Temporary and steady-state voltage stability control method of photovoltaic synchronous generator-pile power generation system

Abstract

The invention relates to the technical field of operation and maintenance of photovoltaic power generation systems, and provides a transient and steady state voltage stability control method for a photovoltaic synchronous machine pile power generation system, which comprises the following steps: training a random forest model by acquiring historical data, establishing a voltage stability prediction model, and collecting data of a photovoltaic power station, a power grid, a load, a synchronous machine and a power system in real time. And comparing the data with expected data output by the prediction model, screening and correcting data exceeding the error range to enable the data to accord with the voltage stability error range, and finally substituting the corrected data into the voltage stability control model, and outputting control parameters to adjust the operation state of the system and keep the voltage stable. According to the invention, the problem of voltage stability caused by data acquisition and transmission errors is effectively solved, and the operation stability of the photovoltaic synchronous generator pile power generation system is remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic power generation system operation and maintenance, and in particular to a method for controlling transient steady-state voltage stability of a photovoltaic synchronous machine stack power generation system. Background Art

[0002] The photovoltaic synchronous machine stack power generation system is a new type of power generation system. By introducing the control strategy of the synchronous machine, the stable output and grid-connected operation of the photovoltaic power generation system are achieved. In the photovoltaic synchronous machine stack power generation system, the photovoltaic cell assembly converts solar energy into DC power, which is then converted into AC power by the inverter and supplied to the grid or load. At the same time, the system uses the control strategy of the synchronous machine to accurately adjust the output of the photovoltaic power generation system so that it is consistent with the frequency, phase and voltage of the grid, thereby achieving stable operation of the photovoltaic power generation system.

[0003] The transient voltage stability control data processing method of the photovoltaic synchronous machine stack power generation system mainly involves real-time monitoring and adjustment of the system voltage. Specifically, by collecting data from photovoltaic power stations, power grids and loads, and combining the control method of the photovoltaic synchronous machine stack power generation system, the system voltage is dynamically adjusted. In the data processing process, algorithms and analysis models, such as voltage stability margin calculation methods, are used to evaluate the voltage stability of the system in real time. When the voltage fluctuates or becomes unstable, the system immediately starts the control strategy, such as adjusting the output power of the photovoltaic power generation system, changing the grid connection method, etc., to keep the system voltage within a stable range.

[0004] The technical pain points of the transient voltage stability control data processing method of the photovoltaic synchronous machine stack power generation system mainly lie in that, since the photovoltaic synchronous machine stack power generation system involves many devices and is widely distributed, there are errors in the data collection and transmission process. The errors in the data collection and transmission process will lead to voltage stability and affect the transient voltage stability control of the photovoltaic synchronous machine stack power generation system. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides a transient voltage stability control method and an extraction method for a photovoltaic synchronous machine stack power generation system, so as to solve the problem that the photovoltaic synchronous machine stack power generation system involves many devices and is widely distributed, and errors exist in the data collection and transmission process, and the errors in the data collection and transmission process will lead to voltage stability.

[0006] In order to solve the above technical problems, the specific technical solutions of the present invention are as follows:

[0007] The transient voltage stability control method of the photovoltaic synchronous machine stack power generation system provided by the present invention comprises:

[0008] Step S101, obtaining historical photovoltaic power station data, historical power grid data, historical load data, historical synchronous machine data, historical power system operation data and historical weather forecast data, using the historical photovoltaic power station data, historical power grid data, historical load data, historical synchronous machine data, historical power system operation data and historical weather forecast data to train a random forest model to obtain a voltage stability prediction model, receiving a voltage stability control task and a voltage stability error range value, the voltage stability error range value including a photovoltaic power station error range value, a real-time power grid error range value, a real-time load error range value, a real-time synchronous machine error range value and a real-time power system operation error range value, substituting the voltage stability control task into the voltage stability prediction model, the voltage stability prediction model outputs the expected data of the photovoltaic synchronous machine stack power generation system corresponding to the voltage stability control task, and the expected data of the photovoltaic synchronous machine stack power generation system corresponding to the voltage stability control task is used as the standard expected data;

[0009] Step S102, collecting real-time photovoltaic power station data, real-time power grid data, real-time load data, real-time synchronous machine data and real-time power system operation data, comparing the real-time photovoltaic power station data, real-time power grid data, real-time load data, real-time synchronous machine data and real-time power system operation data with standard expected data, and obtaining photovoltaic power station error data, real-time power grid error data, real-time load error data, real-time synchronous machine error data and real-time power system operation error data;

[0010] Step S103, comparing the photovoltaic power station error data with the photovoltaic power station error range value, comparing the real-time grid error data with the real-time grid error range value, comparing the real-time load data with the real-time load error range value, comparing the real-time synchronous machine data with the real-time synchronous machine error range value, and comparing the real-time power system operation data with the real-time power system operation error range value, to obtain an error range comparison result;

[0011] Step S104, filtering out data beyond the error range in the error range comparison result to obtain data beyond the error range, extracting data features of the data beyond the error range to obtain data features of the error range, obtaining error data attribute information according to the error range data features, retrieving historical data corresponding to the error data attribute information to obtain historical data corresponding to the error data, using the historical data corresponding to the error data to train a random forest model to obtain a data simulation model;

[0012] Step S105, substitute the error data attribute information into the data simulation model to generate error simulation data, replace the real-time photovoltaic power station data, real-time power grid data, real-time load data, real-time synchronous machine data and real-time power system operation data with the error simulation data according to the error data attribute information, and obtain the corrected real-time photovoltaic power station data, the corrected real-time power grid data, the corrected real-time load data, the corrected real-time synchronous machine data and the corrected real-time power system operation data, and compare the corrected real-time photovoltaic power station data, the corrected real-time power grid data, the corrected real-time load data, the corrected real-time synchronous machine data and the corrected real-time power system operation data with the standard expected data. If the comparison result meets the voltage stability error range value, the corrected real-time power grid data, the corrected real-time load data, the corrected real-time synchronous machine data and the corrected real-time power system operation data are used as the photovoltaic synchronous machine stack power generation system to-be-processed data, and the photovoltaic synchronous machine stack power generation system to-be-processed data are substituted into the preset voltage stability control model to output the voltage stability control parameters.

[0013] Furthermore, the transient voltage stability control method of the photovoltaic synchronous machine stack power generation system of the present invention, the step S101 comprises:

[0014] Obtain historical photovoltaic power station data, historical power grid data, historical load data, historical synchronous machine data, historical power system operation data and historical weather forecast data to train the random forest model and establish a voltage stability prediction model;

[0015] Receiving a voltage stability control task, the voltage stability control task includes predicting the voltage stability of a photovoltaic synchronous machine pile power generation system;

[0016] Receive a voltage stability error range value, where the voltage stability error range value includes a photovoltaic power station error range value, a real-time power grid error range value, a real-time load error range value, a real-time synchronous machine error range value, and a real-time power system operation error range value;

[0017] Substituting the voltage stability control task into the voltage stability prediction model, the voltage stability prediction model performs prediction calculation according to the input voltage stability control task;

[0018] After the voltage stability prediction model is calculated, the expected data of the photovoltaic synchronous machine stack power generation system corresponding to the output voltage stability control task is output. The expected data of the photovoltaic synchronous machine stack power generation system corresponding to the voltage stability control task includes the expected power generation of the photovoltaic power station, the expected grid voltage, the expected grid frequency, the expected grid load and the expected synchronous machine operating status.

[0019] Furthermore, the transient voltage stability control method of the photovoltaic synchronous machine stack power generation system of the present invention, the step S102 comprises:

[0020] Through the sensor equipment arranged in the photovoltaic power station, power grid, load point, synchronous machine and power system, the photovoltaic power station data, real-time power grid data, real-time load data, real-time synchronous machine data and real-time power system operation data are collected in real time. The real-time photovoltaic power station data includes power generation, voltage and current, the real-time power grid data includes power grid voltage, frequency and phase, the real-time load data includes load power and current, the real-time synchronous machine data includes speed, power output and excitation current, and the real-time power system operation data includes the operation status of the power station and the transmission capacity of the transmission line;

[0021] The real-time collected photovoltaic power station data, real-time grid data, real-time load data, real-time synchronous machine data and real-time power system operation data are aligned with the standard expected data according to the timestamp, so that the real-time collected photovoltaic power station data, real-time grid data, real-time load data, real-time synchronous machine data and real-time power system operation data are synchronized with the standard expected data;

[0022] Compare the real-time grid data with the expected grid data in the standard expected data, calculate the errors of grid voltage, frequency and other parameters, and obtain the real-time grid error data;

[0023] Compare the real-time load data with the expected load data in the standard expected data, calculate the errors of load power, current and other parameters, and obtain the real-time load error data;

[0024] Compare the real-time synchronous machine data with the expected data of the synchronous machine in the standard expected data, calculate the errors of parameters such as the rotation speed and the power output, and obtain the real-time synchronous machine error data;

[0025] The real-time power system operation data is compared with the expected power system operation data in the standard expected data, and the errors of parameters such as the operating status of other power stations and the transmission capacity of transmission lines are calculated to obtain the real-time power system operation error data.

[0026] Furthermore, the method for controlling transient voltage stability of a photovoltaic synchronous machine stack power generation system according to the present invention, step S103 comprises:

[0027] The photovoltaic power station error data calculated in step S102 is compared with the photovoltaic power station error range value to determine whether the photovoltaic power station error data falls within the error range value, that is, whether the error is within an acceptable range, and the real-time power grid error data is compared with the real-time power grid error range value;

[0028] Compare the real-time load error data with the real-time load error range value to check whether the errors of load power, current and other parameters are within the given error range; compare the real-time synchronous machine error data with the real-time synchronous machine error range value to verify whether the errors of synchronous machine speed, power output and other parameters meet the error range;

[0029] Compare the real-time power system operation error data with the real-time power system operation error range value to evaluate whether the real-time power system operation error data is within an acceptable range;

[0030] If the error data is within the error range, it is recorded as "qualified" or "meets requirements";

[0031] If the error data exceeds the error range, it will be recorded as "unqualified" or "out of range" and the exceeded value or proportion will be noted.

[0032] Furthermore, the method for controlling transient voltage stability of a photovoltaic synchronous machine stack power generation system according to the present invention, step S104 comprises:

[0033] Extract features of the out-of-error range data screened out in step S103, the features including the size of the error, the frequency of the error, the duration of the error, and the environmental conditions of the error;

[0034] According to the characteristics of the error range data, the attribute information of the error data is determined, and the attribute information of the error data includes the error type including high voltage and low current, the error occurrence time including early morning and evening, and the error environment conditions including sunny and cloudy days;

[0035] According to the determined error data attribute information, a matching data record is searched in a historical database, and a data segment matching the current error data attribute information is screened out from the historical data searched in the database;

[0036] The selected historical data are sorted out, and the error data is used to train the random forest model. The sorted historical data are divided into a training set and a test set. The training set is used to train the random forest model, and the test set is used to evaluate the performance of the model.

[0037] Determine input features and output targets. Input features include time, environmental conditions, and equipment status in historical data, and output targets are error values ​​or error types.

[0038] The random forest model is trained using the training set data. During the training process, a decision tree is built based on these features. Each decision tree divides the data set and gives the final prediction value.

[0039] Furthermore, the transient voltage stability control method of the photovoltaic synchronous machine stack power generation system of the present invention, the step S105 comprises:

[0040] Compare the corrected real-time PV power station data with the expected PV power station data in the standard expected data;

[0041] comparing the corrected real-time power grid data with the expected power grid data in the standard expected data;

[0042] Compare the corrected real-time load data with the load expectation data in the standard expectation data;

[0043] comparing the corrected real-time synchronous machine data with the synchronous machine expected data in the standard expected data;

[0044] comparing the corrected real-time power system operation data with the power system operation expected data in the standard expected data;

[0045] For each comparison, the error between the corrected real-time data and the standard expected data is calculated;

[0046] If the comparison result between the corrected real-time data and the standard expected data meets the voltage stability error range value, that is, the error is within an acceptable range, the next step is carried out;

[0047] If the comparison result does not meet the voltage stability error range value, that is, the error is beyond the acceptable range, it is necessary to reanalyze the data and adjust the data simulation model;

[0048] If the comparison result is consistent with the voltage stability error range value, the corrected real-time grid data, the corrected real-time load data, the corrected real-time synchronous machine data and the corrected real-time power system operation data are determined as the photovoltaic synchronous machine stack power generation system data to be processed.

[0049] Furthermore, the transient voltage stability control method of the photovoltaic synchronous machine stack power generation system of the present invention, the step S105 comprises:

[0050] Substituting the data to be processed of the photovoltaic synchronous machine stack power generation system into a preset voltage stability control model, the voltage stability control model outputs corresponding voltage stability control parameters based on the input data;

[0051] The control parameters output by the voltage stability control model are used to adjust the operating state of the photovoltaic synchronous machine stack power generation system.

[0052] By adjusting the system's output power and grid-connected mode, the system voltage can be kept within a stable range.

[0053] Beneficial effects of the present invention;

[0054] By collecting the operating data of photovoltaic power stations, power grids, loads, synchronous machines and power systems in real time and comparing them with standard expected data, the present invention can timely discover and process data errors, thereby ensuring the voltage stability of the system. By establishing a voltage stability prediction model, the present invention can predict the voltage stability of the photovoltaic synchronous machine stack power generation system in advance, providing a reliable basis for taking control measures in a timely manner, and effectively avoiding the occurrence of voltage fluctuations or instability.

[0055] In view of the fact that the photovoltaic synchronous machine pile power generation system has numerous and widely distributed equipment, the present invention significantly improves the accuracy of data collection and transmission through a series of steps such as data preprocessing, error comparison, and data correction. The data is trained using a random forest model, which further improves the accuracy and robustness of the prediction model and provides a strong guarantee for the accurate collection and transmission of data.

[0056] The present invention generates error simulation data through a data simulation model and corrects the real-time data, thereby obtaining more accurate data as the data to be processed. The data to be processed is substituted into a preset voltage stability control model to output more accurate control parameters for adjusting the operating state of the photovoltaic synchronous machine stack power generation system, thereby optimizing the control strategy and improving the operating efficiency and stability of the system.

[0057] By real-time monitoring and adjusting the voltage stability of the photovoltaic synchronous machine pile power generation system, the present invention can effectively avoid system failures or accidents caused by voltage fluctuations or instability, significantly improving the reliability of the system. At the same time, the present invention also provides the ability to deeply mine and analyze historical data, which helps to discover potential problems and risks in the system and provides strong support for the long-term stable operation of the system.

[0058] The present invention reduces the need for manual intervention through automated data collection, processing and control processes, thereby reducing operation and maintenance costs. By predicting and promptly addressing voltage stability issues in advance, the present invention effectively avoids additional costs such as downtime and maintenance due to system failures or accidents, further reducing operation and maintenance costs. In summary, the present invention has shown significant beneficial effects in many aspects, providing strong support for the stable operation and efficient management of photovoltaic synchronous machine stack power generation systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the technical solution of the present invention, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, for ordinary technicians in this field, other drawings can be obtained based on the drawings without paying any creative labor.

[0060] Figure 1A schematic flow chart of a transient voltage stability control method for a photovoltaic synchronous machine stack power generation system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described in conjunction with the specific embodiments of the present invention and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. The technical solutions provided by the embodiments of the present invention are described in detail below in conjunction with the drawings.

[0062] In order to better understand the purpose of the present invention, the present invention is described in further detail below.

[0063] The transient voltage stability control method of the photovoltaic synchronous machine stack power generation system provided by the present invention comprises:

[0064] Step S101, obtaining historical photovoltaic power station data, historical power grid data, historical load data, historical synchronous machine data, historical power system operation data and historical weather forecast data, using the historical photovoltaic power station data, historical power grid data, historical load data, historical synchronous machine data, historical power system operation data and historical weather forecast data to train a random forest model to obtain a voltage stability prediction model, receiving a voltage stability control task and a voltage stability error range value, the voltage stability error range value including a photovoltaic power station error range value, a real-time power grid error range value, a real-time load error range value, a real-time synchronous machine error range value and a real-time power system operation error range value, substituting the voltage stability control task into the voltage stability prediction model, the voltage stability prediction model outputs the expected data of the photovoltaic synchronous machine stack power generation system corresponding to the voltage stability control task, and the expected data of the photovoltaic synchronous machine stack power generation system corresponding to the voltage stability control task is used as the standard expected data;

[0065] Obtain historical photovoltaic power station data, historical power grid data, historical load data, historical synchronous machine data, historical power system operation data, and historical weather forecast data. These data are the basis for training the random forest model and are used to establish a voltage stability prediction model in the future.

[0066] The random forest model is trained using the collected historical data. Random forest is an integrated learning method that improves the accuracy of prediction by building multiple decision trees and integrating their output results. After the training is completed, a voltage stability prediction model is obtained, which can predict the voltage stability of the photovoltaic synchronous machine stack power generation system based on the input data.

[0067] Receive voltage stability control tasks, clarify control objectives and requirements, and receive voltage stability error range values, including photovoltaic power station error range values, real-time grid error range values, real-time load error range values, real-time synchronous machine error range values, and real-time power system operation error range values. These error range values ​​are used to subsequently evaluate the accuracy of the prediction results.

[0068] Substitute the voltage stability control task into the trained voltage stability prediction model. The model outputs the corresponding expected data of the photovoltaic synchronous machine stack power generation system according to the input control task. The expected data of the photovoltaic synchronous machine stack power generation system corresponding to the voltage stability control task is used as the standard expected data. These data will be used for subsequent comparison with the real-time collected data to evaluate the voltage stability of the system and guide the formulation of the control strategy.

[0069] In summary, step S101 provides a basis for transient voltage stability control of the photovoltaic synchronous machine stack power generation system by collecting historical data, training the random forest model, receiving control tasks and error range values, and substituting the control tasks into the model to obtain standard expected data.

[0070] Step S102, collecting real-time photovoltaic power station data, real-time power grid data, real-time load data, real-time synchronous machine data and real-time power system operation data, comparing the real-time photovoltaic power station data, real-time power grid data, real-time load data, real-time synchronous machine data and real-time power system operation data with standard expected data, and obtaining photovoltaic power station error data, real-time power grid error data, real-time load error data, real-time synchronous machine error data and real-time power system operation error data;

[0071] Through sensors or monitoring equipment, real-time data of photovoltaic power station operation, real-time status data of power grid, real-time demand data of load, real-time operation data of synchronous machine and overall operation data of power system are collected. These data reflect the current actual situation of the system and are the basis for subsequent analysis and control.

[0072] The collected real-time data is compared with the standard expected data obtained in step S101. The standard expected data is obtained based on historical data and random forest model prediction, and represents the operation of the system under ideal conditions. The purpose of the comparison is to find the difference between the real-time data and the expected data, that is, the error.

[0073] By comparison, the photovoltaic power station error data, real-time grid error data, real-time load error data, real-time synchronous machine error data and real-time power system operation error data are calculated. These error data reflect the degree of deviation between the actual operation of the system and the expected operation.

[0074] Step S102 can timely discover abnormal conditions or deviations in system operation through real-time data collection and comparison with standard expected data, providing a basis for subsequent error processing and control strategy formulation. At the same time, the calculation of error data also provides quantitative indicators for evaluating the voltage stability of the system, which helps to more accurately judge the operating status of the system and take corresponding control measures.

[0075] Step S103, comparing the photovoltaic power station error data with the photovoltaic power station error range value, comparing the real-time grid error data with the real-time grid error range value, comparing the real-time load data with the real-time load error range value, comparing the real-time synchronous machine data with the real-time synchronous machine error range value, and comparing the real-time power system operation data with the real-time power system operation error range value, to obtain an error range comparison result;

[0076] Step S103 is a method for controlling the transient voltage stability of a photovoltaic synchronous machine stack power generation system, which is intended to evaluate whether the difference between the real-time collected data and the expected data is within an acceptable error range. The specific operations are as follows:

[0077] Compare the real-time collected photovoltaic power station error data with the pre-set photovoltaic power station error range value. This includes comparing whether the error values ​​of key parameters such as voltage, current, and power exceed the set threshold.

[0078] Compare the real-time grid error data with the real-time grid error range value. This involves error analysis of grid voltage, frequency, phase and other parameters to determine whether the grid status is stable and in line with expectations.

[0079] Compare the real-time load data with the real-time load error range value. This is mainly to evaluate whether the fluctuation of load demand is within the normal range and whether it has an adverse impact on the voltage stability of the system.

[0080] Compare the real-time synchronous machine operation data with the real-time synchronous machine error range value. This includes error analysis of key indicators such as the synchronous machine's speed, power output, and voltage regulation capability to ensure that the synchronous machine can stably support the grid voltage.

[0081] Compare the real-time power system operation data with the real-time power system operation error range value. This is a comprehensive assessment of the operation status of the entire power system, including error analysis in multiple aspects such as power balance, voltage distribution, and current flow.

[0082] Step S104, filtering out data beyond the error range in the error range comparison result to obtain data beyond the error range, extracting data features of the data beyond the error range to obtain data features of the error range, obtaining error data attribute information according to the error range data features, retrieving historical data corresponding to the error data attribute information to obtain historical data corresponding to the error data, using the historical data corresponding to the error data to train a random forest model to obtain a data simulation model;

[0083] Step S104 is a method for controlling transient voltage stability of a photovoltaic synchronous machine stack power generation system, which is intended to process data beyond the error range and optimize and update the model through these data. The following is a detailed explanation of this step:

[0084] In the error range comparison results, identify and filter out the data that exceeds the preset error range. These data may indicate that the system is abnormal or unstable and require special attention.

[0085] Feature extraction is performed on the filtered out-of-error range data. This includes analyzing the statistical characteristics of the data (such as mean, variance, maximum value, minimum value, etc.), time characteristics (such as trend, periodicity, etc.), and correlation with other variables. Through feature extraction, the key features describing these out-of-error range data, namely the error range data features, can be obtained.

[0086] Based on the extracted error range data features, the attribute information of these data is further determined. For example, they may belong to a specific time period, weather conditions, equipment status or load level, etc. The attribute information helps to understand the reasons for the generation of out-of-error range data and provides valuable reference for subsequent data simulation and model training.

[0087] According to the determined error data attribute information, the corresponding historical data is retrieved from the historical database. These data should have similar attribute characteristics and be collected under similar conditions. The retrieved historical data is used to train the random forest model to obtain a data simulation model. This model can simulate the behavior or state that may occur in the system under similar conditions. By training the data simulation model, the generation mechanism of out-of-error range data can be better understood, and stronger support can be provided for subsequent prediction and control.

[0088] Step S105, substitute the error data attribute information into the data simulation model to generate error simulation data, replace the real-time photovoltaic power station data, real-time power grid data, real-time load data, real-time synchronous machine data and real-time power system operation data with the error simulation data according to the error data attribute information, and obtain the corrected real-time photovoltaic power station data, the corrected real-time power grid data, the corrected real-time load data, the corrected real-time synchronous machine data and the corrected real-time power system operation data, and compare the corrected real-time photovoltaic power station data, the corrected real-time power grid data, the corrected real-time load data, the corrected real-time synchronous machine data and the corrected real-time power system operation data with the standard expected data. If the comparison result meets the voltage stability error range value, the corrected real-time power grid data, the corrected real-time load data, the corrected real-time synchronous machine data and the corrected real-time power system operation data are used as the photovoltaic synchronous machine stack power generation system to-be-processed data, and the photovoltaic synchronous machine stack power generation system to-be-processed data are substituted into the preset voltage stability control model to output the voltage stability control parameters.

[0089] Step S105 is to use the data simulation model in the transient voltage stability control method of the photovoltaic synchronous machine pile power generation system to correct the real-time data and ensure that the data meets the expected error range in terms of voltage stability. The following is a detailed description of this step:

[0090] Substitute the error data attribute information determined in the previous step into the data simulation model. This model is trained based on historical data and can simulate the errors that may occur in the system under similar conditions. By running the model, error simulation data with the same attribute characteristics as the real-time data is generated. These data represent the errors that may occur in the system under different conditions.

[0091] Using the generated error simulation data, the real-time photovoltaic power station data, real-time power grid data, real-time load data, real-time synchronous machine data and real-time power system operation data are replaced according to the error data attribute information. The purpose of the replacement is to replace the error part in the real-time data with the simulated error data, so as to obtain the corrected real-time data that is closer to the actual situation.

[0092] The corrected real-time photovoltaic power station data, corrected real-time grid data, corrected real-time load data, corrected real-time synchronous machine data and corrected real-time power system operation data are compared with the standard expected data. The purpose of the comparison is to check whether the corrected data is within the error range of voltage stability, that is, whether it meets the stability requirements of the system.

[0093] If the comparison results show that the corrected data conform to the voltage stability error range, then these data are considered valid and used as the pending data for the photovoltaic synchronous machine stack power generation system. These pending data are substituted into the preset voltage stability control model, and the model will calculate the voltage stability control parameters based on these data. The control parameters are then used to adjust the operating state of the photovoltaic synchronous machine stack power generation system to ensure the voltage stability of the system.

[0094] The completion of step S105 marks the end of the correction and processing of real-time data, and also provides an accurate data basis for subsequent voltage stability control. By continuously correcting and processing real-time data, the photovoltaic synchronous machine stack power generation system can maintain a stable operating state under any conditions.

[0095] Specifically, the method for controlling transient voltage stability of a photovoltaic synchronous machine stack power generation system according to the present invention, step S101 comprises:

[0096] Obtain historical photovoltaic power station data, historical power grid data, historical load data, historical synchronous machine data, historical power system operation data and historical weather forecast data to train the random forest model and establish a voltage stability prediction model;

[0097] Receiving a voltage stability control task, the voltage stability control task includes predicting the voltage stability of a photovoltaic synchronous machine pile power generation system;

[0098] Receive a voltage stability error range value, where the voltage stability error range value includes a photovoltaic power station error range value, a real-time power grid error range value, a real-time load error range value, a real-time synchronous machine error range value, and a real-time power system operation error range value;

[0099] Substituting the voltage stability control task into the voltage stability prediction model, the voltage stability prediction model performs prediction calculation according to the input voltage stability control task;

[0100] After the voltage stability prediction model is calculated, the expected data of the photovoltaic synchronous machine stack power generation system corresponding to the output voltage stability control task is output. The expected data of the photovoltaic synchronous machine stack power generation system corresponding to the voltage stability control task includes the expected power generation of the photovoltaic power station, the expected grid voltage, the expected grid frequency, the expected grid load and the expected synchronous machine operating status.

[0101] Obtain historical operating data of photovoltaic power plants, power grids, loads, synchronous machines, and power systems from historical databases, and collect corresponding historical weather forecast data, because these weather factors (such as light intensity, temperature, wind speed, etc.) may affect the power generation of photovoltaic power plants and the operating status of power systems. Use these historical data to train the random forest model. Random forest is a powerful ensemble learning algorithm that can handle high-dimensional data, has good robustness to outliers and noise, and can provide variable importance assessment.

[0102] After the training is completed, a voltage stability prediction model is obtained. The model can predict the voltage stability of the photovoltaic synchronous machine pile power generation system based on the input feature data (such as historical photovoltaic power station data, historical power grid data, etc.).

[0103] Receive voltage stability control tasks from the system or user. This task usually includes predicting the voltage stability of the photovoltaic synchronous machine stack power generation system in the future, so as to take timely measures to maintain the stable operation of the system. At the same time, receive voltage stability error range values, which define the acceptable deviation range between the predicted results and the actual results. Specifically, it includes the prediction error range of photovoltaic power station power generation, the prediction error range of grid voltage, the prediction error range of grid frequency, the prediction error range of grid load, and the prediction error range of synchronous machine operation status.

[0104] Substitute the voltage stability control task into the trained voltage stability prediction model. The model will perform prediction calculations based on the input task requirements (such as predicting voltage stability at a certain point in the future) and the provided feature data (such as historical data).

[0105] After the model calculation is completed, the expected data of the photovoltaic synchronous machine stack power generation system corresponding to the voltage stability control task is output. These data include the expected photovoltaic power station power generation, the expected grid voltage value, the expected grid frequency value, the expected grid load value, and the expected synchronous machine operating status (such as power output, speed, etc.). These expected data will be used as a benchmark for comparison with actual data in subsequent steps to evaluate the voltage stability of the system and guide the formulation of control strategies.

[0106] Through the detailed implementation of step S101, the present invention provides a solid data foundation and prediction model support for the transient steady-state voltage stability control of the photovoltaic synchronous machine stack power generation system.

[0107] Specifically, the method for controlling transient voltage stability of a photovoltaic synchronous machine stack power generation system according to the present invention, step S102 comprises:

[0108] Through the sensor equipment arranged in the photovoltaic power station, power grid, load point, synchronous machine and power system, the photovoltaic power station data, real-time power grid data, real-time load data, real-time synchronous machine data and real-time power system operation data are collected in real time. The real-time photovoltaic power station data includes power generation, voltage and current, the real-time power grid data includes power grid voltage, frequency and phase, the real-time load data includes load power and current, the real-time synchronous machine data includes speed, power output and excitation current, and the real-time power system operation data includes the operation status of the power station and the transmission capacity of the transmission line;

[0109] The real-time collected photovoltaic power station data, real-time grid data, real-time load data, real-time synchronous machine data and real-time power system operation data are aligned with the standard expected data according to the timestamp, so that the real-time collected photovoltaic power station data, real-time grid data, real-time load data, real-time synchronous machine data and real-time power system operation data are synchronized with the standard expected data;

[0110] Compare the real-time grid data with the expected grid data in the standard expected data, calculate the errors of grid voltage, frequency and other parameters, and obtain the real-time grid error data;

[0111] Compare the real-time load data with the expected load data in the standard expected data, calculate the errors of load power, current and other parameters, and obtain the real-time load error data;

[0112] Compare the real-time synchronous machine data with the expected data of the synchronous machine in the standard expected data, calculate the errors of parameters such as the rotation speed and the power output, and obtain the real-time synchronous machine error data;

[0113] The real-time power system operation data is compared with the expected power system operation data in the standard expected data, and the errors of parameters such as the operating status of other power stations and the transmission capacity of transmission lines are calculated to obtain the real-time power system operation error data.

[0114] Step S102 is the transient voltage stability control method of the photovoltaic synchronous machine stack power generation system of the present invention, which involves the collection and alignment of real-time data and the comparison with standard expected data, thereby obtaining various error data. The following is a detailed description of this step:

[0115] Using sensor equipment distributed in photovoltaic power plants, power grids, load points, synchronous machines and key locations of the power system, various types of data are collected comprehensively and in real time. Real-time photovoltaic power plant data covers key indicators such as power generation, voltage and current, which directly reflect the real-time power generation status of photovoltaic power plants. Real-time power grid data includes grid voltage, frequency and phase, which together depict the real-time operating status of the power grid. Real-time load data includes load power and current, reflecting the current load demand. Real-time synchronous machine data records the speed, power output and excitation current of the synchronous machine, which are crucial for evaluating the operating status of the synchronous machine. Real-time power system operation data integrates the operating status of the power station and the transmission capacity of the transmission line, providing a basis for a comprehensive understanding of the operating status of the power system.

[0116] In order to accurately compare the real-time data with the standard expected data, the two types of data need to be aligned according to the timestamp. This step ensures the temporal consistency of the real-time data and the expected data, laying a solid foundation for the subsequent error calculation.

[0117] The real-time grid data is compared with the expected grid data in the standard expected data one by one, and the deviation of the grid voltage, frequency and other parameters is calculated to obtain the real-time grid error data. Similarly, the real-time load data is compared with the expected load data, and the errors of load power, current and other parameters are calculated to obtain the real-time load error data. For the real-time synchronous machine data, it is also necessary to compare it with the expected synchronous machine data, and the real-time synchronous machine error data is obtained by calculating the errors of parameters such as speed and power output.

[0118] Finally, the real-time power system operation data is compared with the expected power system operation data, and the real-time power system operation error data is obtained by calculating the errors of parameters such as the operating status of other power stations and the transmission capacity of transmission lines.

[0119] Through the careful implementation of step S102, the present invention can grasp the differences between various operating parameters of the photovoltaic synchronous machine stack power generation system and expected values ​​in real time and accurately, providing strong support for error analysis, model correction and formulation of control strategies in subsequent steps.

[0120] Specifically, the method for controlling transient voltage stability of a photovoltaic synchronous machine stack power generation system according to the present invention, step S103 includes:

[0121] The photovoltaic power station error data calculated in step S102 is compared with the photovoltaic power station error range value to determine whether the photovoltaic power station error data falls within the error range value, that is, whether the error is within an acceptable range, and the real-time power grid error data is compared with the real-time power grid error range value;

[0122] Compare the real-time load error data with the real-time load error range value to check whether the errors of load power, current and other parameters are within the given error range; compare the real-time synchronous machine error data with the real-time synchronous machine error range value to verify whether the errors of synchronous machine speed, power output and other parameters meet the error range;

[0123] Compare the real-time power system operation error data with the real-time power system operation error range value to evaluate whether the real-time power system operation error data is within an acceptable range;

[0124] If the error data is within the error range, it is recorded as "qualified" or "meets requirements";

[0125] If the error data exceeds the error range, it will be recorded as "unqualified" or "out of range" and the exceeded value or proportion will be noted.

[0126] Step S103 is the core evaluation link in the transient voltage stability control method of the photovoltaic synchronous machine stack power generation system of the present invention. By comparing the actual error data with the preset error range value, it is determined whether the various parameters of the system are within the acceptable range. The following is a detailed explanation of this step:

[0127] The photovoltaic power station error data calculated in step S102 is compared with the preset photovoltaic power station error range value one by one to evaluate whether the errors of parameters such as power generation, voltage, current, etc. are within an acceptable range, that is, whether they meet the preset error range value.

[0128] If all parameter errors are within the range, the PV power plant data is recorded as "qualified" or "meets requirements".

[0129] If any parameter error exceeds the range, it will be recorded as "unqualified" or "out of range", and the exceeded value or proportion will be specifically noted.

[0130] Compare the real-time grid error data with the real-time grid error range value. Evaluate whether the errors of grid voltage, frequency, phase and other parameters are within the preset range. Based on the evaluation results, record the grid data as "qualified" or "unqualified" and mark the details of the out-of-range.

[0131] Compare the real-time load error data with the real-time load error range value. Check whether the error of load power, current and other parameters meets the requirements. Record the evaluation results of the load data and mark any parameters that are out of range.

[0132] Compare the real-time synchronous machine error data with the real-time synchronous machine error range value. Verify whether the errors of the synchronous machine's speed, power output, excitation current and other parameters are within the acceptable range. Based on the verification results, record the evaluation status of the synchronous machine data.

[0133] Compare the real-time power system operation error data with the real-time power system operation error range value. Evaluate whether the errors of parameters such as power station operation status and transmission line transmission capacity meet system requirements. Record the evaluation results of power system operation data and mark any parameters that do not meet the requirements.

[0134] For all parameters evaluated as "unqualified" or "out of range", the specific values ​​or proportions that exceed the range should be recorded in detail for subsequent analysis and improvement. The evaluation results can be used as a basis for system optimization, model adjustment or control strategy formulation to help improve the transient voltage stability of the photovoltaic synchronous machine pile power generation system.

[0135] Through the careful implementation of step S103, the present invention can timely discover and record the deviations between various system parameters and expected values, and provide powerful data support for subsequent control and optimization.

[0136] Specifically, the method for controlling transient voltage stability of a photovoltaic synchronous machine stack power generation system according to the present invention, step S104 comprises:

[0137] Extract features of the out-of-error range data screened out in step S103, the features including the size of the error, the frequency of the error, the duration of the error, and the environmental conditions of the error;

[0138] According to the characteristics of the error range data, the attribute information of the error data is determined, and the attribute information of the error data includes the error type including high voltage and low current, the error occurrence time including early morning and evening, and the error environment conditions including sunny and cloudy days;

[0139] According to the determined error data attribute information, a matching data record is searched in a historical database, and a data segment matching the current error data attribute information is screened out from the historical data searched in the database;

[0140] The selected historical data are sorted out, and the error data is used to train the random forest model. The sorted historical data are divided into a training set and a test set. The training set is used to train the random forest model, and the test set is used to evaluate the performance of the model.

[0141] Determine input features and output targets. Input features include time, environmental conditions, and equipment status in historical data, and output targets are error values ​​or error types.

[0142] The random forest model is trained using the training set data. During the training process, a decision tree is built based on these features. Each decision tree divides the data set and gives the final prediction value.

[0143] Step S104 is the in-depth analysis of the out-of-error range data, feature extraction, historical data matching and random forest model retraining in the transient steady-state voltage stability control method of the photovoltaic synchronous machine stack power generation system of the present invention. The following is a detailed description of this step:

[0144] The out-of-range data selected in step S103 are analyzed in depth to extract key features, including but not limited to the size of the error (i.e., the specific value that exceeds the range), the frequency of the error (i.e., the number of times the error exceeds the range per unit time), the duration of the error (i.e., the duration of the out-of-range state), and the error environment conditions (such as sunny, cloudy, temperature, humidity, etc.).

[0145] Based on the extracted features, the attribute information of the error data is further determined, including the error type (such as high voltage, low current, etc.), the time when the error occurs (such as specific time periods such as early morning and evening), and the error environment conditions (such as sunny and cloudy weather conditions).

[0146] By using the determined error data attribute information, the matching data records are searched in the historical database. By comparing the attribute information of the historical data and the current error data, the historical data segments similar to the current error situation are screened out.

[0147] The selected historical data are sorted to ensure the accuracy and completeness of the data. Subsequently, the data are divided into training sets and test sets. The training set is used to train the random forest model, while the test set is used to evaluate the performance of the model to ensure the effectiveness and reliability of the model.

[0148] Before training the random forest model, it is necessary to clarify the input features and output targets of the model. Input features usually include time, environmental conditions (such as weather, temperature, etc.), equipment status (such as synchronous machine speed, power output, etc.) in historical data. These features can reflect the operating status of the system and the impact of the external environment on the system. The output target is the error value or error type, that is, what the model needs to predict.

[0149] The random forest model is trained using the training set data. During the training process, the model builds multiple decision trees based on the input features. Each decision tree partitions the data set, makes decisions based on the importance of the features, and ultimately gives a prediction. By integrating the predictions of multiple decision trees, the random forest model can provide more accurate and stable predictions.

[0150] Through the implementation of step S104, the present invention can deeply analyze the characteristics of data beyond the error range, explore its relationship with historical data, and improve the ability to predict errors by training a random forest model, which helps to timely discover and solve voltage stability problems in photovoltaic synchronous machine stack power generation systems and improve system stability.

[0151] Specifically, the method for controlling transient voltage stability of a photovoltaic synchronous machine stack power generation system according to the present invention, step S105 comprises:

[0152] Compare the corrected real-time PV power station data with the expected PV power station data in the standard expected data;

[0153] comparing the corrected real-time power grid data with the expected power grid data in the standard expected data;

[0154] Compare the corrected real-time load data with the load expectation data in the standard expectation data;

[0155] comparing the corrected real-time synchronous machine data with the synchronous machine expected data in the standard expected data;

[0156] comparing the corrected real-time power system operation data with the power system operation expected data in the standard expected data;

[0157] For each comparison, the error between the corrected real-time data and the standard expected data is calculated;

[0158] If the comparison result between the corrected real-time data and the standard expected data meets the voltage stability error range value, that is, the error is within an acceptable range, the next step is carried out;

[0159] If the comparison result does not meet the voltage stability error range value, that is, the error is beyond the acceptable range, it is necessary to reanalyze the data and adjust the data simulation model;

[0160] If the comparison result is consistent with the voltage stability error range value, the corrected real-time grid data, the corrected real-time load data, the corrected real-time synchronous machine data and the corrected real-time power system operation data are determined as the photovoltaic synchronous machine stack power generation system data to be processed.

[0161] Step S105 is a key verification step in the transient voltage stability control method of the photovoltaic synchronous machine stack power generation system of the present invention, which aims to ensure that the corrected real-time data can be consistent with the standard expected data and meet the voltage stability requirements. The following is a detailed explanation of this step:

[0162] Compare the corrected real-time PV power station data with the expected data of the PV power station in the standard expected data to check whether the key parameters such as power generation, voltage, current, etc. are consistent. Compare the corrected real-time grid data, real-time load data, real-time synchronous machine data and real-time power system operation data with their respective standard expected data.

[0163] For each comparison, the error between the corrected real-time data and the standard expected data is calculated. This includes calculating the error values ​​of parameters such as voltage, frequency, phase, load power, current, synchronous machine speed, power output, and power system operating status.

[0164] Determine whether the calculated error is consistent with the voltage stability error range. If the error is within the acceptable range, it means that the corrected real-time data is basically consistent with the standard expected data and meets the voltage stability requirements.

[0165] If the error exceeds the acceptable range, it is necessary to reanalyze the data, check whether there are errors or omissions in the data correction process, and adjust the data simulation model accordingly to ensure the accuracy and reliability of the data.

[0166] If the comparison result meets the voltage stability error range, the corrected real-time grid data, real-time load data, real-time synchronous machine data and real-time power system operation data will be determined as the data to be processed by the photovoltaic synchronous machine stack power generation system. These data will serve as the basis for the formulation and implementation of subsequent control strategies.

[0167] Through the careful implementation of step S105, the present invention can ensure the consistency between the corrected real-time data and the standard expected data, and provide an accurate and reliable data basis for the transient voltage stability control of the photovoltaic synchronous machine stack power generation system. At the same time, through the error range judgment and data determination links, the present invention can timely discover and correct problems in the data correction process, and improve the accuracy of the control method.

[0168] Specifically, the method for controlling transient voltage stability of a photovoltaic synchronous machine stack power generation system according to the present invention, step S105 comprises:

[0169] Substituting the data to be processed of the photovoltaic synchronous machine stack power generation system into a preset voltage stability control model, the voltage stability control model outputs corresponding voltage stability control parameters based on the input data;

[0170] The control parameters output by the voltage stability control model are used to adjust the operating state of the photovoltaic synchronous machine stack power generation system. By adjusting the system output power and grid connection mode, the system voltage is kept within a stable range.

[0171] Step S105 in the transient voltage stability control method of the photovoltaic synchronous machine stack power generation system of the present invention involves applying the verified data to be processed to the actual voltage stability control process to ensure the stable operation of the system. The following is a detailed description of this step:

[0172] The data to be processed for the photovoltaic synchronous machine stack power generation system determined in the previous step (including corrected real-time grid data, real-time load data, real-time synchronous machine data and real-time power system operation data) are substituted into the preset voltage stability control model.

[0173] The voltage stability control model is established based on the in-depth understanding and analysis of the photovoltaic synchronous machine pile power generation system. After receiving the input data, it will calculate the corresponding voltage stability control parameters based on these data. These parameters may include but are not limited to the output power adjustment of the system, the switching instructions of the grid connection mode, the input or removal of the reactive compensation device, etc.

[0174] According to the control parameters output by the voltage stability control model, the operating state of the photovoltaic synchronous machine stack power generation system is adjusted in real time. This may involve adjusting the system's output power to match the current load demand; or changing the grid connection method to optimize the system's voltage stability; or inputting or removing reactive compensation devices to maintain the system's voltage level within a stable range.

[0175] After adjusting the system operation, it is necessary to continuously monitor the system voltage stability and respond promptly to any possible voltage fluctuations or instability. If it is found that there are still problems with voltage stability or the system operation status does not meet expectations, it is necessary to perform data analysis, model adjustment and control parameter optimization again to ensure the continued stable operation of the system.

[0176] Through the implementation of step S105, the present invention can convert the verified data to be processed into actual voltage stability control actions, thereby effectively maintaining the transient voltage stability of the photovoltaic synchronous machine stack power generation system, which not only improves the stability of the system, but also provides strong support for the optimized operation and efficient management of the system.

[0177] The transient voltage stability control method of the photovoltaic synchronous machine stack power generation system provided by the present invention solves the problem of voltage stability caused by errors in data collection and transmission due to the fact that the photovoltaic synchronous machine stack power generation system involves many devices and is widely distributed, through the following technical solutions:

[0178] Obtain historical photovoltaic power station data, historical power grid data, historical load data, historical synchronous machine data, historical power system operation data, and historical weather forecast data, use these data to train the random forest model, and establish a voltage stability prediction model. Receive voltage stability control tasks and voltage stability error range values, substitute the control tasks into the prediction model, and obtain expected data as standard expected data.

[0179] The sensors are used to collect real-time operating data of photovoltaic power plants, power grids, loads, synchronous machines, and power systems. The real-time data is compared with the standard expected data, and various error data are calculated. The error data is compared with the corresponding error range values, and the data exceeding the error range is screened out. The features of the data exceeding the error range are extracted, and the random forest model is trained using the corresponding historical data to obtain the data simulation model.

[0180] Use the data simulation model to generate error simulation data, replace the data beyond the error range in the real-time data, and obtain the corrected real-time data. Compare the corrected real-time data with the standard expected data. If it meets the error range value, it is used as the data to be processed. Substitute the data to be processed into the preset voltage stability control model, and output the voltage stability control parameters, which are used to adjust the operating state of the photovoltaic synchronous machine stack power generation system to ensure voltage stability.

[0181] The present invention effectively solves the error problem existing in the data collection and transmission process, and improves the transient steady-state voltage stability of the photovoltaic synchronous machine stack power generation system.

Claims

1. A method for controlling transient voltage stability of a photovoltaic synchronous machine stack power generation system, characterized in that: include: Step S101, obtaining historical photovoltaic power station data, historical power grid data, historical load data, historical synchronous machine data, historical power system operation data and historical weather forecast data, using the historical photovoltaic power station data, historical power grid data, historical load data, historical synchronous machine data, historical power system operation data and historical weather forecast data to train a random forest model to obtain a voltage stability prediction model, receiving a voltage stability control task and a voltage stability error range value, the voltage stability error range value including a photovoltaic power station error range value, a real-time power grid error range value, a real-time load error range value, a real-time synchronous machine error range value and a real-time power system operation error range value, substituting the voltage stability control task into the voltage stability prediction model, the voltage stability prediction model outputs the expected data of the photovoltaic synchronous machine stack power generation system corresponding to the voltage stability control task, and the expected data of the photovoltaic synchronous machine stack power generation system corresponding to the voltage stability control task is used as the standard expected data; Step S102, collecting real-time photovoltaic power station data, real-time power grid data, real-time load data, real-time synchronous machine data and real-time power system operation data, comparing the real-time photovoltaic power station data, real-time power grid data, real-time load data, real-time synchronous machine data and real-time power system operation data with standard expected data, and obtaining photovoltaic power station error data, real-time power grid error data, real-time load error data, real-time synchronous machine error data and real-time power system operation error data; Step S103, comparing the photovoltaic power station error data with the photovoltaic power station error range value, comparing the real-time grid error data with the real-time grid error range value, comparing the real-time load data with the real-time load error range value, comparing the real-time synchronous machine data with the real-time synchronous machine error range value, and comparing the real-time power system operation data with the real-time power system operation error range value, to obtain an error range comparison result; Step S104, filtering out data beyond the error range in the error range comparison result to obtain data beyond the error range, extracting data features of the data beyond the error range to obtain data features of the error range, obtaining error data attribute information according to the error range data features, retrieving historical data corresponding to the error data attribute information to obtain historical data corresponding to the error data, using the historical data corresponding to the error data to train a random forest model to obtain a data simulation model; Step S105, substitute the error data attribute information into the data simulation model to generate error simulation data, replace the real-time photovoltaic power station data, real-time power grid data, real-time load data, real-time synchronous machine data and real-time power system operation data with the error simulation data according to the error data attribute information, and obtain the corrected real-time photovoltaic power station data, the corrected real-time power grid data, the corrected real-time load data, the corrected real-time synchronous machine data and the corrected real-time power system operation data, and compare the corrected real-time photovoltaic power station data, the corrected real-time power grid data, the corrected real-time load data, the corrected real-time synchronous machine data and the corrected real-time power system operation data with the standard expected data. If the comparison result meets the voltage stability error range value, the corrected real-time power grid data, the corrected real-time load data, the corrected real-time synchronous machine data and the corrected real-time power system operation data are used as the photovoltaic synchronous machine stack power generation system to-be-processed data, and the photovoltaic synchronous machine stack power generation system to-be-processed data are substituted into the preset voltage stability control model to output the voltage stability control parameters.

2. The transient voltage stability control method of the photovoltaic synchronous machine stack power generation system according to claim 1, characterized in that: The step S101 includes: Obtain historical photovoltaic power station data, historical power grid data, historical load data, historical synchronous machine data, historical power system operation data and historical weather forecast data to train the random forest model and establish a voltage stability prediction model; Receiving a voltage stability control task, the voltage stability control task includes predicting the voltage stability of a photovoltaic synchronous machine pile power generation system; Receive a voltage stability error range value, where the voltage stability error range value includes a photovoltaic power station error range value, a real-time power grid error range value, a real-time load error range value, a real-time synchronous machine error range value, and a real-time power system operation error range value; Substituting the voltage stability control task into the voltage stability prediction model, the voltage stability prediction model performs prediction calculation according to the input voltage stability control task; After the voltage stability prediction model is calculated, the expected data of the photovoltaic synchronous machine stack power generation system corresponding to the output voltage stability control task is output. The expected data of the photovoltaic synchronous machine stack power generation system corresponding to the voltage stability control task includes the expected power generation of the photovoltaic power station, the expected grid voltage, the expected grid frequency, the expected grid load and the expected synchronous machine operating status.

3. The transient voltage stability control method of the photovoltaic synchronous machine stack power generation system according to claim 1, characterized in that: The step S102 includes: Through the sensor equipment arranged in the photovoltaic power station, power grid, load point, synchronous machine and power system, the photovoltaic power station data, real-time power grid data, real-time load data, real-time synchronous machine data and real-time power system operation data are collected in real time. The real-time photovoltaic power station data includes power generation, voltage and current, the real-time power grid data includes power grid voltage, frequency and phase, the real-time load data includes load power and current, the real-time synchronous machine data includes speed, power output and excitation current, and the real-time power system operation data includes the operation status of the power station and the transmission capacity of the transmission line; The real-time collected photovoltaic power station data, real-time grid data, real-time load data, real-time synchronous machine data and real-time power system operation data are aligned with the standard expected data according to the timestamp, so that the real-time collected photovoltaic power station data, real-time grid data, real-time load data, real-time synchronous machine data and real-time power system operation data are synchronized with the standard expected data; Compare the real-time grid data with the expected grid data in the standard expected data, calculate the errors of grid voltage and frequency parameters, and obtain real-time grid error data; Compare the real-time load data with the expected load data in the standard expected data, calculate the errors of the load power and current parameters, and obtain the real-time load error data; Compare the real-time synchronous machine data with the expected data of the synchronous machine in the standard expected data, calculate the errors of the speed and power output parameters, and obtain the real-time synchronous machine error data; The real-time power system operation data is compared with the expected power system operation data in the standard expected data, and the errors of the operating status of other power stations and the transmission capacity parameters of the transmission lines are calculated to obtain the real-time power system operation error data.

4. The transient voltage stability control method of the photovoltaic synchronous machine stack power generation system according to claim 1, characterized in that: The step S103 includes: The photovoltaic power station error data calculated in step S102 is compared with the photovoltaic power station error range value to determine whether the photovoltaic power station error data falls within the error range value, that is, whether the error is within an acceptable range, and the real-time power grid error data is compared with the real-time power grid error range value; Compare the real-time load error data with the real-time load error range value to check whether the errors of the load power and current parameters are within the given error range; compare the real-time synchronous machine error data with the real-time synchronous machine error range value to verify whether the errors of the synchronous machine speed and power output parameters meet the error range; Compare the real-time power system operation error data with the real-time power system operation error range value to evaluate whether the real-time power system operation error data is within an acceptable range; If the error data is within the error range, it is recorded as qualified or meets the requirements; If the error data exceeds the error range, it will be recorded as unqualified or out of range, and the exceeded value or proportion will be noted.

5. The transient voltage stability control method of the photovoltaic synchronous machine stack power generation system according to claim 1, characterized in that: The step S104 includes: Extract features of the out-of-error range data screened out in step S103, the features including the size of the error, the frequency of the error, the duration of the error, and the environmental conditions of the error; According to the characteristics of the error range data, the attribute information of the error data is determined, and the attribute information of the error data includes the error type including high voltage and low current, the error occurrence time including early morning and evening, and the error environment conditions including sunny and cloudy days; According to the determined error data attribute information, a matching data record is searched in a historical database, and a data segment matching the current error data attribute information is screened out from the historical data searched in the database; The selected historical data are sorted out, and the error data is used to train the random forest model. The sorted historical data are divided into a training set and a test set. The training set is used to train the random forest model, and the test set is used to evaluate the performance of the model. Determine input features and output targets. Input features include time, environmental conditions, and equipment status in historical data, and output targets are error values ​​or error types. The random forest model is trained using the training set data. During the training process, a decision tree is built based on these features. Each decision tree divides the data set and gives the final prediction value.

6. The transient voltage stability control method of the photovoltaic synchronous machine stack power generation system according to claim 1, characterized in that: The step S105 includes: Compare the corrected real-time PV power station data with the expected PV power station data in the standard expected data; comparing the corrected real-time power grid data with the expected power grid data in the standard expected data; Compare the corrected real-time load data with the load expectation data in the standard expectation data; comparing the corrected real-time synchronous machine data with the synchronous machine expected data in the standard expected data; comparing the corrected real-time power system operation data with the power system operation expected data in the standard expected data; For each comparison, the error between the corrected real-time data and the standard expected data is calculated; If the comparison result between the corrected real-time data and the standard expected data meets the voltage stability error range value, that is, the error is within an acceptable range, the next step is carried out; If the comparison result does not meet the voltage stability error range value, that is, the error is beyond the acceptable range, it is necessary to reanalyze the data and adjust the data simulation model; If the comparison result is consistent with the voltage stability error range value, the corrected real-time grid data, the corrected real-time load data, the corrected real-time synchronous machine data and the corrected real-time power system operation data are determined as the photovoltaic synchronous machine stack power generation system data to be processed.

7. The transient voltage stability control method of the photovoltaic synchronous machine stack power generation system according to claim 1, characterized in that: The step S105 includes: Substituting the data to be processed of the photovoltaic synchronous machine stack power generation system into a preset voltage stability control model, the voltage stability control model outputs corresponding voltage stability control parameters based on the input data; The control parameters output by the voltage stability control model are used to adjust the operating state of the photovoltaic synchronous machine stack power generation system.

Images