Photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering

Abstract

The invention discloses a photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering, and relates to the technical field of inverter control of a photovoltaic power generation system, and the method comprises the steps: sampling a photovoltaic grid-connected inverter, carrying out the system modeling, defining a control parameter needing to be identified, and selecting a proper state variable and observation variable design; performing linearization processing on the nonlinear system to obtain a linearized state transition and observation matrix, constructing a prediction and update equation of a filter, and setting initial state estimation and a covariance matrix; actual operation data of the photovoltaic grid-connected inverter are collected, the improved Kalman filter is operated, a control signal is adjusted, and finally, the converter is controlled through modulation, so that the required effect is achieved. By introducing the improved Kalman filtering algorithm, the control parameters of the inverter can be estimated more accurately, so that the operation performance of the photovoltaic system is optimized, and the identification precision and the overall performance of the system are improved through the EKF-SA algorithm.

Description

Technical Field

[0001] The invention relates to the technical field of inverter control of photovoltaic power generation systems, and in particular to a photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering. Background Art

[0002] With the global energy structure transformation and the emphasis on renewable energy, photovoltaic power generation, as the main clean energy, is expanding its application, and photovoltaic grid-connected inverters play a key role in the system. Accurate control parameters can improve the dynamic response speed and stability of the system and ensure smooth grid connection. However, traditional parameter identification methods often face challenges when dealing with changes in photovoltaic system and grid conditions, resulting in reduced identification accuracy and affecting system performance.

[0003] After years of development, traditional parameter identification methods (such as the least squares method) have a mature theoretical basis and are widely used in various control systems. However, they are easily affected by factors such as nonlinear dynamic characteristics of the system, noise interference, and model uncertainty, and it is difficult to provide satisfactory results. In this context, Kalman filtering, as a classic parameter estimation method, has gradually been introduced into the control parameter identification of photovoltaic inverters. Although the standard Kalman filter performs well in dealing with linear system state estimation problems, the applicability of traditional methods is limited due to the nonlinear characteristics of photovoltaic inverters. Therefore, an improved Kalman filter algorithm for these nonlinear dynamic characteristics is particularly important. The key to improving the Kalman filter algorithm is how to effectively deal with the nonlinear characteristics of the system.

[0004] By introducing an improved Kalman filter algorithm, the present invention can more accurately estimate the control parameters of the inverter, thereby optimizing the operating performance of the photovoltaic system. By improving the identification accuracy and the overall performance of the system through the EKF-SA algorithm, this technology will promote technological progress in the field of photovoltaic power generation and provide strong support for global energy transformation. Summary of the invention

[0005] In view of the above-mentioned problems, the present invention is proposed.

[0006] Therefore, the present invention provides a photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering, which can solve the problems mentioned in the background technology and aims to deal with the situation when the system model is incomplete or there is a large uncertainty. The improved Kalman filtering can continuously update the system model through actual measurement data to improve the accuracy of parameter identification.

[0007] To solve the above technical problems, the present invention provides the following technical solutions: a photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering, comprising: sampling the photovoltaic grid-connected inverter, performing system modeling, setting the initial parameters of the inverter controller, defining the control parameters to be identified, and selecting appropriate state variables and observation variables for design; linearizing the nonlinear system to obtain linearized state transfer and observation matrices, constructing the prediction and update equations of the filter, setting the initial state estimation and covariance matrix, and further optimizing the state estimation through the disturbance and acceptance criteria of simulated annealing on the basis of the EKF algorithm; collecting actual operation data of the photovoltaic grid-connected inverter, running the improved Kalman filter, identifying the control parameters, weighting the predicted value and the measured value, using the weighted estimated value as the input of the controller, adjusting the control signal, and finally realizing the control of the converter through modulation to achieve the desired effect.

[0008] As a preferred solution of the photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering described in the present invention, the photovoltaic grid-connected inverter is sampled, system modeling is performed, initial parameters of the inverter controller are set, control parameters to be identified are defined, and appropriate state variables and observation variables are selected for design, including the following steps:

[0009] First, key operating data of the photovoltaic grid-connected inverter is collected;

[0010] Based on the solar radiation intensity and temperature changes, the output characteristic model is established, including output current, voltage, active power and DC capacitance, as shown in the following formula;

[0011]

[0012] Where: I pv is the output current of the photovoltaic array; U dc is the DC bus voltage; P ac is the active power; C is the DC side capacitance; t refers to the time variable;

[0013] Construct a mathematical model of the inverter based on the photovoltaic panel input voltage and grid-connected output voltage;

[0014] According to the output characteristics and load of the photovoltaic panel, the initial gain is set, the proportional gain and the integral gain are defined, and the output voltage of the inverter and the integral variable of the controller are selected as state variables.

[0015] As a preferred solution of the photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering described in the present invention, the nonlinear system is linearized to obtain linearized state transfer and observation matrices, the prediction and update equations of the filter are constructed, the initial state estimation and covariance matrix are set, and on the basis of the EKF algorithm, the state estimation is further optimized through the disturbance and acceptance criteria of simulated annealing, including the following steps:

[0016] Using known system states and input data, calculate the predicted values ​​of the system's state variables at the next moment;

[0017]

[0018] Where T is the sampling period, Yes k-1 The updated value at the moment, Yes k-1 The estimated state value at time, is the input variable;

[0019] Based on the discretized state transfer matrix and system noise covariance matrix, the state error covariance generated in the prediction process is calculated;

[0020]

[0021] In the formula, F k is the discretized state transfer matrix, P k-1 is at k-1 The state error covariance matrix at time , P k is at k The prediction covariance matrix at time, Q is the system noise covariance matrix; is the discretized state transfer matrix F k The transposed matrix of

[0022] Combine the prediction covariance matrix and the observation noise covariance matrix to calculate the Kalman gain;

[0023]

[0024] In the formula, K k is at k The Kalman filter gain matrix at time t, H k is the observation matrix corresponding to the predicted values, and R is the measurement noise covariance matrix; is the Kalman filter gain matrix K k The transposed matrix of

[0025] After obtaining new observations, the state covariance matrix is ​​updated using the Kalman gain;

[0026] P k =P k-1 -K k H k P k-1 ;

[0027] According to the current measurement value and Kalman gain, the state estimate is corrected;

[0028]

[0029] In the formula, Yes k The measured value at the moment.

[0030] The simulated annealing algorithm is introduced to further optimize the state estimation of the EKF algorithm based on state estimation and covariance update.

[0031] As a preferred solution of the photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering described in the present invention, a simulated annealing algorithm is introduced to further optimize the state estimation of the EKF algorithm on the basis of state estimation and covariance update, including the following steps:

[0032] Calculate the objective function value of the current state estimate to evaluate the performance of the model;

[0033]

[0034] Among them, E k is the objective function value, which is used to measure the quality of the current estimated state, z k is the actual measured value of k at the current moment, is estimated by the state The predicted value obtained by observing model h.

[0035] Generate new candidate states by randomly perturbing the current state estimate;

[0036]

[0037] In the formula, Δx is a randomly generated disturbance; is the new candidate state estimate after the disturbance;

[0038] Calculate the objective function value of the new solution and evaluate the quality of the solution;

[0039]

[0040] In the formula, E new is the new objective function value, which is used to measure the quality of the new candidate state estimate after the disturbance;

[0041] Decide whether to accept a new solution based on the acceptance probability, even if the new solution performs poorly, to ensure that the algorithm does not fall into a local optimum;

[0042]

[0043] In the formula, P refers to the acceptance probability, which is used to determine whether to accept the new state, T k is the temperature at time k;

[0044] Update the current state and objective function value to reflect whether the new solution is accepted.

[0045]

[0046] As a preferred solution of the photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering described in the present invention, the actual operation data of the photovoltaic grid-connected inverter is collected, the improved Kalman filter is run, the control parameters are identified, the predicted value and the measured value are weighted, the weighted estimated value is used as the input of the controller, the control signal is adjusted, and finally the control of the converter is realized by modulation to achieve the desired effect, including the following steps:

[0047] The dynamic characteristics near the grid connection point are simplified, and the phase of the three-phase voltage vector is locked using a phase-locked loop, which is converted into values ​​in the d and q axis coordinate systems to achieve phase synchronization operation of the control system.

[0048]

[0049] Where ΔP ac =P ac -P ac0 , Δx 1 =x 1 -x 10 , ΔU t =U t -U t0 , ΔU dc =U dc -U dc0 , ΔU dc =U dc -U dc0 , and a 11 ,a 12 ,b 11 ,c 11 ,c 12 ,d 11 The coefficients are as follows:

[0050]

[0051] c 11 =U t0 K p,c 12 =-U t0 K i

[0052]

[0053] Where U t is the voltage at the grid common point, x 1 is the variable of the PI control integrator, U t0 is the voltage of the grid common point in steady state, x 10 is the variable of the steady-state PI control integrator. 0 refers to the ideal state value of the system when there is no disturbance. 0 is the steady-state power reference value;

[0054] Accurate estimation of control parameters is achieved through four stages: data collection, initialization and iterative checking, EKF execution and objective function calculation, and acceptance of new solutions and update of status.

[0055] As a preferred solution of the photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering described in the present invention, wherein: the state of the system is estimated using the improved Kalman filtering to obtain a predicted value, that is, the predicted state value at time k, the predicted value and the actual measured value are weighted, and the obtained weighted value is used as the input of the controller to adjust the control signal;

[0056] After the voltage outer loop parameters are adjusted, the control loop output updates the control input v α and v β , and finally the converter is controlled through modulation.

[0057] As a preferred solution of the photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering described in the present invention, the key operating data includes but is not limited to photovoltaic panel input voltage, grid-connected output voltage, current and power.

[0058] To further solve the above technical problems, the present invention provides the following technical solutions: A photovoltaic grid-connected inverter control parameter identification system based on improved Kalman filtering, comprising:

[0059] The model building module is used to sample the photovoltaic grid-connected inverter, perform system modeling, set the initial parameters of the inverter controller, define the control parameters that need to be identified, and select appropriate state variables and observation variables for design;

[0060] The processing optimization module is used to linearize the nonlinear system, obtain the linearized state transfer and observation matrix, construct the prediction and update equations of the filter, set the initial state estimate and covariance matrix, and further optimize the state estimate based on the EKF algorithm through the disturbance and acceptance criteria of simulated annealing;

[0061] The control module is used to collect the actual operating data of the photovoltaic grid-connected inverter, run the improved Kalman filter, identify the control parameters, weight the predicted value and the measured value, use the weighted estimated value as the input of the controller, adjust the control signal, and finally control the converter through modulation to achieve the desired effect.

[0062] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of the photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering are implemented as described above.

[0063] A computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of the photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering as described above are implemented.

[0064] Beneficial effects of the present invention: By introducing the simulated annealing algorithm into the EKF algorithm, the present invention can further optimize the state estimation based on the state estimation and covariance update through the perturbation and acceptance criteria, and enhance the adaptability of the EKF to uncertainty and nonlinear dynamics. This method can help find a better state estimate in a high-dimensional space. It solves the problem of accuracy in the identification of control parameters of photovoltaic grid-connected inverters. This method effectively overcomes the shortcomings of traditional identification methods in dynamic environments, significantly improves the accuracy and robustness of parameter identification, and achieves better control performance in practical applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0066] Figure 1 It is a control principle structure block diagram of a photovoltaic grid-connected inverter based on an improved Kalman filter proposed by the present invention;

[0067] Figure 2 It is a parameter identification algorithm flow chart of the photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering proposed by the present invention;

[0068] Figure 3 It is a schematic diagram of waveform requirements of a photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering proposed by the present invention;

[0069] Figure 4 It is another waveform requirement schematic diagram of the photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering proposed by the present invention. DETAILED DESCRIPTION

[0070] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0071] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0072] Example 1, reference Figure 1 , which is an embodiment of the present invention, provides a photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering.

[0073] S1: Sample the photovoltaic grid-connected inverter, perform system modeling, set the initial parameters of the inverter controller, define the control parameters that need to be identified, and select appropriate state variables and observation variables for design.

[0074] S1.1: First, collect the key operating data of the photovoltaic grid-connected inverter, including the photovoltaic panel input voltage, grid-connected output voltage, current and power, etc.

[0075] S1.2: Photovoltaic array model: Based on the solar radiation intensity and temperature changes, the output characteristic model is established, including output current, voltage, active power and DC capacitance, as shown in the following formula:

[0076]

[0077] Where: I pv is the output current of the photovoltaic array; U dc is the DC bus voltage; P ac is the active power; C is the DC side capacitance; t refers to the time variable.

[0078] S1.3: Inverter model design: Construct the mathematical model of the inverter based on the photovoltaic panel input voltage and grid-connected output voltage.

[0079] S1.4: Gain setting and control: According to the output characteristics and load of the photovoltaic panel, set the initial gain, define the proportional gain and integral gain, and select the output voltage of the inverter and the integral variable of the controller as the state variable.

[0080] S2: Linearize the nonlinear system to obtain the linearized state transfer and observation matrix, construct the prediction and update equations of the filter, set the initial state estimate and covariance matrix, and further optimize the state estimate based on the EKF algorithm through the disturbance and acceptance criteria of simulated annealing.

[0081] S2.1: Using the known system state and input data, calculate the predicted value of the system state variable at the next moment;

[0082]

[0083] Where T is the sampling period, Yes k-1 The updated value at the moment, Yes k-1 The estimated state value at time, is the input variable.

[0084] S2.2: Based on the discretized state transfer matrix and the system noise covariance matrix, calculate the state error covariance generated in the prediction process;

[0085]

[0086] In the formula, F k is the discretized state transfer matrix, P k-1 is at k-1 The state error covariance matrix at time , P k is at k The prediction covariance matrix at time, Q is the system noise covariance matrix; is the discretized state transfer matrix F k The transposed matrix of .

[0087] S2.3: Combine the prediction covariance matrix and the observation noise covariance matrix to calculate the Kalman gain;

[0088]

[0089] In the formula, K k is at k The Kalman filter gain matrix at time t, H k is the observation matrix corresponding to the predicted values, and R is the measurement noise covariance matrix; is the Kalman filter gain matrix K k The transposed matrix of .

[0090] S2.4: After obtaining new observations, the state covariance matrix is ​​updated using the Kalman gain;

[0091] P k =P k-1 -K k H k P k-1 ;

[0092] S2.5: Correct the state estimate based on the current measurement value and Kalman gain.

[0093]

[0094] In the formula, Yes k The measured value at the moment.

[0095] S2.6: Introduce the simulated annealing algorithm to further optimize the state estimation of the EKF algorithm based on state estimation and covariance update.

[0096] S2.6.1: Calculate the objective function value of the current state estimate to evaluate the performance of the model;

[0097]

[0098] Among them, E k is the objective function value, which is used to measure the quality of the current estimated state, z k is the actual measured value of k at the current moment, is estimated by the state The predicted value obtained by observing model h.

[0099] S2.6.2: Generate new candidate states by randomly perturbing the current state estimate;

[0100]

[0101] In the formula, Δx is a randomly generated disturbance; is the new candidate state estimate after the perturbation.

[0102] S2.6.3: Calculate the objective function value of the new solution and evaluate the quality of the solution;

[0103]

[0104] In the formula, E new It is the new objective function value, which is used to measure the quality of the new candidate state estimate after the disturbance.

[0105] S2.6.4: Decide whether to accept a new solution based on the acceptance probability, even if the new solution performs poorly, to ensure that the algorithm does not fall into a local optimum;

[0106]

[0107] In the formula, P refers to the acceptance probability, which is used to determine whether to accept the new state, T k is the temperature at time k.

[0108] S2.6.5: Update the current state and objective function value to reflect whether the new solution is accepted.

[0109]

[0110] S3: Collect the actual operating data of the photovoltaic grid-connected inverter, run the improved Kalman filter, identify the control parameters, weight the predicted value and the measured value, use the weighted estimated value as the input of the controller, adjust the control signal, and finally control the converter through modulation to achieve the desired effect.

[0111] S3.1: Simplify the dynamic characteristics near the grid connection point, use a phase-locked loop to lock the phase of the three-phase voltage vector, and convert it into values ​​in the d and q axis coordinate system to achieve phase synchronization operation of the control system;

[0112]

[0113] Where ΔP ac =P ac -P ac0 , Δx 1 =x 1 -x 10 , ΔU t =U t -U t0 , ΔU dc =U dc -U dc0 , ΔU dc =U dc -U dc0 , and a 11 ,a 12 ,b 11 ,c 11 ,c 12 ,d 11 The coefficients are as follows:

[0114]

[0115] c 11 =U t0 K p ,c 12 =-Ut0 K i

[0116]

[0117] Where U t is the voltage at the grid common point, x 1 is the variable of the PI control integrator, U t0 is the voltage of the grid common point in steady state, x 10 is the variable of the steady-state PI control integrator. 0 refers to the ideal state value of the system when there is no disturbance. 0 is the steady-state power reference value.

[0118] S3.2: Based on S3.1, accurate estimation of control parameters is achieved through four stages: data collection, initialization and iterative checking, EKF execution and objective function calculation, and acceptance of new solutions and update of status.

[0119] S3.3: Based on S3.2, the improved Kalman filter is used to estimate the state of the system and obtain the predicted value (the predicted state value at time k). In order to improve the quality of the input signal of the controller, the predicted value and the actual measured value are weighted (the waveform result is as follows Figure 3 shown).

[0120]

[0121] In the formula, is the predicted state value at time k-1.

[0122] The obtained weighted value is used as the input of the controller to adjust the control signal.

[0123] S3.4: After the voltage outer loop parameters are adjusted, the control loop output updates the control input v α and v β , and finally the converter is controlled through modulation.

[0124] In summary, the present invention, by introducing the simulated annealing algorithm into the EKF algorithm, can further optimize the state estimation on the basis of state estimation and covariance update through the perturbation and acceptance criteria, and enhance the adaptability of EKF to uncertainty and nonlinear dynamics. This method can help find a better state estimate in a high-dimensional space. The problem of accuracy in the identification of control parameters of photovoltaic grid-connected inverters is solved. This method effectively overcomes the shortcomings of traditional identification methods in dynamic environments, significantly improves the accuracy and robustness of parameter identification, and achieves better control performance in practical applications.

[0125] Embodiment 2 is an embodiment of the present invention, which provides a photovoltaic grid-connected inverter control parameter identification system based on improved Kalman filtering, including:

[0126] The model building module is used to sample the photovoltaic grid-connected inverter, perform system modeling, set the initial parameters of the inverter controller, define the control parameters that need to be identified, and select appropriate state variables and observation variables for design;

[0127] The processing optimization module is used to linearize the nonlinear system, obtain the linearized state transfer and observation matrix, construct the prediction and update equations of the filter, set the initial state estimate and covariance matrix, and further optimize the state estimate based on the EKF algorithm through the disturbance and acceptance criteria of simulated annealing;

[0128] The control module is used to collect the actual operating data of the photovoltaic grid-connected inverter, run the improved Kalman filter, identify the control parameters, weight the predicted value and the measured value, use the weighted estimated value as the input of the controller, adjust the control signal, and finally control the converter through modulation to achieve the desired effect.

[0129] Embodiment 3 is an embodiment of the present invention, which is different from the previous embodiment in that: if the function is implemented in the form of a software function unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0130] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0131] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0132] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0133] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering, characterized in that: include: Sampling photovoltaic grid-connected inverters, modeling the system, setting the initial parameters of the inverter controller, defining the control parameters that need to be identified, and selecting the state variables and observation variables for design; Linearize the nonlinear system to obtain the linearized state transfer and observation matrix, construct the prediction and update equations of the filter, set the initial state estimate and covariance matrix, and further optimize the state estimate based on the EKF algorithm through the disturbance and acceptance criteria of simulated annealing; Collect the actual operating data of the photovoltaic grid-connected inverter, run the improved Kalman filter, identify the control parameters, weight the predicted value and the measured value, use the weighted estimated value as the input of the controller, adjust the control signal, and finally control the converter through modulation to achieve the desired effect.

2. The photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering according to claim 1, characterized in that: Sampling the photovoltaic grid-connected inverter, modeling the system, setting the initial parameters of the inverter controller, defining the control parameters that need to be identified, and selecting appropriate state variables and observation variables for design include the following steps: First, key operating data of the photovoltaic grid-connected inverter is collected; Based on the solar radiation intensity and temperature changes, the output characteristic model is established, including output current, voltage, active power and DC capacitance, as shown in the following formula; Where: I pv is the output current of the photovoltaic array; U dc is the DC bus voltage; P ac is the active power; C is the DC side capacitance, and t refers to the time variable; Construct a mathematical model of the inverter based on the photovoltaic panel input voltage and grid-connected output voltage; According to the output characteristics and load of the photovoltaic panel, the initial gain is set, the proportional gain and the integral gain are defined, and the output voltage of the inverter and the integral variable of the controller are selected as state variables.

3. The photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering according to claim 2, characterized in that: The nonlinear system is linearized to obtain the linearized state transfer and observation matrix, the prediction and update equations of the filter are constructed, the initial state estimate and covariance matrix are set, and the state estimate is further optimized based on the EKF algorithm through the disturbance and acceptance criteria of simulated annealing, including the following steps: Using known system states and input data, calculate the predicted values ​​of the system's state variables at the next moment; Where T is the sampling period, Yes k-1 The updated value at the moment, Yes k-1 The estimated state value at time is the input variable; Based on the discretized state transfer matrix and system noise covariance matrix, the state error covariance generated in the prediction process is calculated; In the formula, F k is the discretized state transfer matrix, P k-1 is at k-1 The state error covariance matrix at the moment, P k is at k The prediction covariance matrix at time, Q is the system noise covariance matrix, is the discretized state transfer matrix F k The transposed matrix of Combine the prediction covariance matrix and the observation noise covariance matrix to calculate the Kalman gain; In the formula, K k is at k The Kalman filter gain matrix at time t, H k is the observation matrix corresponding to the predicted values, and R is the measurement noise covariance matrix, is the Kalman filter gain matrix K k The transposed matrix of After obtaining new observations, the state covariance matrix is ​​updated using the Kalman gain; P k =P k-1 -K k H k P k-1 ; Correct the state estimate based on the current measurement value and Kalman gain; In the formula, Yes k The measured value at the moment. The simulated annealing algorithm is introduced to further optimize the state estimation of the EKF algorithm based on state estimation and covariance update.

4. The photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering according to claim 3, characterized in that: The simulated annealing algorithm is introduced to further optimize the state estimation of the EKF algorithm based on state estimation and covariance update, including the following steps: Calculate the objective function value of the current state estimate to evaluate the performance of the model; Among them, E k is the objective function value, which is used to measure the quality of the current estimated state, z k is the actual measured value of k at the current moment, is estimated by the state The predicted value obtained by observing model h. Generate new candidate states by randomly perturbing the current state estimate; In the formula, Δx is a randomly generated disturbance, is the new candidate state estimate after the disturbance; Calculate the objective function value of the new solution and evaluate the quality of the solution; In the formula, E new is the new objective function value, which is used to measure the quality of the new candidate state estimate after the disturbance; Decide whether to accept a new solution based on the acceptance probability, even if the new solution performs poorly, to ensure that the algorithm does not fall into a local optimum; In the formula, P refers to the acceptance probability, which is used to determine whether to accept the new state, T k is the temperature at time k; Update the current state and objective function value to reflect whether the new solution is accepted.

5. The photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering according to claim 4, characterized in that: Collect the actual operation data of the photovoltaic grid-connected inverter, run the improved Kalman filter, identify the control parameters, weight the predicted value and the measured value, use the weighted estimated value as the input of the controller, adjust the control signal, and finally control the converter through modulation to achieve the desired effect, including the following steps: The dynamic characteristics near the grid connection point are simplified, and the phase of the three-phase voltage vector is locked using a phase-locked loop, which is converted into values ​​in the d and q axis coordinate systems to achieve phase synchronization operation of the control system. Where ΔP ac =P ac -P ac0 , Δx1=x1-x 10 , ΔU t =U t -U t0 , ΔU dc =U dc -U dc0 , ΔU dc =U dc -U dc0 , and a 11 ,a 12 ,b 11 ,c 11 ,c 12 ,d 11 The coefficients are as follows: c 11 =U t0 K p ,c 12 =-U t0 K i Where U t is the voltage at the grid common point, x1 is the variable of the PI control integrator, U t0 is the voltage at the grid common point in steady state, x 10 It is the variable of the steady-state PI control integrator, 0 refers to the ideal state value of the system when there is no disturbance, and P0 is the steady-state power reference value; Accurate estimation of control parameters is achieved through four stages: data collection, initialization and iterative checking, EKF execution and objective function calculation, and acceptance of new solutions and update of status.

6. The photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering according to claim 5, characterized in that: The improved Kalman filter is used to estimate the state of the system to obtain a predicted value, that is, the predicted state value at time k. The predicted value and the actual measured value are weighted, and the obtained weighted value is used as the input of the controller to adjust the control signal; After the voltage outer loop parameters are adjusted, the control loop output updates the control input v α and v β , and finally the converter is controlled through modulation.

7. The photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering according to claim 6, characterized in that: The key operating data include but are not limited to photovoltaic panel input voltage, grid-connected output voltage, current and power.

8. A photovoltaic grid-connected inverter control parameter identification system based on improved Kalman filtering, based on the photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering according to any one of claims 1 to 7, characterized in that: include, The model building module is used to sample the photovoltaic grid-connected inverter, perform system modeling, set the initial parameters of the inverter controller, define the control parameters that need to be identified, and select appropriate state variables and observation variables for design; The processing optimization module is used to linearize the nonlinear system, obtain the linearized state transfer and observation matrix, construct the prediction and update equations of the filter, set the initial state estimate and covariance matrix, and further optimize the state estimate based on the EKF algorithm through the disturbance and acceptance criteria of simulated annealing; The control module is used to collect the actual operating data of the photovoltaic grid-connected inverter, run the improved Kalman filter, identify the control parameters, weight the predicted value and the measured value, use the weighted estimated value as the input of the controller, adjust the control signal, and finally control the converter through modulation to achieve the desired effect.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the photovoltaic grid-connected inverter control parameter identification method based on improved Kalman filtering described in any one of claims 1 to 7 are implemented.

Images