Photovoltaic inverter test system and method thereof

Abstract

The invention discloses a photovoltaic inverter test system and method, and the system comprises a photovoltaic simulator module which is used for simulating the output characteristics of a photovoltaic array; the adjustable load module is used for adjusting different types of loads; the data acquisition and monitoring module is used for monitoring and recording operation parameters of the photovoltaic inverter in real time; the control and protection module is used for controlling the test process and providing protection; the environment simulation module is used for simulating working states under different environment conditions; and the test report generation module is used for generating a test report and performing data analysis. Compared with the prior art, the photovoltaic inverter test system has the advantages that remarkable technical breakthroughs are realized in the aspects of improving test precision, enhancing load regulation capability, expanding environment simulation conditions and improving data acquisition and monitoring functions, various defects in the prior art are overcome, the quality and efficiency of photovoltaic inverter test are greatly improved, and the photovoltaic inverter test system has important application value and popularization prospect.

Description

Technical Field

[0001] The present invention relates to the field of photovoltaic technology, and in particular to a photovoltaic inverter testing system and method thereof. Background Art

[0002] As the core component of the photovoltaic power generation system, the photovoltaic inverter is responsible for converting the direct current generated by photovoltaic cells into alternating current. The performance of the photovoltaic inverter directly affects the efficiency and stability of the entire photovoltaic power generation system.

[0003] The existing photovoltaic inverter test system has some problems in practical application, which are mainly manifested in the following aspects:

[0004] Low test accuracy: Traditional photovoltaic inverter test systems are difficult to meet high-standard test requirements in terms of accuracy. This is mainly because the hardware and software configurations of the test system cannot effectively reduce errors, especially in high-frequency and high-voltage test scenarios, which are prone to data deviations.

[0005] Inflexible load regulation: Many existing test systems have limited load regulation capabilities and are unable to simulate the various load changes in the actual operation of the photovoltaic system. This results in the test results not being able to fully reflect the performance of the photovoltaic inverter under different load conditions, affecting the reliability and effectiveness of the test results.

[0006] Limited environmental simulation conditions: The performance of photovoltaic systems is greatly affected by environmental conditions (such as temperature and humidity), but the environmental simulation functions of existing test systems are usually simple and cannot accurately control and simulate various extreme environmental conditions. This limits the comprehensive evaluation of the adaptability of photovoltaic inverters during the research and development process.

[0007] Data collection and monitoring limitations: Traditional test systems have certain limitations in data collection and monitoring, making it difficult to achieve comprehensive, real-time monitoring and analysis of the operating status of photovoltaic inverters. The data collection rate and processing capacity of existing systems are insufficient, resulting in data lag and loss, affecting the accuracy and availability of test results. Summary of the invention

[0008] Aiming at the problems of low test accuracy, inflexible load regulation, limited environmental simulation conditions, and data acquisition and monitoring limitations, the present invention proposes a photovoltaic inverter test system and method to improve test accuracy, enhance load regulation capability, expand environmental simulation conditions, and improve data acquisition and monitoring functions. By adopting advanced pulse width modulation (PWM) technology, digital signal processing (DSP) controller, high-precision analog-to-digital / digital-to-analog converter (ADC / DAC), artificial intelligence-based adaptive control algorithm and other technical means, the defects and deficiencies in the prior art are effectively solved, providing a more efficient and reliable test solution.

[0009] In order to achieve the above object, the present invention is implemented by the following technical solutions:

[0010] A photovoltaic inverter testing system and method thereof, the system comprising:

[0011] Photovoltaic simulator module, used to simulate the output characteristics of the photovoltaic array, using pulse width modulation (PWM) technology and digital signal processing (DSP) controller, with high-precision analog-to-digital / digital-to-analog converter (ADC / DAC) and self-calibration function;

[0012] Adjustable load modules are used to adjust different types of loads. Based on silicon carbide (SiC) and gallium nitride (GaN) power electronic devices, high switching frequency and low loss load regulation circuits are designed;

[0013] Data acquisition and monitoring module, used to monitor and record the operating parameters of the photovoltaic inverter in real time, integrates a high sampling rate analog-to-digital converter (ADC) and a field programmable gate array (FPGA), and uses an improved fusion algorithm for data preprocessing and anomaly detection;

[0014] The control and protection module is used to control the test process and provide protection. It adopts an adaptive control algorithm based on artificial intelligence and has overvoltage protection, overcurrent protection and overtemperature protection functions.

[0015] Environmental simulation module, used to simulate the working state under different environmental conditions, using high-precision temperature and humidity control system and adopting accurate temperature and humidity sensors;

[0016] The test report generation module is used to generate test reports and perform data analysis.

[0017] Optionally, the photovoltaic simulator module comprises the following steps:

[0018] Configure the operating parameters of the DSP controller, including clock frequency and interrupt settings;

[0019] Initialize the ADC / DAC module and set the sampling rate and accuracy;

[0020] Use the ADC module to read the analog input signal and convert the environmental parameters of the photovoltaic array into digital signals;

[0021] The DSP controller processes the collected digital signals and calculates the output characteristic curve of the photovoltaic array;

[0022] Adjust PWM output parameters according to preset photovoltaic model and real-time input data;

[0023] The DSP controller generates corresponding PWM signals based on the calculation results to control the power electronic devices;

[0024] By adjusting the duty cycle and frequency of the PWM signal, the output characteristics of the photovoltaic array can be accurately simulated;

[0025] Monitor the output voltage and current in real time, and convert the digital signal into analog signal through the DAC module and feed it back to the controller;

[0026] The DSP controller performs closed-loop regulation based on the feedback signal;

[0027] Periodically trigger the self-calibration procedure to adjust the accuracy of the ADC / DAC through internal reference signals and external calibration equipment;

[0028] Record operation data to storage devices for subsequent analysis and optimization;

[0029] Data exchange and remote monitoring are carried out with the host computer through the communication interface.

[0030] Optionally, the adjustable load module has the following specific steps:

[0031] Select silicon carbide (SiC) and gallium nitride (GaN) power electronics;

[0032] Design a high-frequency switching circuit to ensure that the switching frequency is above 1MHz;

[0033] Develop and solidify the control algorithm to ensure that the control algorithm can adjust the load in real time;

[0034] Design a control unit including a central processing unit (CPU), memory, and input and output interfaces;

[0035] Integrate SiC and GaN power electronic devices, high-frequency switching circuits, control algorithm modules and control units, and perform debugging and testing;

[0036] Design cooling system.

[0037] Optionally, the control algorithm includes:

[0038] Design PID controller;

[0039] Introducing an adaptive adjustment mechanism to adjust PID parameters according to the real-time performance of the system, including incremental adjustment and model reference adaptive control;

[0040] In the main loop of the control system, the system status is acquired in real time, the error is calculated, and the PID parameters are adjusted according to the adaptive adjustment algorithm to calculate the PID controller output to control the system load.

[0041] Optionally, the data acquisition and monitoring module includes the following steps:

[0042] Initialize the system, configure the ADC sampling rate and FPGA working parameters;

[0043] Use the ADC module to collect the operating parameters of the photovoltaic inverter at a high sampling rate and convert the collected analog signals into digital signals;

[0044] Transfer the digital signal to FPGA for preliminary processing, including filtering and data format conversion;

[0045] Monitor system status in real time, collect new data and feed it back to ANFIS and DRL modules to continuously optimize control strategies.

[0046] Optionally, the data acquisition and monitoring module uses an improved fusion algorithm to perform data preprocessing and anomaly detection, including the following steps:

[0047] 1) Collect multi-source data of photovoltaic inverters, including image data, time series data and sensor data, and clean and normalize the data;

[0048] 2) Use convolutional neural network (CNN) to extract features from the preprocessed image data to obtain the image feature set F cnn ;

[0049] Use the long short-term memory network (LSTM) to extract features from the preprocessed time series data to obtain the time series feature set F lstm ;

[0050] Use Random Forest to extract features from the preprocessed sensor data and obtain the sensor feature set F rf ;

[0051] 3) The image feature set F cnn , time series feature set F lstm and sensor feature set F rf Merge to get the comprehensive feature set F fusion ;

[0052] 4) Use the weighted average method to fuse the comprehensive feature set to obtain the fusion model output result s fusions , the formula is as follows:

[0053] s fusion =α·s cnn +β·s lstm +γ·s rf

[0054] Among them, s cnn 、s lstm 、s rfare the output results of CNN, LSTM and random forest models respectively, α, β and γ are weight coefficients, satisfying α+β+γ=1;

[0055] 5) Clean, normalize and extract features of the photovoltaic inverter data collected in real time, use the fusion model to predict the comprehensive feature set, and obtain real-time anomaly detection results fusion , and judge whether the data is abnormal according to the preset threshold θ.

[0056] Optionally, the control and protection module includes the following steps:

[0057] Real-time collection of voltage, current and temperature data of photovoltaic inverters;

[0058] Normalize and filter the collected data:

[0059] The normalization formula is:

[0060]

[0061] Where: x is the original data, min(x) is the minimum value of the data, max(x) is the maximum value of the data, and x′ is the normalized data;

[0062] The filtering formula is:

[0063]

[0064] Where: y[n] is the filtered data, x[nk] is the nkth data sample, and N is the size of the filter window;

[0065] Generate initial control instructions using the trained ANFIS model:

[0066] The ANFIS formula is:

[0067]

[0068] Where: y is the output of the ANFIS model, w i is the weight of the i-th membership function, f i (x) is the output function of the ith rule, and N is the number of membership functions;

[0069] The ANFIS model provides preliminary control strategies to quickly respond to system changes;

[0070] The DRL module continuously learns and optimizes the control strategy in the background;

[0071] By interacting with the actual system, the parameters of the policy network are continuously adjusted:

[0072] The policy network parameter update formula is:

[0073]

[0074] Where: θ new is the updated parameter, θ old is the parameter before updating, α is the learning rate, is the gradient of the loss function with respect to the parameter θ;

[0075] The output of the DRL module serves as a supplement and correction to the ANFIS control strategy;

[0076] Use the ANFIS model to predict the future behavior of the system:

[0077] The prediction formula is:

[0078]

[0079] In the formula, is the prediction for the future time t+1, x t is the input data at the current moment;

[0080] The MPC algorithm is used to perform rolling optimization in each control cycle to generate the optimal control sequence:

[0081] MPC optimization objective function:

[0082] min(t*Qt+uf*Ru)

[0083] Where: t is a variable, Qt is a state weight matrix, uf is the amount to be optimized, and Ru is a control input weight matrix; the state update formula is:

[0084] x t+1 =Ax t +Bu t

[0085] Where: x t+1 is the state at the next moment, x t is the state at the current moment, A is the state transfer matrix, and B is the control matrix;

[0086] According to the MPC optimization results, execute the optimal control instructions:

[0087] The control instruction calculation formula is:

[0088]

[0089] Where: u t is the optimal control input at time t, Q is the state weight matrix, R is the control input weight matrix, is the quadratic form of the state vector weighted by the weight matrix Q, which represents the weighted sum of the state error. is the quadratic form of the control input vector weighted by the weight matrix R, which represents the weighted sum of the control inputs. u To solve the control input u that minimizes the objective function;

[0090] Monitor system status in real time, collect new data and feed it back to ANFIS and DRL modules to continuously optimize control strategies.

[0091] Optionally, the environment simulation module includes:

[0092] Temperature control unit, using high-precision temperature sensors and PID control algorithms to adjust and maintain the temperature of the test environment;

[0093] The humidity control unit uses a high-precision humidity sensor and a humidity adjustment device to simulate environmental conditions under different humidity conditions.

[0094] Optionally, the test report generating module includes:

[0095] Data analysis unit, which performs trend analysis and fault prediction based on historical test data and machine learning algorithms;

[0096] The report generation unit can automatically generate detailed reports containing test results, data analysis and optimization suggestions, and supports export in multiple formats.

[0097] A photovoltaic inverter testing method, the method comprising:

[0098] Step S1: The photovoltaic simulator module includes setting the working parameters of the DSP controller, initializing the ADC / DAC module, reading and processing the analog input signal, adjusting the PWM output parameters, generating a PWM signal, monitoring and feeding back the output voltage and current, performing closed-loop regulation and self-calibration, recording the operating data and exchanging data with the host computer;

[0099] Step S2: The adjustable load module includes selecting SiC and GaN power electronic devices, designing high-frequency switching circuits, developing control algorithms, designing control units, integrating various components and performing debugging and testing;

[0100] Step S3: the data acquisition and monitoring module includes configuring the working parameters of the ADC and FPGA, acquiring the operating parameters at a high sampling rate, performing preliminary processing and real-time monitoring, and using a fusion algorithm for data preprocessing and anomaly detection;

[0101] Step S4: The control and protection module includes real-time acquisition of voltage, current, and temperature data, data normalization and filtering, generating initial control instructions, optimizing control strategies using the DRL module, predicting system behavior using the ANFIS model, performing rolling optimization through the MPC algorithm, and executing optimal control instructions;

[0102] Step S5: The environmental simulation module includes adjusting the test environment using high-precision temperature and humidity sensors and a control system;

[0103] Step S6: The test report generation module includes performing data analysis and fault prediction based on historical data and machine learning algorithms to generate a report containing test results, data analysis and optimization suggestions.

[0104] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention significantly improves the accuracy of test data by adopting high-precision analog-to-digital / digital-to-analog converters and self-calibration functions, as well as improved fusion algorithms and high-sampling rate data acquisition devices. Silicon carbide and gallium nitride power electronic devices are introduced, and a load regulation circuit with high switching frequency and low loss is designed, so that the system can flexibly simulate various load conditions. By comprehensively evaluating the performance of photovoltaic inverters under different loads, its stability and reliability in practical applications are ensured. A high-precision temperature and humidity control system is adopted, combined with precise sensors, which can simulate extreme environmental conditions (such as high temperature, high humidity, low temperature, etc.). The reliability and stability of photovoltaic inverters in various environments are ensured, which helps to improve their adaptability in practical applications. A high-sampling rate analog-to-digital converter and a field programmable gate array are integrated, and advanced data preprocessing and anomaly detection algorithms are adopted to realize real-time monitoring and comprehensive analysis of the operating status of photovoltaic inverters. This improves the accuracy of data acquisition and the real-time nature of monitoring, and ensures that the test system can detect and handle abnormal situations in a timely manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0105] 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 labor.

[0106] in:

[0107] Figure 1 is an overall structural block diagram of a system according to an embodiment of the present invention;

[0108] Figure 2 is a detailed implementation diagram of a photovoltaic simulator module in an embodiment of the present invention;

[0109] Figure 3 The figure is a flow chart of a method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0110] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention.

[0111] Example 1

[0112] like Figure 1 As shown, it is an embodiment of the present invention, which provides a photovoltaic inverter testing system, including:

[0113] Photovoltaic simulator module, used to simulate the output characteristics of the photovoltaic array, using pulse width modulation (PWM) technology and digital signal processing (DSP) controller, with high-precision analog-to-digital / digital-to-analog converter (ADC / DAC) and self-calibration function;

[0114] Among them, Figure 2 As shown, the photovoltaic simulator module includes the following steps:

[0115] Configure the operating parameters of the DSP controller, including clock frequency and interrupt settings;

[0116] Initialize the ADC / DAC module and set the sampling rate and accuracy;

[0117] Use the ADC module to read the analog input signal and convert the environmental parameters of the photovoltaic array into digital signals;

[0118] The DSP controller processes the collected digital signals and calculates the output characteristic curve of the photovoltaic array;

[0119] Adjust PWM output parameters according to preset photovoltaic model and real-time input data;

[0120] The DSP controller generates corresponding PWM signals based on the calculation results to control the power electronic devices;

[0121] By adjusting the duty cycle and frequency of the PWM signal, the output characteristics of the photovoltaic array can be accurately simulated;

[0122] Monitor the output voltage and current in real time, and convert the digital signal into analog signal through the DAC module and feed it back to the controller;

[0123] The DSP controller performs closed-loop regulation based on the feedback signal;

[0124] Periodically trigger the self-calibration procedure to adjust the accuracy of the ADC / DAC through internal reference signals and external calibration equipment;

[0125] Record operation data to storage devices for subsequent analysis and optimization;

[0126] Data exchange and remote monitoring are carried out with the host computer through the communication interface.

[0127] Using PWM technology and DSP controller, the output characteristics of photovoltaic arrays can be accurately simulated, ensuring the simulation accuracy of the test system under various lighting conditions. High-precision ADC / DAC and self-calibration functions further improve the accuracy and stability of the simulation.

[0128] Furthermore, an adjustable load module is used to adjust different types of loads, and a high switching frequency and low loss load regulation circuit is designed based on silicon carbide (SiC) and gallium nitride (GaN) power electronic devices;

[0129] The specific steps of the adjustable load module are as follows:

[0130] Select silicon carbide (SiC) and gallium nitride (GaN) power electronics;

[0131] Design a high-frequency switching circuit to ensure that the switching frequency is above 1MHz;

[0132] Develop and solidify the control algorithm to ensure that the control algorithm can adjust the load in real time;

[0133] Design a control unit including a central processing unit (CPU), memory, and input and output interfaces;

[0134] Integrate SiC and GaN power electronic devices, high-frequency switching circuits, control algorithm modules and control units, and perform debugging and testing;

[0135] Design cooling system.

[0136] Wherein, the control algorithm includes:

[0137] Design PID controller;

[0138] Introducing an adaptive adjustment mechanism to adjust PID parameters according to the real-time performance of the system, including incremental adjustment and model reference adaptive control;

[0139] In the main loop of the control system, the system status is acquired in real time, the error is calculated, and the PID parameters are adjusted according to the adaptive adjustment algorithm to calculate the PID controller output to control the system load.

[0140] SiC and GaN power electronic devices have the characteristics of high switching frequency and low loss, which enables the load regulation module to respond quickly to load changes, simulate the performance of photovoltaic inverters under different load conditions, and improve the reliability and effectiveness of test results.

[0141] Furthermore, the data acquisition and monitoring module is used to monitor and record the operating parameters of the PV inverter in real time, integrates a high sampling rate analog-to-digital converter (ADC) and a field programmable gate array (FPGA), and uses an improved fusion algorithm for data preprocessing and anomaly detection;

[0142] Wherein, the data acquisition and monitoring module includes the following steps:

[0143] Initialize the system, configure the ADC sampling rate and FPGA working parameters;

[0144] Use the ADC module to collect the operating parameters of the photovoltaic inverter at a high sampling rate and convert the collected analog signals into digital signals;

[0145] Transfer the digital signal to FPGA for preliminary processing, including filtering and data format conversion;

[0146] Monitor system status in real time, collect new data and feed it back to ANFIS and DRL modules to continuously optimize control strategies.

[0147] Specifically, the data acquisition and monitoring module uses the improved fusion algorithm to perform data preprocessing and anomaly detection, including the following steps:

[0148] 1) Collect multi-source data of photovoltaic inverters, including image data, time series data and sensor data, and clean and normalize the data;

[0149] 2) Use convolutional neural network (CNN) to extract features from the preprocessed image data to obtain the image feature set F cnn ;

[0150] Use the long short-term memory network (LSTM) to extract features from the preprocessed time series data to obtain the time series feature set F lstm ;

[0151] Use Random Forest to extract features from the preprocessed sensor data and obtain the sensor feature set F rf ;

[0152] 3) The image feature set F cnn , time series feature set F lstm and sensor feature set F rf Merge to get the comprehensive feature set Ffusion ;

[0153] 4) Use the weighted average method to fuse the comprehensive feature set to obtain the fusion model output result s fusions , the formula is as follows:

[0154] s fusion =α·s cnn +β·s lstm +γ·s rf

[0155] Among them, s cnn 、s lstm 、s rf are the output results of CNN, LSTM and random forest models respectively, α, β and γ are weight coefficients, satisfying α+β+γ=1;

[0156] 5) Clean, normalize and extract features of the photovoltaic inverter data collected in real time, use the fusion model to predict the comprehensive feature set, and obtain real-time anomaly detection results fusion , and judge whether the data is abnormal according to the preset threshold θ.

[0157] The high sampling rate ADC and FPGA enable the data acquisition and monitoring module to achieve real-time monitoring and comprehensive analysis of the operating status of the photovoltaic inverter. The improved fusion algorithm improves the accuracy and reliability of the data, ensuring that the test system can detect and handle abnormal situations in a timely manner.

[0158] Furthermore, the control and protection module is used to control and provide protection for the test process, using an adaptive control algorithm based on artificial intelligence, and has overvoltage protection, overcurrent protection, and overtemperature protection functions;

[0159] Wherein, the control and protection module includes the following steps:

[0160] Real-time collection of voltage, current and temperature data of photovoltaic inverters;

[0161] Normalize and filter the collected data:

[0162] The normalization formula is:

[0163]

[0164] Where: x is the original data, min(x) is the minimum value of the data, max(x) is the maximum value of the data, and x′ is the normalized data;

[0165] The filtering formula is:

[0166]

[0167] Where: y[n] is the filtered data, x[nk] is the nkth data sample, and N is the size of the filter window;

[0168] Generate initial control instructions using the trained ANFIS model:

[0169] The ANFIS formula is:

[0170]

[0171] Where: y is the output of the ANFIS model, w i is the weight of the i-th membership function, f i (x) is the output function of the ith rule, and N is the number of membership functions;

[0172] The ANFIS model provides preliminary control strategies to quickly respond to system changes;

[0173] The DRL module continuously learns and optimizes the control strategy in the background;

[0174] By interacting with the actual system, the parameters of the policy network are continuously adjusted:

[0175] The policy network parameter update formula is:

[0176]

[0177] Where: θ new is the updated parameter, θ old is the parameter before updating, α is the learning rate, is the gradient of the loss function with respect to the parameter θ;

[0178] The output of the DRL module serves as a supplement and correction to the ANFIS control strategy;

[0179] Use the ANFIS model to predict the future behavior of the system:

[0180] The prediction formula is:

[0181]

[0182] In the formula, is the prediction for the future time t+1, x t is the input data at the current moment;

[0183] The MPC algorithm is used to perform rolling optimization in each control cycle to generate the optimal control sequence:

[0184] MPC optimization objective function:

[0185] min(t*Qt+uf*Ru)

[0186] Where: t is a variable, Qt is a state weight matrix, uf is the amount to be optimized, and Ru is a control input weight matrix; the state update formula is:

[0187] x t+1 =Ax t +Bu t

[0188] Where: x t+1 is the state at the next moment, x t is the state at the current moment, A is the state transfer matrix, and B is the control matrix. According to the MPC optimization results, execute the optimal control instructions:

[0189] The control instruction calculation formula is:

[0190]

[0191] Where: u t is the optimal control input at time t, Q is the state weight matrix, R is the control input weight matrix, is the quadratic form of the state vector weighted by the weight matrix Q, which represents the weighted sum of the state error. is the quadratic form of the control input vector weighted by the weight matrix R, which represents the weighted sum of the control inputs. u To solve the control input u that minimizes the objective function;

[0192] Monitor system status in real time, collect new data and feed it back to ANFIS and DRL modules to continuously optimize control strategies.

[0193] The adaptive control algorithm dynamically adjusts the parameters of the test system according to real-time data, improving the intelligence level and response speed of the test. At the same time, overvoltage, overcurrent and overtemperature protection functions ensure the safe operation of the test system under various conditions.

[0194] Furthermore, the environmental simulation module is used to simulate the working state under different environmental conditions, using a high-precision temperature and humidity control system and adopting accurate temperature and humidity sensors;

[0195] Wherein, the environmental simulation module includes:

[0196] Temperature control unit, using high-precision temperature sensors and PID control algorithms to adjust and maintain the temperature of the test environment;

[0197] The humidity control unit uses a high-precision humidity sensor and a humidity adjustment device to simulate environmental conditions under different humidity conditions.

[0198] High-precision temperature and humidity control systems combined with precise sensors can simulate various extreme environmental conditions, ensure the reliability and stability of photovoltaic inverters in different environments, and comprehensively evaluate their performance.

[0199] Furthermore, a test report generation module is used to generate a test report and perform data analysis.

[0200] Wherein, the test report generation module includes:

[0201] Data analysis unit, which performs trend analysis and fault prediction based on historical test data and machine learning algorithms;

[0202] The report generation unit can automatically generate detailed reports containing test results, data analysis and optimization suggestions, and supports export in multiple formats.

[0203] Automatic generation of test reports and data analysis functions improves the efficiency and accuracy of the test system, reduces human intervention, and ensures the objectivity and reliability of test results.

[0204] Example 2

[0205] like Figure 3 As shown in FIG. 1 , another embodiment of the present invention is provided. The example provides a photovoltaic inverter testing method, the method comprising:

[0206] Step S1: The photovoltaic simulator module includes setting the working parameters of the DSP controller, initializing the ADC / DAC module, reading and processing the analog input signal, adjusting the PWM output parameters, generating a PWM signal, monitoring and feeding back the output voltage and current, performing closed-loop regulation and self-calibration, recording the operating data and exchanging data with the host computer;

[0207] Step S2: The adjustable load module includes selecting SiC and GaN power electronic devices, designing high-frequency switching circuits, developing control algorithms, designing control units, integrating various components and performing debugging and testing;

[0208] Step S3: the data acquisition and monitoring module includes configuring the working parameters of the ADC and FPGA, acquiring the operating parameters at a high sampling rate, performing preliminary processing and real-time monitoring, and using a fusion algorithm for data preprocessing and anomaly detection;

[0209] Step S4: The control and protection module includes real-time acquisition of voltage, current, and temperature data, data normalization and filtering, generating initial control instructions, optimizing control strategies using the DRL module, predicting system behavior using the ANFIS model, performing rolling optimization through the MPC algorithm, and executing optimal control instructions;

[0210] Step S5: The environmental simulation module includes adjusting the test environment using high-precision temperature and humidity sensors and a control system;

[0211] Step S6: The test report generation module includes performing data analysis and fault prediction based on historical data and machine learning algorithms to generate a report containing test results, data analysis and optimization suggestions.

[0212] Example 3

[0213] Another embodiment of the present invention verifies the performance of the photovoltaic inverter under different input voltages and environmental conditions, evaluates its output voltage, current, efficiency and environmental adaptability, and ensures that the system operates stably and efficiently under various conditions.

[0214] Experimental preparation:

[0215] Use a photovoltaic simulator to generate voltages and currents for a simulated photovoltaic array.

[0216] Initialize the data acquisition module (ADC, FPGA) and the control and protection module (ANFIS, DRL model).

[0217] Set the temperature and humidity range of the environmental simulation module

[0218] System initialization:

[0219] Configure the operating parameters of the PV simulator, ADC, and FPGA to ensure the normal operation of each module of the system.

[0220] Start the control module and load the control algorithm.

[0221] Data collection and processing:

[0222] Collect the operating parameters of the photovoltaic inverter under different input voltages, including input voltage, current and output voltage and current.

[0223] The data is transferred to FPGA for preliminary processing, and a fusion algorithm is used for data preprocessing and anomaly detection.

[0224] Control and protection:

[0225] Collect operating parameters in real time, generate initial control instructions, and continuously optimize control strategies.

[0226] The MPC algorithm is used to optimize each control cycle and generate the optimal control sequence.

[0227] Environmental simulation:

[0228] Adjust the temperature and humidity of the experimental environment to maintain a stable experimental environment.

[0229] Record the system's operating data under different temperature and humidity conditions.

[0230] Testing and Recording:

[0231] Record the operating data of the PV inverter under different conditions and generate a test report.

[0232] The experimental data is as follows:

[0233]

[0234] Experimental data analysis

[0235] Input and output voltage: When the input voltage is in the range of 220V to 260V, the output voltage changes accordingly, and the output voltage remains in the high-efficiency range, indicating that the control algorithm is effective.

[0236] Output current and efficiency: The output current increases with the input voltage, and the system always maintains high efficiency, up to 96.5%.

[0237] Temperature and humidity environment: The system can run stably under different temperature and humidity conditions, and the environmental simulation module provides precise environmental control.

[0238] Abnormality detection: In high temperature and high humidity environments, the system detects abnormalities and ensures system safety by optimizing control strategies.

[0239] This experiment verified the high efficiency and stability of the photovoltaic inverter under different input voltages and environmental conditions, proving the beneficial effects of the system. The system can operate stably under various conditions and has significant adaptability and reliability.

[0240] In summary, the present invention significantly improves the accuracy and reliability of photovoltaic inverter testing by adopting high-precision analog-to-digital / digital-to-analog converters (ADC / DAC) and self-calibration functions, combined with advanced fusion algorithms and high-sampling rate data acquisition devices. Silicon carbide (SiC) and gallium nitride (GaN) power electronic devices are introduced, and a high switching frequency and low-loss load regulation circuit is designed, so that the system can flexibly simulate various load conditions and comprehensively evaluate the performance of photovoltaic inverters under different loads.

[0241] The high-precision temperature and humidity control system, combined with precise sensors, can simulate extreme environmental conditions and ensure the reliability and stability of photovoltaic inverters in various environments. The high-sampling rate analog-to-digital converter (ADC) and field programmable gate array (FPGA) are integrated, and advanced data preprocessing and anomaly detection algorithms are used to achieve real-time monitoring and comprehensive analysis of the operating status of the photovoltaic inverter, improving the accuracy of data acquisition and the real-time nature of monitoring. The adaptive control algorithm based on artificial intelligence can dynamically adjust the parameters of the test system according to real-time data, improving the intelligence level and response speed of the test. The present invention has achieved significant technological breakthroughs in improving test accuracy, enhancing load regulation capabilities, expanding environmental simulation conditions, and improving data acquisition and monitoring functions, and provides an efficient, reliable and comprehensive photovoltaic inverter test system and method. It not only solves many deficiencies in the prior art, but also greatly improves the quality and efficiency of photovoltaic inverter testing, and has important application value and promotion prospects.

[0242] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification and the features of different embodiments or examples, unless they are contradictory.

[0243] Any process or method description in the flow chart or otherwise described herein can be understood to represent a module, fragment or portion of a code including one or more executable instructions for implementing the steps of a specific logical function or process. And the scope of the preferred embodiment of the present application includes other implementations, in which the functions may not be performed in the order shown or discussed, including in a substantially simultaneous manner or in a reverse order according to the functions involved.

[0244] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A photovoltaic inverter test system, characterized in that: The system comprises: Photovoltaic simulator module, used to simulate the output characteristics of photovoltaic arrays, adopts pulse width modulation PWM technology and digital signal processing DSP controller, and has high-precision analog-to-digital / digital-to-analog converter ADC / DAC and self-calibration function; Adjustable load module, used to adjust different types of loads, based on silicon carbide SiC and gallium nitride GaN power electronic devices, to design high switching frequency and low loss load regulation circuit; The data acquisition and monitoring module is used to monitor and record the operating parameters of the photovoltaic inverter in real time. It integrates a high-sampling-rate analog-to-digital converter ADC and a field-programmable gate array FPGA, and uses an improved fusion algorithm for data preprocessing and anomaly detection. The control and protection module is used to control the test process and provide protection. It uses an artificial intelligence-based adaptive control algorithm and has overvoltage protection, overcurrent protection, and overtemperature protection functions. Environmental simulation module, used to simulate the working state under different environmental conditions, using high-precision temperature and humidity control system and adopting accurate temperature and humidity sensors; The test report generation module is used to generate test reports and perform data analysis.

2. The photovoltaic inverter test system according to claim 1, characterized in that: The photovoltaic simulator module comprises the following steps: Configure the operating parameters of the DSP controller, including clock frequency and interrupt settings; Initialize the ADC / DAC module and set the sampling rate and accuracy; Use the ADC module to read the analog input signal and convert the environmental parameters of the photovoltaic array into digital signals; The DSP controller processes the collected digital signals and calculates the output characteristic curve of the photovoltaic array; Adjust PWM output parameters according to preset photovoltaic model and real-time input data; The DSP controller generates corresponding PWM signals based on the calculation results to control the power electronic devices; By adjusting the duty cycle and frequency of the PWM signal, the output characteristics of the photovoltaic array can be accurately simulated; Monitor the output voltage and current in real time, and convert the digital signal into analog signal through the DAC module and feed it back to the controller; The DSP controller performs closed-loop regulation based on the feedback signal; Periodically trigger the self-calibration procedure to adjust the accuracy of the ADC / DAC through internal reference signals and external calibration equipment; Record operation data to storage devices for subsequent analysis and optimization; Data exchange and remote monitoring are carried out with the host computer through the communication interface.

3. The photovoltaic inverter test system according to claim 1, characterized in that: The adjustable load module comprises the following steps: Choose silicon carbide SiC and gallium nitride GaN power electronic devices; Design a high-frequency switching circuit to ensure that the switching frequency is above 1MHz; Develop and solidify the control algorithm to ensure that the control algorithm can adjust the load in real time; Design a control unit including a central processing unit, memory, and input and output interfaces; Integrate SiC and GaN power electronic devices, high-frequency switching circuits, control algorithm modules and control units, and perform debugging and testing; Design cooling system.

4. The photovoltaic inverter test system according to claim 3, characterized in that: The control algorithm includes: Design PID controller; Introducing an adaptive adjustment mechanism to adjust PID parameters according to the real-time performance of the system, including incremental adjustment and model reference adaptive control; In the main loop of the control system, the system status is acquired in real time, the error is calculated, and the PID parameters are adjusted according to the adaptive adjustment algorithm to calculate the PID controller output to control the system load.

5. The photovoltaic inverter test system according to claim 1, characterized in that: The data acquisition and monitoring module includes the following steps: Initialize the system, configure the ADC sampling rate and FPGA working parameters; Use the ADC module to collect the operating parameters of the photovoltaic inverter at a high sampling rate and convert the collected analog signals into digital signals; Transfer the digital signal to FPGA for preliminary processing, including filtering and data format conversion; Monitor system status in real time, collect new data and feed it back to ANFIS and DRL modules to continuously optimize control strategies.

6. The photovoltaic inverter test system according to claim 1, characterized in that: The data acquisition and monitoring module uses the improved fusion algorithm to perform data preprocessing and anomaly detection, including the following steps: 1) Collect multi-source data of photovoltaic inverters, including image data, time series data and sensor data, and clean and normalize the data; 2) Use convolutional neural network to extract features from preprocessed image data to obtain image feature set F cnn ; Use the long short-term memory network to extract features from the preprocessed time series data to obtain the time series feature set F lstm ; Use random forest to extract features from the preprocessed sensor data to obtain the sensor feature set F rf ; 3) The image feature set F cnn , time series feature set F lstm and sensor feature set F rf Merge to get the comprehensive feature set F fusion ; 4) Use the weighted average method to fuse the comprehensive feature set to obtain the fusion model output result s fusions , the formula is as follows: s fusion =α·s cnn +β·s lstm +γ·s rf Among them, s cnn 、s lstm 、s rf are the output results of the convolutional neural network, long short-term memory network and random forest model respectively. α, β and γ are weight coefficients, satisfying α+β+γ=1; 5) Clean, normalize and extract features of the photovoltaic inverter data collected in real time, use the fusion model to predict the comprehensive feature set, and obtain real-time anomaly detection results fusion , and judge whether the data is abnormal according to the preset threshold θ.

7. The photovoltaic inverter test system according to claim 1, characterized in that: The control and protection module comprises the following steps: Real-time collection of voltage, current and temperature data of photovoltaic inverters; Normalize and filter the collected data: The normalization formula is: Where: x is the original data, min(x) is the minimum value of the data, max(x) is the maximum value of the data, and x′ is the normalized data; The filtering formula is: Where: y[n] is the filtered data, x[nk] is the nkth data sample, and N is the size of the filter window; Generate initial control instructions using the trained ANFIS model: The ANFIS formula is: Where: y is the output of the ANFIS model, w i is the weight of the i-th membership function, f i (x) is the output function of the ith rule, and N is the number of membership functions; The ANFIS model provides preliminary control strategies to quickly respond to system changes; The DRL module continuously learns and optimizes the control strategy in the background; By interacting with the actual system, the parameters of the policy network are continuously adjusted: The policy network parameter update formula is: i new =θ old -a▽ θ J(θ) Where: θ new is the updated parameter, θ old is the parameter before updating, α is the learning rate, ▽ θ J(θ) is the gradient of the loss function with respect to the parameter θ; The output of the DRL module serves as a supplement and correction to the ANFIS control strategy; Use the ANFIS model to predict the future behavior of the system: The prediction formula is: In the formula, is the prediction for the future time t+1, x t is the input data at the current moment; The MPC algorithm is used to perform rolling optimization in each control cycle to generate the optimal control sequence: MPC optimization objective function: min(t*Qt+uf*Ru) Where: t is a variable, Qt is a state weight matrix, uf is the quantity to be optimized, and Ru is a control input weight matrix; The state update formula is: x t+1 =Ax t +Bu t Where: x t+1 is the state at the next moment, x t is the state at the current moment, A is the state transfer matrix, and B is the control matrix; According to the MPC optimization results, execute the optimal control instructions: The control instruction calculation formula is: Where: u t is the optimal control input at time t, Q is the state weight matrix, R is the control input weight matrix, is the quadratic form of the state vector weighted by the weight matrix Q, which represents the weighted sum of the state error. is the quadratic form of the control input vector weighted by the weight matrix R, which represents the weighted sum of the control inputs. u To solve the control input u that minimizes the objective function; Monitor system status in real time, collect new data and feed it back to ANFIS and DRL modules to continuously optimize control strategies.

8. The photovoltaic inverter test system according to claim 1, characterized in that: The environmental simulation module comprises: Temperature control unit, using high-precision temperature sensors and PID control algorithms to adjust and maintain the temperature of the test environment; The humidity control unit uses a high-precision humidity sensor and a humidity adjustment device to simulate environmental conditions under different humidity conditions.

9. The photovoltaic inverter test system according to claim 1, characterized in that: The test report generating module comprises: Data analysis unit, which performs trend analysis and fault prediction based on historical test data and machine learning algorithms; The report generation unit can automatically generate detailed reports containing test results, data analysis and optimization suggestions, and supports export in multiple formats.

10. A photovoltaic inverter testing method according to any one of claims 1 to 9, characterized in that: The method comprises: Step S1: The photovoltaic simulator module includes setting the working parameters of the DSP controller, initializing the ADC / DAC module, reading and processing the analog input signal, adjusting the PWM output parameters, generating a PWM signal, monitoring and feeding back the output voltage and current, performing closed-loop regulation and self-calibration, recording the operating data and exchanging data with the host computer; Step S2: The adjustable load module includes selecting SiC and GaN power electronic devices, designing high-frequency switching circuits, developing control algorithms, designing control units, integrating various components and performing debugging and testing; Step S3: the data acquisition and monitoring module includes configuring the working parameters of the ADC and FPGA, acquiring the operating parameters at a high sampling rate, performing preliminary processing and real-time monitoring, and using a fusion algorithm for data preprocessing and anomaly detection; Step S4: The control and protection module includes real-time acquisition of voltage, current, and temperature data, data normalization and filtering, generating initial control instructions, optimizing control strategies using the DRL module, predicting system behavior using the ANFIS model, performing rolling optimization through the MPC algorithm, and executing optimal control instructions; Step S5: The environmental simulation module includes adjusting the test environment using high-precision temperature and humidity sensors and a control system; Step S6: The test report generation module includes performing data analysis and fault prediction based on historical data and machine learning algorithms to generate a report containing test results, data analysis and optimization suggestions.

Images