Control method, system and equipment of switching power supply and storage medium

Abstract

The invention relates to a switching power supply control method, system and device and a storage medium, and the method comprises the steps: monitoring the voltage and current of the output end of a switching power supply, and the rate and amplitude of load change, and obtaining the real-time data of the load change; according to the real-time data of the load change, the change of the current load is identified, and a load change mode is obtained; dynamically adjusting operation parameters of the controller according to the load change mode; based on the real-time data of the load change, predicting the load change trend of the switching power supply in real time by using a pre-trained load prediction model to obtain a prediction result, and generating a disturbance compensation signal based on the prediction result; and adjusting the switching state of a switching circuit of the switching power supply according to the control parameter adjusted by the controller and the disturbance compensation signal. The switching power supply has the effect of improving the stability of the switching power supply.

Description

Technical Field

[0001] The present application relates to the technical field of switching power supplies, and in particular to a control method, system, device and storage medium for a switching power supply. Background Art

[0002] At present, traditional switching power supplies are widely used in various electronic devices. They maintain stable output voltage or current by adjusting the switching frequency or duty cycle. Although the design of these traditional switching power supplies is mature and stable, their dynamic response speed is relatively slow when facing load changes, especially sudden changes, which limits their application in some high-demand environments.

[0003] The above-mentioned existing technical solutions have the following defects: in the case of sudden load changes, the existing control strategy often cannot respond quickly, causing the voltage or current to exceed the predetermined range, resulting in system instability and affecting the overall performance. Therefore, there is room for improvement. Summary of the invention

[0004] In order to improve the stability of a switching power supply, the present application provides a control method, system, device and storage medium for a switching power supply.

[0005] The above-mentioned invention objective of the present application is achieved through the following technical solutions: A method for controlling a switching power supply, the method comprising: Monitor the voltage, current and load change rate and amplitude at the output end of the switching power supply, and obtain real-time data of the load change; According to the real-time data of the load change, identifying the change of the current load and obtaining the load change pattern; Dynamically adjusting operating parameters of the controller according to the load change mode; Based on the real-time data of the load change, a pre-trained load prediction model is used to predict the load change trend of the switching power supply in real time to obtain a prediction result, and a disturbance compensation signal is generated based on the prediction result; The switching state of the switching circuit of the switching power supply is adjusted according to the control parameter adjusted by the controller and the disturbance compensation signal.

[0006] By adopting the above technical solution, by real-time monitoring of the voltage, current and load change rate and amplitude at the output end of the switching power supply, real-time data of load changes can be effectively obtained, ensuring a comprehensive grasp of the power supply output state and timely capturing load fluctuations; by analyzing the real-time data of load changes and identifying the current load changes, it is helpful to accurately determine the load change mode, provide a basis for the subsequent adjustment of the control strategy, and improve the load adaptability; by dynamically adjusting the operating parameters of the controller according to the load change mode, the operating state of the switching power supply is effectively optimized to ensure stable output under load fluctuations and improve the adaptability and reliability of the system; by using a pre-trained load prediction model, the load change trend is predicted based on real-time load data, and a disturbance compensation signal is effectively generated to respond to load fluctuations in advance, reduce system response delays, and improve the dynamic response capability of the power supply; by combining the adjusted control parameters and disturbance compensation signals, the switching circuit state of the switching power supply is dynamically adjusted to effectively achieve precise control of the power supply output, ensure the stability of the output voltage and current, and reduce power supply fluctuations and interference.

[0007] In a preferred example, the present application may be further configured as follows: identifying the change of the current load according to the real-time data of the load change, and obtaining the load change mode includes: Extracting load current data from the real-time data of load changes by using a short-time Fourier transform algorithm; According to the load current data, the load current change rate formula is: , calculate the rate of change of the load current, where R is the rate of change of the load current, ΔI is the amount of change of the load current, and Δt is the time period during which the load current changes; Setting a first set threshold and a second set threshold to distinguish the load change mode, wherein the load change mode includes a rapid change mode, a slow change mode and a sudden disturbance mode; When the change rate of the load current is higher than the first set threshold, the load change mode is determined to be the rapid change mode; When the change rate of the load current is between the first set threshold and the second set threshold, the load change mode is determined to be the slow change mode; The frequency components in the load current data are analyzed using Fourier transform to generate a spectrum of the load current data. When periodic peaks or random fluctuations appear in the spectrum of the load current data, the load change mode is determined to be the sudden disturbance mode.

[0008] By adopting the above technical scheme, the real-time data of load changes are processed by applying the short-time Fourier transform algorithm, the load current data is effectively extracted, and the accuracy and real-time performance of signal processing are improved; by using the load current change rate formula to calculate the load current change rate, it is helpful to quantify the speed of load change and provide a basis for subsequent classification; by setting the first and second thresholds to distinguish different load change modes, the sharp change, slow change and sudden disturbance modes are effectively identified, and the recognition accuracy of the system is improved; setting thresholds to distinguish change modes, and using Fourier transform to analyze frequency components to determine sudden disturbance modes, the recognition accuracy and response speed of load change modes are significantly improved, the system's adaptability to different load change conditions is enhanced, the stability and reliability of the power supply system are ensured, and the risk of system failure is reduced.

[0009] In a preferred example, the present application may be further configured as follows: the setting of the first set threshold and the second set threshold specifically includes: Acquire historical operation data of the switching power supply, perform statistical analysis on the historical operation data of the switching power supply, and obtain a threshold range; The load characteristics and response requirements of the switching power supply are obtained, and based on the load characteristics and response requirements of the switching power supply, a simulation experiment is performed in a virtual environment, and adjustments are made in combination with the threshold range to obtain a set threshold, wherein the set threshold includes the first set threshold and the second set threshold.

[0010] By adopting the above technical solution, by acquiring and statistically analyzing the historical operating data of the switching power supply, conducting simulation experiments in a virtual environment in combination with the load characteristics and response requirements, and adjusting the first and second set thresholds, it is helpful to accurately set the judgment criteria for load changes, improve the system's recognition ability and response efficiency for different load change modes, ensure the stable operation of the switching power supply under various working conditions, and enhance the reliability and adaptability of the overall system.

[0011] In a preferred example, the present application may be further configured as follows: dynamically adjusting the operating parameters of the controller according to the load change mode includes: When the load change mode is determined to be the rapid change mode, increasing the gain parameter of the controller and shortening the integration time according to the load current rate; When the load change mode is determined to be the slow change mode, reducing the gain parameter of the controller and shortening the differential time according to the load current rate; When the load change mode is determined to be the sudden disturbance mode, the frequency and amplitude of the disturbance signal are extracted from the frequency spectrum of the load current data, and the regulation frequency and switch duty cycle of the controller are optimized in combination with feedforward control.

[0012] By adopting the above technical solution, by dynamically determining the load change mode and adjusting the controller parameters according to different modes, the system's response speed and stability to various load changes can be improved, the system's steady-state performance and energy efficiency can be optimized, and the system's anti-interference ability can be enhanced. This adaptive control strategy ensures that the switching power supply can maintain an efficient and stable operating state when facing rapid changes, slow changes and sudden disturbances, significantly improving the reliability and performance of the overall system.

[0013] In a preferred example, the present application may be further configured as follows: before the load change trend of the switching power supply is predicted in real time based on the real-time data of the load change by using a pre-trained load prediction model to obtain a prediction result, and before a disturbance compensation signal is generated based on the prediction result, the control method of the switching power supply further includes: Collecting historical data of the load change, extracting key features from the historical data of the load change, and constructing an input feature vector, wherein the key features include a load current change rate, a load voltage fluctuation amplitude, and a power spectrum characteristic; A deep learning model is constructed by using a long short-term memory network and a convolutional neural network, the deep learning model is trained based on the input feature vector, and a back propagation algorithm is used to optimize the weight of the deep learning model to obtain a trained load prediction model; The hyperparameters of the trained load prediction model are optimized by combining a genetic algorithm and a gradient descent algorithm to obtain the pre-trained load prediction model.

[0014] By adopting the above technical solution, by collecting and extracting the key features of load changes, building and training a deep learning model combining LSTM and CNN, and using genetic algorithm and gradient descent algorithm to optimize the model hyperparameters, a high-precision and high-robustness load prediction model is realized. This model can accurately predict the operating status of the switching power supply under different load conditions, thereby optimizing the control strategy, improving the response speed and stability of the system, and ultimately improving the overall performance and energy efficiency of the switching power supply system.

[0015] In a preferred example, the present application can be further configured as follows: based on the real-time data of the load change, using a pre-trained load prediction model to predict the load change trend of the switching power supply in real time, obtaining a prediction result, and generating a disturbance compensation signal based on the prediction result includes: Inputting the real-time data of the load change into the pre-trained load prediction model for reasoning and analysis, predicting the load change trend of the switching power supply in a future time period, and obtaining the prediction result; Based on the prediction result, the disturbance compensation signal is calculated and input to the control end, wherein the disturbance compensation signal includes a voltage adjustment amount and a conduction timing adjustment amount.

[0016] By adopting the above technical solution, by inputting real-time load change data into a pre-trained load prediction model for prediction, it is possible to accurately foresee future load change trends, and calculate and input disturbance compensation signals based on the prediction results, effectively adjust the voltage and conduction timing, thereby ensuring the stability of the switching power supply output and the efficient operation of the system, and improving the system's response speed and anti-interference ability.

[0017] In a preferred example, the present application may be further configured as follows: the control method of a switching power supply further includes: during the operation of the switching power supply, continuously monitoring actual operation data of the switching power supply, comparing the actual operation data with the data in the prediction result, and updating the parameters of the pre-trained load prediction model; When it is detected that the operating state of the switching power supply is abnormal, the protection mechanism is triggered, the abnormal situation is analyzed, and a maintenance processing strategy is generated and transmitted to the user end.

[0018] By adopting the above technical solution, by continuously monitoring the actual operation data and comparing it with the prediction results, dynamically updating the load prediction model parameters, and triggering the protection mechanism and generating the maintenance strategy when the operation abnormality is detected, the real-time monitoring and adaptive optimization of the operating status of the switching power supply are realized, which not only improves the accuracy of load prediction and the adaptability of the model, but also enhances the responsiveness and safety of the system under abnormal conditions, ensures the stable operation and efficient maintenance of the switching power supply, and ultimately improves the reliability of the overall system and user experience.

[0019] The second object of the invention is achieved by the following technical solutions: A control system for a switching power supply, the control system for the switching power supply comprising: A real-time monitoring module is used to monitor the voltage, current and load change rate and amplitude at the output end of the switching power supply, and obtain real-time data of the load change; A mode determination module, used to identify the change of the current load according to the real-time data of the load change, and obtain the load change mode; A parameter adjustment module, used for dynamically adjusting the operating parameters of the controller according to the load change mode; A compensation generation module is used to predict the load change trend of the switching power supply in real time based on the real-time data of the load change and using a pre-trained load prediction model to obtain a prediction result, and generate a disturbance compensation signal based on the prediction result; an adjustment state module is used to adjust the switching state of the switching circuit of the switching power supply according to the control parameters adjusted by the controller and the disturbance compensation signal.

[0020] By adopting the above technical solution, by real-time monitoring of the voltage, current and load change rate and amplitude at the output end of the switching power supply, real-time data of load changes can be effectively obtained, ensuring a comprehensive grasp of the power supply output state and timely capturing load fluctuations; by analyzing the real-time data of load changes and identifying the current load changes, it is helpful to accurately determine the load change mode, provide a basis for the subsequent adjustment of the control strategy, and improve the load adaptability; by dynamically adjusting the operating parameters of the controller according to the load change mode, the operating state of the switching power supply is effectively optimized to ensure stable output under load fluctuations and improve the adaptability and reliability of the system; by using a pre-trained load prediction model, the load change trend is predicted based on real-time load data, and a disturbance compensation signal is effectively generated to respond to load fluctuations in advance, reduce system response delays, and improve the dynamic response capability of the power supply; by combining the adjusted control parameters and disturbance compensation signals, the switching circuit state of the switching power supply is dynamically adjusted to effectively achieve precise control of the power supply output, ensure the stability of the output voltage and current, and reduce power supply fluctuations and interference.

[0021] The third objective of the present application is achieved through the following technical solutions: A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above-mentioned control method of a switching power supply when executing the computer program.

[0022] The fourth objective of the present application is achieved through the following technical solutions: A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the control method of a switching power supply are implemented.

[0023] In summary, the present application includes at least one of the following beneficial technical effects: 1. By real-time monitoring of the voltage, current and load change rate and amplitude at the output end of the switching power supply, real-time data of load changes can be effectively obtained to ensure a comprehensive grasp of the power output status and timely capture of load fluctuations; by analyzing the real-time data of load changes and identifying the current load changes, it is helpful to accurately determine the load change mode, provide a basis for the subsequent adjustment of the control strategy, and improve the load adaptability; 2. By dynamically adjusting the operating parameters of the controller according to the load change pattern, the operating state of the switching power supply can be effectively optimized to ensure stable output under load fluctuations and improve the adaptability and reliability of the system; by using the pre-trained load prediction model, the load change trend is predicted based on real-time load data, and the disturbance compensation signal is effectively generated to respond to load fluctuations in advance, reduce system response delays, and improve the dynamic response capability of the power supply; by combining the adjusted control parameters and disturbance compensation signals, the switching circuit state of the switching power supply is dynamically adjusted to effectively achieve precise control of the power supply output, ensure the stability of the output voltage and current, and reduce power supply fluctuations and interference. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 is a flow chart of a control method of a switching power supply in an embodiment of the present application; Figure 2 is a flowchart for implementing step S20 in a method for controlling a switching power supply in an embodiment of the present application; Figure 3 is a flowchart for implementing step S23 in a method for controlling a switching power supply in an embodiment of the present application; Figure 4 is a flowchart for implementing step S30 in a method for controlling a switching power supply in an embodiment of the present application; Figure 5 is another implementation flow chart of step S40 in a method for controlling a switching power supply in an embodiment of the present application; Figure 6 is a flowchart for implementing step S40 in a method for controlling a switching power supply in an embodiment of the present application; Figure 7 is another implementation flow chart of a control method for a switching power supply in one embodiment of the present application; Figure 8 It is a principle block diagram of a control system of a switching power supply in one embodiment of the present application; Fig. 9 It is a schematic diagram of a device in an embodiment of the present application. DETAILED DESCRIPTION

[0025] The present application is further described in detail below in conjunction with the accompanying drawings.

[0026] In one embodiment, if Figure 1 As shown, the present application discloses a control method for a switching power supply, which specifically includes the following steps: S10: Monitor the voltage, current, and load change rate and amplitude at the output end of the switching power supply to obtain real-time data of load changes.

[0027] Specifically, a high-precision voltage sensor and current sensor are installed at the output end of the switching power supply to monitor the changes in the output voltage and current in real time. A data acquisition module, such as an ADC converter, is used to convert the analog signal collected by the sensor into a digital signal. The load change rate and amplitude are calculated based on the collected data, and an embedded processor or microcontroller is used to process and analyze the collected data in real time to generate real-time load change data.

[0028] S20: According to the real-time data of load changes, the changes of the current load are identified to obtain the load change pattern.

[0029] Specifically, pattern recognition algorithms, such as threshold detection, Fourier transform, wavelet transform, etc., are used to analyze real-time load data and identify specific patterns of load changes, such as sudden changes, periodic changes, or random changes. The identified load change patterns are classified and marked for subsequent control strategy adjustments. For example, load changes can be divided into different categories such as rapid rise, rapid fall, and slow change.

[0030] S30: Dynamically adjust the operating parameters of the controller according to the load change mode.

[0031] Specifically, a database containing the best control parameters under different load change modes is established. According to the currently identified load change mode, the corresponding control parameters are extracted from the parameter library, or the control parameters most suitable for the current mode are dynamically calculated through interpolation and optimization algorithms. The adjusted control parameters are applied to the controller in real time to optimize the performance of the switching power supply, such as improving the response speed, reducing oscillation or reducing steady-state error.

[0032] S40: Based on the real-time data of load changes, a pre-trained load prediction model is used to predict the load change trend of the switching power supply in real time to obtain a prediction result, and a disturbance compensation signal is generated based on the prediction result.

[0033] Specifically, a prediction model, such as a neural network, is trained in advance using historical load data to predict the load change trend in the future. The real-time load data of the current and past period of time is input into the prediction model to generate a predicted value of the future load. Based on the prediction result, the impact of the expected load change on the output of the switching power supply is calculated, and a corresponding disturbance compensation signal is generated. The disturbance compensation signal is fed back to the control system, and other control strategies are coordinated to jointly optimize the stability and response speed of the switching power supply.

[0034] S50: adjusting the switching state of the switching circuit of the switching power supply according to the control parameters adjusted by the controller and the disturbance compensation signal.

[0035] Specifically, the adjusted control parameters and disturbance compensation signals are input into the controller of the switching power supply. According to the control parameters and the compensation signal, the corresponding PWM waveform is generated, and the on and off time of the switching elements, such as MOSFET or IGBT, are adjusted to control the output voltage and current. By changing the switching state of the switching elements, such as the frequency, duty cycle, etc., the working state of the switching power supply is adjusted in real time to adapt to load changes and maintain a stable output.

[0036] By adopting the above technical solution, by real-time monitoring of the voltage, current and load change rate and amplitude at the output end of the switching power supply, real-time data of load changes can be effectively obtained, ensuring a comprehensive grasp of the power supply output state and timely capturing load fluctuations; by analyzing the real-time data of load changes and identifying the current load changes, it is helpful to accurately determine the load change mode, provide a basis for the subsequent adjustment of the control strategy, and improve the load adaptability; by dynamically adjusting the operating parameters of the controller according to the load change mode, the operating state of the switching power supply is effectively optimized to ensure stable output under load fluctuations and improve the adaptability and reliability of the system; by using a pre-trained load prediction model, the load change trend is predicted based on real-time load data, and a disturbance compensation signal is effectively generated to respond to load fluctuations in advance, reduce system response delays, and improve the dynamic response capability of the power supply; by combining the adjusted control parameters and disturbance compensation signals, the switching circuit state of the switching power supply is dynamically adjusted to effectively achieve precise control of the power supply output, ensure the stability of the output voltage and current, and reduce power supply fluctuations and interference.

[0037] In one embodiment, if Figure 2 As shown, in step S20, the change of the current load is identified according to the real-time data of the load change, and the load change mode is obtained, which specifically includes: S21: extracting load current data from real-time data of load changes through a short-time Fourier transform algorithm.

[0038] Specifically, the STFT algorithm is used to perform time-frequency analysis on the collected real-time load change data. Through STFT processing, the collected current time series data is divided into multiple time windows, each window corresponds to the data in a short period of time, and Fourier transform is applied to the data in each time window to convert it from the time domain to the frequency domain, and the frequency component and amplitude information in the time period are extracted. The instantaneous value of the load current and its frequency characteristics can be accurately extracted, providing basic data for subsequent change rate calculation and pattern recognition.

[0039] S22: According to the load current data, the load current change rate formula is: , the rate of change of the load current is calculated, where R is the rate of change of the load current, ΔI is the amount of change of the load current, and Δt is the time period during which the load current changes.

[0040] Specifically, according to the load current data, the load current change rate formula is: , at each sampling point, the current change rate between the current moment and the previous moment is calculated, the change rate of the load current is calculated, and the calculation result is stored as a current change rate sequence for subsequent threshold judgment.

[0041] S23: Setting a first set threshold and a second set threshold to distinguish load change modes, where the load change modes include a rapid change mode, a slow change mode, and a sudden disturbance mode.

[0042] Specifically, according to the load characteristics and historical operation data of the switching power supply, a first set threshold and a second set threshold are set to distinguish load change modes, which include a rapid change mode, a slow change mode and a sudden disturbance mode.

[0043] S24: When the change rate of the load current is higher than the first set threshold, the load change mode is determined to be a rapid change mode.

[0044] Specifically, when the load current change rate is higher than the first set threshold, it is considered that the load changes sharply, and the system determines the load change mode as a sharp change mode. S25: When the change rate of the load current is between the first set threshold and the second set threshold, it is considered that the load changes slowly, and the load change mode is determined to be a slow change mode.

[0045] Specifically, when the load current change rate is between the first set threshold and the second set threshold, the system determines the load change mode as a slow change mode. S26: Use Fourier transform to analyze the frequency components in the load current data to generate a spectrum of the load current data. When periodic peaks or random fluctuations appear in the spectrum of the load current data, the load change mode is determined to be a sudden disturbance mode.

[0046] Specifically, the extracted load current data is subjected to Fourier transform to analyze its frequency components. Through Fourier transform, a spectrum diagram of the load current is generated to show the intensity of different frequency components. If obvious periodic peaks appear in the load current spectrum, it means that the load has significant fluctuations at a specific frequency, which may be caused by periodic disturbances; if there are random fluctuations in the spectrum, it means that the load current has been irregularly disturbed; when any of the above situations appears in the spectrum, the system determines the load change mode as a sudden disturbance mode.

[0047] In one embodiment, if Figure 3 As shown, in step S23, the first set threshold and the second set threshold are set, which specifically includes: S231: Acquire historical operation data of the switching power supply, perform statistical analysis on the historical operation data of the switching power supply, and obtain a threshold range.

[0048] Specifically, a data recording system is deployed, such as a data acquisition module or a storage system, at the output end of the switching power supply to record the operating data for a long time to obtain the operating conditions of the switching power supply under different load conditions, and the collected historical operating data of the switching power supply is statistically analyzed, such as using data visualization tools to draw data distribution diagrams, box plots, etc. for analysis to obtain the threshold range.

[0049] S232: Obtain the load characteristics and response requirements of the switching power supply, perform a simulation experiment in a virtual environment based on the load characteristics and response requirements of the switching power supply, and adjust the threshold value in combination with the threshold range to obtain a set threshold value, wherein the set threshold value includes a first set threshold value and a second set threshold value.

[0050] Specifically, identify the type of load driven by the switching power supply, such as constant resistance, constant capacitance, dynamic load, etc., analyze the load's operating current, voltage requirements and changing characteristics, clarify the switching power supply's response time, stability requirements and accuracy requirements under different load changes, determine the system's tolerance to load disturbances, and facilitate subsequent adjustment of control parameters. Use simulation software to establish a mathematical model of the switching power supply, including circuit topology, controller model and load model, incorporate actual load characteristics and response requirements into the simulation model, design different load change scenarios, such as sharp changes, slow changes and sudden disturbances, run the simulation, observe the response behavior of the switching power supply under different working conditions, and adjust the simulation experiment parameters based on the threshold range to obtain the set threshold.

[0051] In one embodiment, if Figure 4 As shown, in step S30, the operating parameters of the controller are dynamically adjusted according to the load change mode, specifically including: S31: When the load change mode is determined to be a rapid change mode, the gain parameter of the controller is increased and the integration time is shortened according to the load current rate.

[0052] Specifically, when the load change rate exceeds the first set threshold, the system determines that it is currently in a rapid change mode. For example, when the load current increases or decreases rapidly, the system output is quickly adjusted according to the calculated load current rate, the controller gain parameter is increased and the integration time is shortened to follow the drastic changes in the load and avoid large fluctuations.

[0053] S32: When the load change mode is determined to be a slow change mode, the gain parameter of the controller is reduced and the differential time is shortened according to the load current rate.

[0054] Specifically, when the load current change rate is between a first set threshold value and a second set threshold value, the system determines that it is currently in a slow change mode. For example, when the load current increases or decreases slowly, the gain parameter can be lowered to avoid excessive response to slowly changing errors, thereby reducing oscillation and over-adjustment according to the load current rate. Therefore, when the load change mode is determined to be a slow change mode, the gain parameter of the controller is lowered and the differential time is shortened according to the load current rate.

[0055] S33: When the load change mode is determined to be a sudden disturbance mode, the frequency and amplitude of the disturbance signal are extracted from the frequency spectrum of the load current data, and the adjustment frequency and switch duty cycle of the controller are optimized in combination with the feedforward control.

[0056] Specifically, when the spectrum analysis shows that there are periodic peaks or random high-frequency components in the load current, the system judges it as a sudden disturbance mode. For example, when high-frequency noise or periodic fluctuations appear in the load, the frequency and amplitude of the disturbance signal are extracted from the spectrum of the load current data, and the main characteristics of the noise or fluctuations are analyzed. The adjustment frequency and switching duty cycle of the controller are optimized in combination with feedforward control. The frequency of the switching circuit is adjusted according to the disturbance frequency to avoid resonance with the disturbance frequency, and the on-time and off-time of the switch tube are adjusted according to the disturbance amplitude to reduce the impact of noise on the output, effectively suppress the interference of high-frequency disturbances on the output, and enhance the stability of the system.

[0057] In one embodiment, if Figure 5 As shown, before step S40, that is, based on the real-time data of load changes, using a pre-trained load prediction model to predict the load change trend of the switching power supply in real time, obtaining a prediction result, and generating a disturbance compensation signal based on the prediction result, the control method of the switching power supply further includes: S401: Collect historical data of load changes, extract key features from the historical data of load changes, and construct an input feature vector. The key features include load current change rate, load voltage fluctuation amplitude, and power spectrum characteristics.

[0058] Specifically, the load change data of the switching power supply during operation is collected. These data include records of load current, voltage, power, etc. changing over time. Signal processing technology is used to extract features that have an important impact on load prediction from the collected raw data, and an input feature vector is constructed. The key features include the load current change rate, load voltage fluctuation amplitude, and power spectrum characteristics.

[0059] S402: Build a deep learning model using a long short-term memory network and a convolutional neural network, train the deep learning model based on an input feature vector, and use a back propagation algorithm to optimize the weight of the deep learning model to obtain a trained load prediction model.

[0060] Specifically, a hybrid model that can capture temporal and spatial features is constructed by combining long short-term memory networks and convolutional neural networks. The extracted input feature vectors are used to train the deep learning model. By continuously adjusting the model parameters, the model can accurately predict the load change trend. The back propagation algorithm is used to calculate the gradient of the loss function relative to the model weights, and the optimization algorithms such as the gradient descent method are used to update the weights to gradually reduce the prediction error, and finally the trained load prediction model is obtained.

[0061] S403: Optimize the hyperparameters of the trained load prediction model by combining the genetic algorithm and the gradient descent algorithm to obtain a pre-trained load prediction model.

[0062] Specifically, based on the trained load prediction model, the hyperparameters of the model are further optimized to improve the performance of the model. The hyperparameters of the trained load prediction model are optimized by combining the genetic algorithm and the gradient descent algorithm. The search space of the hyperparameters is defined, such as the learning rate is between 0.001 and 0.01, the number of network layers is 2 to 5, etc. A population is initialized, and each individual represents a set of hyperparameter combinations. A new generation of population is iteratively generated through operations such as selection, crossover, and mutation. The fitness of each individual is evaluated, and the iterative process is repeated to gradually eliminate individuals with poor performance, retain and optimize individuals with good performance. In the better hyperparameter region found by the genetic algorithm, the gradient descent method is further used to refine the hyperparameters, such as fine-tuning the learning rate and adjusting the weight decay coefficient. By calculating the gradient of the loss function, the hyperparameters are gradually adjusted to optimize the performance of the model on the validation set. The optimization results of the genetic algorithm and the gradient descent algorithm are combined to select the final hyperparameter combination. The model is retrained using the optimized hyperparameters to ensure the stability and prediction accuracy of the model, and finally a pre-trained load prediction model is obtained.

[0063] In one embodiment, if Figure 6 As shown, in step S40, based on the real-time data of load changes, a pre-trained load prediction model is used to predict the load change trend of the switching power supply in real time to obtain a prediction result, and based on the prediction result, a disturbance compensation signal is generated, which specifically includes: S41: The real-time data of load changes are input into a pre-trained load prediction model for reasoning and analysis, so as to predict the load change trend of the switching power supply in a future time period and obtain a prediction result.

[0064] Specifically, the real-time data of load changes are input into a pre-trained load prediction model for inference analysis, and the real-time data of load changes are filtered, denoised, normalized, and other processing is performed on the real-time data after preprocessing, and key features are extracted from the pre-processed real-time data, such as the load current change rate, load voltage fluctuation amplitude, power spectrum characteristics, etc. The key features in the pre-processed real-time data are compared and analyzed with the data in the pre-trained load prediction model to predict the load change trend of the switching power supply in the future time period and obtain the prediction result.

[0065] S42: Based on the prediction result, a disturbance compensation signal is calculated and input to the control end, where the disturbance compensation signal includes a voltage adjustment amount and a conduction timing adjustment amount.

[0066] Specifically, based on the predicted load change trend, the type and amplitude of disturbances that may occur in the future are determined. For example, if a sharp increase in load is predicted, the output voltage may need to be increased accordingly or the conduction timing may need to be adjusted to cope with it. Based on the predicted load change, the output voltage amplitude that needs to be adjusted is calculated to ensure that the output voltage can still be maintained within the set range after the load changes; based on the predicted load change rate and amplitude, the on and off timing of the switching element is adjusted, and the calculated voltage adjustment amount and conduction timing adjustment amount are transmitted to the control end of the switching power supply through the control interface to adjust the working state of the switching circuit in advance.

[0067] In one embodiment, if Figure 7 As shown, the control method of a switching power supply also includes: S60: During the operation of the switching power supply, the actual operation data of the switching power supply is continuously monitored, the actual operation data is compared with the data in the prediction result, and the parameters of the pre-trained load prediction model are updated.

[0068] Specifically, during the operation of the switching power supply, the actual operation data of the switching power supply is continuously monitored, and the actual operation data is compared with the data in the prediction results. By comparing the prediction results and the actual data, the accuracy of the prediction model is evaluated, and based on the comparison results, the parameters of the pre-trained load prediction model are updated.

[0069] S70: When it is detected that the operating state of the switching power supply is abnormal, a protection mechanism is triggered, the abnormal situation is analyzed, and a maintenance processing strategy is generated and transmitted to the user end.

[0070] Specifically, when the system detects abnormal operating status of the switching power supply, such as excessive voltage or sudden load change beyond the predicted range, the system triggers a protection mechanism to prevent further damage. After detecting the abnormality, the system analyzes the specific cause and generates a maintenance strategy. Through the communication module, the maintenance strategy is sent to the user's device end, such as a mobile phone app or control panel, to help users quickly locate problems and improve equipment maintenance efficiency.

[0071] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0072] In one embodiment, a control system of a switching power supply is provided, and the control system of the switching power supply corresponds one-to-one to a control method of a switching power supply in the above embodiment. Figure 8 As shown, the control system of the switching power supply includes a real-time monitoring module, a mode determination module, a parameter adjustment module, a compensation generation module and a state adjustment module. The detailed description of each functional module is as follows: The real-time monitoring module is used to monitor the voltage, current and load change rate and amplitude at the output end of the switching power supply, and obtain real-time data of load changes.

[0073] The mode determination module is used to identify the changes in the current load according to the real-time data of the load change and obtain the load change mode.

[0074] The parameter adjustment module is used to dynamically adjust the operating parameters of the controller according to the load change mode.

[0075] A compensation module is generated, which is used to predict the load change trend of the switching power supply in real time based on the real-time data of load change by using a pre-trained load prediction model, obtain the prediction result, and generate a disturbance compensation signal based on the prediction result.

[0076] The state adjustment module is used to adjust the switching state of the switching circuit of the switching power supply according to the control parameters adjusted by the controller and the disturbance compensation signal.

[0077] Optionally, the determination mode module includes: A short-time Fourier transform submodule is used to extract load current data from real-time data of load changes through a short-time Fourier transform algorithm; The load current rate calculation submodule is used to calculate the load current change rate formula based on the load current data: , calculate the rate of change of the load current, where R is the rate of change of the load current, ΔI is the amount of change of the load current, and Δt is the time period during which the load current changes; A threshold setting and mode division submodule, used to set a first set threshold and a second set threshold, used to distinguish load change modes, the load change modes including a rapid change mode, a slow change mode and a sudden disturbance mode; A rapid change mode determination submodule, configured to determine the load change mode as a rapid change mode when the change rate of the load current is higher than a first set threshold; A slow change mode determination submodule, used for determining the load change mode as the slow change mode when the change rate of the load current is between a first set threshold and a second set threshold; The burst disturbance mode determination submodule is used to analyze the frequency components in the load current data using Fourier transform to generate the spectrum of the load current data. When periodic peaks or random fluctuations appear in the spectrum of the load current data, the load change mode is determined to be a burst disturbance mode.

[0078] Optionally, the threshold setting and mode division submodule includes: A historical operation data acquisition unit is used to acquire historical operation data of the switching power supply, perform statistical analysis on the historical operation data of the switching power supply, and obtain a threshold range; The simulation experiment unit is used to obtain the load characteristics and response requirements of the switching power supply. Based on the load characteristics and response requirements of the switching power supply, a simulation experiment is performed in a virtual environment, and adjustments are made in combination with a threshold range to obtain a set threshold, which includes a first set threshold and a second set threshold.

[0079] Optionally, the parameter adjustment module includes: A rapid change mode adjustment submodule is used to increase the gain parameter of the controller and shorten the integration time according to the load current rate when the load change mode is determined to be a rapid change mode; A slow change mode adjustment submodule is used to reduce the gain parameter of the controller and shorten the differential time according to the load current rate when the load change mode is determined to be a slow change mode; The burst disturbance optimization submodule is used to extract the disturbance signal frequency and amplitude from the spectrum of the load current data when the load change mode is determined to be a burst disturbance mode, and optimize the adjustment frequency and switch duty cycle of the controller in combination with feedforward control.

[0080] Optionally, the control device of the switching power supply further includes: The historical data collection module is used to collect historical data of load changes, extract key features from the historical data of load changes, and construct an input feature vector. The key features include the load current change rate, load voltage fluctuation amplitude, and power spectrum characteristics. The model construction and training module is used to construct a deep learning model using a long short-term memory network and a convolutional neural network, train the deep learning model based on the input feature vector, and use the back propagation algorithm to optimize the weight of the deep learning model to obtain a trained load prediction model. The hyperparameter optimization module is used to optimize the hyperparameters of the trained load prediction model by combining the genetic algorithm and the gradient descent algorithm to obtain a pre-trained load prediction model.

[0081] Optionally, generating a compensation module includes: The load trend prediction submodule is used to input the real-time data of load changes into the pre-trained load prediction model for reasoning and analysis, predict the load change trend of the switching power supply in the future time period, and obtain the prediction result; The compensation signal generation submodule is used to calculate a disturbance compensation signal based on the prediction result, and input the disturbance compensation signal to the control end. The disturbance compensation signal includes a voltage adjustment amount and a conduction timing adjustment amount.

[0082] Optionally, the control device of the switching power supply further includes: An operation data monitoring module is used to continuously monitor the actual operation data of the switching power supply during its operation, compare the actual operation data with the data in the prediction results, and update the parameters of the pre-trained load prediction model; The abnormality detection module is used to trigger the protection mechanism when it detects that the operating status of the switching power supply is abnormal, analyze the abnormal situation, generate a maintenance processing strategy and transmit it to the user end.

[0083] For the specific definition of a control system of a switching power supply, please refer to the definition of a control method of a switching power supply in the above text, which will not be repeated here. Each module in the above-mentioned control system of a switching power supply can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0084] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Fig. 9As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a control method for a switching power supply is implemented.

[0085] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program: Monitor the voltage, current, load change rate and amplitude at the output end of the switching power supply to obtain real-time data of load changes; identify the changes in the current load based on the real-time data of load changes to obtain the load change pattern; Dynamically adjust the operating parameters of the controller according to the load change mode; Based on the real-time data of load changes, a pre-trained load prediction model is used to predict the load change trend of the switching power supply in real time, obtain the prediction result, and generate a disturbance compensation signal based on the prediction result; According to the control parameters adjusted by the controller and the disturbance compensation signal, the switching state of the switching circuit of the switching power supply is adjusted.

[0086] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented: Monitor the voltage, current, load change rate and amplitude at the output end of the switching power supply to obtain real-time data of load changes; identify the changes in the current load based on the real-time data of load changes to obtain the load change pattern; Dynamically adjust the operating parameters of the controller according to the load change mode; Based on the real-time data of load changes, a pre-trained load prediction model is used to predict the load change trend of the switching power supply in real time, obtain the prediction result, and generate a disturbance compensation signal based on the prediction result; According to the control parameters adjusted by the controller and the disturbance compensation signal, the switching state of the switching circuit of the switching power supply is adjusted.

[0087] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0088] Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above.

[0089] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for controlling a switching power supply, characterized in that: The control method of a switching power supply comprises: Monitor the voltage, current and load change rate and amplitude at the output end of the switching power supply, and obtain real-time data of the load change; According to the real-time data of the load change, identifying the change of the current load and obtaining the load change pattern; Dynamically adjusting operating parameters of the controller according to the load change mode; Based on the real-time data of the load change, a pre-trained load prediction model is used to predict the load change trend of the switching power supply in real time to obtain a prediction result, and a disturbance compensation signal is generated based on the prediction result; The switching state of the switching circuit of the switching power supply is adjusted according to the control parameter adjusted by the controller and the disturbance compensation signal.

2. A method for controlling a switching power supply according to claim 1, characterized in that: The step of identifying the change of the current load according to the real-time data of the load change and obtaining the load change mode includes: Extracting load current data from the real-time data of load changes by using a short-time Fourier transform algorithm; According to the load current data, the load current change rate formula is: , calculate the rate of change of the load current, where R is the rate of change of the load current, ΔI is the amount of change of the load current, and Δt is the time period during which the load current changes; Setting a first set threshold and a second set threshold to distinguish the load change mode, wherein the load change mode includes a rapid change mode, a slow change mode and a sudden disturbance mode; When the change rate of the load current is higher than the first set threshold, the load change mode is determined to be the rapid change mode; When the change rate of the load current is between the first set threshold and the second set threshold, the load change mode is determined to be the slow change mode; The frequency components in the load current data are analyzed using Fourier transform to generate a spectrum of the load current data. When periodic peaks or random fluctuations appear in the spectrum of the load current data, the load change mode is determined to be the sudden disturbance mode.

3. A method for controlling a switching power supply according to claim 2, characterized in that: The setting of the first set threshold and the second set threshold specifically includes: Acquiring historical operation data of the switching power supply, performing statistical analysis on the historical operation data of the switching power supply, and obtaining a threshold range; The load characteristics and response requirements of the switching power supply are obtained, and based on the load characteristics and response requirements of the switching power supply, a simulation experiment is performed in a virtual environment, and adjustments are made in combination with the threshold range to obtain a set threshold, wherein the set threshold includes the first set threshold and the second set threshold.

4. A method for controlling a switching power supply according to claim 2, characterized in that: The dynamically adjusting the operating parameters of the controller according to the load change mode includes: When the load change mode is determined to be the rapid change mode, increasing the gain parameter of the controller and shortening the integration time according to the load current rate; When the load change mode is determined to be the slow change mode, reducing the gain parameter of the controller and shortening the differential time according to the load current rate; When the load change mode is determined to be the sudden disturbance mode, the frequency and amplitude of the disturbance signal are extracted from the frequency spectrum of the load current data, and the regulation frequency and switch duty cycle of the controller are optimized in combination with feedforward control.

5. The control method of a switching power supply according to claim 1, characterized in that: Before the real-time data of the load change is used to predict the load change trend of the switching power supply in real time using a pre-trained load prediction model to obtain a prediction result, and before a disturbance compensation signal is generated based on the prediction result, the control method of the switching power supply further includes: Collecting historical data of the load change, extracting key features from the historical data of the load change, and constructing an input feature vector, wherein the key features include a load current change rate, a load voltage fluctuation amplitude, and a power spectrum characteristic; A deep learning model is constructed by using a long short-term memory network and a convolutional neural network, the deep learning model is trained based on the input feature vector, and a back propagation algorithm is used to optimize the weight of the deep learning model to obtain a trained load prediction model; The hyperparameters of the trained load prediction model are optimized by combining a genetic algorithm and a gradient descent algorithm to obtain the pre-trained load prediction model.

6. The control method of a switching power supply according to claim 1, characterized in that: The method of predicting the load change trend of the switching power supply in real time based on the real-time data of the load change by using a pre-trained load prediction model to obtain a prediction result, and generating a disturbance compensation signal based on the prediction result includes: Inputting the real-time data of the load change into the pre-trained load prediction model for reasoning and analysis, predicting the load change trend of the switching power supply in a future time period, and obtaining the prediction result; Based on the prediction result, the disturbance compensation signal is calculated and input to the control end, wherein the disturbance compensation signal includes a voltage adjustment amount and a conduction timing adjustment amount.

7. The method for controlling a switching power supply according to claim 1, characterized in that: The control method of a switching power supply further includes: During the operation of the switching power supply, the actual operation data of the switching power supply is continuously monitored, the actual operation data is compared with the data in the prediction result, and the parameters of the pre-trained load prediction model are updated; When it is detected that the operating state of the switching power supply is abnormal, the protection mechanism is triggered, the abnormal situation is analyzed, and a maintenance processing strategy is generated and transmitted to the user end.

8. A control system for a switching power supply, characterized in that: The control system of a switching power supply comprises: A real-time monitoring module is used to monitor the voltage, current and load change rate and amplitude at the output end of the switching power supply, and obtain real-time data of the load change; A mode determination module, used to identify the change of the current load according to the real-time data of the load change, and obtain the load change mode; A parameter adjustment module, used to dynamically adjust the operating parameters of the controller according to the load change mode; A compensation generation module is used to predict the load change trend of the switching power supply in real time based on the real-time data of the load change and using a pre-trained load prediction model to obtain a prediction result, and generate a disturbance compensation signal based on the prediction result; an adjustment state module is used to adjust the switching state of the switching circuit of the switching power supply according to the control parameters adjusted by the controller and the disturbance compensation signal.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method for controlling a switching power supply as claimed in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for controlling a switching power supply as claimed in any one of claims 1 to 7 are implemented.

Images