Multifunctional inverter control method and system based on improved recursive least square method
By improving the recursive least squares method to identify inverter parameters in real time, the accuracy reduction problem of traditional methods under parameter changes and interference is solved, high-precision current prediction and control are achieved, and the power quality management effect is improved.
Patent Information
- Application Number
- CN202410821318.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-24
- Publication Date
- 2024-11-15
AI Technical Summary
Traditional parameter identification methods are difficult to maintain high accuracy in the face of system parameter changes and external interference, resulting in a decrease in the control accuracy of the inverter and the inability to respond quickly and deal with faults.
Using a multifunctional inverter control method based on improved recursive least squares method, the state space equation is established by collecting operation data, the correction gain matrix and error covariance matrix are updated, the inductance and parasitic resistance parameters are identified in real time, and the switching signal is optimized using the value function. , realize high-precision current prediction and control.
It improves the control accuracy of model prediction control, improves the effectiveness of power quality management, can quickly respond to parameter changes and faults, and reduces the impact of parameter mismatch.
Smart Images

Figure CN118970964A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power quality management, and in particular to a multifunctional inverter control method and system based on an improved recursive least squares method. Background Art
[0002] In recent years, with the rapid development of power electronics technology, new energy sources such as photovoltaics and wind power have been vigorously developed. However, there are a series of differences in characteristics between the power system with a high proportion of new energy access and the traditional power system. In particular, renewable energy sources such as photovoltaics and wind power are connected to the power system through power electronic equipment as an interface. The widespread application of this access method has not only promoted the popularization of new energy, but also injected a large amount of harmonics into the power system.
[0003] The Multi-Function Grid-Connected Inverter (MFGTI) provides a new harmonic control idea. It uses the remaining capacity of the grid-connected inverter to control harmonics, thereby maximizing the utilization of new energy access equipment; at the same time, by injecting a dynamic compensation current with the same harmonic amplitude as the load current but opposite phase into the grid, the harmonic content of the grid current is significantly reduced. To this end, control strategies such as Proportional Integral Control (PI), Proportional Resonance Control (PR), Repetitive Control (RC) and Model Predictive Current Control (MPCC) are widely adopted. Among them, MPCC has been widely used due to its flexible control, simple design and convenient parameter adjustment.
[0004] However, model predictive control relies heavily on accurate mathematical models, and the accuracy of the parameters in the mathematical model directly affects the control performance. In the actual operation of the multifunctional inverter, the filter works in a high dV / dt state for a long time, which increases the electrical stress on the filter inductance, accelerates insulation aging, and may cause the coil to short-circuit. The inductance value is proportional to the square of the number of coil turns, so the inductance parameter mismatch phenomenon is serious. In addition, the access of new energy is often achieved by adding a filter after the multifunctional inverter. When new energy is accessed on a large scale, harmonic resonance is easily generated, causing the inductance to be more easily damaged. When the inductance parameter mismatch is serious, there is a deviation between the predicted current value and the actual current value, thereby reducing the compensation performance of the multifunctional inverter.
[0005] Aiming at the problem of parameter mismatch leading to decreased control accuracy during the control process, a parameter online identification method based on improved recursive least squares method is proposed. This method can identify parameters in real time during the operation of the multifunctional inverter, and can control harmonics with high accuracy even in the case of filter mismatch. In addition, parameter identification is fast enough when a sudden fault occurs in the inductor. Summary of the invention
[0006] In view of the above-mentioned problems, the present invention is proposed.
[0007] Therefore, the technical problem solved by the present invention is that the traditional parameter identification method is difficult to maintain high accuracy when facing system parameter changes and external interference. The inverter control system requires accurate current prediction to achieve the best control effect. The stable operation of the power system requires the control system to quickly respond to and handle various faults and interferences.
[0008] In order to solve the above technical problems, the present invention provides the following technical solutions: a multifunctional inverter control method based on improved recursive least squares method, comprising:
[0009] Collect the operation data related to the grid-connected inverter and establish the state space equation of the grid-connected inverter;
[0010] According to the state space equation of the grid-connected inverter, an improved recursive least squares mathematical model for inductor parameter identification is established;
[0011] Update the correction gain matrix and error covariance matrix of the improved recursive least squares method, update the estimated parameter values and update the adaptive forgetting factor according to the residual of the identification result;
[0012] Substitute the estimated value of inductance and the estimated value of parasitic resistance at the current moment into the MPC function to obtain the one-step predicted value and the two-step predicted value of current respectively;
[0013] The optimal output voltage is obtained according to the current prediction value, and the action signal of the multifunctional inverter switch tube at the next moment is obtained by using the value function optimization method;
[0014] After a delay of one cycle, the switch action signal of the multifunctional inverter at the next moment is output to realize the switch action of the multifunctional inverter at the next moment, assign the next moment to the current moment, and return to continue parameter update and control.
[0015] As a preferred solution of the multifunctional inverter control method and system based on the improved recursive least squares method described in the present invention, wherein: the operation data related to the grid-connected inverter is collected, including current data, voltage data, power data, environmental data, load conditions, switch signals, frequency data, fault records and historical operation data;
[0016] Establishing the state space equation of the grid-connected inverter includes establishing a mathematical model for the multifunctional inverter and obtaining its state space equation, which is expressed as follows:
[0017]
[0018] Where L represents the filter inductance, R represents the filter equivalent resistance, i L Indicates the inductor current u i mν is the voltage on the inverter side, u pcc Indicates the grid voltage at the MFGCI access point.
[0019] As a preferred solution of the multifunctional inverter control method and system based on the improved recursive least squares method described in the present invention, the improved recursive least squares method mathematical model for inductance parameter identification is established, and the improved recursive least squares method identification mathematical model formula is expressed as:
[0020]
[0021] Among them, y(k) represents the output measurement value of the model at time k, a i represents the coefficient of the autoregressive model, y(ki) represents the output measurement value at time ki, and b i represents the coefficient of the input-output model, u(ki) represents the input measurement value at time ki, ε(k) represents the measurement noise of the system at time k, Φ T (k represents the data vector consisting of input measurement values and output measurement values, θ represents the parameter vector to be identified, Φ T (k)θ represents the estimated value of the model output;
[0022] According to the state space equation of the multifunctional inverter:
[0023]
[0024] Substitute it into the improved recursive least squares identification mathematical model, that is, let the output measurement value y(k) = u imν (k)-u pcc (k), data vector Φ T (k)=[i L (k)-i L (k-1)i L (k)], the vector to be identified θ(k) = [L / T s R];
[0025] Among them, u inv (k) represents the inverter side voltage at time k, u pcc (k) represents the grid connection point voltage at time k, L represents the filter inductance, T srepresents the sampling time interval, i L (k) represents the inductor current at time k, i L (k-1) represents the inductor current at time k-1, and R represents the filter equivalent resistance.
[0026] As a preferred solution of the multifunctional inverter control method and system based on the improved recursive least squares method described in the present invention, wherein: updating the correction gain matrix and error covariance matrix of the improved recursive least squares method, updating the estimated parameter value and updating the adaptive forgetting factor according to the residual of the identification result includes: the formula for updating the correction gain matrix of the improved recursive least squares method is expressed as:
[0027]
[0028] Among them, K(k+1) represents the correction gain matrix at time k+1, P(k) represents the error covariance matrix at time k, Φ9k+1) represents the data vector at time k+1, λ represents the forgetting factor, which is generally between 0.95 and 1.0, and Φ T (k+1) represents the transpose of the data vector at time k+1, Φ T (k+1)P(k)Φ(k+1) represents the product of the data vector at time k+1, the error covariance matrix and the data vector;
[0029] Update the error covariance matrix of the improved recursive least squares method, and the formula is expressed as:
[0030] P(k+1)=[IK(k+1)Φ T (k+1)P(k)] / λ
[0031] Where P(k+1) represents the error covariance matrix at time k+1, I represents the identity matrix, K(k+1) represents the correction gain matrix at time k+1, Φ T (k+1) represents the transpose of the data vector at time k+1, P(k) represents the error covariance matrix at time k, and λ represents the forgetting factor, which generally ranges from 0.95 to 1.0.
[0032] As a preferred solution of the multifunctional inverter control method and system based on the improved recursive least squares method described in the present invention, wherein: updating the correction gain matrix and error covariance matrix of the improved recursive least squares method, updating the estimated parameter value and updating the adaptive forgetting factor according to the residual of the identification result also includes:
[0033] The formula for updating the estimated parameter value is expressed as:
[0034]
[0035] in, represents the estimated parameter vector at time k+1, represents the estimated parameter vector at time k, K(k+1) represents the correction gain matrix at time k+1, y(k+1) represents the output measurement value at time k+1, Φ T (k+1) represents the transpose of the data vector at time k+1;
[0036] The formula for updating the forgetting factor is expressed as:
[0037]
[0038] Among them, (k) represents the residual at time k, y(k) represents the output measurement value at time k, Φ T (k) represents the transpose of the data vector at time k, Φ(k) represents the vector containing the system input and output, represents the estimated parameter vector at time k.
[0039] As a preferred solution of the multifunctional inverter control method and system based on the improved recursive least squares method described in the present invention, wherein: updating the estimated parameter value and updating the adaptive forgetting factor according to the residual of the identification result includes calculating the performance index α(k);
[0040] If α(k)>90, set λ=0.9999, if not, set λ=0.99999;
[0041] The performance index calculation formula is expressed as:
[0042]
[0043] Among them, α(k) represents the performance index at time k, M represents the number of considered time points, and r(i) T r(i) represents the square of the residual at time i, r(i) represents the residual at time i, Φ T (i)P(i)Φ(i) represents the product of the data vector Φ(i) at time i and the error covariance matrix P(i) and its transpose, Φ(i) represents the data vector at time i, P(i) represents the error covariance matrix at time i, and λ represents the forgetting factor, which is generally between 0.95 and 1.0.
[0044] As a preferred solution of the multifunctional inverter control method and system based on the improved recursive least squares method described in the present invention, wherein: substituting the estimated inductance value and the estimated parasitic resistance value at the current moment into the MPC function to obtain the one-step prediction value and the two-step prediction value of the current respectively includes obtaining the inductance parameter identification value according to the parameter identification result;
[0045] The inductance parameter identification formula is expressed as:
[0046] L(k)=T s ·θ1(k)
[0047] R(k)=θ2(k)
[0048] Among them, T s represents the sampling time interval, L(k) represents the filter inductance at time k, and θ1(k) represents the parameter vector at time k. The first term, R(k) represents the filter equivalent resistance at time k, and θ2(k) represents the parameter vector at time k The second item;
[0049] The current one-step prediction value formula is expressed as:
[0050]
[0051] Among them, i L (k+1): the inductor current at time k+1, i L (k) represents the inductor current at time k, T s represents the sampling time interval, L represents the filter inductance value, u inv (k) represents the inverter side voltage at time k, u pcc (k) represents the grid connection point voltage at time k, R represents the filter equivalent resistance; the two-step current prediction value formula is expressed as:
[0052]
[0053] Among them, i L (k+2) represents the inductor current at time k+2, i L (k+1) represents the inductor current at time k+1, T s represents the sampling time interval, L represents the filter inductance value, u inv (k+1 represents the inverter side voltage at time k+1, u pcc (k+1) represents the grid connection point voltage at time k+1, and R represents the filter equivalent resistance.
[0054] A multifunctional inverter control system based on improved recursive least squares method, wherein:
[0055] A data collection module collects operation data related to the grid-connected inverter and establishes the state space equation of the grid-connected inverter;
[0056] The state modeling module establishes an improved recursive least squares mathematical model for inductor parameter identification based on the state space equation of the grid-connected inverter;
[0057] An updating module updates the correction gain matrix and error covariance matrix of the improved recursive least squares method, updates the estimated parameter values, and updates the adaptive forgetting factor according to the residual of the identification result;
[0058] The prediction module substitutes the estimated value of inductance and parasitic resistance at the current moment into the MPC function to obtain the one-step prediction value and the two-step prediction value of the current respectively;
[0059] The optimization module obtains the optimal output voltage according to the current prediction value, and obtains the action signal of the multifunctional inverter switch tube at the next moment by using the value function optimization method;
[0060] The execution module outputs the action signal of the multifunctional inverter switch tube at the next moment after a delay of one cycle, realizes the switching action of the multifunctional inverter at the next moment, assigns the next moment to the current moment, and returns to continue to execute parameter update and control.
[0061] A computer device comprises: a memory and a processor; the memory stores a computer program, wherein the processor implements the steps of any one of the methods of the present invention when executing the computer program.
[0062] A computer-readable storage medium stores a computer program, wherein the computer program, when executed by a processor, implements the steps of any one of the methods of the present invention.
[0063] Beneficial effects of the present invention: The multifunctional inverter control method and system invention based on the improved recursive least squares method provided by the present invention is based on the finite set model predictive control. By improving the recursive least squares method, the inductance parameters and parasitic resistance values of the inductor are identified, the influence of parameter mismatch is eliminated, the control accuracy of the model predictive control is improved, and the effect of power quality management is improved. On the basis of the original recursive least squares method based on the forgetting factor, an online parameter identification method of the improved recursive least squares method is proposed, which improves the shortcomings of the traditional forgetting factor that is difficult to formulate and it is difficult to take into account both fast identification and small jitter of the identification result. BRIEF DESCRIPTION OF THE DRAWINGS
[0064] 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.
[0065] Figure 1 An overall flow chart of a multifunctional inverter control method and system based on an improved recursive least squares method provided in the first embodiment of the present invention;
[0066] Figure 2 A multifunctional inverter control method based on improved recursive least squares method provided by the first embodiment of the present invention is a multifunctional inverter system structure diagram; and system
[0067] Figure 3 A parameter estimation effect diagram of an inductor of a multifunctional inverter control method and system based on an improved recursive least squares method provided in the second embodiment of the present invention;
[0068] Figure 4 A multifunctional inverter control method based on an improved recursive least squares method and a system inductance parasitic resistance parameter estimation effect diagram provided in the second embodiment of the present invention;
[0069] Figure 5 A multifunctional inverter control method and system parameter identification algorithm based on improved recursive least squares method provided by the second embodiment of the present invention, and the grid-connected current FFT analysis results when the inductance parameters do not change;
[0070] Figure 6 A parameter identification algorithm of a multifunctional inverter control method and system based on an improved recursive least squares method provided in the second embodiment of the present invention; FFT analysis results of the grid-connected current after the inductance parameter changes for the first time;
[0071] Figure 7 A multifunctional inverter control method and system parameter identification algorithm based on improved recursive least squares method provided by the second embodiment of the present invention, and the grid-connected current FFT analysis results when the inductance parameters do not change;
[0072] Figure 8 A multifunctional inverter control method and system parameter identification algorithm based on an improved recursive least squares method are provided in the second embodiment of the present invention, and the FFT analysis result of the grid-connected current after the inductance parameter changes for the first time. DETAILED DESCRIPTION
[0073] 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.
[0074] Example 1
[0075] Reference Figure 1 , which is an embodiment of the present invention, provides a multifunctional inverter control method and system based on an improved recursive least squares method, including:
[0076] S1: Collect operating data related to the multifunctional inverter and establish the state space equation of the multifunctional inverter.
[0077] The operation data collected related to the multi-function inverter include current data, voltage data, power data, environmental data, load conditions, switching signals, frequency data, fault records and historical operation data.
[0078] Establishing the state space equation of the multifunctional inverter includes establishing a mathematical model for the multifunctional inverter and obtaining its state space equation, which is expressed as follows:
[0079]
[0080] Where L represents the filter inductance, R represents the filter equivalent resistance, i L Indicates the inductor current u i mν is the voltage on the inverter side, u pcc Indicates the grid voltage at the MFGCI access point.
[0081] By collecting detailed operating data, including current data, voltage data, power data, environmental data, load conditions, switching signals, frequency data, fault records, and historical operating data, the operating status of the inverter can be fully understood. These data help to build an accurate mathematical model that reflects the true dynamic behavior of the system. Collecting data in real time and building state-space equations can help monitor the real-time operating status of the inverter. This is essential for timely detection and response to abnormal conditions to ensure stable operation of the system. With an accurate mathematical model, the control strategy can be designed and optimized. The state-space equations provide the dynamic characteristics of the system, allowing the control strategy to be adjusted based on the real-time status of the system to achieve the best control effect.
[0082] Establishing accurate models and state-space equations can help improve system performance, including improving control accuracy, response speed, and system stability. With accurate models and state-space equations, the control system can more accurately predict and adjust the operating status of the inverter, improving control accuracy. Real-time data collection and analysis can detect and handle faults in a timely manner, improving the robustness and reliability of the system. Accurate control strategies can optimize the energy efficiency of the inverter, reduce losses, and improve overall energy efficiency. Detailed operating data and accurate models can help with fault diagnosis and preventive maintenance, reducing downtime and maintenance costs. With real-time data and dynamic models, the system can adapt to different operating conditions and environmental changes, maintaining stable and efficient operation.
[0083] S2: Based on the state space equation of the multifunctional inverter, an improved recursive least squares mathematical model for inductor parameter identification is established.
[0084] During operation, the multifunctional inverter will be affected by various factors such as load changes and grid fluctuations, which will cause changes in inductance parameters. The improved recursive least squares method can be used to identify inductance parameters in real time under dynamic conditions to ensure the accuracy of the model. Accurate inductance parameters are the basis for achieving high-performance control strategies. By identifying inductance parameters in real time, accurate model parameters can be provided for the control algorithm, thereby improving the control accuracy of the system. The inverter system may encounter various disturbances and uncertainties during operation. The improved recursive least squares method can dynamically adjust parameter estimates based on real-time data and improve the robustness of the system to disturbances and uncertainties.
[0085] S3: Update the correction gain matrix and error covariance matrix of the improved recursive least squares method, update the estimated parameter values and update the adaptive forgetting factor according to the residual of the identification result.
[0086] The improved recursive least squares method mathematical model for inductor parameter identification is established, and the improved recursive least squares method identification mathematical model formula is expressed as:
[0087]
[0088] Among them, y(k) represents the output measurement value of the model at time k, a i represents the coefficient of the autoregressive model, y(ki) represents the output measurement value at time ki, and b i represents the coefficient of the input-output model, u(ki) represents the input measurement value at time ki, ε(k) represents the measurement noise of the system at time k, Φ T (k represents the data vector consisting of input measurement values and output measurement values, θ represents the parameter vector to be identified, Φ T (k)θ represents the estimated value of the model output;
[0089] According to the state space equation of the multifunctional inverter:
[0090]
[0091] Substitute it into the improved recursive least squares identification mathematical model, that is, let the output measurement value y(k) = u imν (k)-u pcc (k), data vector Φ T (k)=[i L (k)-i L (k-1)i L (k)], the vector to be identified θ(k) = [L / T s R].
[0092] Among them, u inv(k) represents the inverter side voltage at time k, u pcc (k) represents the grid connection point voltage at time k, L represents the filter inductance, T s represents the sampling time interval, i L (k) represents the inductor current at time k, i L (k-1) represents the inductor current at time k-1, and R represents the filter equivalent resistance.
[0093] The correction gain matrix and error covariance matrix of the improved recursive least squares method are updated, the estimated parameter values are updated, and the adaptive forgetting factor is updated according to the residual of the identification result. The formula for updating the correction gain matrix of the improved recursive least squares method is expressed as follows:
[0094]
[0095] Among them, K(k+1) represents the correction gain matrix at time k+1, P(k) represents the error covariance matrix at time k, Φ(k+1) represents the data vector at time k+1, λ represents the forgetting factor, which is generally between 0.95 and 1.0, and Φ T (k+1) represents the transpose of the data vector at time k+1, Φ T (k+1)P(k)Φ(k+1) represents the product of the data vector at time k+1, the error covariance matrix and the data vector.
[0096] Update the error covariance matrix of the improved recursive least squares method, and the formula is expressed as:
[0097] P(k+1)=[IK(k+1)Φ T (k+1)P(k)] / λ
[0098] Where P(k+1) represents the error covariance matrix at time k+1, I represents the identity matrix, K(k+1) represents the correction gain matrix at time k+1, Φ T (k+1) represents the transpose of the data vector at time k+1, P(k) represents the error covariance matrix at time k, and λ represents the forgetting factor, which generally ranges from 0.95 to 1.0.
[0099] The formula for updating the estimated parameter value is expressed as:
[0100]
[0101] in, represents the estimated parameter vector at time k+1, represents the estimated parameter vector at time k, K(k+1) represents the correction gain matrix at time k+1, y(k+1) represents the output measurement value at time k+1, ΦT (k+1) represents the transpose of the data vector at time k+1;
[0102] The formula for updating the forgetting factor is expressed as:
[0103]
[0104] Among them, (k) represents the residual at time k, y(k) represents the output measurement value at time k, Φ T (k) represents the transpose of the data vector at time k, Φ(k) represents the vector containing the system input and output, represents the estimated parameter vector at time k.
[0105] Updating the estimated parameter values and updating the adaptive forgetting factor according to the residual of the identification result includes calculating the performance index α(k);
[0106] If α(k)>90, set λ=0.9999, if not, set λ=0.99999;
[0107] The performance index calculation formula is expressed as:
[0108]
[0109] Among them, α(k) represents the performance index at time k, M represents the number of considered time points, and r(i) T r(i) represents the square of the residual at time i, r(i) represents the residual at time i, Φ T (i)P(i)Φ(i) represents the product of the data vector Φ(i) at time i and the error covariance matrix P(i) and its transpose, Φ(i) represents the data vector at time i, P(i) represents the error covariance matrix at time i, and λ represents the forgetting factor, which is generally between 0.95 and 1.0.
[0110] The improved recursive least squares method can dynamically adjust the weights in the identification process according to real-time data, so that the system can quickly adapt to changes in the operating environment and maintain high-precision parameter estimation. By obtaining accurate inductance parameters through real-time identification, the control algorithm can more accurately predict and control the output of the inverter, optimize system performance, and reduce errors. The recursive least squares method is an efficient online identification algorithm that reduces the amount of calculation through recursive calculation. Compared with traditional batch processing methods, it improves calculation efficiency and is suitable for real-time applications. By real-time monitoring and identification of inductance parameters, abnormal changes can be discovered in time, fault diagnosis and predictive maintenance can be performed, system downtime can be reduced, and system reliability and stability can be improved. The dynamically adjusted parameter estimation method can enhance the stability and robustness of the system under various working conditions, ensuring that the system can operate stably even in complex environments.
[0111] S4: Substitute the estimated value of inductance and the estimated value of parasitic resistance at the current moment into the MPC function to obtain a one-step predicted value and a two-step predicted value of the current, respectively.
[0112] Obtaining an inductance parameter identification value according to the parameter identification result;
[0113] The inductance parameter identification formula is expressed as:
[0114] L(k)=T s ·θ1(k)
[0115] R(k)=θ2(k)
[0116] Among them, T s represents the sampling time interval, L(k) represents the filter inductance at time k, and θ1(k) represents the parameter vector at time k. The first term, R(k) represents the filter equivalent resistance at time k, and θ2(k) represents the parameter vector at time k The second item;
[0117] The current one-step prediction value formula is expressed as:
[0118]
[0119] Among them, i L (k+1): the inductor current at time k+1, i L (k) represents the inductor current at time k, T s represents the sampling time interval, L represents the filter inductance value, u inv (k) represents the inverter side voltage at time k, u pcc (k) represents the grid connection point voltage at time k, R represents the filter equivalent resistance; the two-step current prediction value formula is expressed as:
[0120]
[0121] Among them, i L (k+2) represents the inductor current at time k+2, i L (k+1) represents the inductor current at time k+1, T s represents the sampling time interval, L represents the filter inductance value, u inv (k+1 represents the inverter side voltage at time k+1, u pcc (k+1) represents the grid connection point voltage at time k+1, and R represents the filter equivalent resistance.
[0122] By identifying and updating the inductor and resistor parameters in real time, the actual state of the system can be accurately reflected. These updated parameters are substituted into the model predictive control (MPC) function to obtain more accurate current prediction values, thereby improving the prediction accuracy of the control system. Accurate current prediction values help optimize the control strategy so that the inverter system can achieve optimal performance under various operating conditions. Through one-step and two-step prediction values, the control input can be better adjusted and the dynamic response of the system can be optimized. Dynamically updating parameters and prediction values can enhance the robustness of the system, enabling it to better cope with load changes and external disturbances, and improve the stability and reliability of the system.
[0123] S5: Obtain the optimal output voltage according to the current prediction value, and use the value function optimization method to obtain the action signal of the multifunctional inverter switch tube at the next moment.
[0124] S6: After a delay of one cycle, the switch action signal of the multifunctional inverter at the next moment is output to realize the switch action of the multifunctional inverter at the next moment, assign the next moment to the current moment, and return to continue parameter update and control.
[0125] On the other hand, this embodiment also provides a multifunctional inverter control system based on an improved recursive least squares method, which includes:
[0126] A data collection module collects operation data related to the grid-connected inverter and establishes the state space equation of the grid-connected inverter;
[0127] The state modeling module establishes an improved recursive least squares mathematical model for inductor parameter identification based on the state space equation of the grid-connected inverter;
[0128] An updating module updates the correction gain matrix and error covariance matrix of the improved recursive least squares method, updates the estimated parameter values, and updates the adaptive forgetting factor according to the residual of the identification result;
[0129] The prediction module substitutes the estimated value of inductance and parasitic resistance at the current moment into the MPC function to obtain the one-step prediction value and the two-step prediction value of the current respectively;
[0130] The optimization module obtains the optimal output voltage according to the current prediction value, and obtains the action signal of the multifunctional inverter switch tube at the next moment by using the value function optimization method;
[0131] The execution module outputs the action signal of the multifunctional inverter switch tube at the next moment after a delay of one cycle, realizes the switching action of the multifunctional inverter at the next moment, assigns the next moment to the current moment, and returns to continue to execute parameter update and control.
[0132] If the above functions are implemented in the form of software functional units and sold or used as independent products, they 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. The computer software product is stored in a storage medium, including several instructions for enabling a computer device (which can be a personal computer, server, or 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, etc., which can store program codes.
[0133] 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.
[0134] 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.
[0135] 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.
[0136] Example 2
[0137] To verify the effectiveness of the parameter identification method proposed in this paper, we built an active power filter (multi-function inverter) model in the Matlab / Simulink environment for verification. The nominal values of the inductor and resistor were set to 8mH and 1.2Ω, and then the parameter mismatch was simulated by changing the inductor and resistor values to 6mH and 1.0Ω respectively. We used the nominal values and the results obtained by the identification algorithm proposed in this study as the parameter values of the model predictive control. Figure 3 and Figure 4 The identification results of the inductance and parasitic resistance values are shown.
[0138] The results show that the inductor parameter identification algorithm proposed in this study can quickly and accurately identify the inductance value and parasitic resistance value. Figures 5 and 6 The FFT analysis results of the grid-connected current without the identification algorithm at each state switching are shown, and the total harmonic distortion (THD) is 2.44% and 4.13% respectively. Figures 7 and 8 The FFT analysis results of the grid-connected current after adding the identification algorithm proposed in this study at each state switching are shown, and the THD values are 2.51% and 2.78% respectively. It can be observed that after the inductor parameter mismatch, the algorithm proposed in this study can effectively reduce the control error and improve the power quality management effect of the multifunctional inverter.
[0139] 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 multifunctional inverter control method and system based on improved recursive least squares method, characterized in that: include: Collecting operation data related to the multifunctional inverter and establishing the state space equation of the multifunctional inverter; According to the state space equation of the multifunctional inverter, an improved recursive least squares mathematical model for inductor parameter identification is established; Update the correction gain matrix and error covariance matrix of the improved recursive least squares method, update the estimated parameter values and update the adaptive forgetting factor according to the residual of the identification result; Substitute the estimated value of inductance and the estimated value of parasitic resistance at the current moment into the MPC function to obtain the one-step predicted value and the two-step predicted value of current respectively; The optimal output voltage is obtained according to the current prediction value, and the action signal of the multifunctional inverter switch tube at the next moment is obtained by using the value function optimization method; After a delay of one cycle, the switch action signal of the multifunctional inverter at the next moment is output to realize the switch action of the multifunctional inverter at the next moment, assign the next moment to the current moment, and return to continue parameter update and control.
2. The multifunctional inverter control method and system based on improved recursive least squares method according to claim 1, characterized in that: Collecting operation data related to the multi-function inverter, including current data, voltage data, power data, environmental data, load conditions, switching signals, frequency data, fault records and historical operation data; Establishing the state space equation of the multifunctional inverter includes establishing a mathematical model for the multifunctional inverter and obtaining its state space equation, which is expressed as follows: Where L represents the filter inductance, R represents the filter equivalent resistance, i L Indicates the inductor current u i mν is the voltage on the inverter side, u pcc Indicates the grid voltage at the MFGCI access point.
3. The multifunctional inverter control method and system based on improved recursive least squares method as claimed in claim 2, characterized in that: The improved recursive least squares method mathematical model for inductor parameter identification is established, and the improved recursive least squares method identification mathematical model formula is expressed as: Among them, y(k) represents the output measurement value of the model at time k, a i represents the coefficient of the autoregressive model, y(ki) represents the output measurement value at time ki, and b i represents the coefficient of the input-output model, u(ki) represents the input measurement value at time ki, ε(k) represents the measurement noise of the system at time k, Φ T (k) represents the data vector consisting of input measurement values and output measurement values, θ represents the parameter vector to be identified, Φ T (k)θ represents the estimated value of the model output; According to the state space equation of the multifunctional inverter: Substitute it into the improved recursive least squares identification mathematical model, that is, let the output measurement value y(k) = u imν (k)-u pcc (k), data vector ΦT(k)=[i L (k)-i L (k-1)i L (k)], the vector to be identified θ(k) = [L / T s R]; Among them, u inv (k) represents the inverter side voltage at time k, u pcc (k) represents the grid connection point voltage at time k, L represents the filter inductance, T s represents the sampling time interval, i L (k) represents the inductor current at time k, i L (k-1) represents the inductor current at time k-1, and R represents the filter equivalent resistance.
4. The multifunctional inverter control method and system based on improved recursive least squares method as claimed in claim 3, characterized in that: The correction gain matrix and error covariance matrix of the improved recursive least squares method are updated, the estimated parameter values are updated, and the adaptive forgetting factor is updated according to the residual of the identification result. The formula for updating the correction gain matrix of the improved recursive least squares method is expressed as follows: Among them, K(k+1) represents the correction gain matrix at time k+1, P(k) represents the error covariance matrix at time k, Φ(k+1) represents the data vector at time k+1, λ represents the forgetting factor, which is generally between 0.95 and 1.0, and Φ T (k+1) represents the transpose of the data vector at time k+1, Φ T (k+1)P(k)Φ(k+1) represents the product of the data vector at time k+1, the error covariance matrix and the data vector; Update the error covariance matrix of the improved recursive least squares method, and the formula is expressed as: P(k+1)=[I-K(k+1)Φ T (k+1)P)k)] / λ Where P(k+1) represents the error covariance matrix at time k+1, I represents the identity matrix, K(k+1) represents the correction gain matrix at time k+1, Φ T )k+1) represents the transpose of the data vector at time k+1, P(k) represents the error covariance matrix at time k, and λ represents the forgetting factor, which generally ranges from 0.95 to 1.
0.
5. The multifunctional inverter control method and system based on improved recursive least squares method as claimed in claim 4, characterized in that: Updating the correction gain matrix and error covariance matrix of the improved recursive least squares method, updating the estimated parameter values and updating the adaptive forgetting factor according to the residual of the identification result also includes, The formula for updating the estimated parameter value is expressed as: in, represents the estimated parameter vector at time k+1, represents the estimated parameter vector at time k, K(k+1) represents the correction gain matrix at time k+1, y(k+1) represents the output measurement value at time k+1, Φ T (k+1) represents the transpose of the data vector at time k+1; The formula for updating the forgetting factor is expressed as: Among them, (k) represents the residual at time k, y(k) represents the output measurement value at time k, Φ T (k) represents the transpose of the data vector at time k, Φ(k) represents the vector containing the system input and output, represents the estimated parameter vector at time k.
6. The multifunctional inverter control method and system based on improved recursive least squares method as claimed in claim 5, characterized in that: Updating the estimated parameter values and updating the adaptive forgetting factor according to the residual of the identification result includes calculating the performance index α(k); If α(k)>90, set λ=0.9999, if not, set λ=0.99999; The performance index calculation formula is expressed as: Among them, α(k) represents the performance index at time k, M represents the number of considered time points, and r(i) T r(i) represents the square of the residual at time i, r(i) represents the residual at time i, Φ T (i)P(i)Φ(i) represents the product of the data vector Φ(i) at time i and the error covariance matrix P(i) and its transpose, Φ(i) represents the data vector at time i, P(i) represents the error covariance matrix at time i, and λ represents the forgetting factor, which is generally between 0.95 and 1.
0.
7. The multifunctional inverter control method and system based on improved recursive least squares method according to claim 6, characterized in that: Substitute the estimated value of inductance and the estimated value of parasitic resistance at the current moment into the MPC function to obtain the one-step predicted value and the two-step predicted value of current respectively, and obtain the inductance parameter identification value according to the parameter identification result; The inductance parameter identification formula is expressed as: L(k)=T s ·θ1(k) R(k)=θ2(k) Among them, T s represents the sampling time interval, L(k) represents the filter inductance at time k, and θ1(k) represents the parameter vector at time k The first term, R(k) represents the filter equivalent resistance at time k, and θ2(k) represents the parameter vector at time k The second item; The current one-step prediction value formula is expressed as: Among them, i L (k+1): the inductor current at time k+1, i L (k) represents the inductor current at time k, T s represents the sampling time interval, L represents the filter inductance value, u inv (k) represents the inverter side voltage at time k, u pcc (k) represents the grid connection point voltage at time k, R represents the filter equivalent resistance; the two-step current prediction value formula is expressed as: Among them, i L (k+2) represents the inductor current at time k+2, i L (k+1) represents the inductor current at time k+1, T s represents the sampling time interval, L represents the filter inductance value, u inv (k+1) represents the inverter side voltage at time k+1, u pcc (k+1) represents the grid connection point voltage at time k+1, and R represents the filter equivalent resistance.
8. A multifunctional inverter control system based on improved recursive least squares method using the method according to any one of claims 1 to 7, characterized in that: A data collection module collects operation data related to the multifunctional inverter and establishes a state space equation of the multifunctional inverter; The state modeling module establishes an improved recursive least squares mathematical model for inductor parameter identification based on the state space equation of the grid-connected inverter; An updating module updates the correction gain matrix and error covariance matrix of the improved recursive least squares method, updates the estimated parameter values, and updates the adaptive forgetting factor according to the residual of the identification result; The prediction module substitutes the estimated value of inductance and parasitic resistance at the current moment into the MPC function to obtain the one-step prediction value and the two-step prediction value of the current respectively; The optimization module obtains the optimal output voltage according to the current prediction value, and obtains the action signal of the multifunctional inverter switch tube at the next moment by using the value function optimization method; The execution module outputs the action signal of the multifunctional inverter switch tube at the next moment after a delay of one cycle, realizes the switching action of the multifunctional inverter at the next moment, assigns the next moment to the current moment, and returns to continue to execute parameter update and control.
9. A computer device comprising: Memory and processor; The memory stores a computer program, characterized in that: when the processor executes the computer program, the steps of the multifunctional inverter control method based on the improved recursive least squares method 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 a multifunctional inverter control method and system based on an improved recursive least squares method are implemented.
Citation Information
Cited By
Multi-factor coupling energy storage power station real-time power compensation method and device
CN120955697A
High-performance file hash calculation method based on double-engine self-adaptive switching
CN121009058A
Distribution transformer no-load and load loss on-line measurement method and system
CN122063369A
A method and system for on-line measurement of no-load and load losses of distribution transformers
CN122063369B
Inverter control method
CN122824010A
