Adaptive control method, system and equipment of switching power supply and storage medium

Abstract

The invention relates to an adaptive control method, system and device for a switching power supply and a storage medium, and the method comprises the steps: obtaining power supply parameter data in real time, and carrying out the preprocessing of the power supply parameter data, and obtaining standard state data; performing load analysis on the standard state data to obtain real-time load data, and determining a corresponding power supply modulation mode according to the real-time load data; obtaining standard state data in a preset time period, and inputting the standard state data into a pre-trained load prediction model to obtain a load prediction result; and determining a corresponding prediction modulation mode based on the load prediction result, and pre-inputting the prediction modulation mode by adopting a smooth switching algorithm so as to improve the response speed and stability of the system. The invention 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 power supply control, and in particular to an adaptive control method, system, device and storage medium for a switching power supply. Background Art

[0002] Existing switching power supply control methods mostly use PWM (pulse width modulation) or PFM (pulse frequency modulation) technology. However, since PWM needs to respond by adjusting the duty cycle when the load changes, it leads to response delay and low efficiency. In particular, fixed high-frequency switching brings unnecessary losses and electromagnetic interference under light load. PFM controls the output voltage through frequency conversion. Although it is more efficient under light load conditions, it has low control accuracy and is prone to output voltage fluctuations and instability when the load changes suddenly.

[0003] The above-mentioned existing technical solutions have the following defects: the existing switching power supply control technology is difficult to balance between fast response and high efficiency and stability. Especially when the load changes rapidly, problems such as power supply fluctuations and response delays often occur, so 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 an adaptive 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 adaptively controlling a switching power supply, the method comprising: Acquire power supply parameter data in real time, and pre-process the power supply parameter data to obtain standard state data; Performing load analysis on the standard state data to obtain real-time load data, and determining a corresponding power supply modulation mode according to the real-time load data; Acquire standard state data within a preset time period, and input the standard state data into a pre-trained load prediction model to obtain a load prediction result; A corresponding prediction modulation mode is determined based on the load prediction result, and a smooth switching algorithm is used to pre-input the prediction modulation mode to improve the response speed and stability of the system.

[0006] By adopting the above technical solution, by acquiring power supply parameter data in real time and preprocessing it, the accuracy and consistency of the input data can be ensured, thereby improving the accuracy of subsequent analysis and processing; by performing load analysis on the standard state data, the current load condition of the power supply can be timely understood, and the corresponding power supply modulation method can be determined according to the real-time load data, which can effectively improve the response speed and stability of the power supply; by acquiring the standard state data within a preset time period and inputting the pre-trained load prediction model, the load change can be predicted in advance, the corresponding modulation method can be determined according to the prediction results, and a smooth switching algorithm can be used for pre-input, ensuring that the system can quickly adapt to load changes and improving the response speed and stability of the system.

[0007] In one example, the present application may be further configured as follows: the preprocessing of the power supply parameter data to obtain standard state data specifically includes: Performing noise filtering on the power parameter data to remove invalid data in the power parameter data to obtain continuous parameter data; The continuous parameter data is subjected to data standardization processing to obtain the standard state data, so as to improve the uniformity of data input.

[0008] By adopting the above technical solution, by filtering the noise of the power supply parameter data, invalid data can be removed to ensure the purity and reliability of the data; by standardizing the continuous parameter data, the uniformity of data input can be improved, thereby providing a stable foundation for subsequent analysis and processing, and ensuring the accuracy of load analysis and modulation method determination.

[0009] In one example, the present application may be further configured as follows: performing load analysis on the standard state data to obtain real-time load data specifically includes: Extracting features from the standard state data to obtain corresponding power state fluctuation data, including voltage fluctuation data and load current change rate; A power supply load model is constructed based on the power supply state fluctuation data, and the real-time load data is determined by the power supply load model.

[0010] By adopting the above technical solution, by extracting features from standard state data, it is possible to accurately obtain the fluctuation data of the power supply state, including voltage fluctuations and load current change rates; by building a power supply load model based on these fluctuation data, it is possible to effectively simulate the load conditions of the power supply, and determine the real-time load data through the model, thereby improving the accuracy and real-time performance of load analysis, thereby being able to more accurately adjust the power supply modulation method.

[0011] In one example, the present application may be further configured as follows: the construction of the load prediction model specifically includes: Acquire historical standard state data and historical load data corresponding to the historical standard state data, and classify the historical standard state data according to the historical load data to obtain a positive training data set and a negative training data set; The load prediction model is obtained by performing positive training on the preset machine learning algorithm according to the positive training data set, and performing negative training on the preset machine learning algorithm according to the negative training data set.

[0012] By adopting the above technical solution, a comprehensive training set can be formed by acquiring historical standard state data and corresponding historical load data. Positive and negative training of the machine learning algorithm based on these data can effectively improve the accuracy and reliability of the load prediction model. Through classification training of historical data, the law of power load changes can be better understood, thereby improving the accuracy of load prediction and providing a reliable basis for adjusting the real-time modulation method.

[0013] In one example, the present application may be further configured as follows: the historical standard state data is classified according to the historical load data to obtain a positive training data set and a negative training data set, specifically including: Generate a corresponding load change image according to the historical load data, select abnormal change data from the load change image, determine the historical load data corresponding to the abnormal change data as negative load data, and determine other historical load data as positive load data; The negative load data and the historical standard state data corresponding to the negative load data are encapsulated as the negative training data set, and the positive load data and the historical standard state data corresponding to the positive load data are encapsulated as the positive training data set.

[0014] By adopting the above technical solution, by generating load change images and selecting abnormal change data therefrom, the abnormal situations in the historical load data can be effectively identified and classified. By using these abnormal data for negative training, the model's ability to recognize sudden load changes can be improved. By classifying and encapsulating positive and negative training data sets, the load prediction model can be trained more accurately, thereby improving its prediction accuracy and reliability under different load conditions.

[0015] In one example, the present application may be further configured as follows: the step of pre-inputting the predicted modulation mode by using a smooth switching algorithm specifically includes: Adopt interpolation algorithm to gradually adjust modulation frequency and duty cycle to ensure smooth transition of modulation mode; The output voltage and the load current in the modulation process are monitored in real time, and when the output voltage and the load current in the modulation process are abnormal, the modulation frequency and the duty cycle are reduced.

[0016] By adopting the above technical solution and gradually adjusting the modulation frequency and duty cycle by using an interpolation algorithm, a smooth transition of the modulation mode can be achieved, avoiding power supply instability caused by sudden changes; by real-time monitoring of the output voltage and load current during the modulation process, abnormal conditions during the modulation process can be discovered and corrected in time, and by reducing the modulation frequency and duty cycle, the stability and response speed of the power supply system can be further improved, thereby ensuring the stability and reliability of the power supply output.

[0017] The second object of the invention is achieved by the following technical solutions: An adaptive control system for a switching power supply, the adaptive control system for the switching power supply comprising: A data processing module, used for acquiring power supply parameter data in real time, and preprocessing the power supply parameter data to obtain standard state data; A load analysis module, used to perform load analysis on the standard state data to obtain real-time load data, and determine a corresponding power supply modulation mode according to the real-time load data; A model prediction module is used to obtain standard state data within a preset time period and input the standard state data into a pre-trained load prediction model to obtain a load prediction result; The modulation determination module is used to determine the corresponding predicted modulation mode based on the load prediction result, and pre-input the predicted modulation mode using a smooth switching algorithm to improve the response speed and stability of the system.

[0018] By adopting the above technical solution, by acquiring power supply parameter data in real time and preprocessing it, the accuracy and consistency of the input data can be ensured, thereby improving the accuracy of subsequent analysis and processing; by performing load analysis on the standard state data, the current load condition of the power supply can be timely understood, and the corresponding power supply modulation method can be determined according to the real-time load data, which can effectively improve the response speed and stability of the power supply; by acquiring the standard state data within a preset time period and inputting the pre-trained load prediction model, the load change can be predicted in advance, the corresponding modulation method can be determined according to the prediction results, and a smooth switching algorithm can be used for pre-input, ensuring that the system can quickly adapt to load changes and improving the response speed and stability of the system.

[0019] 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 adaptive control method of the switching power supply when executing the computer program.

[0020] 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 adaptive control method of the switching power supply are implemented.

[0021] In summary, this application includes the following beneficial technical effects: 1. By acquiring power parameter data in real time and preprocessing it, the accuracy and consistency of input data can be ensured, thereby improving the accuracy of subsequent analysis and processing; by performing load analysis on standard state data, the current load condition of the power supply can be timely understood, and the corresponding power modulation method can be determined according to the real-time load data, which can effectively improve the response speed and stability of the power supply; by acquiring standard state data within a preset time period and inputting the pre-trained load prediction model, the load change can be predicted in advance, the corresponding modulation method can be determined according to the prediction results, and the smooth switching algorithm is used for pre-input, ensuring that the system can quickly adapt to load changes, thereby improving the response speed and stability of the system; 2. By extracting features from standard state data, the fluctuation data of the power supply state can be accurately obtained, including voltage fluctuation and load current change rate. The power supply load model is constructed based on these fluctuation data, which can effectively simulate the load condition of the power supply, and the real-time load data can be determined through the model, which improves the accuracy and real-time performance of load analysis, so that the power supply modulation method can be adjusted more accurately. 3. By using the interpolation algorithm to gradually adjust the modulation frequency and duty cycle, a smooth transition of the modulation mode can be achieved to avoid power instability caused by sudden changes; by real-time monitoring of the output voltage and load current during the modulation process, abnormal conditions in the modulation process can be discovered and corrected in time, and by reducing the modulation frequency and duty cycle, the stability and response speed of the power supply system can be further improved, thereby ensuring the stability and reliability of the power supply output. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is a flow chart of an adaptive control method of a switching power supply in an embodiment of the present application; Figure 2 is a flowchart for implementing step S10 in the adaptive control method of a switching power supply in one embodiment of the present application; Figure 3 is a flowchart for implementing step S20 in the adaptive control method of a switching power supply in one embodiment of the present application; Figure 4 is a flowchart for implementing step S30 in the adaptive control method of a switching power supply in one embodiment of the present application; Figure 5is a flowchart for implementing step S301 in the adaptive control method of a switching power supply in one embodiment of the present application; Figure 6 is a flowchart for implementing step S40 in the adaptive control method of a switching power supply in one embodiment of the present application; Figure 7 It is a principle block diagram of an adaptive control system of a switching power supply in one embodiment of the present application; Figure 8 It is a schematic diagram of a device in an embodiment of the present application. DETAILED DESCRIPTION

[0023] The present application is further described in detail below in conjunction with the accompanying drawings.

[0024] In one embodiment, if Figure 1 As shown, the present application discloses an adaptive control method for a switching power supply, which specifically includes the following steps: S10: acquiring power parameter data in real time, and preprocessing the power parameter data to obtain standard state data.

[0025] Specifically, high-precision sensors are used to collect real-time parameter data such as the output voltage, load current and ambient temperature of the power supply, and the data are transmitted to the data preprocessing module. The collected power parameter data is filtered through a denoising algorithm to remove invalid data to ensure the purity and continuity of the data. The continuous parameter data is then normalized to standardize data of different dimensions into data of unified dimensions, so that the data is consistent and comparable, and finally standard state data is obtained.

[0026] S20: Perform load analysis on the standard state data to obtain real-time load data, and determine a corresponding power supply modulation method according to the real-time load data.

[0027] Specifically, the standard state data obtained by preprocessing is input into the load analysis module. First, feature extraction is performed to extract key features in the power supply state, including voltage fluctuation data and load current change rate, etc. Then, based on the extracted feature data, a power supply load model is constructed using statistical analysis methods and pattern recognition algorithms. The real-time data is analyzed through the model to identify the current load state, such as light load, heavy load or sudden load, etc. According to the identified load state, a suitable power supply modulation method is selected. For example, when the load is light, the PFM modulation method is selected to improve efficiency; when the load is heavy, the PWM modulation method is selected to ensure stability.

[0028] S30: Acquire standard state data within a preset time period, and input the standard state data into a pre-trained load prediction model to obtain a load prediction result.

[0029] Specifically, within a preset time period, power supply parameter data is continuously collected and preprocessed to obtain standard state data within the corresponding time period. These data are input into a pre-trained load prediction model as input data to predict future load change trends and generate corresponding load prediction results, indicating possible future load changes and providing a basis for subsequent adjustments to the modulation method.

[0030] S40: Determine a corresponding prediction modulation mode based on the load prediction result, and pre-input the prediction modulation mode using a smooth switching algorithm to improve the response speed and stability of the system.

[0031] Specifically, based on the load prediction results, the load state of the power supply in the future is determined, and the appropriate power modulation method is selected. In order to avoid sudden changes and instability during the switching of the modulation method, a smooth switching algorithm is used to gradually adjust the modulation parameters to ensure a smooth transition of the switching process.

[0032] In one embodiment, if Figure 2 As shown, in step S10, the power supply parameter data is preprocessed to obtain standard state data, which specifically includes: S11: performing noise filtering on the power parameter data to remove invalid data in the power parameter data to obtain continuous parameter data.

[0033] Specifically, a filtering algorithm is used to process the collected power parameter data, such as mean filtering, median filtering and Kalman filtering, and according to the characteristics of the power parameter data, a suitable filtering algorithm is selected for noise filtering to remove high-frequency noise and random interference in the data, retain valid power parameter information, obtain continuous and clean parameter data, and ensure the accuracy and stability of subsequent data processing.

[0034] S12: Perform data standardization processing on the continuous parameter data to obtain standard state data to improve the uniformity of data input.

[0035] Specifically, data standardization methods are used to process continuous parameter data, such as minimum-maximum standardization and Z-score standardization, to convert power parameter data of different dimensions into a unified dimension range, eliminate the dimensional differences between different parameters, and make the standardized data have a consistent scale, which is convenient for subsequent feature extraction and analysis, ensuring the uniformity and comparability of data input, and finally obtaining standard state data.

[0036] In one embodiment, if Figure 3 As shown, in step S20, load analysis is performed on the standard state data to obtain real-time load data, which specifically includes: S21: Extract features from the standard state data to obtain corresponding power state fluctuation data, including voltage fluctuation data and load current change rate.

[0037] Specifically, signal processing technology is used to extract features from standard state data. Through methods such as Fourier transform and wavelet transform, the frequency domain characteristics and time domain characteristics of voltage fluctuations are extracted from the standard state data. At the same time, the rate of change of load current is calculated, and the amplitude and frequency of current fluctuations are analyzed to obtain key characteristic data of the power supply state. This describes the fluctuations of voltage and current during power supply operation and provides basic data for subsequent load analysis.

[0038] S22: Building a power load model based on the power state fluctuation data, and determining real-time load data through the power load model.

[0039] Specifically, the extracted power state fluctuation data is used to construct a power load model using statistical analysis and pattern recognition algorithms, and a mapping relationship between the power state fluctuation data and the actual load state is established. The real-time power state fluctuation data is then input into the model to predict the current load state and obtain real-time load data.

[0040] In one embodiment, if Figure 4 As shown, in step S30, i.e., construction of a load prediction model, specifically includes: S301: Acquire historical standard state data and historical load data corresponding to the historical standard state data, and classify the historical standard state data according to the historical load data to obtain a positive training data set and a negative training data set.

[0041] Specifically, standard state data and corresponding load data are extracted from the historical operation records of the power supply to establish a historical data set. According to the changing trends and characteristics of the historical load data, they are classified into positive load data and negative load data. Positive load data represents a normal or expected load state, and negative load data represents an abnormal or sudden load state. Through data classification, positive training data sets and negative training data sets are formed as training samples for the machine learning model, providing basic data for model training.

[0042] S302: Perform positive training on a preset machine learning algorithm according to a positive training data set, and perform negative training on a preset machine learning algorithm according to a negative training data set to obtain a load prediction model.

[0043] Specifically, a machine learning algorithm suitable for load prediction is selected, such as regression analysis, decision tree or neural network, and the positive training data set is input into the machine learning algorithm for positive training to learn the characteristic patterns and change laws under normal load conditions. Similarly, the negative training data set is input into the machine learning algorithm for negative training to learn the characteristic patterns and change laws under abnormal load conditions, so as to fully understand and predict different load conditions and finally obtain a load prediction model.

[0044] In one embodiment, if Figure 5 As shown, in step S301, the historical standard state data is classified according to the historical load data to obtain a positive training data set and a negative training data set, which specifically include: S3011: Generate a corresponding load change image according to historical load data, select abnormal change data from the load change image, determine the historical load data corresponding to the abnormal change data as negative load data, and determine other historical load data as positive load data.

[0045] Specifically, historical load data is used to generate load change images in time series, and image processing technology is used to analyze the changing trends and patterns of the load data. In the load change images, abnormal points of load change are identified, such as protruding points that appear during a smooth load process, and the historical load data corresponding to these abnormal points are extracted as negative load data. Other normally changing data are determined as positive load data, thereby effectively distinguishing between normal and abnormal load states and ensuring the accuracy and effectiveness of the training data.

[0046] S3012: Encapsulate the negative load data and the historical standard state data corresponding to the negative load data into a negative training data set, and encapsulate the positive load data and the historical standard state data corresponding to the positive load data into a positive training data set.

[0047] Specifically, the negative load data is paired with its corresponding historical standard state data and encapsulated into a negative training data set. The positive load data is paired with its corresponding historical standard state data and encapsulated into a positive training data set to ensure a one-to-one correspondence between the load data and the standard state data in the training data set, providing complete and consistent sample data for the training of the machine learning model.

[0048] In one embodiment, if Figure 6 As shown, in step S40, the predicted modulation mode is pre-input using a smooth switching algorithm, which specifically includes: S41: Use interpolation algorithm to gradually adjust the modulation frequency and duty cycle to ensure smooth transition of the modulation mode.

[0049] Specifically, based on the difference between the predicted modulation mode and the current modulation mode, interpolation algorithms such as linear interpolation and cubic spline interpolation are used to calculate the intermediate values ​​of the modulation frequency and duty cycle during the transition period. During the modulation mode switching process, the modulation parameters are gradually adjusted to gradually approach the target values, avoiding instability caused by sudden changes in parameters.

[0050] S42: monitor the output voltage and load current in real time during the modulation process, and reduce the modulation frequency and duty cycle when the output voltage and load current are abnormal during the modulation process.

[0051] Specifically, during the switching of modulation modes, sensors are used to monitor the data changes of output voltage and load current in real time. If abnormal fluctuations in output voltage and load current are detected, timely adjustments are made, such as lowering the modulation frequency or reducing the duty cycle, to alleviate the impact of load changes on the power supply and ensure the stability and reliability of the system during the switching of modulation modes.

[0052] 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.

[0053] In one embodiment, an adaptive control system of a switching power supply is provided, and the adaptive control system of the switching power supply corresponds one-to-one to the adaptive control method of the switching power supply in the above embodiment. Figure 7 As shown, the adaptive control system of the switching power supply includes a data processing module, a load analysis module, a model prediction module, and a modulation determination module. The detailed description of each functional module is as follows: A data processing module is used to obtain power supply parameter data in real time and pre-process the power supply parameter data to obtain standard state data; The load analysis module is used to perform load analysis on the standard state data, obtain real-time load data, and determine the corresponding power supply modulation mode according to the real-time load data; The model prediction module is used to obtain standard state data within a preset time period and input the standard state data into a pre-trained load prediction model to obtain a load prediction result; The modulation determination module is used to determine the corresponding predicted modulation mode based on the load prediction result, and pre-input the predicted modulation mode using a smooth switching algorithm to improve the response speed and stability of the system.

[0054] Optionally, the data processing module specifically includes: A noise filtering submodule is used to filter the power parameter data for noise, remove invalid data in the power parameter data, and obtain continuous parameter data; The data standardization submodule is used to perform data standardization processing on continuous parameter data to obtain standard state data to improve the uniformity of data input.

[0055] Optionally, the load analysis module specifically includes: The feature extraction submodule is used to extract features from the standard state data to obtain corresponding power state fluctuation data, including voltage fluctuation data and load current change rate; The load model building submodule is used to build a power load model based on the power state fluctuation data, and determine the real-time load data through the power load model.

[0056] Optionally, the construction of a load prediction model includes: A historical data acquisition submodule is used to acquire historical standard state data and historical load data corresponding to the historical standard state data, and classify the historical standard state data according to the historical load data to obtain a positive training data set and a negative training data set; The model training submodule is used to perform positive training on the preset machine learning algorithm according to the positive training data set, and to perform negative training on the preset machine learning algorithm according to the negative training data set to obtain a load prediction model.

[0057] Optionally, the historical data acquisition submodule specifically includes: a classification processing unit, configured to generate a corresponding load change image according to the historical load data, select abnormal change data from the load change image, determine the historical load data corresponding to the abnormal change data as negative load data, and determine other historical load data as positive load data; The training data encapsulation unit is used to encapsulate the negative load data and the historical standard state data corresponding to the negative load data into a negative training data set, and to encapsulate the positive load data and the historical standard state data corresponding to the positive load data into a positive training data set.

[0058] Optionally, the modulation determination module specifically includes: The smooth switching algorithm submodule is used to gradually adjust the modulation frequency and duty cycle using an interpolation algorithm to ensure a smooth transition of the modulation mode; The real-time monitoring submodule is used to monitor the output voltage and load current in the modulation process in real time, and reduce the modulation frequency and duty cycle when the output voltage and load current in the modulation process are abnormal.

[0059] The specific definition of the adaptive control system of the switching power supply can be found in the definition of the adaptive control method of the switching power supply in the above text, which will not be repeated here. Each module in the adaptive control system of the switching power supply can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above 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 each of the above modules.

[0060] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 8 As 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, an adaptive control method for a switching power supply is implemented.

[0061] 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: Acquire power supply parameter data, and pre-process the power supply parameter data to obtain standard state data; Perform load analysis on standard state data to obtain real-time load data, and determine the corresponding power modulation method based on the real-time load data; Obtaining standard state data within a preset time period, and inputting the standard state data into a pre-trained load prediction model to obtain a load prediction result; The corresponding prediction modulation mode is determined based on the load prediction result, and the prediction modulation mode is pre-input using a smooth switching algorithm to improve the response speed and stability of the system.

[0062] 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: Acquire power supply parameter data, and pre-process the power supply parameter data to obtain standard state data; Perform load analysis on standard state data to obtain real-time load data, and determine the corresponding power modulation method based on the real-time load data; Obtaining standard state data within a preset time period, and inputting the standard state data into a pre-trained load prediction model to obtain a load prediction result; The corresponding prediction modulation mode is determined based on the load prediction result, and the prediction modulation mode is pre-input using a smooth switching algorithm to improve the response speed and stability of the system.

[0063] 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 may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may 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).

[0064] 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.

[0065] 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. An adaptive control method for a switching power supply, characterized in that: The adaptive control method of the switching power supply comprises: Acquire power supply parameter data in real time, and pre-process the power supply parameter data to obtain standard state data; Performing load analysis on the standard state data to obtain real-time load data, and determining a corresponding power supply modulation mode according to the real-time load data; Acquire standard state data within a preset time period, and input the standard state data into a pre-trained load prediction model to obtain a load prediction result; A corresponding prediction modulation mode is determined based on the load prediction result, and a smooth switching algorithm is used to pre-input the prediction modulation mode to improve the response speed and stability of the system.

2. The adaptive control method of the switching power supply according to claim 1, characterized in that: The preprocessing of the power supply parameter data to obtain standard state data specifically includes: Performing noise filtering on the power parameter data to remove invalid data in the power parameter data to obtain continuous parameter data; The continuous parameter data is subjected to data standardization processing to obtain the standard state data, so as to improve the uniformity of data input.

3. The adaptive control method of the switching power supply according to claim 1, characterized in that: The load analysis of the standard state data to obtain real-time load data specifically includes: Extracting features from the standard state data to obtain corresponding power state fluctuation data, including voltage fluctuation data and load current change rate; A power supply load model is constructed based on the power supply state fluctuation data, and the real-time load data is determined by the power supply load model.

4. The adaptive control method of a switching power supply according to claim 1, characterized in that: The construction of the load prediction model specifically includes: Acquire historical standard state data and historical load data corresponding to the historical standard state data, and classify the historical standard state data according to the historical load data to obtain a positive training data set and a negative training data set; The load prediction model is obtained by performing positive training on the preset machine learning algorithm according to the positive training data set, and performing negative training on the preset machine learning algorithm according to the negative training data set.

5. The adaptive control method of the switching power supply according to claim 4, characterized in that: The classifying the historical standard state data according to the historical load data to obtain a positive training data set and a negative training data set specifically includes: Generate a corresponding load change image according to the historical load data, select abnormal change data from the load change image, determine the historical load data corresponding to the abnormal change data as negative load data, and determine other historical load data as positive load data; The negative load data and the historical standard state data corresponding to the negative load data are encapsulated as the negative training data set, and the positive load data and the historical standard state data corresponding to the positive load data are encapsulated as the positive training data set.

6. The adaptive control method of a switching power supply according to claim 1, characterized in that: The adopting a smooth switching algorithm to pre-input the predicted modulation mode specifically includes: Adopt interpolation algorithm to gradually adjust modulation frequency and duty cycle to ensure smooth transition of modulation mode; The output voltage and the load current in the modulation process are monitored in real time, and when the output voltage and the load current in the modulation process are abnormal, the modulation frequency and the duty cycle are reduced.

7. An adaptive control system for a switching power supply, characterized in that: The adaptive control system of the switching power supply comprises: a data processing module, which is used to obtain power supply parameter data in real time and pre-process the power supply parameter data to obtain standard state data; A load analysis module, used to perform load analysis on the standard state data to obtain real-time load data, and determine a corresponding power supply modulation mode according to the real-time load data; A model prediction module is used to obtain standard state data within a preset time period and input the standard state data into a pre-trained load prediction model to obtain a load prediction result; The modulation determination module is used to determine the corresponding predicted modulation mode based on the load prediction result, and pre-input the predicted modulation mode using a smooth switching algorithm to improve the response speed and stability of the system.

8. The adaptive control system of the switching power supply according to claim 7, characterized in that: The data processing module specifically includes: A noise filtering submodule, used for performing noise filtering on the power parameter data, removing invalid data in the power parameter data, and obtaining continuous parameter data; The data standardization submodule is used to perform data standardization processing on the continuous parameter data to obtain the standard state data to improve the uniformity of data input.

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 adaptive control method of the switching power supply according to any one of claims 1 to 6 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 adaptive control method of the switching power supply according to any one of claims 1 to 6 are implemented.

Images