A control method for improving grid stability using photovoltaic synchronous reactor devices

Abstract

This invention relates to the field of photovoltaic equipment operation and maintenance technology. The invention provides a control method for improving grid stability using photovoltaic synchronous machine (PVM) stacks. This method acquires data on the PVM stack, the power grid, meteorological conditions, and faults to construct a PV output prediction model. It then combines this data with real-time grid data to predict grid load. By comparing the predicted load with actual demand, the error value is calculated, and the output power of the PVM stack is adjusted. Real-time monitoring of grid operation data is conducted. If data is abnormal, a genetic algorithm is used to optimize the output power control model. This invention effectively solves the problems of insufficient voltage stability, inaccurate reactive power compensation, and slow control speed in photovoltaic power grids, improving grid stability and operating efficiency, and promoting the utilization of renewable energy and green, low-carbon development.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic equipment operation and maintenance technology, and in particular to a control method for improving grid stability using a photovoltaic synchronous machine stack device. Background Technology

[0002] A photovoltaic (PV) synchronous generator (SSG) stack is a device that combines a photovoltaic (PV) power generation system with synchronous generator technology. Like a traditional synchronous generator, the SSG stack not only provides active power and voltage support in the power system, but also leverages the advantages of PV power generation systems to achieve clean and renewable energy utilization. Through control algorithms and power electronic conversion technology, the SSG stack converts the direct current (DC) generated by PV modules into alternating current (AC) synchronized with the power grid, and possesses advanced functions such as rapid response to grid changes and participation in grid frequency and voltage regulation. The SSG stack plays a significant role in improving the grid connection performance of PV power generation systems, enhancing grid stability, and promoting the large-scale application of renewable energy.

[0003] The main steps to improve grid stability using photovoltaic (PV) synchronous generator (SRG) systems include the following: First, by using precise PV output forecasting and weather forecasting systems, the output power of the SRG system can be planned in advance to match grid demand and reduce the impact of power fluctuations on the grid. Second, by utilizing the built-in control algorithms of the SRG system, the grid status can be monitored in real time, and the output power and voltage of the device can be quickly adjusted according to changes in grid frequency, voltage, and other parameters, providing dynamic reactive power compensation and voltage support to the grid and enhancing grid stability. Furthermore, the SRG system can also participate in grid frequency regulation control, assisting the grid in maintaining frequency stability by adjusting its own output. Finally, by optimizing the interface technology between the SRG system and the grid, harmonic pollution can be reduced, power quality improved, and the safe and stable operation of the grid further guaranteed.

[0004] In actual operation, power grids using photovoltaic synchronous PV (PV) stacks face the following technical challenges: the output power of PV power generation is significantly affected by natural factors such as sunlight intensity and temperature, exhibiting intermittency and fluctuations. This directly leads to insufficient voltage stability in the PV power grid, hindering timely responses to changes in grid conditions. This further exacerbates voltage stability issues and inaccurate reactive power compensation, impacting the stable operation of the PV power grid. To address this technical problem, this invention provides... Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a control method for improving grid stability using a photovoltaic synchronous machine stack device, solving the problems of insufficient voltage stability and inaccurate reactive power compensation in photovoltaic grids, which result in slow control speed.

[0006] To solve the above-mentioned technical problems, the specific technical solution of the present invention is as follows:

[0007] This invention provides a control method for improving grid stability using a photovoltaic synchronous rectifier stack, comprising:

[0008] Step S101: Obtain the operating data of the photovoltaic synchronous machine stack device, the operating data of the power grid, the meteorological environment data, and the fault data. The operating data of the photovoltaic synchronous machine stack device includes the basic operating parameters of the synchronous machine stack device and the status information of the synchronous machine stack device. The operating data of the power grid includes the power grid voltage data and the power grid current data. The meteorological environment data includes the light intensity, temperature data, and weather data. Store the obtained operating data of the photovoltaic synchronous machine stack device, the operating data of the power grid, the meteorological environment data, and the fault data in the database.

[0009] Step S102: Retrieve historical operating data and meteorological environment data of the photovoltaic synchronous machine stack device from the database, construct a photovoltaic power output prediction model, obtain real-time operating data of the photovoltaic synchronous machine stack device and real-time meteorological environment data, substitute the real-time operating data of the photovoltaic synchronous machine stack device and real-time meteorological environment data into the photovoltaic power output prediction model, output the photovoltaic power output prediction result, store the photovoltaic power output prediction result data set.

[0010] Step S103: Obtain real-time grid operation data and real-time grid demand data; retrieve the time information of the real-time grid operation data to obtain the time information of the data to be retrieved; retrieve the time information of the data to be retrieved from the photovoltaic power output prediction result data set to obtain the prediction result corresponding to the time information of the data to be retrieved; substitute the prediction result corresponding to the time information of the data to be retrieved, the real-time grid demand data, and the real-time grid operation data into the preset grid load prediction model to output the real-time grid load prediction result.

[0011] Step S104: Obtain the real-time operating data of the photovoltaic synchronous machine stack device, compare the real-time grid load forecast result with the real-time grid demand data to obtain the real-time grid load forecast error value, and substitute the real-time grid load forecast error value, the real-time grid operating data and the real-time operating data of the photovoltaic synchronous machine stack device into the output power control model of the photovoltaic synchronous machine stack device to output the output power of the photovoltaic synchronous machine stack device.

[0012] Step S105: After the photovoltaic synchronous machine stack device outputs power to the grid, the grid operation data is monitored in real time to obtain the real-time grid operation data. The real-time grid operation data is compared with the preset standard value. If the real-time grid operation data is higher than the preset standard value, the output power control model of the photovoltaic synchronous machine stack device is optimized using a preset genetic algorithm to obtain the optimized output power of the photovoltaic synchronous machine stack device. The optimized output power of the photovoltaic synchronous machine stack device is then transmitted to the grid.

[0013] Furthermore, in the control method for improving grid stability using a photovoltaic synchronous synchrotron device described in this invention, step S101 is also used for:

[0014] The system acquires operational data from the photovoltaic synchronous machine (PVSM) unit, the power grid, meteorological data, and fault data. This data is then preprocessed to obtain preprocessed PVSM unit operational data, power grid operational data, meteorological data, and fault data. Integrity checks are performed on the preprocessed PVSM unit operational data, power grid operational data, meteorological data, and fault data. Data that passes the integrity check is stored in a database. Data that fails the integrity check is matched against a pre-defined data simulation model knowledge base to obtain the corresponding data simulation model. The data simulation model is then used to simulate the data that fails the integrity check, resulting in simulated data corresponding to the data that fails the integrity check. This simulated data is then stored in the database. Finally, the simulated data corresponding to the data that fails the integrity check is used to replace the data that fails the integrity check in the database.

[0015] Furthermore, in the control method for improving grid stability using a photovoltaic synchronous synchrotron device described in this invention, step S102 is also used for:

[0016] Historical operating data and meteorological data of the photovoltaic synchronous machine stack are retrieved from the database. Data features are extracted using the historical operating data and meteorological data to obtain the data features to be processed. The data features to be processed are matched in a preset algorithm knowledge base to obtain the algorithm model corresponding to the data features to be processed. The algorithm model corresponding to the data features to be processed is trained using the historical operating data and meteorological data of the photovoltaic synchronous machine stack to obtain the photovoltaic power output prediction model.

[0017] Furthermore, in the control method for improving grid stability using a photovoltaic synchronous synchrotron device described in this invention, step S103 is also used for:

[0018] The real-time power grid operation data and the data in the photovoltaic output prediction result dataset are aligned based on time to obtain aligned data. The aligned data is then inspected to obtain the alignment data inspection results, which include successfully aligned data and unsuccessfully aligned data. The time information of the unsuccessfully aligned data is collected.

[0019] Receive the time information for retrieving real-time power grid operation data, match the time information for retrieving real-time power grid operation data with the time information for data that failed to align, and obtain the time information matching result, which includes data that successfully matched the time and data that failed to match the time.

[0020] Retrieve the time information corresponding to the time information of the unsuccessful time matching data, and re-predict the photovoltaic output through step S102 to obtain the photovoltaic output prediction result corresponding to the time information of the unsuccessful time matching data.

[0021] Furthermore, in the control method for improving grid stability using a photovoltaic synchronous synchrotron device described in this invention, step S104 is also used for:

[0022] The operation data of the photovoltaic synchronous machine stack is acquired in real time through monitoring systems or sensors. The operation data of the photovoltaic synchronous machine stack includes the output power, voltage, current, temperature and fault status of the synchronous machine stack.

[0023] The real-time power grid load forecast result is obtained from step S103, and the real-time power grid demand data is obtained from the power grid dispatch system. The real-time power grid demand data includes the total load demand of the power grid and the load allocation of each region.

[0024] By comparing the real-time power grid load forecast results with the real-time power grid demand data and calculating the error value, the degree of deviation between the real-time power grid load forecast results and the real-time power grid demand data is obtained.

[0025] The calculated real-time power grid load prediction error value is stored, and an error value monitoring mechanism is established. When the error value exceeds the preset threshold, an alarm is triggered.

[0026] Furthermore, in the control method for improving grid stability using a photovoltaic synchronous synchrotron device described in this invention, step S105 is also used for:

[0027] The system acquires real-time power grid operation data from the power grid monitoring system, preprocesses the real-time power grid operation data, and compares the preprocessed real-time power grid operation data with preset standard values ​​one by one. The data to be compared includes the voltage fluctuation range, current over-limit situation, frequency stability, and power factor qualification rate.

[0028] Based on the comparison results, the error value is calculated to obtain the difference between the real-time power grid operation data and the preset standard value, and the result of the error analysis is obtained.

[0029] Based on the results of error analysis, it is determined whether the power grid's operating status is within the normal range. If the real-time data exceeds the preset standard value, an alarm mechanism is triggered.

[0030] Furthermore, in the control method for improving grid stability using a photovoltaic synchronous synchrotron device described in this invention, step S105 is also used for:

[0031] The target of the optimization and the population size are received. The optimization target is to minimize the deviation between the power grid operation data and the preset standard value.

[0032] Based on the population size, a set of control model parameter combinations is randomly generated as the initial population, and each control parameter combination of the photovoltaic synchronous machine stack device represents a solution.

[0033] Calculate the deviation between the power grid operation data and the preset standard value under each parameter combination, and evaluate the merits and demerits of each parameter combination in the initial population;

[0034] Based on the evaluation results, a subset of control parameter combinations for photovoltaic synchronous machine stack devices were selected as parent generation for crossover and mutation operations.

[0035] By cross-operation, the control parameters of the photovoltaic synchronous machine stack device in the parent parameter combination are exchanged to generate a new control parameter combination of the child photovoltaic synchronous machine stack device. By mutation operation, the parameters in the control parameter combination of the child photovoltaic synchronous machine stack device are randomly adjusted to increase the diversity of the control parameters of the photovoltaic synchronous machine stack device.

[0036] The deviation between the grid operation data and the preset standard value under each parameter combination is calculated. The quality of the control combination of the newly generated generation of photovoltaic synchronous machine stack devices is evaluated. Full cross-operation and mutation operations are performed until the set number of generations is reached or the deviation between the grid operation data and the preset standard value under each parameter combination reaches the threshold. The parameter combination with the smallest deviation between the grid operation data and the preset standard value under each parameter combination is selected from the final population as the optimal solution, that is, the optimized output power control model parameters of the photovoltaic synchronous machine stack device.

[0037] The beneficial effects of this invention are:

[0038] By accurately predicting photovoltaic power output and grid load, this invention enables advance planning of the output power of photovoltaic synchronous generator sets, matching it with grid demand. This effectively reduces the impact of power fluctuations on the grid and enhances grid stability.

[0039] By utilizing the built-in control algorithm of the photovoltaic synchronous machine stack, this invention can monitor the grid status in real time and quickly adjust the output power and voltage of the device according to changes in grid frequency, voltage and other parameters, providing dynamic reactive power compensation for the grid and further improving the voltage stability of the grid.

[0040] This invention achieves rapid response and adjustment to grid conditions by monitoring grid operation data in real time and optimizing the output power control model of the photovoltaic synchronous machine stack device using a genetic algorithm when necessary. This significantly improves control response speed. By optimizing the operation of the photovoltaic synchronous machine stack device, this invention maximizes the utilization of the photovoltaic power generation system's output power, reduces dependence on traditional energy sources, and promotes the large-scale application of renewable energy.

[0041] This invention fully considers the integrity and accuracy of data during data acquisition, processing, and model building. Through data preprocessing, integrity detection, and data simulation, it effectively improves the system's robustness. By integrating advanced data processing algorithms and control strategies, this invention achieves intelligent management of photovoltaic synchronous machine stack devices, reducing operation and maintenance costs and improving management efficiency.

[0042] The application of this invention helps to promote the development of clean energy, reduce greenhouse gas emissions, and promote green and low-carbon development, which is in line with the global trend of energy transition and sustainable development. Attached Figure Description

[0043] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on the drawings without creative effort.

[0044] Figure 1 This is a schematic flowchart of a control method for improving power grid stability using a photovoltaic synchronous generator stack, provided as an embodiment of the present invention. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention. The technical solutions provided by various embodiments of this invention will be described in detail below with reference to the accompanying drawings.

[0046] To better understand the purpose of this invention, the invention will now be described in further detail.

[0047] This invention provides a control method for improving grid stability using a photovoltaic synchronous rectifier stack, comprising:

[0048] Step S101: Obtain the operating data of the photovoltaic synchronous machine stack device, the operating data of the power grid, the meteorological environment data, and the fault data. The operating data of the photovoltaic synchronous machine stack device includes the basic operating parameters of the synchronous machine stack device and the status information of the synchronous machine stack device. The operating data of the power grid includes the power grid voltage data and the power grid current data. The meteorological environment data includes the light intensity, temperature data, and weather data. Store the obtained operating data of the photovoltaic synchronous machine stack device, the operating data of the power grid, the meteorological environment data, and the fault data in the database.

[0049] In step S101, the present invention provides a comprehensive and systematic data acquisition and storage method, which aims to provide complete data support for subsequent photovoltaic power output prediction, grid load prediction, and output power control. The specific process is as follows:

[0050] Operating data of photovoltaic synchronous machine stack: including basic operating parameters of the synchronous machine stack (such as power, voltage, current, frequency, etc.) and status information of the synchronous machine stack (such as operating status, fault status, etc.).

[0051] Power grid operation data: including grid voltage and current data. Meteorological environment data: including sunlight intensity, temperature, and weather data (such as sunny, cloudy, rainy, snowy, etc.). Since the output power of photovoltaic power generation is greatly affected by meteorological conditions, meteorological environment data is crucial for photovoltaic power output forecasting.

[0052] Fault data: Records any fault information that occurs in the photovoltaic synchronous machine stack or the power grid, including fault type, occurrence time, and scope of impact.

[0053] Data storage: Store all the data obtained above in a database. The selection and design of the database should take into account the integrity of the data. Through the database, the data can be easily managed and queried, which will facilitate subsequent analysis and prediction.

[0054] Before storing data in a database, it is necessary to preprocess the data to improve its accuracy. Preprocessing operations include data cleaning (such as removing duplicate data and handling missing values), data format conversion (such as converting data from different sources into a unified format), and data standardization (such as converting data into a dimensionless form to facilitate comparison and analysis).

[0055] Step S102: Retrieve historical operating data and meteorological environment data of the photovoltaic synchronous machine stack device from the database, construct a photovoltaic power output prediction model, obtain real-time operating data of the photovoltaic synchronous machine stack device and real-time meteorological environment data, substitute the real-time operating data of the photovoltaic synchronous machine stack device and real-time meteorological environment data into the photovoltaic power output prediction model, output the photovoltaic power output prediction result, store the photovoltaic power output prediction result data set.

[0056] Historical operating data and meteorological data of the photovoltaic synchronous generator unit were retrieved from the database. This data forms the basis for building the photovoltaic power output prediction model, and historical data can reflect the complex relationship between photovoltaic power output and meteorological environment.

[0057] A photovoltaic (PV) power output prediction model is constructed using historical operating data and meteorological data retrieved from the PV synchronous generator unit. This model can be based on machine learning, deep learning, or other statistical methods to predict the output power of the PV synchronous generator unit under given meteorological conditions.

[0058] The system acquires real-time operating data of the photovoltaic synchronous machine (PVSM) unit and real-time meteorological data, which serve as inputs for model prediction and are used to generate real-time PV power output prediction results.

[0059] The real-time operating data of the photovoltaic synchronous machine and the real-time meteorological environment data are substituted into the photovoltaic power output prediction model. The photovoltaic power output prediction model outputs real-time photovoltaic power output prediction results based on the input data and the historical patterns it has learned.

[0060] The photovoltaic power output prediction results are stored to form a photovoltaic power output prediction result data set. The photovoltaic power output prediction result data set records the photovoltaic power output prediction values ​​at different time points, providing an important reference for subsequent grid load prediction and output power control.

[0061] Through step S102, the present invention realizes real-time prediction of the output power of the photovoltaic synchronous machine stack device. The prediction result not only takes into account the operating status of the photovoltaic synchronous machine stack device itself, but also fully considers the influence of meteorological environmental factors, thereby improving the accuracy of the prediction.

[0062] Step S103: Obtain real-time grid operation data and real-time grid demand data; retrieve the time information of the real-time grid operation data to obtain the time information of the data to be retrieved; retrieve the time information of the data to be retrieved from the photovoltaic power output prediction result data set to obtain the prediction result corresponding to the time information of the data to be retrieved; substitute the prediction result corresponding to the time information of the data to be retrieved, the real-time grid demand data, and the real-time grid operation data into the preset grid load prediction model to output the real-time grid load prediction result.

[0063] Real-time power grid operation data: Real-time acquisition of power grid operation data such as voltage, current, and frequency from the power grid monitoring system.

[0064] Real-time power grid demand data: Obtain real-time power grid demand data from the power grid dispatching system, including the total load demand of the power grid and the load distribution in each region.

[0065] The time information of real-time power grid operation data is retrieved, that is, the timestamp of the power grid operation data at the current time point or the most recent time point is obtained. Based on the retrieved time information, the photovoltaic output prediction result for the corresponding time point is searched in the photovoltaic output prediction result data set obtained in step S102.

[0066] The photovoltaic output prediction results, real-time grid demand data, and real-time grid operation data corresponding to the data time information to be retrieved are substituted into the preset grid load prediction model. The grid load prediction model is a mathematical model based on historical data, real-time data, and prediction data, which is used to predict the load changes of the grid in the future.

[0067] The power grid load forecasting model outputs real-time power grid load forecasting results based on the input data and its internal algorithms and parameters. The real-time power grid load forecasting results take into account various factors such as photovoltaic output, power grid demand, and the current power grid operating status, and have high accuracy and reliability.

[0068] Through step S103, this invention achieves real-time prediction of grid load, providing a reference for subsequent output power control of photovoltaic synchronous generator units, which helps to achieve grid supply and demand balance and improve grid stability and reliability. Simultaneously, by monitoring and predicting grid load in real time, potential problems in grid operation can be identified in a timely manner, providing decision support for grid dispatching and operation management.

[0069] Step S104: Obtain the real-time operating data of the photovoltaic synchronous machine stack device, compare the real-time grid load forecast result with the real-time grid demand data to obtain the real-time grid load forecast error value, and substitute the real-time grid load forecast error value, the real-time grid operating data and the real-time operating data of the photovoltaic synchronous machine stack device into the output power control model of the photovoltaic synchronous machine stack device to output the output power of the photovoltaic synchronous machine stack device.

[0070] Real-time operational data acquisition involves using monitoring systems or sensors to acquire real-time operational data of the photovoltaic synchronous machine (PVM) stack, including the stack's output power, voltage, current, temperature, and fault status. This data reflects the current operational status of the PVM stack.

[0071] Load forecasting and demand comparison: The real-time grid load forecasting results obtained in step S103 are compared with the real-time grid demand data. The real-time grid demand data is obtained from the grid dispatching system and includes the total load demand of the grid and the load allocation of each region.

[0072] By comparing the forecasts, the real-time power grid load prediction error is calculated. This error reflects the degree of deviation between the predicted load and the actual demand, and is an important indicator for measuring the accuracy of load forecasting.

[0073] The real-time grid load prediction error, real-time grid operating data (such as voltage and current), and real-time operating data of the photovoltaic synchronous machine (PVSM) stack are substituted into the output power control model of the PVSM stack. The output power control model is a model based on physical principles, mathematical algorithms, and experimental data, used to calculate and output the optimal output power of the PVSM stack based on the input data.

[0074] The output power control model calculates the optimal output power of the photovoltaic synchronous machine stack based on the input data, combined with the algorithm and parameters inside the model, and outputs this output power to the actuator of the photovoltaic synchronous machine stack.

[0075] Through step S104, this invention achieves real-time control of the output power of the photovoltaic synchronous machine stack. The real-time control process fully considers grid load forecasts, actual grid demand, and the operating status of the photovoltaic synchronous machine stack itself, contributing to supply and demand balance in the photovoltaic grid and improving grid stability and reliability. Simultaneously, by adjusting the output power of the photovoltaic synchronous machine stack in real time, it can maximize the utilization of renewable energy, reduce dependence on traditional energy sources, and promote green and low-carbon development.

[0076] Step S105: After the photovoltaic synchronous machine stack device outputs power to the grid, the grid operation data is monitored in real time to obtain the real-time grid operation data. The real-time grid operation data is compared with the preset standard value. If the real-time grid operation data is higher than the preset standard value, the output power control model of the photovoltaic synchronous machine stack device is optimized using a preset genetic algorithm to obtain the optimized output power of the photovoltaic synchronous machine stack device. The optimized output power of the photovoltaic synchronous machine stack device is then transmitted to the grid.

[0077] Step S105 is a crucial step in this invention for real-time monitoring of the power grid's operating status and for evaluating and optimizing the output power control model of the photovoltaic synchronous accelerator unit. The following is a detailed explanation of this step:

[0078] After the photovoltaic synchronous generator stack transmits its output power to the grid, the grid monitoring system acquires real-time operating data of the grid, including parameters such as voltage, current, frequency, and power factor.

[0079] The operational data is compared with preset standard values. Real-time monitored power grid operational data is compared with preset standard values. These preset standard values ​​are formulated based on power grid operation safety specifications, performance indicators, and historical experience, and are used to assess the stability and security of the current power grid operation.

[0080] The judgment and optimization conditions are as follows: if the real-time power grid operating data is higher than the preset standard value, it indicates that the current operating state of the power grid is unstable or has safety risks, and adjustments and optimizations are needed. At this time, the genetic algorithm is triggered to optimize the output power control model of the photovoltaic synchronous machine stack device.

[0081] Genetic algorithm optimization: A genetic algorithm is an optimization algorithm based on the principles of natural selection and genetics, which searches for the optimal solution by simulating the natural evolutionary process. In this step, the genetic algorithm is used to optimize the output power control model of the photovoltaic synchronous machine stack to improve its adaptability and stability to the grid operating conditions.

[0082] The specific optimization process includes steps such as determining the optimization objective (e.g., minimizing the deviation between power grid operation data and preset standard values), preparing the parameters required for the genetic algorithm (e.g., population size, number of generations, mutation rate, crossover rate, etc.), initializing the population, evaluating fitness, selecting parents, performing crossover and mutation operations, iterative optimization, and determining the optimal solution.

[0083] The optimized output power is then further optimized using a genetic algorithm to obtain the optimized output power control model parameters for the photovoltaic synchronous machine (PVSM) stack. These parameters are applied to the output power control of the PVSM stack to calculate the optimized output power, which is then transmitted to the grid.

[0084] Through step S105, this invention achieves real-time monitoring and evaluation of the power grid's operating status, as well as dynamic optimization of the output power control model for the photovoltaic synchronous machine stack. This step helps improve the stability and security of the power grid, enabling the photovoltaic synchronous machine stack to operate efficiently and stably in complex and ever-changing power grid environments, providing strong support for the large-scale application of renewable energy.

[0085] Specifically, in the control method for improving grid stability using a photovoltaic synchronous synchrotron device described in this invention, step S101 is further used for:

[0086] The system acquires operational data from the photovoltaic synchronous machine (PVSM) unit, the power grid, meteorological data, and fault data. This data is then preprocessed to obtain preprocessed PVSM unit operational data, power grid operational data, meteorological data, and fault data. Integrity checks are performed on the preprocessed PVSM unit operational data, power grid operational data, meteorological data, and fault data. Data that passes the integrity check is stored in a database. Data that fails the integrity check is matched against a pre-defined data simulation model knowledge base to obtain the corresponding data simulation model. The data simulation model is then used to simulate the data that fails the integrity check, resulting in simulated data corresponding to the data that fails the integrity check. This simulated data is then stored in the database. Finally, the simulated data corresponding to the data that fails the integrity check is used to replace the data that fails the integrity check in the database.

[0087] Operational data of the photovoltaic synchronous machine (PV) stack, power grid operation data, meteorological environmental data, and fault data are obtained from various data sources. The PV stack's operational data includes basic operating parameters and status information; the power grid's operational data covers voltage and current data; and the meteorological environmental data includes solar irradiance, temperature, and weather data.

[0088] The acquired data undergoes preprocessing. Preprocessing steps include data cleaning (such as removing outliers and filling in missing values), data transformation (such as converting data from different units to a unified unit), and data reduction (such as reducing the complexity of the dataset through dimensionality reduction techniques) to ensure the accuracy and consistency of the data.

[0089] Perform integrity checks on the preprocessed data. Integrity checks aim to verify the completeness of the data, that is, to check for missing or inconsistencies. By setting certain integrity check criteria, we can identify which data records are complete and which are incomplete. Data that passes the integrity check is then stored in the database for subsequent analysis and model building.

[0090] For data that fails integrity checks, this invention proposes an innovative solution. First, a matching process is performed in a pre-defined data simulation model knowledge base to find the data simulation model corresponding to the incomplete data. Then, these data simulation models are used to simulate the incomplete data, generating simulated data. Finally, these simulated data are stored in a database, and the simulated data replaces the incomplete data within the database.

[0091] The database is continuously updated through ongoing data acquisition and preprocessing. Simultaneously, the database undergoes regular integrity checks to ensure its accuracy and completeness. For newly discovered incomplete data, the aforementioned data simulation and replacement process is repeated.

[0092] Through the series of operations in step S101, this invention effectively solves the problem of incomplete data, providing a reliable data foundation for subsequent photovoltaic power output prediction, grid load prediction, and control of the output power of photovoltaic synchronous generator units. At the same time, this data preprocessing and integrity detection method also improves the robustness and accuracy of the entire control method.

[0093] Specifically, in the control method for improving grid stability using a photovoltaic synchronous synchrotron device described in this invention, step S102 is further used for:

[0094] Historical operating data and meteorological data of the photovoltaic synchronous machine stack are retrieved from the database. Data features are extracted using the historical operating data and meteorological data to obtain the data features to be processed. The data features to be processed are matched in a preset algorithm knowledge base to obtain the algorithm model corresponding to the data features to be processed. The algorithm model corresponding to the data features to be processed is trained using the historical operating data and meteorological data of the photovoltaic synchronous machine stack to obtain the photovoltaic power output prediction model.

[0095] In step S102, the present invention describes in detail the process of constructing a photovoltaic power output prediction model. This process makes full use of the historical operating data and meteorological environmental data of the photovoltaic synchronous generator unit. The following is a detailed interpretation of this step:

[0096] Historical operating data of the photovoltaic synchronous generator unit and historical meteorological data were retrieved from the database. This data forms the basis for building the photovoltaic output prediction model, as it contains historical relationships between photovoltaic output and various factors (such as light intensity and temperature).

[0097] Data feature extraction is performed on the retrieved historical data. This step aims to extract useful feature information for photovoltaic power output prediction from the raw data. For example, features such as light intensity, temperature, and humidity can be extracted from historical meteorological data; and features such as output power, voltage, and current can be extracted from the historical operating data of photovoltaic synchronous generator units.

[0098] The extracted data features to be processed are matched against a pre-defined algorithm knowledge base. This knowledge base is a database containing various algorithm models designed to handle different types of data features and output prediction results. Through matching, the most suitable algorithm model for processing the current data features can be found.

[0099] The algorithm model corresponding to the characteristics of the data to be processed is trained using historical operating data of the photovoltaic synchronous machine stack and historical meteorological environmental data. The training process involves continuously adjusting the parameters of the algorithm model to better fit the patterns in the historical data, thereby improving the accuracy of predicting future photovoltaic power output.

[0100] After training, a photovoltaic power output prediction model is obtained. The photovoltaic power output prediction model can receive real-time photovoltaic synchronous machine stack operation data and meteorological environment data as input, and output the predicted photovoltaic power output results.

[0101] The constructed photovoltaic power output prediction model is stored in a database for later use and retrieval. Simultaneously, as new data accumulates, the model can be periodically updated and retrained to improve its prediction accuracy and adaptability.

[0102] This invention presents a photovoltaic power output prediction model based on historical data. This model can predict the output power of photovoltaic synchronous machine stacks in real time, providing an important basis for subsequent grid load prediction and control of the output power of photovoltaic synchronous machine stacks. This not only helps to improve the stability of the grid, but also promotes the efficient utilization of renewable energy.

[0103] Specifically, in the control method for improving grid stability using a photovoltaic synchronous synchrotron device described in this invention, step S103 is further used for:

[0104] The real-time power grid operation data and the data in the photovoltaic output prediction result dataset are aligned based on time to obtain aligned data. The aligned data is then inspected to obtain the alignment data inspection results, which include successfully aligned data and unsuccessfully aligned data. The time information of the unsuccessfully aligned data is collected.

[0105] Receive the time information for retrieving real-time power grid operation data, match the time information for retrieving real-time power grid operation data with the time information for data that failed to align, and obtain the time information matching result, which includes data that successfully matched the time and data that failed to match the time.

[0106] Retrieve the time information corresponding to the time information of the unsuccessful time matching data, and re-predict the photovoltaic output through step S102 to obtain the photovoltaic output prediction result corresponding to the time information of the unsuccessful time matching data.

[0107] In step S103, the present invention describes in detail how to handle the time alignment problem between the real-time grid operation data and the photovoltaic output prediction result data set, and provides a solution for data that fails to align. The following is a detailed interpretation of this step:

[0108] The real-time grid operation data and the data in the photovoltaic power output prediction dataset are aligned based on time. This step is to ensure that the grid operation data and the corresponding photovoltaic power output prediction results are matched in time, so that accurate grid load forecasting can be performed subsequently.

[0109] The aligned data is then inspected to obtain the alignment data inspection results. These results include successfully aligned data and unaligned data. Successfully aligned data refers to grid operation data that successfully matches the corresponding photovoltaic power output prediction results, while unaligned data refers to data that cannot match the corresponding prediction results.

[0110] For data that failed to align, its time information is collected, and the time information of real-time grid operation data is received and retrieved. This time information is then matched with the time information of the unaligned data. The matching results include data that successfully matched in time and data that failed to match in time. Data that successfully matched in time refers to data that, although it failed to match in the alignment stage, successfully found the corresponding photovoltaic power output prediction result in the subsequent time information matching stage; while data that failed to match in time refers to data that could not find the corresponding prediction result in either the alignment stage or the time information matching stage.

[0111] For data that fails to match the time, retrieve the corresponding time information and re-predict the photovoltaic output through step S102. This step is to fill the data gaps and ensure that the power grid operation data at each time point has corresponding photovoltaic output prediction results, thereby improving the accuracy and reliability of power grid load prediction.

[0112] Through this series of operations in step S103, the present invention effectively solves the time alignment problem between the real-time grid operation data and the photovoltaic power output prediction result data set, and provides a solution for data that fails to align. This not only improves the accuracy of grid load prediction, but also provides more accurate data support for the output power control of subsequent photovoltaic synchronous stack devices.

[0113] Specifically, in the control method for improving grid stability using a photovoltaic synchronous synchrotron device described in this invention, step S104 is further used for:

[0114] The operation data of the photovoltaic synchronous machine stack is acquired in real time through monitoring systems or sensors. The operation data of the photovoltaic synchronous machine stack includes the output power, voltage, current, temperature and fault status of the synchronous machine stack.

[0115] The real-time power grid load forecast result is obtained from step S103, and the real-time power grid demand data is obtained from the power grid dispatch system. The real-time power grid demand data includes the total load demand of the power grid and the load allocation of each region.

[0116] By comparing the real-time power grid load forecast results with the real-time power grid demand data and calculating the error value, the degree of deviation between the real-time power grid load forecast results and the real-time power grid demand data is obtained.

[0117] The calculated real-time power grid load prediction error value is stored, and an error value monitoring mechanism is established. When the error value exceeds the preset threshold, an alarm is triggered.

[0118] The operation data of the photovoltaic synchronous machine stack is acquired in real time by monitoring systems or sensors. The operation data of the photovoltaic synchronous machine stack includes parameters such as output power, voltage, current, temperature and fault status of the synchronous machine stack. The operation data of the photovoltaic synchronous machine stack reflects the current operating status of the photovoltaic synchronous machine stack.

[0119] The real-time grid load forecast result is obtained from step S103. The real-time grid load forecast result is calculated based on multiple factors such as photovoltaic power output forecast and historical grid data. The real-time grid load forecast result represents the load demand forecast of the grid in the future.

[0120] At the same time, real-time grid demand data is obtained from the grid dispatch system. This real-time grid demand data includes information such as the total load demand of the grid and the load distribution in each region. The real-time grid demand data reflects the current actual load demand of the grid.

[0121] The accuracy of the prediction results is evaluated by comparing the real-time power grid load forecast results with the real-time power grid demand data and calculating the error value. The error value reflects the degree of deviation between the prediction results and the actual demand, and is an important indicator for evaluating the performance of the prediction model.

[0122] The calculated real-time power grid load forecast error value is stored, and a monitoring mechanism for the error value is established. The monitoring mechanism can track the changes in the error value in real time. When the error value exceeds a preset threshold, an alarm mechanism will be triggered to remind operators to pay attention to the accuracy of the forecast results.

[0123] When the error value exceeds the preset threshold, the alarm mechanism will issue a warning signal to remind operators to take corresponding measures. These measures include re-predicting photovoltaic power output, adjusting the grid load allocation plan, and strengthening the operation and maintenance management of photovoltaic synchronous generator units to ensure the stable operation of the grid and the balance between supply and demand.

[0124] Specifically, in the control method for improving grid stability using a photovoltaic synchronous synchrotron device described in this invention, step S105 is further used for:

[0125] The system acquires real-time power grid operation data from the power grid monitoring system, preprocesses the real-time power grid operation data, and compares the preprocessed real-time power grid operation data with preset standard values ​​one by one. The data to be compared includes the voltage fluctuation range, current over-limit situation, frequency stability, and power factor qualification rate.

[0126] Based on the comparison results, the error value is calculated to obtain the difference between the real-time power grid operation data and the preset standard value, and the result of the error analysis is obtained.

[0127] Based on the results of error analysis, it is determined whether the power grid's operating status is within the normal range. If the real-time data exceeds the preset standard value, an alarm mechanism is triggered.

[0128] Real-time acquisition of power grid operation data from the power grid monitoring system is a direct indicator of the current operating status of the power grid, including but not limited to parameters such as voltage, current, frequency, and power factor.

[0129] The acquired real-time power grid operation data is preprocessed, including data cleaning, noise reduction, and formatting, to improve the accuracy of the data and provide a reliable foundation for subsequent analysis.

[0130] The preprocessed real-time power grid operation data is compared one by one with the preset standard values. The preset standard values ​​are determined comprehensively based on factors such as power grid operation safety regulations, technical requirements, and historical experience. The preset standard values ​​represent the threshold range for normal operation of the power grid.

[0131] The data for comparison includes indicators such as voltage fluctuation range, current over-limit situation, frequency stability, and power factor qualification rate.

[0132] Based on the comparison results, the error value is calculated to obtain the difference between the real-time power grid operation data and the preset standard value. The magnitude of the error value reflects the degree of deviation between the current operating state of the power grid and the normal state.

[0133] Based on the results of error analysis, it is determined whether the power grid is operating within the normal range. If the real-time data fluctuates within the allowable range of the preset standard value, the power grid is considered to be operating normally. If the real-time data exceeds the preset standard value, it is determined that the power grid is operating abnormally.

[0134] If real-time data exceeds preset standard values, an alarm mechanism is triggered. The alarm mechanism includes various methods such as sound alarm, light signal alarm, and sending alarm information to the mobile devices of relevant personnel, so that operators can obtain information about abnormal power grid operation in a timely manner.

[0135] After an alarm is triggered, operators need to take corresponding measures based on the alarm information. These measures include adjusting the output power of the photovoltaic synchronous generator, switching to backup power, and performing emergency repairs on the power grid to ensure the safe and stable operation of the power grid.

[0136] At the same time, it is also necessary to conduct in-depth analysis of the abnormal data, find out the root cause of the abnormal power grid operation, and take targeted improvement measures to prevent similar problems from happening again.

[0137] Specifically, in the control method for improving grid stability using a photovoltaic synchronous synchrotron device described in this invention, step S105 is further used for:

[0138] The target of the optimization and the population size are received. The optimization target is to minimize the deviation between the power grid operation data and the preset standard value.

[0139] Based on the population size, a set of control model parameter combinations is randomly generated as the initial population, and each control parameter combination of the photovoltaic synchronous machine stack device represents a solution.

[0140] Calculate the deviation between the power grid operation data and the preset standard value under each parameter combination, and evaluate the merits and demerits of each parameter combination in the initial population;

[0141] Based on the evaluation results, a subset of control parameter combinations for photovoltaic synchronous machine stack devices were selected as parent generation for crossover and mutation operations.

[0142] By cross-operation, the control parameters of the photovoltaic synchronous machine stack device in the parent parameter combination are exchanged to generate a new control parameter combination of the child photovoltaic synchronous machine stack device. By mutation operation, the parameters in the control parameter combination of the child photovoltaic synchronous machine stack device are randomly adjusted to increase the diversity of the control parameters of the photovoltaic synchronous machine stack device.

[0143] The deviation between the grid operation data and the preset standard value under each parameter combination is calculated. The quality of the control combination of the newly generated generation of photovoltaic synchronous machine stack devices is evaluated. Full cross-operation and mutation operations are performed until the set number of generations is reached or the deviation between the grid operation data and the preset standard value under each parameter combination reaches the threshold. The parameter combination with the smallest deviation between the grid operation data and the preset standard value under each parameter combination is selected from the final population as the optimal solution, that is, the optimized output power control model parameters of the photovoltaic synchronous machine stack device.

[0144] Given the optimization objective and population size, the first step is to define the optimization objective: minimizing the deviation between the power grid operating data and the preset standard value. This is the core pursuit of the genetic algorithm optimization process, aiming to make the power grid operating data closer to the preset ideal state by adjusting the control model parameters.

[0145] Based on the set population size, a set of control model parameter combinations is randomly generated as the initial population. Each control parameter combination of a photovoltaic synchronous machine stack represents a solution, forming the initial set of the genetic algorithm's search space.

[0146] Initial population assessment involves calculating the deviation between grid operation data and preset standard values ​​for each parameter combination. This deviation serves as the basis for evaluating the performance of each parameter combination in the initial population. The smaller the deviation, the closer the grid operation data under that parameter combination is to the preset standard value, and therefore, the better the performance of that parameter combination.

[0147] Based on the evaluation results, a subset of high-performing photovoltaic synchronous synchrotron reactor (PSRC) control parameter combinations are selected as parent generations. These parent parameter combinations will play a crucial role in subsequent crossover and mutation operations, providing a foundation for generating new and superior offspring parameter combinations.

[0148] Crossover and mutation operations are used to exchange control parameters in parent parameter combinations, generating new parameter combinations in offspring. This process simulates gene recombination in nature and helps explore new parameter combination spaces. Simultaneously, mutation operations randomly adjust certain parameters in offspring parameter combinations to increase population diversity. Mutation helps to escape the limitations of local optima and explore a broader solution space.

[0149] Iterative optimization involves evaluating newly generated offspring parameter combinations and calculating the deviation between their corresponding power grid operation data and preset standard values. Based on the evaluation results, high-performing offspring parameter combinations are selected as new populations, and crossover, mutation, and evaluation operations are repeated until a set number of generations is reached or the deviation value for each parameter combination reaches a preset threshold.

[0150] To determine the optimal solution, after the iterative optimization process is completed, the best-performing parameter combination is selected from the final population as the optimal solution. This optimal solution represents the optimized output power control model parameters of the photovoltaic synchronous machine stack, which minimizes the deviation between the grid operation data and the preset standard value.

[0151] Through the optimization process of a genetic algorithm, this invention can automatically adjust the output power control model parameters of a photovoltaic synchronous machine (PVSM) stack to adapt to real-time changes in the grid's operating status, thereby improving the grid's stability and operating efficiency. This optimization method has advantages such as strong adaptability and strong global search capability, providing strong technical support for the application of PVSM stacks in the power grid.

[0152] The technical solution of this invention effectively solves the problems of insufficient voltage stability, inaccurate reactive power compensation, and slow control speed of photovoltaic power grids by using a control method to improve grid stability through a photovoltaic synchronous machine stack device. The specific solution is as follows:

[0153] The system acquires operational data from the photovoltaic synchronous machine (PV) unit, grid operation data, meteorological environmental data, and fault data, storing this data in a database. The acquired data undergoes preprocessing, including integrity checks. For data failing integrity checks, simulations are performed using a pre-defined data simulation model knowledge base to obtain simulated data, which then replaces the original non-compliant data, improving data integrity. A PV power output prediction model is constructed using historical operational data from the PV synchronous machine and historical meteorological environmental data. This model can predict PV power output based on real-time operational and meteorological environmental data. The prediction model is then substituted with real-time PV synchronous machine operational data and real-time meteorological environmental data to output the predicted PV power output, which is then stored to form a dataset.

[0154] The system acquires real-time grid operation data and real-time grid demand data, combines them with photovoltaic (PV) output forecast results, and inputs them into a pre-defined grid load forecast model to output real-time grid load forecast results. Time alignment processing is performed on the real-time grid operation data and PV output forecast results; for data that fails to align, PV output forecasting is performed again to ensure data synchronization and integrity. The real-time grid load forecast results are compared with real-time grid demand data to calculate the real-time grid load prediction error. The error value, real-time grid operation data, and real-time operation data of the PV synchronous generator stack are then input into the output power control model to output the output power of the PV synchronous generator stack.

[0155] After the photovoltaic synchronous machine (PVSM) stack output power is transmitted to the grid, the grid's operating data is monitored in real time. The real-time grid operating data is compared with preset standard values. If the data exceeds the standard values, a genetic algorithm is used to optimize the output power control model. The optimization process includes setting optimization objectives, initializing the population, evaluating fitness, selecting parents, performing crossover and mutation operations, and iterative optimization to finally determine the optimal solution, i.e., the optimized output power control model parameters. An error value monitoring mechanism is established; when the error value exceeds a preset threshold, an alarm is triggered to promptly detect problems and take corresponding measures.

[0156] Through the above technical solution, the present invention can achieve precise control of the output power of photovoltaic synchronous machine stack device, improve the voltage stability of photovoltaic grid, enhance the accuracy of reactive power compensation, and accelerate the control response speed, thereby effectively solving the key problems in the operation of photovoltaic grid.

Claims

1. A control method for improving grid stability using a photovoltaic synchronous rectifier stack, characterized in that, include: Step S101: Obtain the operating data of the photovoltaic synchronous machine stack device, the operating data of the power grid, the meteorological environment data, and the fault data. The operating data of the photovoltaic synchronous machine stack device includes the basic operating parameters of the synchronous machine stack device and the status information of the synchronous machine stack device. The operating data of the power grid includes the power grid voltage data and the power grid current data. The meteorological environment data includes the light intensity, temperature data, and weather data. Store the obtained operating data of the photovoltaic synchronous machine stack device, the operating data of the power grid, the meteorological environment data, and the fault data in the database. Step S102: Retrieve historical operating data and meteorological environment data of the photovoltaic synchronous machine stack device from the database, construct a photovoltaic power output prediction model, obtain real-time operating data of the photovoltaic synchronous machine stack device and real-time meteorological environment data, substitute the real-time operating data of the photovoltaic synchronous machine stack device and real-time meteorological environment data into the photovoltaic power output prediction model, output the photovoltaic power output prediction result, store the photovoltaic power output prediction result data set. Step S103: Obtain real-time grid operation data and real-time grid demand data; retrieve the time information of the real-time grid operation data to obtain the time information of the data to be retrieved; retrieve the time information of the data to be retrieved from the photovoltaic power output prediction result data set to obtain the prediction result corresponding to the time information of the data to be retrieved; substitute the prediction result corresponding to the time information of the data to be retrieved, the real-time grid demand data, and the real-time grid operation data into the preset grid load prediction model to output the real-time grid load prediction result. Step S104: Obtain the real-time operating data of the photovoltaic synchronous machine stack device, compare the real-time grid load forecast result with the real-time grid demand data to obtain the real-time grid load forecast error value, and substitute the real-time grid load forecast error value, the real-time grid operating data and the real-time operating data of the photovoltaic synchronous machine stack device into the output power control model of the photovoltaic synchronous machine stack device to output the output power of the photovoltaic synchronous machine stack device. Step S105: After the photovoltaic synchronous machine stack device outputs power to the grid, the grid operation data is monitored in real time to obtain the real-time grid operation data. The real-time grid operation data is compared with the preset standard value. If the real-time grid operation data is higher than the preset standard value, the output power control model of the photovoltaic synchronous machine stack device is optimized using a preset genetic algorithm to obtain the optimized output power of the photovoltaic synchronous machine stack device. The optimized output power of the photovoltaic synchronous machine stack device is then transmitted to the grid.

2. The control method for improving grid stability using a photovoltaic synchronous accelerator stack as described in claim 1, characterized in that, Step S101 is further used for: The system acquires operational data from the photovoltaic synchronous machine (PVSM) unit, the power grid, meteorological data, and fault data. This data is then preprocessed to obtain preprocessed PVSM unit operational data, power grid operational data, meteorological data, and fault data. Integrity checks are performed on the preprocessed PVSM unit operational data, power grid operational data, meteorological data, and fault data. Data that passes the integrity check is stored in a database. Data that fails the integrity check is matched against a pre-defined data simulation model knowledge base to obtain the corresponding data simulation model. The data simulation model is then used to simulate the data that fails the integrity check, resulting in simulated data corresponding to the data that fails the integrity check. This simulated data is then stored in the database. Finally, the simulated data corresponding to the data that fails the integrity check is used to replace the data that fails the integrity check in the database.

3. The control method for improving grid stability using a photovoltaic synchronous accelerator stack as described in claim 1, characterized in that, Step S102 is further used for: Historical operating data and meteorological data of the photovoltaic synchronous machine stack are retrieved from the database. Data features are extracted using the historical operating data and meteorological data to obtain the data features to be processed. The data features to be processed are matched in a preset algorithm knowledge base to obtain the algorithm model corresponding to the data features to be processed. The algorithm model corresponding to the data features to be processed is trained using the historical operating data and meteorological data of the photovoltaic synchronous machine stack to obtain the photovoltaic power output prediction model.

4. The control method for improving grid stability using a photovoltaic synchronous accelerator stack as described in claim 1, characterized in that, Step S103 is further used for: The real-time power grid operation data and the data in the photovoltaic output prediction result dataset are aligned based on time to obtain aligned data. The aligned data is then inspected to obtain the alignment data inspection results, which include successfully aligned data and unsuccessfully aligned data. The time information of the unsuccessfully aligned data is collected. Receive the time information for retrieving real-time power grid operation data, match the time information for retrieving real-time power grid operation data with the time information for data that failed to align, and obtain the time information matching result, which includes data that successfully matched the time and data that failed to match the time. Retrieve the time information corresponding to the time information of the unsuccessful time matching data, and re-predict the photovoltaic output through step S102 to obtain the photovoltaic output prediction result corresponding to the time information of the unsuccessful time matching data.

5. The control method for improving grid stability using a photovoltaic synchronous accelerator stack as described in claim 1, characterized in that, Step S104 is further used for: The operation data of the photovoltaic synchronous machine stack is acquired in real time through monitoring systems or sensors. The operation data of the photovoltaic synchronous machine stack includes the output power, voltage, current, temperature and fault status of the synchronous machine stack. The real-time power grid load forecast result is obtained from step S103, and the real-time power grid demand data is obtained from the power grid dispatch system. The real-time power grid demand data includes the total load demand of the power grid and the load allocation of each region. By comparing the real-time power grid load forecast results with the real-time power grid demand data and calculating the error value, the degree of deviation between the real-time power grid load forecast results and the real-time power grid demand data is obtained. The calculated real-time power grid load prediction error value is stored, and an error value monitoring mechanism is established. When the error value exceeds the preset threshold, an alarm is triggered.

6. The control method for improving grid stability using a photovoltaic synchronous accelerator stack as described in claim 1, characterized in that, Step S105 is further used for: The system acquires real-time power grid operation data from the power grid monitoring system, preprocesses the real-time power grid operation data, and compares the preprocessed real-time power grid operation data with preset standard values ​​one by one. The data to be compared includes the voltage fluctuation range, current over-limit situation, frequency stability, and power factor qualification rate. Based on the comparison results, the error value is calculated to obtain the difference between the real-time power grid operation data and the preset standard value, and the result of the error analysis is obtained. Based on the results of error analysis, it is determined whether the power grid's operating status is within the normal range. If the real-time data exceeds the preset standard value, an alarm mechanism is triggered.

7. The control method for improving grid stability using a photovoltaic synchronous accelerator stack as described in claim 6, characterized in that, Step S105 is further used for: The target of the optimization and the population size are received. The optimization target is to minimize the deviation between the power grid operation data and the preset standard value. Based on the population size, a set of control model parameter combinations is randomly generated as the initial population, and each control parameter combination of the photovoltaic synchronous machine stack device represents a solution. Calculate the deviation between the power grid operation data and the preset standard value under each parameter combination, and evaluate the merits and demerits of each parameter combination in the initial population; Based on the evaluation results, a subset of control parameter combinations for photovoltaic synchronous machine stack devices were selected as parent generation for crossover and mutation operations. By cross-operation, the control parameters of the photovoltaic synchronous machine stack device in the parent parameter combination are exchanged to generate a new control parameter combination of the child photovoltaic synchronous machine stack device. By mutation operation, the parameters in the control parameter combination of the child photovoltaic synchronous machine stack device are randomly adjusted to increase the diversity of the control parameters of the photovoltaic synchronous machine stack device. The deviation between the grid operation data and the preset standard value under each parameter combination is calculated. The quality of the control combination of the newly generated generation of photovoltaic synchronous machine stack devices is evaluated. Full cross-operation and mutation operations are performed until the set number of generations is reached or the deviation between the grid operation data and the preset standard value under each parameter combination reaches the threshold. The parameter combination with the smallest deviation between the grid operation data and the preset standard value under each parameter combination is selected from the final population as the optimal solution, that is, the optimized output power control model parameters of the photovoltaic synchronous machine stack device.

Images