A photovoltaic maximum power point tracking control method, system, device and storage medium

Abstract

This invention discloses a photovoltaic (PV) maximum power point tracking (MPPT) control method, system, device, and storage medium, comprising: collecting irradiance and ambient temperature data of the environment in which the PV module is located, and inputting them into a support vector regression prediction model after hyperparameter optimization using a chaotic sparrow search algorithm; the prediction model outputs a reference value for the PV module's maximum power point voltage under the current operating conditions; constructing a second-order linear extended state observer, treating model parameter perturbations and external disturbances as the total extended state, and outputting real-time PV observation values; calculating the tracking error based on the voltage reference value and the real-time observed voltage, constructing an error integration stage to obtain an extended state space model, and solving for the optimal state feedback matrix; introducing the total disturbance observation value into the control law for feedforward compensation to generate the optimal control law, and controlling the DC / DC converter according to the optimal control law to adjust the output voltage of the PV module. This method can improve the accuracy and stability of the PV power point tracking control in a PV power generation system.

Description

Technical Field

[0001] This invention relates to the technical field of new energy power generation, and in particular to a photovoltaic maximum power point tracking control method, system, device and storage medium. Background Technology

[0002] The actual output power of a photovoltaic (PV) power generation system is affected by factors such as solar irradiance and ambient temperature. To fully utilize solar energy resources and improve PV power generation efficiency, maximum power point tracking (MPPT) control is needed to ensure that the PV modules always operate at their instantaneous maximum power point. MPPT control achieves PV output voltage tracking by adjusting the duty cycle of the switching transistors in the DC / DC converter, thereby ensuring that the PV output power is always located at the extreme point of the power-voltage (PV) curve.

[0003] Existing mainstream photovoltaic MPPT control methods, such as the traditional perturbation observation method and the incremental conductance method, have obvious technical defects: these methods require repeated perturbation to achieve optimization, resulting in continuous oscillation of photovoltaic output power and high ineffective losses. In particular, they are prone to tracking lag and misjudgment when illumination conditions change abruptly. On the other hand, the method of predicting maximum power voltage using conventional support vector regression (SVR) relies heavily on trial and error to obtain the hyperparameters of the SVR model, which easily leads to weak generalization ability and inability to adapt to complex operating conditions. In addition, the conventional robust optimal tracking control for adjusting DC / DC output voltage does not consider the real-time observation and compensation of model perturbation and external disturbances, which will cause a decrease in robustness when system parameters fluctuate, and it is difficult to balance dynamic response speed and steady-state tracking accuracy. Summary of the Invention

[0004] In view of the aforementioned existing problems, this invention is proposed. Therefore, this invention provides a photovoltaic maximum power point tracking (MPPT) control method, system, device, and storage medium to solve the problems of low efficiency, large oscillations, and weak anti-interference capability of traditional MPPT methods under complex lighting conditions.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, embodiments of the present invention provide a photovoltaic maximum power point tracking control method, comprising: Data on the irradiance and ambient temperature of the environment where the photovoltaic module is located are collected and input into the support vector regression prediction model after the hyperparameters are optimized by the chaotic sparrow search algorithm. The prediction model outputs the reference value of the maximum power point voltage of the photovoltaic module under the current operating conditions. A second-order linear extended state observer is constructed, which treats model parameter perturbations and external disturbances as the total extended state and outputs real-time photovoltaic observations. The real-time observations include photovoltaic output voltage, voltage derivative, and total disturbance observations. The tracking error is calculated based on the voltage reference value and the real-time observed voltage. An extended state space model is obtained by constructing an error integration loop. After verifying the system's controllability using the rank criterion, the matrix Riccati algebraic equation is solved to obtain the optimal state feedback matrix. The total disturbance observation is introduced into the control law for feedforward compensation to generate the optimal control law. The DC / DC converter is controlled according to the optimal control law to adjust the output voltage of the photovoltaic module.

[0006] As a preferred embodiment of the photovoltaic maximum power point tracking control method of the present invention, wherein: the chaotic sparrow search algorithm optimizes the hyperparameters of the support vector regression prediction model, including: Set the population size, maximum number of iterations, and convergence threshold; The sparrow population locations are initialized using the Tent chaotic mapping to generate an initial population; The fitness function is to minimize the mean squared error of support vector regression prediction. The fitness value of each individual in the population is calculated. In each iteration, sparrows are divided into discoverers, joiners and watchers according to their fitness values, and their positions are updated proportionally. When the fitness value changes less than the convergence threshold in consecutive iterations, the optimal hyperparameter penalty coefficient and kernel function width are output. When the fitness value changes more than or equal to the convergence threshold in consecutive iterations, the iteration updates continue. The optimal hyperparameters are used to train a support vector regression model, and the trained model is used to predict the reference value of the maximum power point voltage of photovoltaic power in real time.

[0007] As a preferred embodiment of the photovoltaic maximum power point tracking control method described in this invention, the support vector regression model adopts the RBF kernel function, and the mathematical expression of the support vector regression model is: in, The kernel width parameter. For the input vector, For bias.

[0008] As a preferred embodiment of the photovoltaic maximum power point tracking control method described in this invention, the mathematical model of the second-order linear extended state observer is expressed as follows: in, The observed value of photovoltaic output voltage. The differential observation value of the photovoltaic output voltage. This represents the total system disturbance observations, which include both model perturbations and external disturbances. , , The observer gain coefficient, To control the input gain, For controller input, This represents the actual voltage output of the photovoltaic system. It is a first-order differential operator.

[0009] As a preferred embodiment of the photovoltaic maximum power point tracking control method of the present invention, the tracking error is calculated based on the voltage reference value and the real-time observed voltage, and an extended state-space model is obtained by constructing an error integration element, including: Define tracking error Construct the error integral term The extended state vector is obtained. ; Establish extended system matrix and extended input matrix This makes the extended state-space model satisfy: in, This is the reference value for the maximum power point voltage. Photovoltaic output voltage, This represents the inductor current of the Boost converter.

[0010] As a preferred embodiment of the photovoltaic maximum power point tracking control method described in this invention, the following steps are included: solving the matrix Riccati algebraic equation to obtain the optimal state feedback matrix: Define the performance index of the quadratic form as: in, For performance index functionals, It is a positive definite symmetric state weighting matrix. It is a positive definite control weighting matrix; Solve the matrix Riccati algebraic equation We obtain a symmetric positive definite solution matrix. Based on symmetric positive definite solution matrices Calculate the optimal state feedback matrix .

[0011] As a preferred embodiment of the photovoltaic maximum power point tracking control method of the present invention, wherein: the total disturbance observation value is introduced into the control law for feedforward compensation to generate the optimal control law, including: The optimal control law with extended state observer disturbance compensation is expressed as: in, For robust optimal tracking control, This is the total disturbance feedforward compensation term.

[0012] Secondly, the present invention provides a photovoltaic maximum power point tracking control system, comprising: The prediction module is used to collect data on the irradiance and ambient temperature of the environment where the photovoltaic module is located, and input them into the support vector regression prediction model after the hyperparameters are optimized by the chaotic sparrow search algorithm. The prediction model outputs the reference value of the maximum power point voltage of the photovoltaic module under the current operating conditions. The state observation module is used to construct a second-order linear extended state observer, which treats model parameter perturbations and external disturbances as the total extended state and outputs real-time photovoltaic observation values. The real-time observation values ​​include photovoltaic output voltage, voltage derivative, and total disturbance observation values. The error processing module is used to calculate the tracking error based on the voltage reference value and the real-time observed voltage, construct an error integration loop to obtain an extended state space model, verify the system controllability using the rank criterion, and then solve the matrix Riccati algebraic equation to obtain the optimal state feedback matrix. The control module is used to introduce the total disturbance observation into the control law for feedforward compensation to generate the optimal control law, and control the DC / DC converter according to the optimal control law to adjust the output voltage of the photovoltaic module.

[0013] Thirdly, the present invention provides an electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of the photovoltaic maximum power point tracking control method.

[0014] Fourthly, the present invention provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the photovoltaic maximum power point tracking control method.

[0015] Compared with existing technologies, the advantages of this invention are as follows: This invention optimizes SVR model parameters using a chaotic sparrow search algorithm, improving the model's prediction accuracy and overcoming the limitations of manual parameter tuning. Simultaneously, it utilizes an extended state observer to perform real-time observation and compensation for parameter perturbations and external disturbances, ensuring high robustness of the system under complex operating conditions. Furthermore, the algorithm has moderate complexity and can be embedded into conventional controllers such as STM32 and DSP without requiring additional hardware, resulting in low modification costs and good compatibility. Compared with classic PI controllers, this invention further improves both steady-state tracking accuracy and dynamic response performance, demonstrating better engineering practicality. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a schematic flowchart of a photovoltaic maximum power point tracking control method according to an embodiment of the present invention; Figure 2 is a schematic diagram of the overall photovoltaic MPPT control principle of a photovoltaic maximum power point tracking control method according to an embodiment of the present invention; Figure 3 is a flowchart of the modeling and optimization process of the chaotic sparrow optimization support vector regression prediction model for a photovoltaic maximum power point tracking control method according to an embodiment of the present invention. Figure 4 is a block diagram of the design of the extended state observer robust optimal tracking controller of a photovoltaic maximum power point tracking control method according to an embodiment of the present invention. Figure 5 is a hardware topology diagram of a Boost DC / DC converter for a photovoltaic maximum power point tracking control method according to an embodiment of the present invention. Detailed Implementation

[0017] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0018] Example 1, referring to Figure 1 Figure 5 illustrates an embodiment of the present invention, which provides a photovoltaic maximum power point tracking control method, such as... Figure 1 As shown, it includes: S100: Collects data on the irradiance and ambient temperature of the environment where the photovoltaic module is located, and inputs them into the support vector regression prediction model after the hyperparameters have been optimized by the chaotic sparrow search algorithm. The prediction model outputs the reference value of the maximum power point voltage of the photovoltaic module under the current operating conditions. S200: Construct a second-order linear extended state observer, treating model parameter perturbations and external disturbances as the total extended state, and output real-time photovoltaic observations; the real-time observations include photovoltaic output voltage, voltage derivative, and total disturbance observations; S300: Based on the voltage reference value and the real-time observed voltage, the tracking error is calculated, an error integration element is constructed to obtain an extended state space model, and the system controllability is verified using the rank criterion. Then, the matrix Riccati algebraic equation is solved to obtain the optimal state feedback matrix. S400: The total disturbance observation is introduced into the control law for feedforward compensation to generate the optimal control law. The DC / DC converter is controlled according to the optimal control law to adjust the output voltage of the photovoltaic module.

[0019] It should be noted that existing mainstream photovoltaic MPPT control methods, such as the traditional perturbation observation method and the incremental conductance method, have obvious technical defects: these methods require repeated application of perturbations to achieve optimization, resulting in continuous oscillation of photovoltaic output power and high ineffective losses. In particular, they are prone to tracking lag and misjudgment when illumination conditions change abruptly. On the other hand, the method of predicting the maximum power voltage using conventional support vector regression relies heavily on trial and error to obtain the hyperparameters of the SVR model, which easily leads to weak generalization ability and inability to adapt to complex operating conditions. In addition, the conventional robust optimal tracking control for adjusting the DC / DC output voltage does not consider the real-time observation and compensation of model perturbations and external disturbances, which will cause a decrease in robustness when system parameters fluctuate, and it is difficult to balance dynamic response speed and steady-state tracking accuracy. Referring to Figure 2, this invention uses a chaotic sparrow algorithm to globally optimize the hyperparameters of the SVR model, accurately fitting the nonlinear mapping relationship between irradiance, temperature, and the maximum power point voltage of photovoltaics under different illumination conditions; it uses an extended state observer to observe and compensate for model parameter perturbations and external disturbances in real time; and it combines robust optimal tracking control to achieve precise adjustment of the DC / DC converter, ensuring that the photovoltaic power generation system can operate stably and efficiently at the maximum power point under different operating conditions, further improving the accuracy and stability of MPPT control of the photovoltaic power generation system. It can be widely applied to residential photovoltaic, distributed photovoltaic power stations, photovoltaic microgrids, and other technical scenarios.

[0020] Furthermore, there is a nonlinear relationship between irradiance, ambient temperature, and the output voltage at the photovoltaic maximum power point: in, This is the output voltage corresponding to the maximum power point. For irradiation intensity, The ambient temperature.

[0021] by Represents the input vector, that is ,by Represents output, that is , The nonlinear model to be fitted is represented by the SVR nonlinear regression function: ;in, It is a nonlinear function. w To represent the weight vector, b For bias.

[0022] The construction of the chaotic sparrow optimization support vector regression prediction model in this embodiment of the invention includes: The dataset of output voltage corresponding to the maximum power point of photovoltaic modules under different irradiation intensities and different ambient temperatures was collected as the original dataset; The original dataset is normalized and then divided into a training dataset and a test dataset; let the training set be: Based on the accuracy requirements of the prediction model, the following is introduced The insensitive loss function represents the upper limit of the absolute error between the predicted and the true values. ; parameter The nonlinear regression problem is transformed into an optimization problem, and slack variables are added. With penalty coefficient The optimization objective is expressed as: in, Indicates the confidence range. C This indicates the degree of penalty for samples that exceed the error.

[0023] Furthermore, we introduce the Lagrange multiplier. and The dual form of the above convex optimization problem is obtained, expressed as: Using kernel function Replace the inner product in the feature space Then it transforms into: satisfy and of These are called support vectors, resulting in a support vector regression model.

[0024] In this embodiment of the invention, the support vector regression model in step S100 adopts the RBF kernel function, and the mathematical expression of the support vector regression model is: in, The kernel width parameter. For the input vector, For bias.

[0025] In this embodiment of the invention, step S100, where the chaotic sparrow search algorithm optimizes the hyperparameters of the support vector regression prediction model, includes: Set the population size, maximum number of iterations, and convergence threshold; The sparrow population locations are initialized using the Tent chaotic mapping to generate an initial population; The fitness function is to minimize the mean squared error of support vector regression prediction. The fitness value of each individual in the population is calculated. In each iteration, sparrows are divided into discoverers, joiners and watchers according to their fitness values, and their positions are updated proportionally. When the fitness value changes less than the convergence threshold in consecutive iterations, the optimal hyperparameter penalty coefficient and kernel function width are output. When the fitness value changes more than or equal to the convergence threshold in consecutive iterations, the iteration updates continue. A support vector regression model is trained using optimal hyperparameters, and the trained model is used to predict the reference value of the maximum power point voltage of photovoltaic power in real time.

[0026] Specifically, the steps for optimizing the hyperparameters of a support vector regression model using the Chaotic Sparrow Optimization algorithm include: Set population size N =30, maximum number of iterations Convergence threshold Tent chaotic mapping parameters Hyperparameter optimization range: , ; Calculate individual fitness, update positions according to the division of labor of discoverer-joiner-watcher, introduce chaotic perturbation to avoid premature convergence, where the proportion of discoverers is set to 20% and the proportion of watchers is set to 10%; When the fitness value changes less than the convergence threshold for 10 consecutive iterations, the iteration is terminated, and the optimal hyperparameter penalty coefficient and kernel function width are output. The optimal parameters (penalty coefficient and kernel width) are substituted into the training SVR model, and the model is used as the final prediction model after being verified on the test set.

[0027] It should be noted that during actual operation, by inputting the real-time collected irradiance and temperature signals, the model output is the predicted value of the voltage corresponding to the maximum power point, and this predicted value is used as the reference value for MPPT control.

[0028] As shown in Figure 3, the specific steps of the modeling and optimization process for the chaotic sparrow optimization support vector regression prediction model include: Collect photovoltaic data under different operating conditions, and perform normalization and dataset partitioning; Set algorithm parameters and search range, use chaotic mapping to improve population ergonomics, evaluate individual fitness, and simulate the behavior of sparrow discoverers and joiners to update positions; Chaotic perturbation is introduced to avoid the algorithm getting trapped in local optima. It is then checked whether the convergence condition is met. If it is met, the optimized penalty coefficient and kernel function width are output. If not, the algorithm continues to update. The SVR model is trained using optimal parameters, and the maximum power point voltage is output in real time.

[0029] Furthermore, the fitness function is defined as minimizing the mean squared error of support vector regression predictions. The fitness function is defined as follows: in, This represents the fitness value; a smaller value indicates higher prediction accuracy. For the support vector regression model Predicted voltage values ​​for each sample. For the first Measured voltage values ​​for each sample The total number of samples.

[0030] Furthermore, a second-order linear extended state observer is constructed, which treats the uncertainties of the DC / DC converter model and external disturbances as the total extended state, and observes and feeds forward to compensate for them in real time.

[0031] In this embodiment of the invention, the mathematical model of the second-order linear extended state observer in step S200 is expressed as follows: in, The observed value of photovoltaic output voltage. The differential observation value of the photovoltaic output voltage. This represents the total system disturbance observations, which include both model perturbations and external disturbances. , , The observer gain coefficient, To control the input gain, The controller input is the PWM duty cycle of the DC / DC converter. This represents the actual voltage output of the photovoltaic system. It is a first-order differential operator.

[0032] It should be noted that, in the embodiments of the present invention, the observer gain coefficients are respectively set as follows: , , .

[0033] Furthermore, a robust optimal controller is designed to control the photovoltaic power generation DC / DC converter. Figure 5 The diagram shows the hardware topology of a Boost DC / DC converter. To ensure that the actual output voltage of the photovoltaic system accurately tracks the voltage reference value corresponding to the maximum power, a small-signal linear state-space model of the Boost DC / DC converter is established, represented as follows: in, z for m A ×1 dimensional system state vector , U pv The output voltage of the photovoltaic system. i L This represents the inductor current of the Boost converter. u To control the input vector, y For the system output vector, A For the system matrix, B For the input matrix, C This is the output matrix.

[0034] Furthermore, the rank criterion is used to verify the complete controllability of the system and ensure that optimal control is achievable. The criterion is shown in the following formula: That is, matrix Rank and order of the system model When they are equal, the system is fully controllable.

[0035] In this embodiment of the invention, step S300 involves calculating the tracking error between the voltage reference value and the real-time observed voltage, constructing an error integration stage to obtain an extended state-space model, including: Define tracking error Construct the error integral term The extended state vector is obtained. ; Establish extended system matrix and extended input matrix This makes the extended state-space model satisfy: in, This is the reference value for the maximum power point voltage. Photovoltaic output voltage, This represents the inductor current of the Boost converter.

[0036] It should be noted that in the embodiments of the present invention, an error integral term is constructed to eliminate steady-state error.

[0037] In this embodiment of the invention, step S300, solving the matrix Riccati algebraic equation to obtain the optimal state feedback matrix, includes: Define the performance index of the quadratic form as: in, For performance index functionals, It is a positive definite symmetric state weighting matrix. It is a positive definite control weighting matrix; Solve the matrix Riccati algebraic equation We obtain a symmetric positive definite solution matrix. Based on symmetric positive definite solution matrices Calculate the optimal state feedback matrix .

[0038] In this embodiment of the invention, step S400, which involves introducing the total disturbance observation into the control law for feedforward compensation to generate the optimal control law, includes: The optimal control law including extended state observer disturbance compensation is expressed as: in, For robust optimal tracking control, This is the total disturbance feedforward compensation term.

[0039] like Figure 4 As shown, the specific steps of the robust optimal control law design process based on a second-order linear extended state observer include: A second-order linear extended state observer is constructed to estimate the system state and total disturbance in real time. Establish a linear state-space model of the Boost DC / DC converter, and use the rank criterion to verify whether the model is completely controllable. If it is controllable, proceed to the next step to build an extended state-space model. If it is not controllable, skip the error construction step and proceed to the performance index definition stage. Define the tracking error and construct an extended state-space model, incorporating the tracking error integral term into the state vector to form an extended system.

[0040] Define the mathematical expression of the control target, solve the matrix Riccati algebraic equation, and obtain the symmetric positive definite solution matrix; The optimal state feedback matrix is ​​obtained by solving the solution matrix to obtain the feedback gain matrix; The final output control quantity includes a state feedback term and a disturbance compensation term observed by the ESO, thereby achieving disturbance rejection control.

[0041] It should be noted that this invention uses the chaotic sparrow algorithm to achieve global optimization of the hyperparameters of the support vector regression model, accurately fitting the nonlinear mapping relationship between irradiance, temperature and photovoltaic maximum power point voltage under different illumination conditions; it uses an extended state observer to observe and compensate for model parameter perturbations and external disturbances in real time; and it combines robust optimal tracking control to achieve precise adjustment of the DC / DC converter, ensuring that the photovoltaic power generation system can operate stably and efficiently at the maximum power point under different operating conditions.

[0042] Example 2 is an embodiment of the present invention. This embodiment differs from the first embodiment in that it provides a photovoltaic maximum power point tracking control system, comprising: The prediction module is used to collect data on the irradiance and ambient temperature of the environment where the photovoltaic module is located, and input them into the support vector regression prediction model after the hyperparameters are optimized by the chaotic sparrow search algorithm. The prediction model outputs the reference value of the maximum power point voltage of the photovoltaic module under the current operating conditions. The state observation module is used to construct a second-order linear extended state observer, which treats model parameter perturbations and external disturbances as the total extended state and outputs real-time photovoltaic observations. The real-time observations include photovoltaic output voltage, voltage derivative, and total disturbance observations. The error processing module is used to calculate the tracking error based on the voltage reference value and the real-time observed voltage, construct the error integration link to obtain the extended state space model, verify the controllability of the system using the rank criterion, and then solve the matrix Riccati algebraic equation to obtain the optimal state feedback matrix. The control module is used to introduce the total disturbance observation into the control law for feedforward compensation to generate the optimal control law, and control the DC / DC converter according to the optimal control law to adjust the output voltage of the photovoltaic module.

[0043] Specifically, when each module of the photovoltaic maximum power point tracking control system in this embodiment is executed, it implements the steps of the photovoltaic maximum power point tracking control method in Embodiment 1, for example: In one implementation, a photovoltaic maximum power point tracking control system may perform the following steps: Set the population size, maximum number of iterations, and convergence threshold; The sparrow population locations are initialized using the Tent chaotic mapping to generate an initial population; The fitness function is to minimize the mean squared error of support vector regression prediction. The fitness value of each individual in the population is calculated. In each iteration, sparrows are divided into discoverers, joiners and watchers according to their fitness values, and their positions are updated proportionally. When the fitness value changes less than the convergence threshold in consecutive iterations, the optimal hyperparameter penalty coefficient and kernel function width are output. When the fitness value changes more than or equal to the convergence threshold in consecutive iterations, the iteration updates continue. A support vector regression model is trained using optimal hyperparameters, and the trained model is used to predict the reference value of the maximum power point voltage of photovoltaic power in real time.

[0044] The support vector regression model uses the RBF kernel function, and its mathematical expression is as follows: in, The kernel width parameter. For the input vector, For bias.

[0045] The mathematical model of the second-order linear extended state observer is expressed as follows: in, The observed value of photovoltaic output voltage. The differential observation value of the photovoltaic output voltage. This represents the total system disturbance observations, which include both model perturbations and external disturbances. , , The observer gain coefficient, To control the input gain, For controller input, This represents the actual voltage output of the photovoltaic system. It is a first-order differential operator.

[0046] Define tracking error Construct the error integral term The extended state vector is obtained. ; Establish extended system matrix and extended input matrix This makes the extended state-space model satisfy: in, This is the reference value for the maximum power point voltage. Photovoltaic output voltage, This represents the inductor current of the Boost converter.

[0047] Define the performance index of the quadratic form as: in, For performance index functionals, It is a positive definite symmetric state weighting matrix. It is a positive definite control weighting matrix; Solve the matrix Riccati algebraic equation We obtain a symmetric positive definite solution matrix. Based on symmetric positive definite solution matrices Calculate the optimal state feedback matrix .

[0048] The optimal control law including extended state observer disturbance compensation is expressed as: in, For robust optimal tracking control, This is the total disturbance feedforward compensation term.

[0049] This embodiment also provides an electronic device applicable to photovoltaic maximum power point tracking control methods, including: The system includes a memory and a processor. The memory stores computer-executable instructions, and the processor executes these instructions to implement the photovoltaic maximum power point tracking control method proposed in the above embodiments.

[0050] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the photovoltaic maximum power point tracking control method proposed in the above embodiments.

[0051] The storage medium proposed in this embodiment belongs to the same inventive concept as the photovoltaic maximum power point tracking control method proposed in the above embodiments. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0052] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0053] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A photovoltaic maximum power point tracking control method, characterized in that, include: Data on the irradiance and ambient temperature of the environment where the photovoltaic module is located are collected and input into the support vector regression prediction model after the hyperparameters are optimized by the chaotic sparrow search algorithm. The prediction model outputs the reference value of the maximum power point voltage of the photovoltaic module under the current operating conditions. A second-order linear extended state observer is constructed, which treats model parameter perturbations and external disturbances as the total extended state and outputs real-time photovoltaic observations. The real-time observations include photovoltaic output voltage, voltage derivative, and total disturbance observations. The tracking error is calculated based on the voltage reference value and the real-time observed voltage. An extended state space model is obtained by constructing an error integration loop. After verifying the system's controllability using the rank criterion, the matrix Riccati algebraic equation is solved to obtain the optimal state feedback matrix. The total disturbance observation is introduced into the control law for feedforward compensation to generate the optimal control law. The DC / DC converter is controlled according to the optimal control law to adjust the output voltage of the photovoltaic module.

2. The photovoltaic maximum power point tracking control method as described in claim 1, characterized in that, The chaotic sparrow search algorithm optimizes the hyperparameters of the support vector regression prediction model, including: Set the population size, maximum number of iterations, and convergence threshold; The sparrow population locations are initialized using the Tent chaotic mapping to generate an initial population; The fitness function is to minimize the mean squared error of support vector regression prediction. The fitness value of each individual in the population is calculated. In each iteration, sparrows are divided into discoverers, joiners and watchers according to their fitness values, and their positions are updated proportionally. When the fitness value changes less than the convergence threshold in consecutive iterations, the optimal hyperparameter penalty coefficient and kernel function width are output. When the fitness value changes more than or equal to the convergence threshold in consecutive iterations, the iteration updates continue. The optimal hyperparameters are used to train a support vector regression model, and the trained model is used to predict the reference value of the maximum power point voltage of photovoltaic power in real time.

3. The photovoltaic maximum power point tracking control method as described in claim 2, characterized in that, The support vector regression model uses the RBF kernel function, and the mathematical expression of the support vector regression model is: in, The kernel width parameter. For the input vector, For bias.

4. The photovoltaic maximum power point tracking control method as described in claim 3, characterized in that, The mathematical model of the second-order linear extended state observer is expressed as follows: in, The observed value of photovoltaic output voltage. The differential observation value of the photovoltaic output voltage. This represents the total system disturbance observations, which include both model perturbations and external disturbances. , , The observer gain coefficient, To control the input gain, For controller input, This represents the actual voltage output of the photovoltaic system. It is a first-order differential operator.

5. The photovoltaic maximum power point tracking control method as described in claim 4, characterized in that, Based on the voltage reference value and the real-time observed voltage, the tracking error is calculated, and an error integration stage is constructed to obtain an extended state-space model, including: Define tracking error Construct the error integral term The extended state vector is obtained. ; Establish extended system matrix and extended input matrix This makes the extended state-space model satisfy: in, This is the reference value for the maximum power point voltage. Photovoltaic output voltage, This represents the inductor current of the Boost converter.

6. The photovoltaic maximum power point tracking control method as described in claim 5, characterized in that, Solving the matrix Riccati algebraic equation to obtain the optimal state feedback matrix includes: Define the performance index of the quadratic form as: in, For performance index functionals, It is a positive definite symmetric state weighting matrix. It is a positive definite control weighting matrix; Solve the matrix Riccati algebraic equation We obtain a symmetric positive definite solution matrix. Based on symmetric positive definite solution matrices Calculate the optimal state feedback matrix .

7. The photovoltaic maximum power point tracking control method as described in claim 6, characterized in that, The total disturbance observations are incorporated into the control law for feedforward compensation to generate the optimal control law, including: The optimal control law with extended state observer disturbance compensation is expressed as: in, For robust optimal tracking control, This is the total disturbance feedforward compensation term.

8. A photovoltaic maximum power point tracking control system, applied to the method described in any one of claims 1-7, characterized in that, include: The prediction module is used to collect data on the irradiance and ambient temperature of the environment where the photovoltaic module is located, and input them into the support vector regression prediction model after the hyperparameters are optimized by the chaotic sparrow search algorithm. The prediction model outputs the reference value of the maximum power point voltage of the photovoltaic module under the current operating conditions. The state observation module is used to construct a second-order linear extended state observer, which treats model parameter perturbations and external disturbances as the total extended state and outputs real-time photovoltaic observation values. The real-time observation values ​​include photovoltaic output voltage, voltage derivative, and total disturbance observation values. The error processing module is used to calculate the tracking error based on the voltage reference value and the real-time observed voltage, construct an error integration loop to obtain an extended state space model, verify the system controllability using the rank criterion, and then solve the matrix Riccati algebraic equation to obtain the optimal state feedback matrix. The control module is used to introduce the total disturbance observation into the control law for feedforward compensation to generate the optimal control law, and control the DC / DC converter according to the optimal control law to adjust the output voltage of the photovoltaic module.

9. An electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions, which, when executed by the processor, implement the steps of the photovoltaic maximum power point tracking control method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of the photovoltaic maximum power point tracking control method according to any one of claims 1 to 7.

Images