovered electrical energy, directly supplying it to the feedback circuit. This not only reduces dependence on external power supply but also utilizes the originally dissipated ripple energy to improve the overall energy efficiency of the system, making it particularly suitable for energy recovery in light-load scenarios. After the recovered electrical energy is introduced into the inverting proportional circuit, the gain coefficient and phase offset are dynamically set according to the dynamic change pattern of the load to generate an inverted ripple signal. This provides a precise phase matching basis for ripple cancellation, avoiding compensation deviations caused by traditional fixed parameters. The application dynamically tracks the ripple amplitude changes caused by wide voltage inputs and superimposes the inverted ripple signal onto the feedback voltage to cancel ripple fluctuations, significantly suppressing output voltage overshoot or oscillation and improving power supply stability. After the feedback control signal is input to the adaptive ripple lockout loop, the response speed of the rectifier drive timing is adjusted according to the input voltage fluctuation amplitude to accurately match the trough of the output current ripple, thereby reducing switching losses and avoiding false triggering, and resolving the contradiction between dynamic response and energy efficiency.

[0045] Furthermore, by performing wavelet packet decomposition on the current ripple waveform to generate multiple sub-band signals, the complex time-domain ripple characteristics are transformed into frequency-domain sub-bands, providing a data foundation for refined analysis of load characteristics. The average energy value of the preset target sub-bands is extracted and arranged in a time series to form the energy distribution of characteristic frequency bands, quantifying the migration pattern of energy in different frequency bands over time and revealing the frequency-domain characterization of load changes. Based on the dynamic offset characteristics of the characteristic frequency band energy distribution under wide voltage, load patterns are identified: when the energy increase of the high-frequency sub-band exceeds that of the low-frequency sub-band, the load is determined to be dynamically increasing; when the attenuation of the low-frequency sub-band is less than that of the high-frequency sub-band, the load is determined to be dynamically decreasing. This achieves quantitative classification of load states, providing a decision-making basis for control strategies.

[0046] These or other aspects of this application will become more apparent in the following description of the embodiments. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 A flowchart of a power management adaptive feedback regulation method with a wide voltage input range provided in this application is shown;

[0049] Figure 2 A scenario diagram is shown for a power management adaptive feedback regulation method with a wide voltage input range provided in this application;

[0050] Figure 3 A schematic diagram of the structure of a power management adaptive feedback regulation system with a wide voltage input range provided in this application is shown;

[0051] Figure 4 A schematic diagram of the structure of a computing device provided in this application is shown. Detailed Implementation

[0052] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0053] In some of the processes described in the specification, claims, and accompanying drawings of this application, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a chronological order, nor do they limit "first" and "second" to different types.

[0054] Researchers have discovered a significant trade-off between energy efficiency and dynamic response under light load scenarios in power management systems with wide input voltage ranges. While existing solutions can achieve basic adaptation to wide input voltage ranges, their core flaw lies in their inability to effectively address energy efficiency issues under light load scenarios. Specifically, ripple energy recovery efficiency is extremely low because the frequency band characteristics of the RC absorption circuit do not match the low-frequency energy of the load ripple, causing most of the electromagnetic vibration energy to be converted into heat dissipation, failing to supply the feedback circuit to reduce static power consumption. Furthermore, the ripple amplitude dynamically changes with input voltage fluctuations, further exacerbating energy loss and dynamic response lag. Therefore, a power management method that can efficiently recover energy, adaptively cancel ripple, and optimize dynamic response is urgently needed to improve system efficiency and stability under light load scenarios.

[0055] To address the aforementioned issues, this invention proposes a power management adaptive feedback regulation method with a wide voltage input range. Its core lies in integrating ripple feature recognition, piezoelectric energy recovery, and dynamic parameter adjustment mechanisms to achieve closed-loop adaptive control. Specifically, by acquiring the current ripple waveform of the power supply output, wavelet packet decomposition is performed to extract the energy distribution of characteristic frequency bands. Based on the dynamic offset characteristics of the energy distribution with wide voltage fluctuations, the dynamic change pattern of the load is identified. Simultaneously, an array of piezoelectric cantilever beams set at the edge of the printed circuit board is used to capture the electromagnetic vibration caused by the ripple waveform, directly converting the vibration energy into recovered electrical energy and supplying it to the feedback circuit, providing adaptive power supply energy for the feedback circuit. Subsequently, the recovered electrical energy is introduced into the ripple injection inverting proportional circuit. According to the dynamic change pattern of the load, the gain coefficient and phase offset are dynamically set to generate an inverted ripple signal. Further, the ripple amplitude change caused by the wide voltage input range is dynamically tracked, and the inverted ripple signal is superimposed on the feedback voltage to cancel the ripple amplitude fluctuation, generating a feedback control signal after ripple cancellation. Finally, the feedback control signal is input into the adaptive ripple lockout loop, and the response speed of the rectifier drive timing is adjusted according to the input voltage fluctuation amplitude, so that the drive timing matches the trough of the output current ripple, achieving accelerated matching at high voltage to suppress losses and deceleration at low voltage to avoid false triggering. This method effectively addresses the shortcomings of the prior art: by efficiently recovering low-frequency electromagnetic vibration energy through a piezoelectric cantilever beam array, heat dissipation is avoided and the energy is directly supplied to the feedback circuit, significantly reducing static power consumption; at the same time, the dynamic adjustment mechanism based on ripple characteristics ensures high energy efficiency and response matching under light load scenarios, eliminating the contradiction between energy efficiency and dynamic response, and improving the overall stability and efficiency of the system.

[0056] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0057] Figure 1 A flowchart illustrating a power management adaptive feedback regulation method with a wide voltage input range is provided in this application embodiment, as shown below. Figure 1 As shown, the method includes:

[0058] 101. Acquire the current ripple waveform of the power supply output, perform wavelet packet decomposition on the current ripple waveform, and extract the energy distribution of the characteristic frequency band. Based on the dynamic offset characteristics of the energy distribution of the characteristic frequency band fluctuating with a wide voltage input range, identify the dynamic change pattern of the load.

[0059] Optionally, step 101 may specifically include the following steps:

[0060] 1011. Acquire the current ripple waveform at the power supply output terminal, and perform wavelet packet decomposition on the current ripple waveform to generate multiple sub-frequency band signals;

[0061] 1012. Extract at least one preset target sub-frequency band, calculate the average energy value of the target sub-frequency band within a continuous time window, and arrange the average energy values ​​of all time windows in a time series to form a characteristic frequency band energy distribution;

[0062] 1013. Under wide voltage input range fluctuations, based on the energy distribution of the characteristic frequency band, detect the dynamic offset characteristics of the target sub-frequency band, and identify the dynamic load change pattern according to the dynamic offset characteristics;

[0063] 1014. If the average energy value of the high-frequency sub-band increases more than that of the low-frequency sub-band, the energy distribution shifts to the high frequency, which is determined to be a dynamic load increase mode. If the average energy value of the low-frequency sub-band decreases less than that of the high-frequency sub-band, the energy distribution shifts to the low frequency, which is determined to be a dynamic load decrease mode.

[0064] In the above steps, current ripple waveform refers to the fluctuating component in the current signal at the power supply output terminal, typically reflecting minute vibrations caused by load changes; wavelet packet decomposition is a signal decomposition technique that uses mathematical transformation to break down a signal into different frequency ranges; sub-band signals are multiple signal segments generated after wavelet packet decomposition, each segment representing a specific frequency interval; target sub-band refers to one or more preset frequency intervals of interest, such as high or low frequency ranges; average energy value is the average value of the signal intensity change within a certain frequency band, reflecting the activity level of the signal in that band; characteristic frequency band energy distribution is a sequence of average energy values ​​arranged in chronological order, showing the trend of energy change over time; wide voltage input range fluctuation refers to the input voltage value changing over a large range, such as dynamic adjustment from 90V to 240V; dynamic offset characteristics describe the pattern of characteristic frequency band energy distribution shifting towards higher or lower frequencies when voltage fluctuates; load dynamic change pattern refers to identifying the trend of increasing or decreasing load demand, where high-frequency sub-bands correspond to higher frequency signals, and low-frequency sub-bands correspond to lower frequency signals.

[0065] In this embodiment, firstly, through step 1011, the current ripple waveform at the power supply output terminal is acquired, and wavelet packet decomposition is performed on the waveform to generate multiple sub-frequency band signals. The specific process includes: firstly, using a current sensor to measure the power supply output current and obtaining a current signal sequence that changes over time, which is represented as the amplitude value at discrete time points, and the sampling frequency can be set according to the application, such as 1000Hz; secondly, the signal is processed by applying a wavelet packet decomposition algorithm, which is based on a pre-selected wavelet basis function and a decomposition level, such as using the Daubechies4 wavelet basis function and setting the decomposition level to 3 levels. By iteratively calculating the wavelet coefficients of the signal, the original signal is divided into multiple uniform sub-frequency band signals in multiple frequency ranges, such as a low-frequency sub-band of 0 to 500Hz, a mid-frequency sub-band of 500 to 1000Hz, and a high-frequency sub-band of 1000 to 1500Hz after decomposition; finally, these sub-frequency band signals are output as the basis for subsequent processing. For example, in a power supply system, a 5-millisecond current ripple waveform is acquired when the input voltage is stable, with 500 sampling points. Wavelet packet decomposition is applied, selecting the db4 wavelet basis and decomposing it into 3 layers to generate 4 sub-band signals: sub-band 1 is 0 to 250Hz, representing low-frequency components; sub-band 2 is 250 to 500Hz; sub-band 3 is 500 to 750Hz; and sub-band 4 is 750 to 1000Hz, representing high-frequency components. In this way, the length of each sub-band signal is reduced, but the characteristics of the original signal at specific frequencies are preserved, facilitating subsequent analysis.

[0066] Secondly, through step 1012, at least one preset target sub-frequency band is extracted, the average energy value of these target sub-frequency bands within a continuous time window is calculated, and the average energy values ​​of all time windows are arranged in a time series to form a characteristic frequency band energy distribution. The specific process includes: First, selecting the target sub-frequency band range according to preset rules or application requirements, for example, specifying the high-frequency range 750 to 1000 Hz as the first target sub-frequency band and the low-frequency range 0 to 250 Hz as the second target sub-frequency band; Second, dividing the continuous time window, with each window having a fixed size, for example, a 50-millisecond window containing 50 sampling points; Then, within each window, calculating the average energy value of the target sub-frequency band signal, the average energy value is calculated using a formula. Where E is the average energy value in joules (J), s(k) is the signal amplitude value at the k-th sampling point within the window, and M is the number of sampling points within the window. The average value is obtained by summing the squares of the amplitudes of all sampling points and then dividing by M. This formula reflects the intensity change of the signal in this frequency band and window. Finally, the calculation results of all windows are arranged in chronological order to form a time series, which represents the trend of energy distribution over time, i.e., the energy distribution of the characteristic frequency band. For example, in a specific application, the target sub-frequency band is preset to a high-frequency sub-band of 750 to 1000 Hz, and the continuous time window size is set to 10 windows, each window being 50 milliseconds. In the first window, the sum of squares of the sub-frequency band signal amplitude is 0.5, the number of sampling points M is equal to 50, and the average energy value E is calculated as 0.5 ÷ 50 = 0.01 J. Subsequent windows are calculated sequentially: Window 1 energy 0.01 J, Window 2 energy 0.012 J, Window 3 energy 0.015 J. The 10 window values ​​are arranged in time series as the sequence 0.01, 0.012, 0.03 J, forming the characteristic frequency band energy distribution sequence of the high-frequency sub-frequency band, which is used for subsequent detection.

[0067] Next, through step 1013, under wide voltage input range fluctuations, based on the characteristic frequency band energy distribution, the dynamic offset characteristics of the target sub-band are detected, and based on these dynamic offset characteristics, the dynamic change pattern of the load is preliminarily identified. The specific process includes: First, under an environment of input voltage variation, for example, when the voltage gradually fluctuates from 100V to 220V, the characteristic frequency band energy distribution sequence of the target sub-band is monitored; then, the dynamic offset characteristics are detected, that is, the change law of the energy sequence of multiple target sub-bands is calculated, and the relative differences or trends between energy values ​​at different times are compared, for example, the change value of the slope of the high-frequency sub-band energy sequence is calculated, or the change of the high-low frequency energy ratio over time is analyzed; finally, based on the offset characteristics, if the overall migration direction of the energy distribution is found, it is preliminarily determined whether the load is changing; however, the specific pattern determination details are completed in step 1014. For example, during a 10-second fluctuation of a wide voltage input increasing from 100V to 220V, the high-frequency sub-band energy distribution sequence is 0.01J, 0.015J, 0.02J (increasing trend), while the low-frequency sub-band sequence is 0.05J, 0.06J, 0.065J (slowly increasing). When detecting dynamic offset characteristics, comparing the changes in values ​​in the third window with those in the first window and calculating the slope or difference reveals that the high-frequency energy increases faster, initially indicating a dynamic increase in load. This step, through simple analysis of the offset trend, provides a foundation for subsequent accurate identification.

[0068] Finally, through step 1014, the dynamic load change pattern is identified in detail based on the dynamic offset characteristics. Specifically, if the average energy value increase of the high-frequency sub-band exceeds that of the low-frequency sub-band, the energy distribution migrates to higher frequencies, which is determined to be a dynamic load increase pattern; if the average energy value decrease of the low-frequency sub-band is less than that of the high-frequency sub-band, the energy distribution migrates to lower frequencies, which is determined to be a dynamic load decrease pattern. The specific process includes: first, calculating the increase or decrease magnitude, where the magnitude is defined as the increase rate. or attenuation amplitude The increase represents the percentage increase in energy, and the decrease represents the percentage decrease. For the new window's energy value, The first step is to determine the energy value of the old window. The second step is to compare the amplification or attenuation of different sub-bands. For example, if the high-frequency amplification is greater than the low-frequency amplification, then energy migrates to higher frequencies. Finally, based on the comparison results, the mode is directly determined. For example, during voltage fluctuations, two continuous time windows are taken: the first window has a high-frequency energy of 1.0 J and a low-frequency energy of 2.0 J; the second window has a high-frequency energy of 1.5 J and a low-frequency energy of 2.1 J; the high-frequency growth amplitude is calculated. Calculate the low-frequency growth rate: Due to high-frequency amplification Exceeding the low-frequency increase The energy distribution is determined to shift towards higher frequencies, indicating a dynamic load increase mode. Similarly, if the low-frequency energy decays to 1.8J and the high-frequency energy decays to 0.6J in the second window, the low-frequency decay amplitude is calculated as follows: The high-frequency attenuation amplitude is Low-frequency attenuation amplitude Less than the high-frequency attenuation amplitude The system determines a shift to lower frequencies, indicating a dynamic reduction in load. This determination clearly identifies the load change pattern.

[0069] In a practical application scenario for a portable device's battery management system with a wide voltage input, the system first acquires the current ripple waveform at the power output terminal at a sampling rate of 10kHz, obtaining 1000 data points each time. It then performs three-level wavelet packet decomposition using the db4 wavelet basis function, generating eight sub-frequency bands covering the 0-5kHz range. Subsequently, the 0-100Hz low-frequency band and the 1-5kHz high-frequency band are selected as target sub-frequency bands. The average energy value of each band is calculated using a 50ms time window, and 20 consecutive windows are recorded to form a dynamic energy distribution curve. When the input voltage fluctuates within a wide range of 80V to 240V, the energy distribution shift characteristics are monitored: if the energy growth rate in the high-frequency band significantly exceeds that in the low-frequency band with a ratio difference of 1.5 times, it is determined that energy is migrating to higher frequencies, identifying a dynamic load increase mode; if the energy attenuation in the low-frequency band is significantly less than that in the high-frequency band with a ratio factor below 0.8, it is determined that energy is migrating to lower frequencies, identifying a dynamic load decrease mode. In reduced load mode, a high-frequency ripple energy recovery mechanism is activated, feeding energy in the 1-5kHz frequency band back to the energy storage unit through the Buck-Boost topology of the bidirectional DC / DC converter. Simultaneously, the operating mode is switched to pulse charging and redundant modules are disabled, optimizing energy consumption under light load conditions. When the load increases, a high-current fast charging protocol is employed, combined with low-resistance conduction technology to reduce power loss. This solution significantly improves light-load operating efficiency under a wide voltage input environment, making it particularly suitable for scenarios such as fast-charging power banks and portable medical devices.

[0070] In the overall scheme of step 101 above, multiple sub-frequency band signals are generated by acquiring the current ripple waveform at the power supply output terminal and performing wavelet packet decomposition on the waveform; at least one preset target sub-frequency band is extracted, the average energy value of each target sub-frequency band within a continuous time window is calculated, and the average energy values ​​are arranged in a time sequence to form a characteristic frequency band energy distribution; under the condition of wide voltage input range fluctuation, the dynamic offset characteristics of the target sub-frequency band are detected based on the characteristic frequency band energy distribution, and the load dynamic change mode is identified based on the characteristics; specifically, when the growth rate of the average energy value of the high-frequency sub-frequency band exceeds that of the low-frequency sub-frequency band, the energy distribution migrates to the high frequency, which is determined to be a load dynamic increase mode; while when the decay rate of the average energy value of the low-frequency sub-frequency band is lower than that of the high-frequency sub-frequency band, the energy distribution migrates to the low frequency, which is determined to be a load dynamic decrease mode.

[0071] 102. By using a piezoelectric cantilever beam array set on the edge of the printed circuit board, the electromagnetic vibration caused by the current ripple waveform is captured, the electromagnetic vibration is converted into recovered electrical energy, and the recovered electrical energy is directly supplied to the feedback circuit to provide adaptive power supply energy for the feedback circuit.

[0072] Optionally, step 102 may specifically include the following steps:

[0073] 1021. A ring-shaped array of piezoelectric cantilever beams is set at the edge of the printed circuit board, each cantilever beam containing symmetrically distributed displacement sensing piezoelectric elements and energy conversion piezoelectric elements.

[0074] 1022. The displacement state of the cantilever beam is detected by the displacement sensing piezoelectric element, a displacement detection signal is generated and input to the synchronous switch control circuit, and the synchronous switch control circuit processes the displacement detection signal to generate a trigger pulse that is synchronized with the maximum displacement of the cantilever beam.

[0075] 1023. The electromagnetic vibration caused by the induced current ripple is directly captured by the energy conversion piezoelectric element, and alternating charge is output. The alternating charge is input into the synchronous switching inductor circuit. The digital switch is closed when the displacement of the cantilever beam reaches the maximum value by the trigger pulse, so that the inductor and the energy conversion piezoelectric element form an oscillation circuit to obtain a high-frequency pulse current.

[0076] 1024. The high-frequency pulse current is converted into DC power through a full-bridge rectifier circuit, and the DC power is directly connected to the power input terminal of the feedback circuit as the recovered electrical energy to provide adaptive power supply energy for the feedback circuit.

[0077] In this embodiment of the application, firstly, as Figure 2 As shown, step 1021 involves setting a ring-shaped array of piezoelectric cantilever beams along the edge of the printed circuit board. Specifically, this includes: designing a ring-shaped substrate (e.g., 10 cm in diameter), uniformly welding 6-8 cantilever beams, each beam being 3 cm long and 1 cm wide; and attaching displacement-sensing piezoelectric elements, such as lead zirconate titanate PZT-5H with a piezoelectric constant of 650 × 10⁻⁶, to the upper surface of the beams. - ¹² m / V, with an energy conversion piezoelectric element such as aluminum nitride (ScAlN) bonded to the lower surface, and an electromechanical coupling coefficient of 0.2. The symmetrical layout ensures multi-directional vibration capture; for example, in a 12V power supply system, when the current ripple frequency is 1 kHz, this array can cover a vibration range of ±5 mm, improving the magnetic field disturbance capture rate by 35%. Case study: A floating bridge monitoring system uses an 8-beam ring array. Phosphor bronze mass blocks (10 mm × 20 mm × 4 mm) are added to the free ends of the cantilever beams, reducing the natural frequency to 46 Hz, matching the environmental vibration frequency, and improving the measured vibration energy conversion efficiency.

[0078] Secondly, in step 1022, a displacement detection signal is generated by the displacement sensing piezoelectric element, which triggers the synchronous switch control circuit. Specifically, the beam bending displacement δ is converted into a voltage signal. (d is the piezoelectric constant, g is the elastic modulus 0.001 m) (The unit δ is meters). When δ = 2 millimeters, calculate... The signal is input to a comparator and compared with a preset threshold of 2 volts. If the threshold is exceeded, a pulse generator is triggered to generate a pulse 100 microseconds wide. Key processing: The rising edge of the signal is extracted using a differentiating circuit to ensure synchronization between the pulse and the maximum displacement value, with a synchronization error of <0.1 milliseconds. Example: In a pipeline vibration monitoring device, the displacement sensing signal is processed by a 1000x gain amplifier. When the cantilever beam bends by 1.5 mm, a 1.3V signal is output, with a trigger pulse delay of only 50 microseconds, precisely matching the vibration peak value.

[0079] Then, in step 1023, alternating charges are output from the piezoelectric element using energy conversion, and a high-frequency pulse current is generated through a synchronous switching inductor circuit. The core operation consists of two steps: charge generation: vibration generates alternating charges. Where k is the electromechanical coupling coefficient. It is a 15 nanofarad piezoelectric capacitor. This is the open-circuit voltage. For example, the measured voltage under 12V ripple. volt, calculation Coulomb; Oscillation Conversion: Trigger Pulse Closes MOSFET Switch Model IRF540, Inductor L120 millihenries and Forming an LC circuit, oscillation frequency At kilohertz, it converts charge into a high-frequency pulsed current with a peak value of 50 mA within 2 milliseconds. Example: In a campus bus energy recovery device, vehicle vibration generates an inertial force of 1,430 N, triggering an oscillation circuit that outputs 10 pulses with a total charge of 24 microcoulombs and a peak current of 60 mA.

[0080] Finally, step 1024 rectifies the high-frequency pulsed current into DC power, which is then directly supplied to the feedback circuit. The specific process is as follows: the high-frequency current passes through a full-bridge rectifier circuit, where four 1N4007 diodes form a bridge structure, outputting a pulsating DC voltage. ,in, Where R is the peak current and R is the load resistance. For example milliampere, Calculation at 1000 ohms 35.35 volts; after being filtered and regulated by a 100 microfarad capacitor, the output is a stable 1.8 volt DC, which is directly connected to the power supply of the feedback circuit, such as the TL081 operational amplifier. Pinout. Anti-charge cancellation design: Each cantilever beam output phase is independent, and the phase is unified through full-bridge rectification to avoid energy cancellation. Example: In a wireless sensor node power supply system, recovered electrical energy directly drives the feedback circuit, reducing power consumption and eliminating the need for energy storage batteries.

[0081] In a practical application scenario involving vibration energy recovery from industrial equipment, a ring-shaped array of piezoelectric cantilever beams is deployed along the edge of the printed circuit board on the equipment base. This array comprises 16 cantilever beam units, each containing one low-frequency (80-150 Hz) resonant beam and one high-frequency (200-500 Hz) resonant beam. Each cantilever beam integrates symmetrically arranged displacement-sensing piezoelectric elements and energy-conversion piezoelectric elements. The displacement-sensing element has a sensitivity of 0.5 volts per micrometer of deformation, and the energy-conversion element has a piezoelectric constant of 650 picocous per newton. When the equipment is running, current ripple induces electromagnetic vibration, exciting the cantilever beam to deform. Displacement sensing elements detect the beam's displacement state and output a signal to the synchronous switch control circuit when the displacement reaches its peak, generating a 1-millisecond trigger pulse. This pulse drives a digital switch to close, creating an oscillating circuit between a 10 millihenry inductor and an energy conversion piezoelectric element, converting alternating charge into a 1-10 kHz high-frequency pulse current. After being converted to 3.3V DC by a full-bridge rectifier circuit, this current is directly connected to the power input port of the equipment's feedback circuit, while a 220 microfarad energy storage capacitor buffers voltage fluctuations. Actual measurements show that under a 200Hz vibration environment with 0.5g acceleration, a single cantilever beam can output 2.5 microwatts of recovered energy. The entire array's power supply capacity can meet the 10-microampere static power consumption requirement of the feedback circuit, while maintaining stable operation of the feedback system within a range of input voltage fluctuations, significantly improving energy recovery efficiency and system robustness.

[0082] In the overall scheme of step 102 above, an annular piezoelectric cantilever beam array is arranged at the edge of the printed circuit board. The displacement state of the cantilever beam is detected by the symmetrically distributed displacement sensing piezoelectric elements. The displacement detection signal is generated and input to the synchronous switch control circuit for processing, generating a trigger pulse synchronized with the maximum displacement of the cantilever beam. At the same time, the electromagnetic vibration caused by the current ripple is captured by the energy conversion piezoelectric element and the alternating charge is output. The charge is input to the synchronous switch inductor circuit. The trigger pulse drives the digital switch to close at the moment of maximum displacement, so that the inductor and the piezoelectric element form an oscillation circuit to generate a high-frequency pulse current. The high-frequency pulse current is then converted into DC current by the full-bridge rectifier circuit, which is used as the power input terminal for the recovery of electrical energy directly injected into the feedback circuit, realizing the recovery and reuse of electromagnetic vibration energy and providing self-sustaining operating energy for adaptive drive of the feedback circuit.

[0083] 103. Inject the ripple of the recovered electrical energy into the inverting proportional circuit, and dynamically set the gain coefficient and phase offset according to the load dynamic change mode to generate an inverting ripple signal.

[0084] Optionally, step 103 may specifically include the following steps:

[0085] 1031. The input ripple of the DC power is injected into the inverting amplifier circuit, in which the inverting input terminal of the operational amplifier is connected to the positive terminal of the DC power, and the non-inverting input terminal is grounded.

[0086] 1032. According to the load dynamic change mode switching resistor network, if the load dynamic change mode is the load dynamic increase mode, then a low resistance feedback resistor is connected to set a high gain coefficient; if the load dynamic change mode is the load dynamic decrease mode, then a high resistance feedback resistor is connected to set a low gain coefficient.

[0087] 1033. An adjustable delay line is connected in series at the output of the operational amplifier, and the delay time is proportional to the rate of change of the input voltage to generate an inverted ripple signal with a phase offset.

[0088] In the above steps, the piezoelectric cantilever beam array refers to an elastic beam structure distributed in a ring along the edge of a printed circuit board, composed of multiple cantilever beams, which convert mechanical vibration into electric charge through the piezoelectric effect; the displacement sensing piezoelectric element is a piezoelectric material attached to the upper surface of the cantilever beam, used to detect the bending displacement of the beam and output a proportional voltage signal; the energy conversion piezoelectric element is a piezoelectric material attached to the lower surface of the cantilever beam, specifically used to convert vibration energy into alternating charge; the synchronous switch control circuit generates a trigger pulse synchronized with the vibration peak by processing the displacement signal; the synchronous switch inductor circuit closes the switch under the action of the trigger pulse, so that the inductor and the piezoelectric element form an LC oscillation circuit, converting the charge into a high-frequency pulse current; the high-frequency pulse current refers to the instantaneous large current generated by the LC oscillation; The full-bridge rectifier circuit consists of a bridge structure composed of four diodes, converting AC to DC. Adaptive power supply refers to the direct connection of recovered electrical energy to the power supply terminal of the feedback circuit, eliminating the need for energy storage and matching its power consumption requirements. The inverting amplifier circuit is a signal processing structure based on an operational amplifier, using the ratio of the feedback resistor to the input resistor to set the gain coefficient, making the output signal out of phase with the input signal. The gain coefficient refers to the amplification ratio of the output voltage to the input voltage, dynamically adjusted by the feedback resistor value. The phase offset refers to the time delay applied to the output signal through a delay line, used to match ripple frequency drift. The inverted ripple signal is the signal after inverting amplification and phase adjustment, used to cancel the original power supply ripple. Dynamic setting refers to switching circuit parameters according to load change patterns.

[0089] In this embodiment, firstly, step 1031 inputs the ripple signal (i.e., the small voltage fluctuation) from the power supply output into a circuit capable of generating an opposite waveform. The specific process is as follows: The positive terminal of the DC power supply is connected to the inverting input terminal (marked "-") of the operational amplifier via a resistor called input resistor R1. Simultaneously, the non-inverting input terminal (marked "+") of the operational amplifier is grounded via another resistor called balancing resistor R2. The resistance values ​​of these two resistors need to be matched to ensure stable circuit operation. When the ripple signal enters the operational amplifier through input resistor R1, due to the characteristics of the operational amplifier, the output terminal generates a signal with an amplitude proportionally amplified but opposite to the input ripple waveform. For example, if the input ripple is 100 millivolts, by setting R1 = 1 kiloohm and R2 = 10 kiloohms, the output terminal can generate a 1-volt inverted ripple signal with completely opposite phase.

[0090] Secondly, step 1032 dynamically switches the resistor network to adjust the gain coefficient based on load changes. This is achieved by monitoring changes in the load (i.e., the electrical equipment) and dynamically adjusting the circuit's amplification factor and gain coefficient. Specifically, there are two modes: when a sudden increase in load is detected, such as a server starting a high-power task, the circuit automatically switches to a low-resistance feedback resistor (e.g., Rf = 1 kΩ), raising the gain coefficient to -20 and enhancing ripple suppression. When the load decreases, such as when the device enters standby mode, the circuit switches to a high-resistance feedback resistor (e.g., Rf = 100 kΩ), reducing the gain coefficient to -0.1 to avoid overcompensation and signal distortion. Resistor switching is achieved through an electronic switch in a multiplexer, controlled by the load detection signal.

[0091] Finally, in step 1033, since the frequency of the power supply ripple fluctuates with load changes, simple phase inversion may not completely cancel the original ripple. Therefore, this step connects an adjustable delay line in series at the operational amplifier output. This is a component that controls the signal delay time, dynamically adjusting the delay time of the inverted ripple signal to precisely match the phase of the original ripple. The delay time τ is determined by the rate of change of the input ripple, and is calculated using the formula τ = k × |ΔU / Δt|, where k is a proportionality coefficient, typically taken as 0.1–0.2 milliseconds / volt. For example, when an electric vehicle is charging, if the input voltage changes by 5 volts within 10 milliseconds, setting k = 0.2 results in a delay time τ = 1 millisecond, improving the measured ripple cancellation rate.

[0092] In practical applications, a portable medical device power management system integrates piezoelectric energy recovery and dynamic voltage regulation technologies to achieve efficient energy consumption control. A specific implementation is as follows: Sixteen piezoelectric cantilever beam arrays are arranged in a ring around the edge of the device's printed circuit board. Each array includes one low-frequency beam with a resonant frequency of 80-150Hz and one high-frequency beam with a resonant frequency of 200-500Hz. Each cantilever beam integrates symmetrically distributed displacement-sensing piezoelectric elements (scandium-doped aluminum nitride material, sensitivity 0.5V / μm) and energy-conversion piezoelectric elements (PZT-5H material, piezoelectric constant 650pC / N). When the device is running, the current ripple excites electromagnetic vibration. The displacement-sensing elements detect beam deformation, and when the displacement reaches its peak, they output a signal to the synchronous switch control circuit, generating a 1ms trigger pulse to drive the MOSFET switch to close. This causes the 10mH inductor and the energy-conversion piezoelectric elements to form an LC oscillation circuit, converting alternating charge into a 1-10kHz high-frequency pulse current. This current is then converted into a full-bridge rectifier circuit composed of Schottky diodes. A 3.3V DC power supply is directly connected to the power input of the feedback circuit, while a 220μF energy storage capacitor buffers voltage fluctuations. Recovered energy is further injected into the inverting proportional circuit to reduce ripple. The resistor network is switched according to the load dynamic mode: a 10kΩ feedback resistor is connected when the load increases to achieve a gain coefficient of -3.5, and a 100kΩ resistor is switched when the load decreases to achieve a gain coefficient of -0.8. Simultaneously, a voltage-controlled delay unit is connected in series at the operational amplifier output, dynamically adjusting the delay time according to the rate of change of the input voltage. When the rate of change is ≥50mV / μs, a 200ns delay generates a 45° phase shift; when the rate of change is ≤10mV / μs, a 50ns delay generates a 10° phase shift, ultimately generating an inverted signal that precisely cancels out the original ripple. Under a 200Hz vibration environment with 0.5g acceleration, the system outputs 2.5μW of power from a single cantilever beam. The entire array can meet the 10μA-level static power consumption requirement of the feedback circuit and maintain stable power supply within a ±20% range of input voltage fluctuations, significantly improving light-load energy efficiency and system robustness.

[0093] In the overall scheme of step 103 above, the recovered electrical energy captured by the piezoelectric cantilever beam array is introduced into an inverting proportional circuit. The gain coefficient and phase offset are adaptively adjusted according to the dynamic load change pattern to achieve efficient conversion and cancellation of ripple energy. Specifically, a ripple signal is injected into the inverting input of the operational amplifier at the DC input terminal. The feedback resistor network is automatically switched based on the load dynamic identification result: when the load dynamically increases, a low-resistance resistor is connected to increase the gain coefficient; when the load dynamically decreases, a high-resistance resistor is switched to decrease the gain coefficient. Simultaneously, an adjustable delay line dynamically matched to the input voltage change rate is connected in series at the output terminal to compensate for the phase deviation caused by the electromechanical response lag of the piezoelectric material, ultimately generating a precisely synchronized inverted ripple signal. This design intelligently balances gain amplification and phase delay under wide voltage input fluctuations, effectively suppressing high-frequency oscillations in the energy recovery link, while ensuring amplitude matching and phase alignment between the inverted signal and the original ripple, significantly improving the energy conversion efficiency and output stability of the electromagnetic vibration recovery system.

[0094] 104. Dynamically track the ripple amplitude change caused by a wide voltage input range, and superimpose the inverted ripple signal onto the feedback voltage to dynamically cancel the ripple amplitude change and generate a feedback control signal after ripple cancellation.

[0095] Optionally, step 104 may specifically include the following steps:

[0096] 1041. Dynamically track the ripple amplitude change caused by a wide voltage input range at the power input terminal, and calculate the instantaneous increment of the current ripple amplitude based on the ripple amplitude change;

[0097] 1042. The inverted ripple signal is input to the first input terminal of the adder, and the feedback voltage signal is input to the second input terminal of the adder. When the instantaneous increment of the ripple amplitude exceeds a preset threshold, the adder is immediately triggered to perform a signal superposition operation, so that the inverted ripple signal and the feedback voltage signal are dynamically synthesized to cancel the change in ripple amplitude and generate a feedback control signal after ripple cancellation.

[0098] Specifically, step 1042 may include the following process: when the instantaneous increment of the ripple amplitude exceeds a preset threshold, the triggering unit immediately sends a superposition enable signal to the signal synthesis unit. The signal synthesis unit responds to the superposition enable signal and superimposes the inverted ripple signal of the first input port with the feedback voltage signal of the second input port. According to the load dynamic change mode, the resistance ratio on the inverted ripple signal path is adjusted by a digitally controlled potentiometer so that the amplitude of the inverted ripple signal matches the ripple amplitude in the feedback voltage signal, canceling the change in ripple amplitude and generating a feedback control signal after ripple cancellation.

[0099] In the above steps, ripple amplitude change refers to the change in the amplitude of the periodic fluctuation of the output voltage caused by fluctuations in the power supply input voltage. For example, when the input voltage changes, the output ripple changes from 5 millivolts to 20 millivolts. Instantaneous increment measures the amount of change in ripple amplitude over a short period of time; for example, the ripple increases by 15 millivolts within 10 milliseconds. The inverted ripple signal is a waveform signal generated by special circuitry, and its waveform is completely opposite to the original ripple. The feedback voltage signal is the actual voltage data collected from the power supply output terminal, which includes the ripple component generated when the power supply is operating. An adder is a component in electronic equipment that combines two signals into a single signal. A preset threshold is a pre-set critical value, such as 20 millivolts; when the ripple change exceeds this value, the cancellation function is activated. A digitally controlled potentiometer is an adjusting device that uses electronic signals to control the resistance, used to adjust the signal strength. The superposition enable signal is an electronic command that triggers the signal synthesis function.

[0100] In this embodiment, firstly, step 1041 monitors the change in the power input voltage and calculates the instantaneous ripple increment. Specifically, a high-precision detection element continuously measures the input voltage. When the input voltage changes between 85 volts and 265 volts, the instantaneous change in ripple is calculated using the formula: the instantaneous increment equals a proportional coefficient multiplied by the input voltage change rate, where the proportional coefficient ranges from 0.1 to 0.2, and the input voltage change rate is the voltage change per unit time. For example, if the input voltage of an electric vehicle charging station increases from 200 volts to 300 volts within 5 milliseconds, the input voltage change rate is calculated as 100 volts divided by 0.005 seconds, equaling 20,000 volts per second. With a proportional coefficient of 0.15, the instantaneous increment equals 0.15 multiplied by 20,000, equaling 3,000 millivolts. When this value exceeds a preset threshold of 500 millivolts, subsequent cancellation operations are automatically triggered. The entire process uses a high-speed data acquisition chip to collect voltage data millions of times per second, ensuring timely change detection.

[0101] Next, step 1042 achieves precise superposition of the inverted ripple signal and the feedback voltage. Specifically, when the instantaneous increment exceeds a threshold, the trigger circuit immediately sends a start signal to the adder control terminal. At this time, the inverted ripple signal from the preceding circuit is connected to the first channel of the adder, while the actual feedback voltage from the power supply output is connected to the second channel. The digitally controlled potentiometer adjusts the channel resistance value according to the current load state: if an increase in load is detected, the resistance value is adjusted to 1 kΩ to amplify the inverted signal by 20 times; if the load decreases, it is adjusted to 100 kΩ to reduce the signal by 0.1 times. Finally, the adder merges the two signals, eliminating the ripple component and generating a stable control signal. For example, in a computer server power supply system, when a sudden change in input voltage causes a 600 mV ripple increment, the digitally controlled potentiometer automatically adjusts the resistance value to 2 kΩ, ensuring the inverted signal amplitude precisely matches the original ripple. After superposition, the output voltage ripple decreases from 45 mV to 4 mV, achieving a ripple elimination rate of 91%, ensuring stable power supply operation.

[0102] In practical applications, specifically in an industrial-grade wide-voltage input power supply system, the system tracks ripple amplitude changes caused by input voltage fluctuations within the range of 80V to 240V. A high-speed differential amplifier captures the instantaneous ripple increment at a sampling rate of 1MSPS. When the detected ripple increment exceeds a 50mV threshold, the trigger unit issues a superimposed enable signal within 100 nanoseconds, driving the signal synthesis unit to respond immediately. Simultaneously, the digitally controlled potentiometer is adjusted according to the dynamic load change pattern: during a sudden load increase, the resistance ratio is adjusted to 0.2:1, boosting the gain of the inverting ripple signal path by a factor of 3; during a sudden load decrease, the resistance ratio is switched to 1.5:1, reducing the gain to a factor of 0.8, ensuring that the inverting ripple amplitude accurately matches the current ripple characteristics. In the signal synthesis unit, the phase-calibrated inverted ripple signal has its phase offset dynamically set according to the input voltage change rate: when the change rate exceeds 50mV / μs, a 200-nanosecond delay generates a 45-degree offset; when it is below 10mV / μs, a 50-nanosecond delay generates a 10-degree offset. This offset is then superimposed on the original feedback voltage through a current feedback operational amplifier. When the amplitude error is less than 5%, dynamic cancellation is achieved, generating a feedback control signal with a ripple peak-to-peak value of less than 15mV. This scheme reduces the feedback loop response time to 50 microseconds under input voltage fluctuations of ±20%, while simultaneously controlling the adjustment accuracy of the digitally controlled potentiometer to within 0.5%.

[0103] In the overall scheme of step 104 above, the ripple amplitude change caused by wide voltage input fluctuations is dynamically tracked, the instantaneous increment of the ripple is calculated, and a signal superposition operation is immediately triggered when it exceeds a preset threshold. The inverted ripple signal and the feedback voltage are accurately synthesized in the adder. At the same time, based on the dynamic load change mode, the resistance ratio of the inverted signal path is adaptively tuned by a digitally controlled potentiometer to ensure that the amplitude of the inverted ripple is precisely matched with the ripple component of the feedback voltage. This dynamically cancels the ripple energy fluctuations caused by the wide voltage input range, achieving high-precision ripple suppression of the feedback control signal. This mechanism breaks through the limitations of traditional static compensation. Under the condition of sudden input voltage changes, it can still maintain the amplitude synchronization and phase alignment of the inverted signal and the original ripple, effectively eliminating the superposition effect of high-frequency oscillation and low-frequency interference, significantly improving the anti-interference ability and dynamic response speed of the feedback system, while reducing the dependence on external filtering components. It provides a stable and reliable adaptive ripple cancellation solution for power control systems under complex operating conditions.

[0104] 105. Input the feedback control signal into the adaptive ripple lockout loop, and adjust the response speed of the rectifier drive timing according to the ripple amplitude change caused by the wide voltage input range, so that the rectifier drive timing matches the trough of the output current ripple. The response speed is positively correlated with the input voltage fluctuation amplitude. When the voltage is high, the matching is accelerated to suppress loss, and when the voltage is low, the speed is reduced to avoid false triggering.

[0105] Optionally, step 105 may specifically include the following steps:

[0106] 1051. When the input voltage fluctuation amplitude continuously exceeds the preset range for three ripple cycles, the feedback control signal is input into the adaptive ripple lockout loop.

[0107] 1052. Based on the ripple amplitude change caused by the wide voltage input range, dynamically adjust the response speed of the rectifier drive timing. If the input voltage is higher than the rated value, reduce the hysteresis voltage of the comparator to accelerate the matching of the rectifier drive timing and suppress switching losses. If the input voltage is lower than the rated value, increase the hysteresis voltage of the comparator to reduce the response speed to avoid noise-induced false triggering.

[0108] 1053. The falling edge slope of the output current ripple is detected by the comparator. When the falling edge slope approaches zero, the rectifier drive signal is triggered so that the rectifier drive timing matches the trough of the output current ripple.

[0109] In the above steps, the feedback control signal refers to the stable voltage signal used to control the power switch operation after ripple cancellation; the adaptive ripple lockout loop is a control loop that can automatically adjust the switching timing and optimize response performance according to changes in input voltage; the rectifier drive timing refers to the pulse signal sequence that controls the power rectifier switch to turn on / off; the response speed indicates how quickly the control signal reacts to changes in current ripple, measured in nanoseconds; the trough point is the lowest point of the current ripple waveform, corresponding to the minimum current value; the hysteresis voltage is the range of input voltage differences required for the comparator trigger state to flip, used for anti-interference; the falling edge slope refers to the rate at which the current ripple changes from its peak to its trough (measured in amperes per microsecond); the preset range is the rated safe range of the input voltage, such as ±15% of 220V AC, i.e., 187-253V; the ripple period is the time interval between repeated occurrences of the current ripple waveform, such as 10 microseconds in a 100kHz power supply.

[0110] In this embodiment, firstly, the input voltage fluctuation amplitude is continuously monitored through step 1051. When the voltage value exceeds the preset range three times consecutively, such as 260V > 253V, and the interval between each time is exactly equal to a complete ripple cycle (e.g., 10 microseconds is one cycle in a 100kHz power supply), the feedback control signal is connected to the adaptive ripple lockout loop. Specifically, a high-speed voltage comparator is used to collect input voltage data. When 260V is detected for the first time, a timer is started. If 261V and 259V are measured again within the next two ripple cycles (20 microseconds), the start-up condition is determined to be met. For example, if the industrial frequency converter input voltage records a sequence of 265V, 263V, and 268V within 30 microseconds, with each data point spaced 10 microseconds apart, the loop control function is immediately activated.

[0111] Secondly, response speed optimization is achieved by adjusting the hysteresis voltage in step 1052: When the input voltage is higher than the rated value, such as 240V > 220V, the comparator's hysteresis voltage is reduced from 50mV to 30mV, improving the switching response speed and allowing the rectifier drive signal to match the current trough more quickly. Reducing the hysteresis voltage improves the matching speed by 30%, thereby reducing switching losses. When the input voltage is lower than the rated value, such as 190V < 220V, the hysteresis voltage is increased from 50mV to 70mV, slowing down the response speed and preventing malfunctions. The formula for hysteresis voltage variation is as follows: in The reference hysteresis voltage is 50 millivolts. For every 1 volt increase in the input voltage difference, the hysteresis voltage changes by 2%. For example, when the input voltage of an electric vehicle charging station drops to 200 volts, the hysteresis voltage is adjusted to 68 millivolts to reduce the response speed and avoid increasing the false trigger rate.

[0112] Finally, step 1053 precisely controls the switching timing: the differential circuit calculates the decreasing slope of the output current ripple. When the slope approaches zero (within ±0.1 A / µs), it indicates the current is about to reach its lowest point, at which point the rectifier drive signal is immediately triggered to turn on the rectifier. In a measured case of a server power supply, the current ripple decreased from a peak of 1 A to 0 A in 5 µs. When the slope dropped to -0.15 A / µs, it was determined to be close to the trough, with a triggering error of less than 0.1 µs. The calculation process was as follows: from a slope of -0.8 A / µs at t=0 to -0.12 A / µs at t=4 µs, and at t=4.8 µs, a slope of -0.09 A / µs was detected, triggering the switch immediately. This process reduces switching losses and avoids voltage oscillations.

[0113] In practical applications, in the adaptive rectification timing control scenario of a wide-voltage input power supply system, the system achieves precise trough point matching through the following mechanism: when the input voltage fluctuates continuously within the range of 80V to 240V for more than three consecutive ripple cycles (e.g., ±15%), the feedback control signal immediately activates the adaptive ripple lockout loop. This loop dynamically adjusts the response speed of the rectifier drive timing based on the actual input voltage value: if the input voltage is higher than the rated value of 220V, the hysteresis voltage of the hysteresis comparator is reduced from 200mV to 50mV, accelerating the response speed to the 0.1V / microsecond level and shortening the drive signal trigger time to within 50 nanoseconds, achieving a 15% reduction in switching losses under high voltage conditions; if the input voltage is lower than 180V, the hysteresis voltage is increased to 300mV, and the response speed is reduced to the 0.5V / microsecond level, avoiding false triggering caused by low-voltage noise. Simultaneously, a high-speed comparator detects the falling edge slope of the output current ripple. When the slope decays to below 0.5V / µs, it indicates that the current has entered the trough region, and the rectifier drive signal is triggered, ensuring that more than 90% of the conduction time is precisely aligned with the ripple trough point. Actual measurements show that at a 100kHz switching frequency, this mechanism reduces the peak temperature of the rectifier conduction loss by 8℃, and maintains a trough lock-in error of less than 15 nanoseconds under input voltage fluctuations of ±20%.

[0114] In the overall scheme of step 105 above, by inputting the feedback control signal into the adaptive ripple lock-in loop, the response speed of the rectifier drive timing is adjusted according to the ripple amplitude changes caused by the wide voltage input range, achieving the dual effects of energy efficiency optimization and noise suppression. Specifically, when the input voltage fluctuation amplitude continuously exceeds the preset range, the ripple lock-in loop is activated, and the comparator hysteresis voltage is dynamically adjusted: when the input voltage is higher than the rated value, the hysteresis voltage is reduced to accelerate the rectifier drive timing, effectively suppressing high-frequency switching losses by quickly matching the output current ripple trough; when the input voltage is lower than the rated value, the hysteresis voltage is increased to reduce the response speed and avoid false triggering caused by transient noise interference. By accurately detecting the critical state point where the slope of the falling edge of the output current ripple approaches zero, the rectifier drive signal is triggered to ensure that the drive timing and the current trough are kept synchronously locked, so that the rectifier is always in the minimum conduction impedance state under wide voltage fluctuations. This mechanism adaptively balances loss suppression under high-voltage conditions with noise tolerance under low-voltage conditions, significantly improving power conversion efficiency and system immunity, and providing stable and reliable energy transmission for power systems in variable input environments.

[0115] The following is a complete embodiment for steps 101 to 105:

[0116] In the adaptive feedback regulation scenario of an industrial power supply system, the system executes the following integrated control process: First, the power supply output current ripple waveform is acquired at a sampling rate of 10kHz. Then, three-level wavelet packet decomposition is performed using the db4 wavelet basis function to generate eight sub-frequency band signals from 0 to 5kHz. The low-frequency band of 0-100Hz and the high-frequency band of 1-5kHz are extracted as characteristic frequency bands. The average energy value is calculated within 20 consecutive 50ms time windows to form a dynamic energy distribution curve. When the wide voltage input fluctuates between 80V and 240V, the load dynamic mode is identified based on the energy migration characteristics: when the high-frequency band energy increases by 1.5 times compared to the low-frequency band, a sudden load increase is determined; when the low-frequency attenuation amplitude is less than 0.8 times that of the high-frequency band, a sudden load decrease is determined. Meanwhile, 16 sets of annular piezoelectric cantilever beam arrays arranged on the edge of the printed circuit board contain 16 low-frequency beams (80-150Hz) and 16 high-frequency beams (200-500Hz) to capture electromagnetic vibrations caused by current ripple. The displacement sensing piezoelectric element integrated in each beam has a sensitivity of 0.5V / μm and generates a 1ms trigger pulse at the deformation peak. This pulse drives the MOSFET switch to form an LC oscillation circuit with the 10mH inductor and the piezoelectric element with a piezoelectric constant of 650pC / N. The output 1-10kHz alternating current is converted into 3.3V DC by a full-bridge rectifier and directly supplied to the feedback circuit, supplemented by a 220μF energy storage capacitor for voltage regulation.

[0117] After the recovered electrical energy is injected into the inverting proportional circuit with ripple, the parameters are dynamically configured according to the load mode: when the load suddenly increases, a 10kΩ feedback resistor is connected to achieve a gain coefficient of -3.5; when the load suddenly decreases, a 100kΩ resistor is switched to achieve a gain coefficient of -0.8. At the same time, a voltage control delay line is connected in series at the op-amp output, and the delay change rate is adjusted according to the input voltage change rate to ≥50m. A 200ns time delay generates a 45° phase shift with a rate of change ≤10m. A 50ns time delay generates a 10° phase shift. Subsequently, it tracks the ripple amplitude change caused by input voltage fluctuations. When the ripple increment exceeds a 50mV threshold, a signal synthesis unit is triggered within 100ns. This unit uses a digitally controlled potentiometer to adjust the resistance ratio according to the load mode: 0.2:1 for a sudden increase and 1.5:1 for a sudden decrease, precisely matching the inverting ripple amplitude. This is then superimposed on the feedback voltage to generate a control signal with a ripple peak-to-peak value of less than 15mV. Finally, this signal is input to an adaptive ripple lockout loop: when the input voltage fluctuates by more than ±15% for three consecutive cycles, if the voltage is higher than 220V, the hysteresis comparator's hysteresis voltage is reduced to 50mV, accelerating the response to 0.1. The drive signal is triggered within 50ns; if the voltage is below 180V, the hysteresis voltage is increased to 300mV and the speed is reduced to 0.5. Simultaneously, the falling edge slope of the output current ripple is detected; when the slope is below 0.5... The rectifier drive is triggered instantly, ensuring precise matching of the trough point during 90% conduction. This system achieves a trough lock-in error of less than 15ns under ±20% input voltage fluctuations, and reduces the rectifier peak temperature by 8℃.

[0118] Figure 3 This application provides a schematic diagram of a power management adaptive feedback regulation system with a wide voltage input range, as shown in the embodiment of the present application. Figure 3 As shown, the system includes:

[0119] The identification module 31 is used to acquire the current ripple waveform of the power supply output, perform wavelet packet decomposition on the current ripple waveform, extract the energy distribution of the characteristic frequency band, and identify the dynamic change mode of the load based on the dynamic offset characteristics of the energy distribution of the characteristic frequency band fluctuating with a wide voltage input range.

[0120] The recycling module 32 is used to capture the electromagnetic vibration caused by the current ripple waveform through a piezoelectric cantilever beam array set on the edge of the printed circuit board, convert the electromagnetic vibration into recycled electrical energy, and directly supply the recycled electrical energy to the feedback circuit to provide adaptive power supply energy for the feedback circuit.

[0121] The generation module 33 is used to inject the ripple of the recovered electrical energy into the inverting proportional circuit, and dynamically set the gain coefficient and phase offset according to the load dynamic change mode to generate an inverting ripple signal.

[0122] The cancellation module 34 is used to dynamically track the ripple amplitude change caused by a wide voltage input range, and superimpose the inverted ripple signal onto the feedback voltage to dynamically cancel the ripple amplitude change and generate a feedback control signal after ripple cancellation.

[0123] The adjustment module 35 is used to input the feedback control signal into the adaptive ripple lockout loop, and adjust the response speed of the rectifier drive timing according to the ripple amplitude change caused by the wide voltage input range, so that the rectifier drive timing matches the trough of the output current ripple. The response speed is positively correlated with the input voltage fluctuation amplitude. At high voltage, the matching is accelerated to suppress loss, and at low voltage, the speed is reduced to avoid false triggering.

[0124] Figure 3 The aforementioned wide voltage input range power management adaptive feedback regulation system can perform... Figure 1 The implementation principle and technical effects of the wide voltage input range power management adaptive feedback regulation method described in the illustrated embodiment will not be repeated here. The specific operation methods of each module and unit in the wide voltage input range power management adaptive feedback regulation system in the above embodiments have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0125] In one possible design, Figure 3 The wide voltage input range power management adaptive feedback regulation system of the embodiment shown can be implemented as a computing device, such as... Figure 4 As shown, the computing device may include a storage component 41 and a processing component 42;

[0126] The storage component 41 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 42.

[0127] The processing component 42 is used for the above Figure 1 The embodiment describes a power management adaptive feedback regulation method with a wide voltage input range.

[0128] The processing component 42 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described method.

[0129] Storage component 41 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0130] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0131] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0132] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0133] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0134] This application also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 1 The embodiment shown illustrates a power management adaptive feedback regulation method with a wide voltage input range.

[0135] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0136] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0137] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0138] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A power management adaptive feedback regulation method with a wide voltage input range, characterized in that, include: Acquire the current ripple waveform of the power supply output, perform wavelet packet decomposition on the current ripple waveform, and extract the energy distribution of the characteristic frequency band. Based on the dynamic offset characteristics of the energy distribution of the characteristic frequency band fluctuating with a wide voltage input range, identify the dynamic change pattern of the load. By using a piezoelectric cantilever beam array positioned at the edge of the printed circuit board, the electromagnetic vibrations caused by the current ripple waveform are captured, the electromagnetic vibrations are converted into recovered electrical energy, and the recovered electrical energy is directly supplied to the feedback circuit to provide adaptive power supply energy for the feedback circuit. The ripple of the recovered electrical energy is injected into the inverting proportional circuit. The gain coefficient and phase offset are dynamically set according to the load dynamic change mode to generate an inverting ripple signal. The system dynamically tracks the ripple amplitude changes caused by a wide voltage input range and superimposes the inverted ripple signal onto the feedback voltage to dynamically cancel the ripple amplitude changes, generating a feedback control signal after ripple cancellation. The feedback control signal is input into the adaptive ripple lockout loop. Based on the ripple amplitude change caused by the wide voltage input range, the response speed of the rectifier drive timing is adjusted so that the rectifier drive timing matches the trough of the output current ripple. The response speed is positively correlated with the input voltage fluctuation amplitude. At high voltage, the matching is accelerated to suppress losses, and at low voltage, the speed is reduced to avoid false triggering.

2. The method according to claim 1, characterized in that, Acquire the current ripple waveform of the power supply output, perform wavelet packet decomposition on the current ripple waveform, and extract the energy distribution of the characteristic frequency band. Based on the dynamic offset characteristics of the energy distribution of the characteristic frequency band fluctuating with a wide voltage input range, identify the dynamic change pattern of the load, including: Acquire the current ripple waveform at the power supply output terminal, perform wavelet packet decomposition on the current ripple waveform, and generate multiple sub-frequency band signals; Extract at least one preset target sub-frequency band, calculate the average energy value of the target sub-frequency band within a continuous time window, and arrange the average energy values ​​of all time windows in a time series to form a characteristic frequency band energy distribution; Under wide voltage input range fluctuations, based on the energy distribution of the characteristic frequency band, the dynamic offset characteristics of the target sub-frequency band are detected, and the dynamic load change pattern is identified according to the dynamic offset characteristics. If the average energy value of the high-frequency sub-band increases more than that of the low-frequency sub-band, the energy distribution shifts to higher frequencies, which is determined to be a dynamic load increase mode. If the average energy value of the low-frequency sub-band decreases less than that of the high-frequency sub-band, the energy distribution shifts to lower frequencies, which is determined to be a dynamic load decrease mode.

3. The method according to claim 1, characterized in that, The system dynamically tracks the ripple amplitude changes caused by a wide voltage input range and superimposes the inverted ripple signal onto the feedback voltage to dynamically cancel the ripple amplitude changes, generating a ripple-canceled feedback control signal, including: The system dynamically tracks the ripple amplitude changes caused by a wide voltage input range at the power input terminal and calculates the instantaneous increment of the current ripple amplitude based on the ripple amplitude changes. The inverted ripple signal is input to the first input terminal of the adder, and the feedback voltage signal is input to the second input terminal of the adder. When the instantaneous increment of the ripple amplitude exceeds a preset threshold, the adder is immediately triggered to perform a signal superposition operation, so that the inverted ripple signal and the feedback voltage signal are dynamically synthesized to cancel the change in ripple amplitude and generate a feedback control signal after ripple cancellation.

4. The method according to claim 3, characterized in that, When the instantaneous increment of the ripple amplitude exceeds a preset threshold, the adder is immediately triggered to perform a signal superposition operation, so that the inverted ripple signal and the feedback voltage signal are dynamically synthesized to cancel the change in ripple amplitude and generate a feedback control signal after ripple cancellation, including: When the instantaneous increment of the ripple amplitude exceeds a preset threshold, the triggering unit immediately sends a superposition enable signal to the signal synthesis unit. The signal synthesis unit responds to the superposition enable signal and superimposes the inverted ripple signal of the first input port and the feedback voltage signal of the second input port. According to the load dynamic change mode, the resistance ratio on the inverted ripple signal path is adjusted by a digitally controlled potentiometer so that the amplitude of the inverted ripple signal matches the ripple amplitude in the feedback voltage signal, thereby canceling the ripple amplitude change and generating a feedback control signal after ripple cancellation.

5. The method according to claim 1, characterized in that, The feedback control signal is input into the adaptive ripple lockout loop. Based on the ripple amplitude variation caused by the wide voltage input range, the response speed of the rectifier drive timing is adjusted to match the trough of the output current ripple. This response speed is positively correlated with the input voltage fluctuation amplitude. At high voltage, the timing is accelerated to suppress losses; at low voltage, the timing is reduced to avoid false triggering. This includes: When the input voltage fluctuation amplitude continuously exceeds the preset range for three ripple cycles, the feedback control signal is input into the adaptive ripple lockout loop. Based on the ripple amplitude variation caused by the wide voltage input range, the response speed of the rectifier drive timing is dynamically adjusted. If the input voltage is higher than the rated value, the hysteresis voltage of the comparator is reduced to accelerate the rectifier drive timing matching and suppress switching losses. If the input voltage is lower than the rated value, the hysteresis voltage of the comparator is increased to reduce the response speed to avoid noise-induced false triggering. The comparator detects the falling edge slope of the output current ripple. When the falling edge slope approaches zero, the rectifier drive signal is triggered so that the rectifier drive timing matches the trough of the output current ripple.

6. The method according to claim 1, characterized in that, By using a piezoelectric cantilever beam array positioned at the edge of the printed circuit board, the electromagnetic vibrations caused by the current ripple waveform are captured, the electromagnetic vibrations are converted into recovered electrical energy, and the recovered electrical energy is directly supplied to the feedback circuit to provide adaptive power supply energy to the feedback circuit, including: A ring-shaped array of piezoelectric cantilever beams is set at the edge of the printed circuit board, and each cantilever beam contains symmetrically distributed displacement sensing piezoelectric elements and energy conversion piezoelectric elements. The displacement state of the cantilever beam is detected by the displacement sensing piezoelectric element, a displacement detection signal is generated and input to the synchronous switch control circuit, and the synchronous switch control circuit processes the displacement detection signal to generate a trigger pulse that is synchronized with the maximum displacement of the cantilever beam. The energy conversion piezoelectric element directly captures the electromagnetic vibration caused by the induced current ripple and outputs alternating charge. The alternating charge is input into the synchronous switching inductor circuit. The digital switch is closed when the displacement of the cantilever beam reaches its maximum value through the trigger pulse, so that the inductor and the energy conversion piezoelectric element form an oscillation circuit to obtain a high-frequency pulse current. The high-frequency pulse current is converted into DC power through a full-bridge rectifier circuit, and the DC power is directly connected to the power input terminal of the feedback circuit as the recovered electrical energy, providing adaptive power supply energy for the feedback circuit.

7. The method according to claim 6, characterized in that, The ripple from the recovered electrical energy is injected into the inverting proportional circuit. Based on the load dynamic change mode, the gain coefficient and phase offset are dynamically set to generate the inverting ripple signal, including: The input ripple of the DC power is injected into the inverting amplifier circuit, in which the inverting input terminal of the operational amplifier is connected to the positive terminal of the DC power, and the non-inverting input terminal is grounded. According to the load dynamic change mode switching resistor network, if the load dynamic change mode is the load dynamic increase mode, a low resistance feedback resistor is connected to set a high gain coefficient; if the load dynamic change mode is the load dynamic decrease mode, a high resistance feedback resistor is connected to set a low gain coefficient. An adjustable delay line is connected in series at the output of the operational amplifier, and the delay time is proportional to the rate of change of the input voltage, thereby generating an inverted ripple signal with a phase offset.

8. A power management adaptive feedback regulation system with a wide voltage input range, characterized in that, include: The identification module is used to acquire the current ripple waveform of the power supply output, perform wavelet packet decomposition on the current ripple waveform, extract the energy distribution of the characteristic frequency band, and identify the dynamic change pattern of the load based on the dynamic offset characteristics of the energy distribution of the characteristic frequency band fluctuating with a wide voltage input range. The recycling module is used to capture the electromagnetic vibration caused by the current ripple waveform through a piezoelectric cantilever beam array set on the edge of the printed circuit board, convert the electromagnetic vibration into recycled electrical energy, and directly supply the recycled electrical energy to the feedback circuit to provide adaptive power supply energy for the feedback circuit. The generation module is used to inject the ripple of the recovered electrical energy into the inverting proportional circuit, and dynamically set the gain coefficient and phase offset according to the load dynamic change mode to generate an inverting ripple signal. The cancellation module is used to dynamically track the ripple amplitude change caused by a wide voltage input range, and superimpose the inverted ripple signal onto the feedback voltage to dynamically cancel the ripple amplitude change and generate a feedback control signal after ripple cancellation. The adjustment module is used to input the feedback control signal into the adaptive ripple lockout loop, and adjust the response speed of the rectifier drive timing according to the ripple amplitude change caused by the wide voltage input range, so that the rectifier drive timing matches the trough of the output current ripple. The response speed is positively correlated with the input voltage fluctuation amplitude. At high voltage, the matching is accelerated to suppress loss, and at low voltage, the speed is reduced to avoid false triggering.

9. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement a power management adaptive feedback regulation method with a wide voltage input range as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that, The device contains a computer program that, when executed by a computer, implements a power management adaptive feedback regulation method with a wide voltage input range as described in any one of claims 1 to 7.

Images