Photovoltaic module adaptive MPPT optimization method, system and device and storage medium

Abstract

The invention relates to the technical field of photovoltaic power generation, and discloses a photovoltaic module adaptive MPPT optimization method, system and device and a storage medium, and the method comprises the steps: collecting the multi-dimensional parameter information of a photovoltaic module through adaptive weight distribution, generating a weighted original signal, and generating a filtering signal through dynamic filtering processing; on the basis of the filtering signals, principal component feature vectors are generated through principal component analysis, the illumination intensity is solved, and prediction power is calculated through a multi-parameter coupling prediction model; based on the predicted power, the control action is optimized through model predictive control, the reference voltage is optimized, the multi-parameter anomaly detection weight is adjusted based on the self-adaptive fuzzy logic, predictive fault correction is executed, and maximum power point tracking is achieved. According to the invention, through adaptive weight distribution and dynamic Gaussian filtering, noise interference is effectively reduced; power output is accurately predicted in combination with principal component analysis and a multi-parameter coupling prediction model; the reference voltage is optimized through model prediction control, and rapid and accurate maximum power point tracking is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic power generation, and in particular to a method, system, device and storage medium for adaptive MPPT optimization of photovoltaic modules. Background Art

[0002] Photovoltaic power generation, as an important form of renewable energy, is widely used in distributed power generation and grid-connected systems. Maximum power point tracking (MPPT) technology is the core of improving the efficiency of photovoltaic systems. It aims to dynamically adjust the operating voltage to track the maximum power point. Traditional MPPT methods, such as the perturbation observation method and the conductance increment method, perform well in stable environments. However, under complex conditions such as light fluctuations, temperature changes, or partial shading, they are prone to falling into local maximum power points, resulting in reduced efficiency. In recent years, MPPT methods based on intelligent algorithms such as fuzzy logic and neural networks have gradually emerged, but these methods mostly rely on single parameters such as voltage and current, and it is difficult to fully capture the coupling effects of multi-dimensional parameters such as temperature and impedance. In addition, existing technologies lack a dynamic filtering mechanism in signal processing, making it difficult to effectively deal with noise interference, and their fault detection and correction capabilities are limited, affecting the robustness of the system.

[0003] Therefore, an MPPT optimization method that integrates multi-parameter coupling, dynamic signal processing, and intelligent control is urgently needed to improve the power tracking accuracy and stability of photovoltaic systems in complex environments. This paper proposes an MPPT optimization scheme based on adaptive weight allocation, dynamic Gaussian filtering, principal component analysis, model predictive control, and fuzzy logic to address these issues. Summary of the Invention

[0004] The purpose of the present invention is to provide a photovoltaic module adaptive MPPT optimization method, system, device and storage medium to solve the problem of low power tracking accuracy and poor efficiency in complex environments in the prior art.

[0005] To achieve one of the above-mentioned objects of the invention, an embodiment of the present invention provides an adaptive MPPT optimization method for photovoltaic modules, the method comprising collecting multi-dimensional parameter information of photovoltaic modules through adaptive weight allocation to generate a weighted original signal, and generating a filtered signal through dynamic filtering processing; based on the filtered signal, generating a principal component eigenvector through principal component analysis, solving the light intensity, and calculating the predicted power using a multi-parameter coupling prediction model; based on the predicted power, optimizing the reference voltage through model predictive control to optimize the control action, and adjusting the multi-parameter anomaly detection weights based on adaptive fuzzy logic, performing predictive fault correction, and realizing maximum power point tracking.

[0006] As a further improvement of an embodiment of the present invention, the method further includes collecting multi-dimensional parameter information of the photovoltaic module through adaptive weight distribution, including collecting the voltage, current, temperature and impedance of the photovoltaic module, and generating a weighted original signal through adaptive weight distribution, the formula is: ; in, is the weighted original signal, is the voltage, is the current, is the temperature, is the impedance, t is the time variable, which is used to represent the value of the multi-dimensional parameters of the photovoltaic module at a specific moment. is the corresponding weight, which is dynamically adjusted based on the parameter change rate. , the weight adjustment formula is: ; in, is the corresponding weight; express , , , , is the value of the i-th parameter corresponding to the previous moment; The weight adjustment sensitivity coefficient is used to adjust the weight's response to parameter changes; j is an index that traverses the four parameters of voltage, current, temperature, and impedance, ranging from 1 to 4.

[0007] As a further improvement of one embodiment of the present invention, the method further includes generating a filtered signal by dynamic filtering to reduce noise and enhance the stability of maximum power point tracking; the dynamic filtering uses dynamic Gaussian filtering, and the formula is: ; in, is the filtered signal generated after dynamic Gaussian filtering at time t; n is the index of a specific moment in the sampling window, and N is the sampling window size; is the sampling time interval; is the dynamic variance, which is adjusted based on the light intensity and temperature change rate. The formula is: ; in, is the benchmark variance; is the light intensity, the initial value is the calibration value, is the light intensity at the previous moment, is the standard light intensity; is the temperature change weight coefficient, which is used to balance the impact of temperature on filtering; is the temperature at the previous moment.

[0008] As a further improvement of an embodiment of the present invention, the method further includes that the “generating a principal component eigenvector by principal component analysis, solving the light intensity, and calculating the predicted power using a multi-parameter coupling prediction model” includes that the principal component eigenvector is expressed as: ; in, is the principal component eigenvector at the current moment, is the principal component matrix obtained by offline analysis of historical photovoltaic system operation data. are the voltage and current obtained based on the filtered signal; the calculation formula for the predicted power is: ; in, is the predicted power at the next moment, is the current control action, The multi-parameter coupled prediction function obtained by learning historical photovoltaic system operation data and corresponding control actions can capture the nonlinear relationship between multiple variables and predict the future power output of the system.

[0009] As a further improvement of an embodiment of the present invention, the method further includes that the calculation formula for calculating the light intensity is: ; in, is the instantaneous power, is the impedance correction factor, is the power value under standard test conditions, is the temperature value under standard test conditions, is the temperature compensation factor, which is used to correct the nonlinear effect of temperature on the light intensity solution; the calculation formula of the impedance correction factor is: ; in, is the dynamic impedance influence coefficient, is the reference impedance influence coefficient, is the impedance estimation sensitivity coefficient, is the impedance value of the photovoltaic module under standard test conditions, is the impedance of the photovoltaic module at the previous moment; is the interaction coefficient; the calculation formula of the temperature compensation factor is: ; Among them, α represents the temperature attenuation exponent, which is used to characterize the nonlinear effect of temperature on the light intensity solution; and The solution is achieved through iterative solution, the specific steps include: initialization Light intensity at the previous moment Or preset initial value; based on the current ,calculate Based on the current and ,calculate ; Through iterative calculation, until is less than the preset convergence threshold or reaches the maximum number of iterations; is the initial value of the light intensity during the iterative solution process, is the illumination intensity value obtained in the nth iteration during the iterative solution process, is the impedance correction factor obtained in the nth iteration during the iterative solution process, is the illumination intensity value obtained in the n+1th iteration during the iterative solution process, It is the absolute value of the difference between the illumination intensity values ​​obtained from two adjacent iterations, and is used to determine whether the iteration has converged.

[0010] As a further improvement of an embodiment of the present invention, the method further includes optimizing the reference voltage by optimizing the control action through model predictive control, including maximizing the predicted power by optimizing the objective function, the formula is: ; in, is the optimal control action, To predict the time domain length, is the predicted power at the next k moments; the update formula of the reference voltage is: ; in, is the reference voltage at the next moment, For voltage regulation based on model predictive control optimization, is the optimal operating voltage at the next moment predicted based on the model, is the reference voltage value at the current moment.

[0011] As a further improvement of an embodiment of the present invention, the method further includes adjusting the multi-parameter anomaly detection weight based on adaptive fuzzy logic and performing predictive fault correction, including calculating an anomaly index to detect the system operating status, the formula is: ; in, is the temperature change value, is the impedance change value, is the weight coefficient for anomaly detection, which is used to balance the contribution of power, temperature and impedance changes to anomaly indicators; The adaptive adjustment is achieved through a fuzzy logic controller, which takes the rate of change of light intensity, the rate of change of temperature and the current abnormal index trend as input and outputs the corresponding weight adjustment amount. The fuzzy logic controller achieves adaptive adjustment through the following steps: defining fuzzy sets and membership functions of input variables and output variables, and establishing a fuzzy rule base; performing fuzzy reasoning based on the fuzzification results of the current system state and the fuzzy rule base; defuzzifying the fuzzy reasoning results into clear weight adjustment amounts. ; Update weights: ; in, is the anomaly detection weight value at the previous moment, is the weight adjustment calculated by the adaptive fuzzy logic controller at the current moment, X is the index, corresponding to P, T, Z; when the abnormal index exceeds the preset threshold, the correction is triggered, and the correction formula is: ; in, It is the reference voltage used by the system for fault correction at the next moment after an abnormality occurs and triggers correction. is the reference voltage at the current moment; is the maximum power point voltage value at the previous moment, is the maximum power point voltage value at the first two moments, is the correction weight coefficient, which is used to adjust the correction amplitude.

[0012] In order to achieve one of the above-mentioned purposes of the invention, an embodiment of the present invention also provides a photovoltaic component adaptive MPPT optimization system, which includes an acquisition module, a prediction module and a correction module; the acquisition module is used to acquire multi-dimensional parameter information of the photovoltaic component through adaptive weight distribution, generate a weighted original signal, and generate a filtered signal through dynamic filtering processing; the prediction module is used to generate a principal component eigenvector based on the filtered signal through principal component analysis, and solve the light intensity, and calculate the predicted power using a multi-parameter coupling prediction model; the correction module is used to optimize the reference voltage based on the predicted power through model predictive control optimization control action, and adjust the multi-parameter anomaly detection weight based on adaptive fuzzy logic, and perform predictive fault correction to achieve maximum power point tracking.

[0013] In order to achieve one of the above-mentioned objects of the invention, an embodiment of the present invention further provides an electronic device, including a memory and a processor, characterized in that the memory stores a computer program that can be run on the processor, and when the program is executed on the processor, the steps in the above-mentioned photovoltaic component adaptive MPPT optimization method are implemented.

[0014] To achieve one of the above-mentioned objects of the invention, an embodiment of the present invention further provides a storage medium, wherein the storage medium stores a computer program, and is characterized in that when the computer program is executed by a processor, the steps in the above-mentioned photovoltaic module adaptive MPPT optimization method are implemented.

[0015] Compared to existing technologies, this invention provides a method, system, device, and storage medium for adaptive MPPT optimization of photovoltaic modules. Through adaptive weight allocation and dynamic Gaussian filtering, this method effectively reduces noise interference and improves the stability of multidimensional parameter processing. It also combines principal component analysis with a multi-parameter coupling prediction model to accurately predict power output. It optimizes reference voltage through model predictive control to achieve fast and accurate maximum power point tracking. Adaptive fuzzy logic enhances anomaly detection and fault correction capabilities, improving system robustness. Compared to traditional methods, this solution significantly improves power tracking efficiency and system reliability in complex environments, making it suitable for distributed photovoltaic and microgrid applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is an overall flow chart of the photovoltaic module adaptive MPPT optimization method described in the present invention.

[0017] Figure 2 It is a schematic diagram of the architecture of the photovoltaic module adaptive MPPT optimization system described in the present invention. DETAILED DESCRIPTION

[0018] The present invention will be described in detail below with reference to the specific embodiments shown in the accompanying drawings. However, these embodiments do not limit the present invention, and any structural, methodological, or functional changes made by those skilled in the art based on these embodiments are all within the scope of protection of the present invention.

[0019] The embodiments of the present invention are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention and are not to be construed as limiting the present invention.

[0020] In the first embodiment of the present invention, the present invention provides a method for optimizing the MPPT of photovoltaic modules. Figure 1As shown, the method includes: S1: collecting multi-dimensional parameter information of photovoltaic components through adaptive weight allocation to generate weighted original signals, and generating filtered signals through dynamic filtering processing; S2: based on the filtered signals, generating principal component eigenvectors through principal component analysis, solving the light intensity, and calculating the predicted power using a multi-parameter coupling prediction model; S3: based on the predicted power, optimizing the reference voltage through model predictive control optimization control action, and adjusting the multi-parameter anomaly detection weights based on adaptive fuzzy logic, performing predictive fault correction, and realizing maximum power point tracking.

[0021] In a specific embodiment of the present invention, multi-dimensional parameter information of the photovoltaic module is collected by adaptive weight distribution. Specifically, the voltage, current, temperature and impedance of the photovoltaic module are collected, and a weighted original signal is generated by adaptive weight distribution. The formula is:

[0022] in, is the weighted original signal, is the voltage, is the current, is the temperature, is the impedance, t is the time variable, which is used to represent the value of the multi-dimensional parameters of the photovoltaic module at a specific moment. is the corresponding weight, which is dynamically adjusted based on the parameter change rate. , the weight adjustment formula is: ; in, is the corresponding weight; express , , , , is the value of the i-th parameter corresponding to the previous moment; The weight adjustment sensitivity coefficient is used to adjust the weight's response to parameter changes; j is an index that traverses the four parameters of voltage, current, temperature, and impedance, ranging from 1 to 4.

[0023] It should be noted that the system uses high-precision sensors to monitor the operating status of PV modules in real time, acquiring key parameters such as voltage, current, temperature, and impedance to comprehensively characterize the dynamic characteristics of the PV system. These parameters are collected at fixed intervals (determined through experimental optimization) to form time series data, which provides the basis for subsequent signal processing and optimized control. The acquisition process must ensure high accuracy and low latency to accommodate rapid changes in external conditions such as light intensity and ambient temperature.

[0024] Furthermore, adaptive weight allocation integrates multi-dimensional parameters by weighting the original signal formula to generate a comprehensive signal This signal reflects the relative contribution of each parameter to the system state, and the weight The design aims to dynamically adjust the influence of each parameter to adapt to the operating characteristics of the photovoltaic system under different working conditions. The weight adjustment is based on the parameter change rate. , through exponential function and normalization processing, the weight can quickly respond to the dynamic changes of parameters. For example, when the sudden change of light causes the current change rate to increase, the current weight Automatically increase and enhance the focus on key parameters, thereby improving the quality of the signal.

[0025] It should be noted that the generated weighted original signal As the input of dynamic filtering processing, it provides a high-quality data basis for principal component analysis and power prediction by reducing noise and enhancing signal stability. The value of can be determined by analyzing the response speed of the system to sudden changes in illumination, ensuring that the sensitivity of weight adjustment matches the dynamic characteristics of the system. Usually, it is adjusted based on the dynamic response characteristics of the photovoltaic system (such as the frequency of illumination changes) to ensure that the sensitivity of the weight to parameter changes is moderate. The signal quality (such as signal-to-noise ratio) at these values ​​is appropriately selected to balance response speed and stability. Furthermore, the selection of acquisition parameters (voltage, current, temperature, and impedance) considers the core factors influencing PV system operation. The inclusion of impedance particularly enhances the ability to monitor component aging or connection issues, providing critical information support for subsequent anomaly detection and fault correction.

[0026] In a specific embodiment of the present invention, a filter signal is generated by dynamic filtering to reduce noise and enhance the stability of maximum power point tracking; the dynamic filtering adopts dynamic Gaussian filtering, and the formula is: ; in, is the filtered signal generated after dynamic Gaussian filtering at time t; n is the index of a specific moment in the sampling window, and N is the sampling window size; is the sampling time interval; is the dynamic variance, which is adjusted based on the light intensity and temperature change rate. The formula is: ; in, is the benchmark variance; is the light intensity, the initial value is the calibration value, is the light intensity at the previous moment, is the standard light intensity; is the temperature change weight coefficient, which is used to balance the impact of temperature on filtering; is the temperature at the previous moment.

[0027] It should be noted that the dynamic filtering process is to weight the original signal Smoothing is performed to reduce the impact of environmental noise and measurement errors, thereby enhancing MPPT stability. Gaussian filtering utilizes a weighted average within a time window, combined with a Gaussian kernel function, to effectively suppress high-frequency noise while retaining the main trend of the signal. This is suitable for the dynamic characteristics of photovoltaic systems when light intensity or temperature changes rapidly.

[0028] Furthermore, dynamic variance It is the core parameter of Gaussian filtering, according to the rate of change of light intensity and temperature change rate Dynamically adjust the filter intensity. When the light or temperature changes significantly, Increase to relax the filtering range and retain more dynamic information; in a stable environment, This adaptive mechanism ensures that the filtering process can adapt to the complexity of the photovoltaic system operating environment.

[0029] Furthermore, the baseline variance The noise suppression effect of the filter can be adjusted according to the temperature change weight coefficient The setting can be done by analyzing the degree of influence of temperature changes on the system. In the experiment, the filtering effect (such as signal smoothness and dynamic response) under different lighting and temperature change scenarios can be analyzed to select the value that can balance noise suppression and information retention. Sampling window size N and time interval Adjust according to hardware performance and real-time requirements, and usually determine the optimal value through simulation to meet real-time control needs.

[0030] It should be noted that the system receives the weighted original signal As input, it generates high-quality , providing stable input data for subsequent PCA and power prediction. and temperature The rate of change directly affects the filtering parameters and is linked with the subsequent light intensity calculation steps to ensure that the system responds quickly to environmental changes.

[0031] In one specific implementation of the present invention, the filtering process is implemented on an embedded platform (such as a DSP or ARM microcontroller), using C or MATLAB to develop a real-time algorithm. The hardware must support fast data sampling and floating-point operations to ensure real-time filtering.

[0032] In a specific embodiment of the present invention, principal component eigenvectors are generated by principal component analysis, and the light intensity is solved, and the predicted power is calculated using a multi-parameter coupling prediction model. Specifically, the principal component eigenvectors are expressed as: ; in, is the principal component eigenvector at the current moment, is the principal component matrix obtained by offline analysis of historical photovoltaic system operation data. are the voltage and current obtained based on the filtered signal; the calculation formula for the predicted power is: ; in, is the predicted power at the next moment, is the current control action, The multi-parameter coupled prediction function obtained by learning historical photovoltaic system operation data and corresponding control actions can capture the nonlinear relationship between multiple variables and predict the future power output of the system.

[0033] It should be noted that the principal component analysis and power prediction step aims to extract key features from the filtered multidimensional parameters and predict the future power output of the photovoltaic system based on them to optimize the MPPT performance. This step generates feature vectors through PCA and uses the formula to map the filtered voltage, current, temperature, and impedance to a low-dimensional feature space. The principal component matrix Trained through offline analysis of historical PV system operating data, it effectively extracts correlations between parameters, reduces redundant information, and improves computational efficiency for subsequent prediction and control. PCA implementation requires standardization of the input data (zero mean, unit variance) to ensure accurate feature extraction.

[0034] Furthermore, the principal component matrix It is obtained by offline analysis and training of historical photovoltaic system operation data using principal component analysis method, and is used to extract key features from multidimensional parameters (voltage, current, temperature, impedance) to generate low-dimensional feature vectors. , the specific generation process is:

[0035] Collect historical PV system operating data, including voltage and impedance. This data is collected at regular intervals using high-precision sensors and covers a wide range of operating conditions, such as varying light intensities and temperature ranges, as well as scenarios such as module aging and partial shading. Data collection must ensure sample diversity to reflect the dynamic characteristics of PV systems in real-world environments. Data preprocessing is required, including removing outliers (such as invalid data caused by sensor failure) and normalizing (setting all parameters to zero mean and unit variance) to eliminate dimensionality differences and improve analysis accuracy.

[0036] The principal component analysis method is used to reduce the dimensionality of the pre-processed historical data. The specific steps include: first, constructing a time series data set containing voltage, current, temperature and impedance to form a matrix, in which each row represents a data vector at a time point and each column corresponds to a parameter. Then, the covariance matrix of the data matrix is ​​calculated to analyze the correlation between the parameters. Based on the covariance matrix, the eigenvalue decomposition method is used to extract the eigenvalues ​​and corresponding eigenvectors. The eigenvectors are sorted according to the size of the eigenvalues, and the vectors with larger eigenvalues ​​correspond to the main direction of change of the data. The first several eigenvectors are selected to form the principal component matrix Each column of this matrix is ​​a feature vector that maps the original multi-dimensional parameters to a low-dimensional feature space.

[0037] Preferably, the principal component matrix The training process can be completed in an offline environment, using computing software or programming tools, such as MATLAB's PCA function, Python's scikit-learn library, or R language statistical analysis package. It is stored in the control unit of the photovoltaic system (such as an embedded microcontroller or DSP chip) in the form of a fixed matrix and used to calculate the feature vector in real time. To ensure the applicability of the matrix, it can be retrained regularly based on the newly added operating data. , to adapt to aging of photovoltaic modules or environmental changes.

[0038] Furthermore, the prediction function It is obtained by training historical photovoltaic system operation data and corresponding control actions through machine learning methods. It is used to capture the nonlinear relationship between parameters such as voltage, current, temperature, impedance, light intensity and control actions, and predict the power output at the next moment. The generation process of this function first involves the collection of historical data, including at least several weeks to several months of photovoltaic system operation records, covering voltage, current, temperature, impedance, light intensity, control actions, and the corresponding actual power output. This data is collected in real time by sensors and annotated with environmental conditions to ensure sample diversity. Data preprocessing steps include outlier removal (such as using threshold filtering to remove noisy data), normalization, and feature engineering to improve the model's generalization capabilities.

[0039] Preferably, function The training adopts a supervised learning framework, and can use neural network models (such as multilayer perceptrons or long short-term memory networks) or regression algorithms (such as support vector regression) to deal with the coupling effects between multiple variables. The specific training methods include: dividing the preprocessed historical data into training sets and validation sets, defining the loss function as mean square error (MSE), and iteratively training the model parameters through the gradient descent optimizer. During the training process, the model needs to optimize hyperparameters (such as the number of hidden layer neurons or regularization coefficients) to avoid overfitting. Training can be completed in an offline environment using programming tools such as Python's TensorFlow or scikit-learn library. After the training is completed, The model parameters are stored in the system controller for real-time inference. The model update mechanism includes periodic retraining (e.g., quarterly based on newly added data) to adapt to PV panel aging or environmental changes. This implementation ensures prediction accuracy and supports the dynamic decision-making process for MPPT optimization.

[0040] Furthermore, the current control action It is a key input variable in the photovoltaic module adaptive maximum power point tracking (MPPT) optimization method, representing the control command applied to the photovoltaic system at time t, used to adjust the system's operating point to achieve maximum power output. In the photovoltaic power generation system, the control action Usually refers to specific parameters such as reference voltage adjustment, duty cycle change or inverter control signal, which directly affect the output power curve of photovoltaic modules. , the system can respond to environmental changes (such as light intensity or temperature fluctuations) and guide the operating point to approach the maximum power point. Specifically, Can be represented as a vector or scalar, for example in a DC-DC converter, Can be defined as the incremental value of the duty cycle; in the model predictive control framework, Includes multiple control sequences within the future prediction time domain. The value range of is determined based on the electrical characteristics of the photovoltaic system. For example, when adjusting the reference voltage, The amplitude is usually limited to within 5% of the open circuit voltage of the photovoltaic module to avoid system instability. The generation of depends on the output of upstream modules, such as principal component eigenvectors and light intensity solution results, to ensure seamless integration with the multi-parameter coupling prediction model. Through real-time calculation and application of embedded controller, it supports closed-loop feedback mechanism to improve the accuracy and response speed of MPPT.

[0041] In one specific implementation scenario of the present invention, historical data covering a variety of operating conditions (e.g., sunny, cloudy, and partially shaded) is required. A training set is constructed by collecting at least several months of PV system operating data. Data preprocessing includes outlier removal and normalization. Training utilizes supervised learning methods, with the goal of minimizing the error between predicted and actual power. Model development and training are performed using Python machine learning libraries (such as TensorFlow or Scikit-learn). After training, the model parameters are embedded in the embedded system. PCA and the prediction model must be run on a computing platform with floating-point computing capabilities, such as a DSP or embedded GPU. Software development can be performed in C or Python, and model inference must be optimized to meet real-time requirements.

[0042] It should be noted that the system relies on filtering the signal and As input, ensure the quality of input data for feature extraction. It directly serves the subsequent MPC step and provides a key basis for optimizing the reference voltage. The inclusion of improves the ability to characterize the dynamic characteristics of the system and provides support for anomaly detection and fault correction.

[0043] In a specific embodiment of the present invention, the calculation formula for calculating the light intensity is: ; in, is the instantaneous power, is the impedance correction factor, is the power value under standard test conditions, is the temperature value under standard test conditions, is the temperature compensation factor, which is used to correct the nonlinear effect of temperature on the light intensity solution; the calculation formula of the impedance correction factor is: ; in, is the dynamic impedance influence coefficient, is the reference impedance influence coefficient, is the impedance estimation sensitivity coefficient, is the impedance value of the photovoltaic module under standard test conditions, is the impedance of the photovoltaic module at the previous moment; is the interaction coefficient; the calculation formula of the temperature compensation factor is: ; Among them, α represents the temperature attenuation exponent, which is used to characterize the nonlinear effect of temperature on the light intensity solution; and The solution is achieved through iterative solution, the specific steps include: initialization Light intensity at the previous moment Or preset initial value; based on the current ,calculate Based on the current and ,calculate ; Through iterative calculation, until is less than the preset convergence threshold or reaches the maximum number of iterations; is the initial value of the light intensity during the iterative solution process, is the illumination intensity value obtained in the nth iteration during the iterative solution process, is the impedance correction factor obtained in the nth iteration during the iterative solution process, is the illumination intensity value obtained in the n+1th iteration during the iterative solution process, It is the absolute value of the difference between the illumination intensity values ​​obtained from two adjacent iterations, and is used to determine whether the iteration has converged.

[0044] It should be noted that the light intensity calculation step aims to accurately estimate the current light intensity through the operating parameters of the photovoltaic system. , providing key environmental information for subsequent power prediction and control optimization. In the formula, the instantaneous power The filtered voltage and current as well as the impedance correction factor are combined to reflect the actual operating status of the photovoltaic module. Power under standard test conditions and temperature Provide a benchmark reference to ensure comparability of solution results.

[0045] Furthermore, in the calculation formula of the impedance correction factor, the dynamic impedance influence coefficient takes into account the dynamic impact of impedance changes, enhancing the adaptability to component aging or connection problems. The nonlinear effect of temperature on the illumination solution is corrected exponentially to ensure estimation accuracy.

[0046] Furthermore, due to and There are mutual dependencies, and the solution process is implemented using an iterative method. Based on the light intensity at the previous moment or the preset value, it is updated iteratively and , until convergence or the maximum number of iterations is reached. This iterative strategy effectively handles the nonlinear coupling between parameters and improves the robustness of the estimation.

[0047] Furthermore, the parameters This is determined through experimentation and simulation optimization. Experimentation allows analysis of the impact of different parameter values ​​on lighting solution accuracy (e.g., error relative to actual lighting data), selecting the value that minimizes error. The convergence threshold and maximum number of iterations can be set based on the system's accuracy requirements for light intensity estimation. In this application, the threshold range can be adjusted based on estimation error and real-time requirements.

[0048] In a specific implementation scenario of the present invention, the iterative solution needs to be implemented on a platform that supports fast floating-point operations. By using a DSP or a microcontroller, the software is developed using C language, and the iterative algorithm is optimized to reduce the computational complexity.

[0049] It should be noted that the system relies on filtering to generate and , and multi-dimensional parameter acquisition and , to ensure the reliability of input data. Directly inputting multi-parameter coupling prediction models influences the accuracy of power prediction and provides environmental information support for subsequent MPC and anomaly detection. By dynamically adjusting impedance and temperature effects, this step enhances the system's adaptability to complex environments (such as sudden changes in illumination or temperature fluctuations), providing an accurate environmental parameter foundation for MPPT optimization.

[0050] In a specific embodiment of the present invention, the reference voltage is optimized by optimizing the control action through model predictive control, specifically, by maximizing the predicted power by optimizing the objective function, which is: ; in, is the optimal control action, To predict the time domain length, is the predicted power at the next k moments; the update formula of the reference voltage is: ; in, is the reference voltage at the next moment, For voltage regulation based on model predictive control optimization, is the optimal operating voltage at the next moment predicted based on the model, is the reference voltage value at the current moment.

[0051] It should be noted that the MPC optimization step aims to optimize the reference voltage of the photovoltaic system by dynamically adjusting the control action to achieve the accuracy and stability of MPPT. and by predicting the time domain length Maximize predicted power , the system can proactively evaluate the impact of different control actions on future power output, thereby selecting the optimal control strategy. Based on principal component eigenvector and the current control action , taking into account the coupling effect of multi-dimensional parameters.

[0052] Furthermore, in the reference voltage update formula, the voltage adjustment amount Optimal operating voltage predicted by the model Dynamically adjust the current reference voltage This update mechanism ensures the system gradually approaches the maximum power point while avoiding the risk of traditional MPPT methods (such as the perturbation-and-observe method) falling into local extremes in complex environments. The MPC optimization process can utilize numerical optimization algorithms (such as gradient descent or dynamic programming) to ensure computational efficiency and real-time performance.

[0053] In one specific implementation scenario of the present invention, the prediction time domain length H is determined through simulation optimization, balancing prediction accuracy and computational complexity, and is typically selected based on the system's dynamic response time. MPC runs on a high-performance computing platform, such as a DSP or FPGA. Software development uses C or MATLAB, and algorithms are optimized to meet real-time control requirements.

[0054] It should be noted that the system relies on the power prediction step to provide and principal component analysis , ensuring that the optimization is based on high-quality forecast data. Light intensity By indirectly influencing the optimization process through the prediction model, the system's adaptability to environmental changes is enhanced. and updated reference voltage Directly applied to photovoltaic inverters or control circuits, it provides a stable operating status reference for subsequent anomaly detection and fault correction steps.

[0055] In one embodiment of the present invention, the weights of multi-parameter anomaly detection are adjusted based on adaptive fuzzy logic to perform predictive fault correction. Specifically, an anomaly index is calculated to detect the system operating status. The formula is: ; in, is the temperature change value, is the impedance change value, is the weight coefficient for anomaly detection, which is used to balance the contribution of power, temperature and impedance changes to anomaly indicators; The adaptive adjustment is achieved through a fuzzy logic controller, which takes the rate of change of light intensity, the rate of change of temperature and the current abnormal index trend as input and outputs the corresponding weight adjustment amount. The fuzzy logic controller achieves adaptive adjustment through the following steps: defining fuzzy sets and membership functions of input variables and output variables, and establishing a fuzzy rule base; performing fuzzy reasoning based on the fuzzification results of the current system state and the fuzzy rule base; defuzzifying the fuzzy reasoning results into clear weight adjustment amounts. ; Update weights to ,in, is the anomaly detection weight value at the previous moment, is the weight adjustment calculated by the adaptive fuzzy logic controller at the current moment, X is the index, corresponding to P, T, Z; when the abnormal index exceeds the preset threshold, the correction is triggered, and the correction formula is: ; in, It is the reference voltage used by the system for fault correction at the next moment after an abnormality occurs and triggers correction. is the reference voltage at the current moment; is the maximum power point voltage value at the previous moment, is the maximum power point voltage value at the first two moments, is the correction weight coefficient, which is used to adjust the correction amplitude.

[0056] It should be noted that the adaptive fuzzy logic adjustment and predictive fault correction step aims to ensure the stability and reliability of MPPT by dynamically monitoring the operating status of the photovoltaic system, identifying anomalies and performing corrections. This step comprehensively evaluates the rate of change of power, temperature and impedance through the abnormal index formula to detect abnormal conditions in the system operating status in real time. The abnormal index integrates the changes of multiple parameters in a weighted manner, and the weights are Dynamic adjustment ensures that the system can flexibly cope with the differences in the contribution of different parameters to anomaly detection. Example rules of fuzzy logic include: if the rate of change of light is high and the trend of anomaly indicators is rising, then increase the power weight .

[0057] Furthermore, the weight adjustment is achieved through a fuzzy logic controller, which uses the light intensity change rate , temperature change rate and abnormal indicator trends as input, and generate weight adjustment through fuzzification, fuzzy rule reasoning and defuzzification The fuzzy rule base is designed based on expert experience or historical data and can capture the nonlinear relationship between environmental changes and system status. The updated weights dynamically reflect the system's operating characteristics, improving the sensitivity and accuracy of anomaly detection.

[0058] Furthermore, when the abnormal index exceeds the preset threshold, the predictive fault correction is triggered, and the formula Update the reference voltage. The correction process combines the historical maximum power point voltage and its changing trend to quickly adjust the operating point to restore the system to a state close to the maximum power point, enhancing robustness. The abnormality threshold can be determined by analyzing the distribution of abnormal indicators under normal operating conditions using statistical methods (such as the mean plus a multiple of the standard deviation).

[0059] In a specific implementation scenario of the present invention, the weight coefficient The initial value and The sensitivity of anomaly detection and the effectiveness of corrections are determined through experimental optimization. The fuzzy rule base is designed through historical data analysis or expert consultation to ensure inference accuracy. Fuzzy logic controllers can be implemented on embedded microcontrollers and developed using the C language. The rule base is stored in non-volatile memory.

[0060] It should be noted that the system relies on multi-dimensional parameter collection , , lighting solution The power data processed by the filter ensures the reliability of the abnormal detection input. This step works in conjunction with the MPC step to optimize subsequent control actions while providing feedback to the system and enhancing overall stability. Through dynamic adjustment and predictive correction using fuzzy logic, this step effectively responds to abnormal situations in complex environments, improving the MPPT system's anti-interference capabilities and operational efficiency.

[0061] In a specific implementation scenario of this invention, to verify the effectiveness of the technical solution, simulation and experimental testing were conducted in both the laboratory and in an actual photovoltaic system. The simulations were conducted using the MATLAB / Simulink platform, and a model was constructed based on typical light and temperature variation scenarios (sunny, cloudy, and partially shaded). The MPPT efficiency, response speed, and anomaly detection accuracy were verified. The experiments were conducted on a small-scale photovoltaic array (single or multiple modules in series), using high-precision sensors and a DSP controller. The test results demonstrated that the system can effectively track the maximum power point and quickly correct anomalies in dynamic environments.

[0062] Hardware Configuration: High-precision sensors and DSP controllers or ARM Cortex-M microcontrollers are recommended, supporting real-time floating-point operations and fast data sampling. Storage must be capable of storing at least several months of historical data for model training. Preferably, a processor supporting real-time floating-point operations, such as a hardware platform based on a digital signal processor or microcontroller, is recommended.

[0063] Software Implementation: C or MATLAB is recommended for algorithm development. Predictive models and PCA can be trained on a PC and then embedded in an embedded system. Real-time control software must be optimized to meet millisecond-level response requirements, and an RTOS can be used to support multitasking. Ideally, development should be done in a programming language that supports real-time control, and model training should be performed using data analysis tools.

[0064] In a second embodiment of the present invention, the present invention provides an adaptive MPPT optimization system for photovoltaic components, the system comprising an acquisition module, a prediction module and a correction module; the acquisition module is used to acquire multi-dimensional parameter information of photovoltaic components through adaptive weight distribution, generate a weighted original signal, and generate a filtered signal through dynamic filtering processing; the prediction module is used to generate a principal component eigenvector based on the filtered signal through principal component analysis, and solve the light intensity, and calculate the predicted power using a multi-parameter coupling prediction model; the correction module is used to optimize the reference voltage based on the predicted power through model predictive control optimization control action, and adjust the multi-parameter anomaly detection weight based on adaptive fuzzy logic, and perform predictive fault correction to achieve maximum power point tracking.

[0065] In a third embodiment of the present invention, the present invention provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and when the program is executed on the processor, the steps in the above-mentioned photovoltaic component adaptive MPPT optimization method are implemented.

[0066] In a fourth embodiment of the present invention, the present invention provides a storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps in the above-mentioned photovoltaic module adaptive MPPT optimization method.

[0067] In summary, the present invention provides an adaptive MPPT optimization method and system for photovoltaic modules. Through adaptive weight allocation and dynamic Gaussian filtering, this method effectively reduces noise interference and improves the stability of multidimensional parameter processing. It also combines principal component analysis with a multi-parameter coupling prediction model to accurately predict power output. It optimizes the reference voltage through model predictive control to achieve fast and accurate maximum power point tracking. Adaptive fuzzy logic enhances anomaly detection and fault correction capabilities, improving system robustness. Compared to traditional methods, this solution significantly improves power tracking efficiency and system reliability in complex environments, making it suitable for distributed photovoltaic and microgrid applications.

[0068] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working process of the modules described above can refer to the corresponding process in the aforementioned method implementation, and will not be repeated here.

[0069] Modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical modules, i.e., they may be located in one place or distributed across multiple network modules. Some or all of these modules may be selected to achieve the purpose of this embodiment according to actual needs.

[0070] In addition, the functional modules in each embodiment of the present application may be integrated into a processing module, or each module may exist physically separately, or two or more modules may be integrated into a single module. The above-mentioned integrated modules may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0071] The above-mentioned integrated modules implemented in the form of software functional modules can be stored in a computer-readable storage medium. The above-mentioned software functional modules are stored in a storage medium and include a number of instructions for causing a computer system (which may be a personal computer, server, or network system, etc.) or a processor to execute some of the steps of the methods described in various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0072] Finally, it should be noted that the above embodiments 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, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application.

Claims

1. A photovoltaic module adaptive MPPT optimization method, characterized by: include, The multi-dimensional parameter information of the photovoltaic module is collected through adaptive weight distribution to generate a weighted original signal, and a filtered signal is generated through dynamic filtering processing; Based on the filtered signal, principal component eigenvectors are generated through principal component analysis, and the light intensity is solved, and the predicted power is calculated using a multi-parameter coupling prediction model; Based on the predicted power, the reference voltage is optimized by optimizing the control action through model predictive control, and the multi-parameter anomaly detection weights are adjusted based on adaptive fuzzy logic to perform predictive fault correction to achieve maximum power point tracking.

2. The photovoltaic module adaptive MPPT optimization method according to claim 1, characterized in that: The method of collecting multi-dimensional parameter information of photovoltaic components by adaptive weight distribution includes: The voltage, current, temperature and impedance of the photovoltaic module are collected and the weighted original signal is generated through adaptive weight distribution. The formula is: ； in, is the weighted original signal, is the voltage, is the current, is the temperature, is the impedance, t is the time variable, which is used to represent the value of the multi-dimensional parameters of the photovoltaic module at a specific moment. is the corresponding weight, which is dynamically adjusted based on the parameter change rate. , the weight adjustment formula is: ； in, is the corresponding weight; express , , , , is the value of the i-th parameter corresponding to the previous moment; The weight adjustment sensitivity coefficient is used to adjust the weight's response to parameter changes; j is an index that traverses the four parameters of voltage, current, temperature, and impedance, ranging from 1 to 4.

3. The photovoltaic module adaptive MPPT optimization method according to claim 2, characterized in that: The dynamic filtering process generates a filtered signal for reducing noise and enhancing the stability of maximum power point tracking; The dynamic filtering process adopts dynamic Gaussian filtering, and the formula is: ； in, is the filtered signal generated after dynamic Gaussian filtering at time t; n is the index of a specific moment in the sampling window, and N is the sampling window size; is the sampling time interval; is the dynamic variance, which is adjusted based on the light intensity and temperature change rate. The formula is: ； in, is the benchmark variance; is the light intensity, the initial value is the calibration value, is the light intensity at the previous moment, is the standard light intensity; is the temperature change weight coefficient, which is used to balance the impact of temperature on filtering; is the temperature at the previous moment.

4. The method for optimizing photovoltaic module adaptive MPPT according to claim 3, characterized in that: The principal component eigenvector is generated by principal component analysis, and the light intensity is calculated, and the predicted power is calculated using a multi-parameter coupling prediction model, including: The principal component eigenvector is expressed as: ； in, is the principal component eigenvector at the current moment, is the principal component matrix obtained by offline analysis of historical photovoltaic system operation data. are the voltage and current obtained based on the filtered signal; The calculation formula of the predicted power is: ； in, is the predicted power at the next moment, is the current control action, The multi-parameter coupled prediction function obtained by learning historical photovoltaic system operation data and corresponding control actions can capture the nonlinear relationship between multiple variables and predict the future power output of the system.

5. The photovoltaic module adaptive MPPT optimization method according to claim 4, characterized in that: The calculation formula for calculating the light intensity is: ； in, is the instantaneous power, is the impedance correction factor, is the power value under standard test conditions, is the temperature value under standard test conditions, is the temperature compensation factor, which is used to correct the nonlinear effect of temperature on the light intensity solution; The impedance correction factor is calculated as: ； in, is the dynamic impedance influence coefficient, is the reference impedance influence coefficient, is the impedance estimation sensitivity coefficient, is the impedance value of the photovoltaic module under standard test conditions, is the impedance of the photovoltaic module at the previous moment; is the interaction coefficient; The temperature compensation factor is calculated as: ； Among them, α represents the temperature attenuation exponent, which is used to characterize the nonlinear effect of temperature on the light intensity solution; and The solution is achieved through iterative solution, and the specific steps include: initialization Light intensity at the previous moment Or preset initial value; Based on the current ,calculate ; Based on the current and ,calculate ; By iterative calculation, until Less than the preset convergence threshold or reaching the maximum number of iterations; in, is the initial value of the light intensity during the iterative solution process, is the illumination intensity value obtained in the nth iteration during the iterative solution process, is the impedance correction factor obtained in the nth iteration during the iterative solution process, is the illumination intensity value obtained in the n+1th iteration during the iterative solution process, It is the absolute value of the difference between the illumination intensity values ​​obtained from two adjacent iterations, and is used to determine whether the iteration has converged.

6. The photovoltaic module adaptive MPPT optimization method according to claim 5, characterized in that: The optimizing control action by model predictive control to optimize the reference voltage includes: The predicted power is maximized by optimizing the objective function, the formula is: ； in, is the optimal control action, To predict the time domain length, is the predicted power at k moments in the future; The updating formula of the reference voltage is: ； in, is the reference voltage at the next moment, For voltage regulation based on model predictive control optimization, is the optimal operating voltage at the next moment predicted based on the model, is the reference voltage value at the current moment.

7. The method for optimizing photovoltaic module adaptive MPPT according to claim 6, characterized in that: The method of adjusting the weights of multi-parameter anomaly detection based on adaptive fuzzy logic and performing predictive fault correction includes: Calculate the abnormal index to detect the system operation status. The formula is: ； in, is the temperature change value, is the impedance change value, is the weight coefficient for anomaly detection, which is used to balance the contribution of power, temperature, and impedance changes to anomaly indicators; described The adaptive adjustment is achieved through a fuzzy logic controller, which takes the rate of change of light intensity, the rate of change of temperature, and the current abnormal index trend as input and outputs the adjustment amount of the corresponding weight; the fuzzy logic controller achieves adaptive adjustment through the following steps: Define fuzzy sets and membership functions of input and output variables, and establish a fuzzy rule base; Perform fuzzy reasoning based on the fuzzification results of the current system state and the fuzzy rule base; Defuzzify the fuzzy inference results into clear weight adjustment quantities ; Update the weight to ,in, is the anomaly detection weight value at the previous moment, is the weight adjustment calculated by the adaptive fuzzy logic controller at the current moment, X is the index, corresponding to P, T, Z; When the abnormal indicator exceeds the preset threshold, correction is triggered. The correction formula is: ； in, It is the reference voltage used by the system for fault correction at the next moment after an abnormality occurs and triggers correction. is the reference voltage at the current moment; is the maximum power point voltage value at the previous moment, is the maximum power point voltage value at the first two moments, is the correction weight coefficient, which is used to adjust the correction amplitude.

8. A photovoltaic module adaptive MPPT optimization system, characterized by: It includes acquisition module, prediction module and correction module; The acquisition module is used to collect multi-dimensional parameter information of the photovoltaic module through adaptive weight distribution, generate a weighted original signal, and generate a filtered signal through dynamic filtering processing; The prediction module is used to generate a principal component eigenvector based on the filtered signal through principal component analysis, calculate the light intensity, and calculate the predicted power using a multi-parameter coupling prediction model; The correction module is used to optimize the reference voltage based on the predicted power through model predictive control optimization control action, and adjust the multi-parameter anomaly detection weight based on adaptive fuzzy logic to perform predictive fault correction to achieve maximum power point tracking.

9. An electronic device comprising a memory and a processor, characterized in that: The memory stores a computer program that can be run on the processor, and when the program is executed on the processor, the steps of the photovoltaic component adaptive MPPT optimization method as described in any one of claims 1 to 7 are implemented.

10. A storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the photovoltaic component adaptive MPPT optimization method according to any one of claims 1 to 7 are implemented.

Images