Grid-connected control method for 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 grid-connected control method for a photovoltaic synchronous generator pile power generation system, which comprises the following steps: acquiring photovoltaic system data and photovoltaic power generation task data, training a random forest model to obtain a grid-connected model, acquiring photovoltaic array operation data and natural environment data, and predicting power generation conditions. Real-time photovoltaic array operation data and prediction data are compared, early warning is generated or grid-connected control parameters are output, the grid-connected control parameters are compared with system real-time parameters, errors are calculated, a grid-connected model is optimized through a genetic algorithm, and finally the system executes the optimized real-time parameters to achieve synchronization with a power grid. And the stability and the grid-connected efficiency of the photovoltaic power generation system are improved. According to the invention, the problem that the voltage, frequency and phase are not matched in the grid connection process is effectively solved, and intelligent operation and maintenance of photovoltaic power generation are promoted.

Description

Technical Field

[0001] The present invention relates to the field of photovoltaic power generation system operation and maintenance technology, and in particular to a photovoltaic synchronous machine stack power generation system grid-connected control method. Background Art

[0002] The photovoltaic synchronous machine stack, that is, the photovoltaic energy storage virtual synchronous machine technology, combines photovoltaic power generation with the energy storage system. By introducing the concept of a virtual synchronous generator, it effectively improves the stability of the photovoltaic power generation system. By adjusting the damping coefficient of the energy storage system to suppress the low-frequency oscillation phenomenon in photovoltaic power generation, the photovoltaic synchronous machine stack power generation system can maintain stable operation under different working conditions.

[0003] The grid connection of a synchronous machine stack power generation system usually refers to the connection between a power generation system that uses a synchronous generator as a power generation equipment and a large power grid. The grid connection process requires that the output voltage, frequency and phase of the generator match the voltage, frequency and phase of the grid. During the grid connection process of an existing photovoltaic synchronous machine stack power generation system, the output voltage of the generator is unstable, and the output voltage, frequency and phase of the generator do not match the voltage, frequency and phase of the grid. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a photovoltaic synchronous machine stack power generation system grid-connected control method and an extraction method to solve the problem that during the grid-connected process of the existing photovoltaic synchronous machine stack power generation system, the output voltage of the generator is unstable and the output voltage, frequency and phase of the generator do not match the voltage, frequency and phase of the grid.

[0005] In order to solve the above technical problems, the specific technical solutions of the present invention are as follows:

[0006] The photovoltaic synchronous machine stack power generation system grid-connected control method provided by the present invention comprises:

[0007] Step S101, obtaining photovoltaic system data and photovoltaic power generation task data, the photovoltaic system data including control commands, photovoltaic component data, inverter data, meteorological data, power generation data and grid-connected control data, preprocessing the photovoltaic system data and photovoltaic power generation task data to obtain preprocessed photovoltaic system data and photovoltaic power generation task data, training the preprocessed photovoltaic system data and photovoltaic power generation task data on a random forest model to obtain a photovoltaic synchronous machine stack power generation system grid-connected model;

[0008] Step S102, collecting photovoltaic array operation data and natural environment data, substituting the collected photovoltaic array operation data and natural environment data into a preset photovoltaic array power generation prediction model to obtain photovoltaic array power generation prediction data;

[0009] Step S103, obtaining real-time photovoltaic array operation data, substituting the real-time photovoltaic array operation data into the photovoltaic array power generation prediction data for comparison, and generating photovoltaic array power generation warning information if the real-time photovoltaic array operation data is higher than the photovoltaic array power generation prediction data;

[0010] Step S104, if the real-time photovoltaic array operation data is lower than the photovoltaic array power generation prediction data, the real-time photovoltaic array operation data is substituted into the photovoltaic synchronous machine stack power generation system grid-connected model, and the photovoltaic synchronous machine stack power generation system grid-connected control parameters are output;

[0011] Step S105, compare the grid-connected control parameters of the photovoltaic synchronous machine stack power generation system with the real-time parameters of the photovoltaic synchronous machine stack power generation system to obtain parameter comparison errors, the parameter comparison errors include output voltage errors, frequency errors and phase adjustment errors, obtain historical data of photovoltaic array operation data, use the historical data of photovoltaic array operation data and parameter comparison errors to optimize the grid-connected model of the photovoltaic synchronous machine stack power generation system using a genetic algorithm, use the optimized grid-connected model of the photovoltaic synchronous machine stack power generation system to process the real-time photovoltaic array operation data, output the optimized real-time parameters of the photovoltaic synchronous machine stack power generation system, the photovoltaic synchronous machine stack power generation system executes the optimized real-time parameters of the photovoltaic synchronous machine stack power generation system to achieve synchronization with the power grid, convert the electric energy generated by the photovoltaic synchronous machine stack power generation system into alternating current that meets the power grid standards, and achieve connection with the power grid.

[0012] Furthermore, the photovoltaic synchronous machine stack power generation system grid-connected control method of the present invention, step S101 comprises:

[0013] Volt component data including power, efficiency, and temperature;

[0014] Inverter data includes inverter conversion efficiency and inverter working status;

[0015] Meteorological data include solar irradiance, temperature, and wind speed;

[0016] Grid-connected control data include grid-connected voltage, grid-connected current and grid-connected frequency;

[0017] Photovoltaic power generation mission data includes expected power generation, power generation period, and power regulation requirements;

[0018] The preprocessed photovoltaic system data and photovoltaic power generation task data are used as input to train the random forest model. After the training is completed, the random forest model becomes a grid-connected model of the photovoltaic synchronous machine stack power generation system, which is used to receive new photovoltaic system data and photovoltaic power generation task data and output grid-connected control parameters.

[0019] Furthermore, the photovoltaic synchronous machine stack power generation system grid-connected control method of the present invention, the step S102 comprises:

[0020] The operation data of the photovoltaic array includes the voltage of the photovoltaic module, the current of the photovoltaic module, the power of the photovoltaic module, and the real-time parameters of the temperature of the photovoltaic module;

[0021] The preprocessed photovoltaic array operation data and natural environment data are used as input and substituted into the preset photovoltaic array power generation prediction model. The preset photovoltaic array power generation prediction model is started, the input data is processed, and the power generation prediction data of the photovoltaic array is calculated through the preset photovoltaic array power generation prediction model. The power generation prediction data of the photovoltaic array includes predicted power generation, predicted power generation efficiency and predicted power output.

[0022] Furthermore, the photovoltaic synchronous machine stack power generation system grid-connected control method of the present invention, step S103 includes:

[0023] Obtain the prediction data of the corresponding time period from the photovoltaic array power generation prediction model, and align the real-time photovoltaic array operation data with the photovoltaic array power generation prediction data in time, that is, the real-time photovoltaic array operation data and the photovoltaic array power generation prediction data correspond to the data of the same time point or time period;

[0024] Compare the real-time PV array operation data with the PV array power generation prediction data item by item to check whether the parameters of the real-time PV array are higher than the predicted values;

[0025] When the power output of the real-time photovoltaic array is higher than the predicted power by, for example, 10% for 5 consecutive minutes, an early warning is triggered, which includes the time, parameters and level of the triggering warning.

[0026] Furthermore, the photovoltaic synchronous machine stack power generation system grid-connected control method of the present invention, the step S104 comprises:

[0027] Determine whether the real-time photovoltaic array operation data is lower than the photovoltaic array power generation prediction data, compare the real-time photovoltaic array operation data with the photovoltaic array power generation prediction data in terms of voltage, current and power; if the real-time photovoltaic array operation data is lower than the prediction data, use the voltage, current and power of the real-time photovoltaic array operation data as input information;

[0028] Determine the photovoltaic synchronous machine stack power generation system grid-connected model to be used. The photovoltaic synchronous machine stack power generation system grid-connected model is established based on historical data, system characteristics and control objectives. The photovoltaic synchronous machine stack power generation system grid-connected model is used to output grid-connected control parameters;

[0029] Substitute the real-time photovoltaic array operation data into the photovoltaic synchronous machine pile power generation system grid-connected model, and correspond the voltage, current and timestamp of the data to the input variables of the model;

[0030] The photovoltaic synchronous machine stack power generation system grid-connected model is started, and the photovoltaic synchronous machine stack power generation system grid-connected model processes the input real-time photovoltaic array operation data, and the photovoltaic synchronous machine stack power generation system grid-connected model calculates the grid-connected control parameters;

[0031] The output grid-connected control parameters are obtained from the grid-connected model of the photovoltaic synchronous machine stack power generation system. The grid-connected control parameters include output voltage, frequency and phase adjustment. The grid-connected control parameters are used to adjust the output of the photovoltaic synchronous machine stack power generation system so that the photovoltaic synchronous machine stack power generation system is synchronized with the power grid.

[0032] Furthermore, the photovoltaic synchronous machine stack power generation system grid-connected control method of the present invention, step S105 includes:

[0033] The photovoltaic synchronous machine stack power generation system grid-connected control parameters output by step S104 are obtained. The photovoltaic synchronous machine stack power generation system grid-connected control parameters are calculated based on the photovoltaic synchronous machine stack power generation system grid-connected model, and are intended to synchronize the system output with the power grid.

[0034] Real-time acquisition of actual operating parameters of the photovoltaic synchronous machine stack power generation system, including the current output voltage, frequency and phase;

[0035] The grid-connected control parameters and the real-time system parameters are aligned in time, that is, the grid control parameters and the real-time parameters of the photovoltaic synchronous machine stack power generation system correspond to the same moment;

[0036] Compare the target output voltage, target frequency and target phase in the grid-connected control parameters with the actual output voltage, actual frequency and actual phase in the real-time parameters of the photovoltaic synchronous machine stack power generation system item by item, and calculate the error value of each parameter, that is, the difference between the real-time value of the photovoltaic synchronous machine stack power generation system and the target value in the grid-connected control parameters;

[0037] For the output voltage, the difference between the actual output voltage and the target output voltage is calculated to obtain the output voltage error;

[0038] For frequency, the difference between the actual frequency and the target frequency is calculated to obtain the frequency error;

[0039] For the phase, the difference between the actual phase and the target phase is calculated to obtain the phase adjustment error.

[0040] Furthermore, the photovoltaic synchronous machine stack power generation system grid-connected control method of the present invention, step S105 includes:

[0041] According to the grid connection requirements and control objectives of the photovoltaic synchronous machine stack power generation system, the optimization objectives are determined, and the optimization objectives include minimizing the weighted sum of the output voltage error, frequency error and phase adjustment error and the error threshold;

[0042] Based on the optimization objective, a fitness function is constructed, which is used to evaluate the grid-connected model under given parameters;

[0043] Determine the parameters of the genetic algorithm, such as population size, number of iterations, crossover probability, and mutation probability;

[0044] Initialize the population, that is, generate a set of random grid-connected model parameters as the initial solution.

[0045] Use the fitness function to evaluate the fitness value of each individual in the current population, substitute the individual's parameters into the grid-connected model, calculate the error between the model output and the actual data, and calculate the fitness value based on the error;

[0046] Select excellent individuals as parents according to their fitness values, and perform crossover operations to generate new offspring;

[0047] Perform mutation operations on offspring individuals, that is, randomly change the values ​​of some of their genes, and repeat the selection and crossover mutation process until the predetermined number of iterations is reached;

[0048] After the iteration, the individual with the lowest fitness value is selected from the population as the optimal solution, that is, the optimized grid-connected model parameters;

[0049] The optimized grid-connected model parameters are output and used in the subsequent grid-connected control of the photovoltaic synchronous machine stack power generation system.

[0050] Beneficial effects of the present invention;

[0051] The present invention monitors and adjusts the output voltage, frequency and phase of the photovoltaic synchronous machine stack power generation system in real time. The parameters of the present invention are highly matched with the grid parameters, thereby significantly improving the grid-connected stability of the photovoltaic power generation system. The random forest model and genetic algorithm are used to optimize the grid-connected control parameters. The present invention can reduce the output voltage error, frequency error and phase adjustment error, so that the power quality output by the photovoltaic power generation system is more in line with the grid standards.

[0052] The present invention can collect photovoltaic array operation data and natural environment data in real time, and make predictions through a preset photovoltaic array power generation prediction model, so that the system can adapt to different lighting conditions and environmental changes, and enhance the adaptability and robustness of the system.

[0053] When the real-time photovoltaic array operation data is higher than the predicted data, the system will generate early warning information to remind the operation and maintenance personnel to deal with abnormal situations in time to prevent the system from being overloaded or damaged, thereby achieving effective protection for the photovoltaic power generation system. By accurately controlling the output parameters of the photovoltaic synchronous machine stack power generation system, the present invention enables the photovoltaic power generation system to synchronize with the power grid more efficiently, reduce energy losses during the grid connection process, and improve the overall grid connection efficiency of the system.

[0054] The present invention combines multiple technical means such as data collection, model prediction, real-time monitoring and optimization algorithm, provides strong support for the intelligent operation and maintenance of photovoltaic power generation systems, reduces operation and maintenance costs, and improves operation and maintenance efficiency.

[0055] In summary, the present invention effectively solves the problems existing in the grid-connected process of photovoltaic synchronous machine stack power generation system through a series of innovative technical means, improves the stability, adaptability and grid-connected efficiency of photovoltaic power generation system, and provides a strong guarantee for the sustainable development of photovoltaic power generation industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] 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.

[0057] Figure 1 A schematic flow chart of a method for controlling a photovoltaic synchronous machine stack power generation system connected to the grid provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0058] 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.

[0059] In order to better understand the purpose of the present invention, the present invention is described in further detail below.

[0060] The photovoltaic synchronous machine stack power generation system grid-connected control method provided by the present invention comprises:

[0061] Step S101, obtaining photovoltaic system data and photovoltaic power generation task data, the photovoltaic system data including control commands, photovoltaic component data, inverter data, meteorological data, power generation data and grid-connected control data, preprocessing the photovoltaic system data and photovoltaic power generation task data to obtain preprocessed photovoltaic system data and photovoltaic power generation task data, training the preprocessed photovoltaic system data and photovoltaic power generation task data on a random forest model to obtain a photovoltaic synchronous machine stack power generation system grid-connected model;

[0062] In step S101, the present invention constructs a photovoltaic synchronous machine stack power generation system grid-connected model through the following specific steps to solve the problem that the output voltage, frequency and phase of the generator do not match the power grid during the photovoltaic power generation system grid-connected process:

[0063] Acquire PV system data, including control commands, PV module data (such as power, efficiency, temperature), inverter data (such as inverter conversion efficiency, inverter working status), meteorological data (such as solar irradiance, temperature, wind speed), power generation data, and grid-connected control data (such as grid-connected voltage, grid-connected current, grid-connected frequency).

[0064] Obtain PV power generation mission data, including expected power generation, power generation period, and power regulation requirements.

[0065] The collected photovoltaic system data and photovoltaic power generation task data are preprocessed to remove noise, fill missing values, normalize or standardize the data, and improve the data quality.

[0066] The random forest model is trained using the preprocessed PV system data and PV power generation task data as input. The random forest model is an integrated learning method that improves the accuracy and stability of the model by building multiple decision trees and combining their prediction results. After training, the random forest model becomes a grid-connected model for the PV synchronous machine stack power generation system, which can receive new PV system data and PV power generation task data and output the corresponding grid-connected control parameters.

[0067] Through step S101, the present invention constructs a photovoltaic synchronous machine pile power generation system grid-connected model based on random forest. The model can comprehensively consider various photovoltaic system data and power generation task requirements, output grid-connected control parameters that meet the requirements of the power grid, and thus provide an accurate basis for subsequent photovoltaic power generation system grid-connected control. This not only solves the problem that the output voltage, frequency and phase of the generator do not match the power grid, but also improves the stability of the photovoltaic power generation system.

[0068] Step S102, collecting photovoltaic array operation data and natural environment data, substituting the collected photovoltaic array operation data and natural environment data into a preset photovoltaic array power generation prediction model to obtain photovoltaic array power generation prediction data;

[0069] Collect the operation data of the photovoltaic array, which includes real-time parameters such as the voltage, current, power and temperature of the photovoltaic module. These data reflect the current working status and power generation capacity of the photovoltaic array. At the same time, collect natural environment data such as sunlight irradiance, temperature, wind speed, etc. These data have a direct impact on the power generation efficiency of the photovoltaic array, because the power generation capacity of the photovoltaic module is affected by environmental factors such as light intensity and temperature.

[0070] The collected photovoltaic array operation data and natural environment data are substituted into the preset photovoltaic array power generation prediction model. This prediction model is established based on historical data, system characteristics, environmental factors and other factors, and can accurately predict the power generation of the photovoltaic array. After the prediction model is started, the input data is processed and analyzed to calculate the power generation prediction data of the photovoltaic array. These data include predicted power generation, power generation efficiency and power output, etc., which provide important reference for subsequent grid-connected control.

[0071] Through step S102, the present invention can obtain the power generation prediction data of the photovoltaic array in time, providing an important reference for the subsequent grid-connected control. It helps the system to understand the power generation capacity of the photovoltaic array in advance, so as to better adjust the grid-connected control parameters and ensure that the photovoltaic power generation system is stable and efficient in synchronization with the power grid. At the same time, it also provides strong support for dealing with emergencies (such as sudden changes in the power generation capacity of the photovoltaic array due to sudden weather changes), so that the system can respond quickly and ensure the smooth progress of the grid-connected process.

[0072] Step S103, obtaining real-time photovoltaic array operation data, substituting the real-time photovoltaic array operation data into the photovoltaic array power generation prediction data for comparison, and generating photovoltaic array power generation warning information if the real-time photovoltaic array operation data is higher than the photovoltaic array power generation prediction data;

[0073] The system collects the operating data of the photovoltaic array in real time. These data include but are not limited to the voltage, current, power output and component temperature of the photovoltaic module, reflecting the actual working status and power generation capacity of the photovoltaic array.

[0074] The collected real-time PV array operation data is compared with the PV array power generation prediction data obtained previously through the preset PV array power generation prediction model. This comparison process is aimed at evaluating the difference between the actual power generation and the expected power generation.

[0075] If the real-time PV array operation data is higher than the PV array power generation forecast data, the system determines that the current PV array power generation capacity exceeds the expected range. In this case, the system generates PV array power generation warning information. The warning information usually contains key information such as the specific time of triggering the warning, the specific parameters that exceed expectations (such as power output, voltage, etc.), and the warning level, so that operation and maintenance personnel can understand and respond to abnormal situations in a timely manner.

[0076] Through step S103, the present invention realizes the real-time monitoring and early warning function of real-time photovoltaic array operation data. The generation of early warning information also provides decision support for operation and maintenance personnel, enabling them to take corresponding response measures according to the early warning level and specific circumstances, thereby ensuring the continuous operation of the photovoltaic power generation system.

[0077] Step S104, if the real-time photovoltaic array operation data is lower than the photovoltaic array power generation prediction data, the real-time photovoltaic array operation data is substituted into the photovoltaic synchronous machine stack power generation system grid-connected model, and the photovoltaic synchronous machine stack power generation system grid-connected control parameters are output;

[0078] When the real-time photovoltaic array operation data is lower than the predicted data, the system substitutes these data into the photovoltaic synchronous machine stack power generation system grid-connected model previously trained in step S101. The photovoltaic synchronous machine stack power generation system grid-connected model is constructed based on the random forest algorithm, which can comprehensively consider multiple factors (such as photovoltaic system data, power generation task data, etc.) to output grid-connected control parameters that meet the requirements of the power grid.

[0079] After receiving the real-time photovoltaic array operation data, the photovoltaic synchronous machine stack power generation system grid-connected model will perform a series of calculations and analyses, and finally output the photovoltaic synchronous machine stack power generation system grid-connected control parameters. The grid-connected control parameters include but are not limited to output voltage, frequency, and phase adjustment, which together determine the output characteristics of the photovoltaic synchronous machine stack power generation system so that it can be synchronized with the power grid.

[0080] Through step S104, the present invention can timely adjust the output characteristics of the photovoltaic synchronous machine stack power generation system when the real-time photovoltaic array operation data is lower than expected, so that the photovoltaic synchronous machine stack power generation system and the power grid can operate synchronously. This not only improves the stability and grid-connected efficiency of the photovoltaic power generation system, but also helps to reduce power quality problems caused by system mismatch. At the same time, since the grid-connected control parameters are dynamically generated based on real-time data, the system can better adapt to various working conditions and environmental changes.

[0081] Step S105, compare the grid-connected control parameters of the photovoltaic synchronous machine stack power generation system with the real-time parameters of the photovoltaic synchronous machine stack power generation system to obtain parameter comparison errors, the parameter comparison errors include output voltage errors, frequency errors and phase adjustment errors, obtain historical data of photovoltaic array operation data, use the historical data of photovoltaic array operation data and parameter comparison errors to optimize the grid-connected model of the photovoltaic synchronous machine stack power generation system using a genetic algorithm, use the optimized grid-connected model of the photovoltaic synchronous machine stack power generation system to process the real-time photovoltaic array operation data, output the optimized real-time parameters of the photovoltaic synchronous machine stack power generation system, the photovoltaic synchronous machine stack power generation system executes the optimized real-time parameters of the photovoltaic synchronous machine stack power generation system to achieve synchronization with the power grid, convert the electric energy generated by the photovoltaic synchronous machine stack power generation system into alternating current that meets the power grid standards, and achieve connection with the power grid.

[0082] The system compares the grid-connected control parameters of the photovoltaic synchronous machine stack power generation system (including output voltage, frequency and phase adjustment, etc.) with the real-time parameters of the system (i.e. the current output voltage, frequency and phase). By comparison, the parameter comparison errors are calculated, which include output voltage error, frequency error and phase adjustment error. These errors reflect the difference between the current system output and the expected grid-connected control parameters.

[0083] The system obtains the historical data of the photovoltaic array operation data, including the operation status of the photovoltaic array in the past period of time. Combined with the previously calculated parameter comparison error, the system is ready to use this data to optimize the grid-connected model of the photovoltaic synchronous machine stack power generation system.

[0084] The system optimizes the grid-connected model of the photovoltaic synchronous machine pile power generation system using a genetic algorithm. A genetic algorithm is a search algorithm that simulates natural selection and genetic mechanisms. Through continuous iteration and optimization, it can find the parameter combination that minimizes the error between the model output and the actual data. During the optimization process, the system constructs a fitness function based on the optimization objectives (such as minimizing the weighted sum of the output voltage error, frequency error, and phase adjustment error) to evaluate the performance of the model under different parameter combinations.

[0085] Through operations such as selection, crossover, and mutation, the genetic algorithm gradually optimizes the model parameters until the predetermined number of iterations is reached or the convergence condition is met.

[0086] After the optimization is completed, the system uses the optimized photovoltaic synchronous machine stack power generation system grid-connected model to process the real-time photovoltaic array operation data.

[0087] The model outputs the optimized real-time parameters of the photovoltaic synchronous machine stack power generation system, which more accurately reflects the requirements of the power grid and the actual status of the system.

[0088] The photovoltaic synchronous machine stack power generation system executes these optimized real-time parameters to achieve synchronization with the power grid. By adjusting parameters such as output voltage, frequency and phase, the system can convert the generated electrical energy into AC power that meets the grid standards and achieve a stable connection with the grid.

[0089] Through step S105, the present invention not only realizes real-time monitoring and optimization of the grid-connected parameters of the photovoltaic synchronous machine pile power generation system, but also improves the accuracy of the model through advanced technical means such as genetic algorithms. This helps to ensure the stable operation of the photovoltaic power generation system under different working conditions and environmental conditions, and improves the grid-connected efficiency and power quality.

[0090] Specifically, the photovoltaic synchronous machine stack power generation system grid-connected control method of the present invention, step S101 includes:

[0091] Volt component data including power, efficiency, and temperature;

[0092] Inverter data includes inverter conversion efficiency and inverter working status;

[0093] Meteorological data include solar irradiance, temperature, and wind speed;

[0094] Grid-connected control data include grid-connected voltage, grid-connected current and grid-connected frequency;

[0095] Photovoltaic power generation mission data includes expected power generation, power generation period, and power regulation requirements;

[0096] The preprocessed photovoltaic system data and photovoltaic power generation task data are used as input to train the random forest model. After the training is completed, the random forest model becomes a grid-connected model of the photovoltaic synchronous machine stack power generation system, which is used to receive new photovoltaic system data and photovoltaic power generation task data and output grid-connected control parameters.

[0097] In the detailed description of step S101, the photovoltaic synchronous machine stack power generation system grid-connected control method of the present invention further clarifies the specific content of the photovoltaic system data and the photovoltaic power generation task data, and how to use the data to train the random forest model, so as to construct the photovoltaic synchronous machine stack power generation system grid-connected model. The following is a detailed description of this step:

[0098] Photovoltaic module data:

[0099] Power: The current power output of the PV panel.

[0100] Efficiency: How effectively a PV panel converts light energy into electrical energy.

[0101] Temperature: The operating temperature of PV panels, which affects efficiency and power output.

[0102] Inverter data:

[0103] Inverter conversion efficiency: The efficiency of the inverter in converting DC power to AC power.

[0104] Inverter working status: including whether the inverter is operating normally, whether there is a fault alarm and other status information.

[0105] Weather data:

[0106] Solar irradiance: The intensity of sunlight that hits a photovoltaic module.

[0107] Temperature: Ambient temperature affects the performance of PV panels and inverters.

[0108] Wind speed: may affect the cooling effect of PV panels and the overall heat dissipation of the system.

[0109] Grid-connected control data:

[0110] Grid-connected voltage: The voltage output from the photovoltaic power generation system to the grid.

[0111] Grid-connected current: the current output from the photovoltaic power generation system to the power grid.

[0112] Grid-connected frequency: The frequency of the AC power output from the photovoltaic power generation system to the power grid.

[0113] Photovoltaic power generation task data:

[0114] Expected electricity generation: The total amount of electricity planned to be generated within a certain period of time.

[0115] Power generation period: The specified power generation time range.

[0116] Power regulation requirements: The requirements for regulating output power based on grid demand or system conditions.

[0117] The collected PV system data and PV power generation task data are cleaned, denoised, normalized or standardized to improve the effect of model training. The preprocessed data is used as input to train the random forest model. Random forest is an ensemble learning method that improves the accuracy and stability of the model by building multiple decision trees and combining their prediction results. After training, the random forest model becomes the grid-connected model of the PV synchronous machine stack power generation system. The model can receive new PV system data and PV power generation task data, and output grid-connected control parameters, such as adjusted grid-connected voltage, grid-connected current and grid-connected frequency, to ensure that the PV power generation system can achieve stable and efficient synchronization with the power grid.

[0118] Through the detailed implementation of step S101, the present invention constructs a photovoltaic synchronous machine stack power generation system grid-connected model based on random forest, which provides an accurate basis for the subsequent photovoltaic power generation system grid-connected control.

[0119] Specifically, the photovoltaic synchronous machine stack power generation system grid-connected control method of the present invention, step S102 includes:

[0120] The operation data of the photovoltaic array includes the voltage of the photovoltaic module, the current of the photovoltaic module, the power of the photovoltaic module, and the real-time parameters of the temperature of the photovoltaic module;

[0121] The preprocessed photovoltaic array operation data and natural environment data are used as input and substituted into the preset photovoltaic array power generation prediction model. The preset photovoltaic array power generation prediction model is started, the input data is processed, and the power generation prediction data of the photovoltaic array is calculated through the preset photovoltaic array power generation prediction model. The power generation prediction data of the photovoltaic array includes predicted power generation, predicted power generation efficiency and predicted power output.

[0122] In the detailed description of step S102, the photovoltaic synchronous machine stack power generation system grid-connected control method of the present invention further clarifies the specific content of the photovoltaic array operation data, and how to use these data and environmental data to start and run the preset photovoltaic array power generation prediction model, so as to obtain the photovoltaic array power generation prediction data. The following is a detailed description of this step:

[0123] PV array operating data:

[0124] PV panel voltage: the current output voltage of the PV panel.

[0125] PV panel current: the current value currently output by the PV panel.

[0126] Power of PV panels: The power currently output by a PV panel, usually the product of voltage and current.

[0127] Real-time parameters of photovoltaic module temperature: The operating temperature of photovoltaic modules, which has a direct impact on the efficiency and power output of photovoltaic modules. At the same time, continue to collect natural environment data, including sunlight irradiance, temperature, wind speed, etc.

[0128] Perform necessary preprocessing on the collected photovoltaic array operation data and natural environment data, such as cleaning, denoising, normalization or standardization, to ensure the accuracy of the data. Use the preprocessed photovoltaic array operation data and natural environment data as input and substitute them into the preset photovoltaic array power generation prediction model. Start the preset photovoltaic array power generation prediction model to process and analyze the input data. This model may be built based on machine learning, deep learning or other statistical methods, and can comprehensively consider multiple factors to predict the power generation of the photovoltaic array. The photovoltaic array power generation prediction data is calculated through the preset photovoltaic array power generation prediction model. These data include:

[0129] Forecasted power generation: The amount of power generated by a PV array over a period of time in the future (such as a day, week, or month).

[0130] Predicted Power Generation Efficiency: The predicted efficiency of a photovoltaic array in converting light energy into electrical energy.

[0131] Predicted power output: The predicted power output value of a photovoltaic array at a certain point in the future or during a period of time.

[0132] Through the detailed implementation of step S102, the present invention can accurately predict the power generation of the photovoltaic array, provide an important reference basis for subsequent grid-connected control, and help the system understand the power generation capacity of the photovoltaic array in advance, so as to better adjust the grid-connected control parameters and improve the photovoltaic power generation system to stably and efficiently operate synchronously with the power grid.

[0133] Specifically, the photovoltaic synchronous machine stack power generation system grid-connected control method of the present invention, step S103 includes:

[0134] Obtain the prediction data of the corresponding time period from the photovoltaic array power generation prediction model, and align the real-time photovoltaic array operation data with the photovoltaic array power generation prediction data in time, that is, the real-time photovoltaic array operation data and the photovoltaic array power generation prediction data correspond to the data of the same time point or time period;

[0135] Compare the real-time PV array operation data with the PV array power generation prediction data item by item to check whether the parameters of the real-time PV array are higher than the predicted values;

[0136] When the power output of the real-time photovoltaic array is higher than the predicted power by, for example, 10% for 5 consecutive minutes, an early warning is triggered, which includes the time, parameters and level of the triggering warning.

[0137] The prediction data of the corresponding time period is obtained from the photovoltaic array power generation prediction model, so that the real-time photovoltaic array operation data and the prediction data are aligned in time, and the two correspond to the data of the same time point or time period, so as to facilitate accurate comparison and analysis.

[0138] Compare the real-time PV array operation data with the PV array power generation forecast data item by item, including voltage, current, power, temperature and other parameters. Check whether the real-time PV array parameters are higher than the forecasted values. This comparison helps identify the differences between the actual operation data and the forecasted data, as well as possible problems or optimization space.

[0139] When the real-time PV array power output is higher than the predicted power by 10% for 5 consecutive minutes, the early warning mechanism is triggered. This threshold (10%) and time window (5 minutes) are set based on system characteristics and operating experience to ensure the accuracy and timeliness of the early warning.

[0140] The warning information includes the time when the warning is triggered, parameters (such as power, voltage, etc.) and the warning level. The warning level can be divided according to the actual situation, such as low-level warning, medium-level warning and high-level warning to reflect the severity and urgency of the problem.

[0141] When an early warning is triggered, the system should immediately take appropriate measures, including adjusting the operating parameters of the PV array, checking the system status, troubleshooting potential faults, etc. At the same time, the system should also record the early warning information for subsequent analysis and improvement.

[0142] Through the implementation of step S103, the present invention can timely identify the difference between the actual operation and prediction of the photovoltaic array, and remind the operator or the system to automatically take corresponding processing measures through the early warning mechanism, thereby ensuring the stable operation and efficient grid connection of the photovoltaic power generation system.

[0143] Specifically, the photovoltaic synchronous machine stack power generation system grid-connected control method of the present invention, step S104 includes:

[0144] Determine whether the real-time photovoltaic array operation data is lower than the photovoltaic array power generation prediction data, compare the real-time photovoltaic array operation data with the photovoltaic array power generation prediction data in terms of voltage, current and power; if the real-time photovoltaic array operation data is lower than the prediction data, use the voltage, current and power of the real-time photovoltaic array operation data as input information;

[0145] Determine the photovoltaic synchronous machine stack power generation system grid-connected model to be used. The photovoltaic synchronous machine stack power generation system grid-connected model is established based on historical data, system characteristics and control objectives. The photovoltaic synchronous machine stack power generation system grid-connected model is used to output grid-connected control parameters;

[0146] Substitute the real-time photovoltaic array operation data into the photovoltaic synchronous machine pile power generation system grid-connected model, and correspond the voltage, current and timestamp of the data to the input variables of the model;

[0147] The photovoltaic synchronous machine stack power generation system grid-connected model is started, and the photovoltaic synchronous machine stack power generation system grid-connected model processes the input real-time photovoltaic array operation data, and the photovoltaic synchronous machine stack power generation system grid-connected model calculates the grid-connected control parameters;

[0148] The output grid-connected control parameters are obtained from the grid-connected model of the photovoltaic synchronous machine stack power generation system. The grid-connected control parameters include output voltage, frequency and phase adjustment. The grid-connected control parameters are used to adjust the output of the photovoltaic synchronous machine stack power generation system so that the photovoltaic synchronous machine stack power generation system is synchronized with the power grid.

[0149] Determine whether the real-time PV array operation data (including voltage, current, and power) is lower than the PV array power generation prediction data. If the real-time PV array operation data is lower than the prediction data, it means that the current system power generation capacity has not met expectations and needs to be adjusted. The voltage, current, and power of the real-time PV array operation data are used as input information for subsequent calculations.

[0150] Determine the PV synchronous machine stack power generation system grid-connected model to be used. This model is built based on historical data, system characteristics and control objectives, and can output grid-connected control parameters for adjusting system output. Substitute the real-time PV array operation data into the PV synchronous machine stack power generation system grid-connected model. Make sure that the voltage, current and timestamp of the data correspond to the input variables of the model so that the model can process the data accurately. Start the PV synchronous machine stack power generation system grid-connected model to process the input real-time PV array operation data.

[0151] The photovoltaic synchronous machine stack power generation system grid-connected model processes the input real-time photovoltaic array operation data and calculates the grid-connected control parameters according to the model's algorithm and logic. The output grid-connected control parameters are obtained from the photovoltaic synchronous machine stack power generation system grid-connected model. These parameters include output voltage, frequency, and phase adjustment. These parameters are key information used to adjust the output of the photovoltaic synchronous machine stack power generation system to ensure that it is synchronized with the power grid.

[0152] The calculated grid-connected control parameters are applied to the photovoltaic synchronous machine stack power generation system to adjust the system's output voltage, frequency and phase to keep them consistent with the corresponding parameters of the power grid. According to the actual operation of the system and the needs of the power grid, the grid-connected control parameters are continuously adjusted to ensure that the photovoltaic power generation system can operate stably and efficiently in synchronization with the power grid.

[0153] Through the implementation of step S104, the present invention can calculate accurate grid-connected control parameters based on real-time photovoltaic array operation data and the photovoltaic synchronous machine pile power generation system grid-connected model, thereby realizing the synchronous operation of the photovoltaic power generation system and the power grid. This helps to improve the grid-connected efficiency of the photovoltaic power generation system and reduce the risks and costs in the grid-connected process.

[0154] Specifically, the photovoltaic synchronous machine stack power generation system grid-connected control method of the present invention, step S105 includes:

[0155] The photovoltaic synchronous machine stack power generation system grid-connected control parameters output by step S104 are obtained. The photovoltaic synchronous machine stack power generation system grid-connected control parameters are calculated based on the photovoltaic synchronous machine stack power generation system grid-connected model, and are intended to synchronize the system output with the power grid.

[0156] Real-time acquisition of actual operating parameters of the photovoltaic synchronous machine stack power generation system, including the current output voltage, frequency and phase;

[0157] The grid-connected control parameters and the real-time system parameters are aligned in time, that is, the grid control parameters and the real-time parameters of the photovoltaic synchronous machine stack power generation system correspond to the same moment;

[0158] Compare the target output voltage, target frequency and target phase in the grid-connected control parameters with the actual output voltage, actual frequency and actual phase in the real-time parameters of the photovoltaic synchronous machine stack power generation system item by item, and calculate the error value of each parameter, that is, the difference between the real-time value of the photovoltaic synchronous machine stack power generation system and the target value in the grid-connected control parameters;

[0159] For the output voltage, the difference between the actual output voltage and the target output voltage is calculated to obtain the output voltage error;

[0160] For frequency, the difference between the actual frequency and the target frequency is calculated to obtain the frequency error;

[0161] For the phase, the difference between the actual phase and the target phase is calculated to obtain the phase adjustment error.

[0162] The actual operating parameters of the photovoltaic synchronous machine stack power generation system are obtained in real time. The actual operating parameters include the current output voltage, frequency and phase, which reflect the current operating status of the system. The grid-connected control parameters are aligned with the actual operating parameters of the system in time, that is, the two correspond to the data at the same moment, which helps to accurately evaluate the difference between the current state of the system and the expected state.

[0163] Output voltage comparison and error calculation: compare the target output voltage in the grid-connected control parameters with the actual output voltage in the actual system operation parameters, calculate the difference between the two, and obtain the output voltage error.

[0164] Frequency comparison and error calculation, compare the target frequency in the grid-connected control parameters with the actual frequency in the actual operating parameters of the system, calculate the difference between the two, and obtain the frequency error.

[0165] Phase comparison and error calculation, compare the target phase in the grid-connected control parameters with the actual phase in the actual operating parameters of the system, calculate the difference between the two, and obtain the phase adjustment error.

[0166] The calculated error values ​​(output voltage error, frequency error, phase adjustment error) reflect the difference between the current state of the system and the expected state. The error values ​​can be used for subsequent control strategy adjustments to reduce the difference and achieve synchronization between the system output and the power grid.

[0167] Through the implementation of step S105, the present invention can accurately evaluate the difference between the current state and the expected state of the photovoltaic synchronous machine stack power generation system, and provide an important basis for subsequent control strategy adjustments, which helps to synchronize the system output with the power grid and improve the stability and grid-connected efficiency of the photovoltaic power generation system.

[0168] Specifically, the photovoltaic synchronous machine stack power generation system grid-connected control method of the present invention, step S105 includes:

[0169] According to the grid connection requirements and control objectives of the photovoltaic synchronous machine stack power generation system, the optimization objectives are determined, and the optimization objectives include minimizing the weighted sum of the output voltage error, frequency error and phase adjustment error and the error threshold;

[0170] Based on the optimization objective, a fitness function is constructed, which is used to evaluate the grid-connected model under given parameters;

[0171] Determine the parameters of the genetic algorithm, such as population size, number of iterations, crossover probability, and mutation probability;

[0172] Initialize the population, that is, generate a set of random grid-connected model parameters as the initial solution.

[0173] Use the fitness function to evaluate the fitness value of each individual in the current population, substitute the individual's parameters into the grid-connected model, calculate the error between the model output and the actual data, and calculate the fitness value based on the error;

[0174] Select excellent individuals as parents according to their fitness values, and perform crossover operations to generate new offspring;

[0175] Perform mutation operations on offspring individuals, that is, randomly change the values ​​of some of their genes, and repeat the selection and crossover mutation process until the predetermined number of iterations is reached;

[0176] After the iteration, the individual with the lowest fitness value is selected from the population as the optimal solution, that is, the optimized grid-connected model parameters;

[0177] The optimized grid-connected model parameters are output and used in the subsequent grid-connected control of the photovoltaic synchronous machine stack power generation system.

[0178] According to the grid connection requirements and control objectives of the photovoltaic synchronous machine stack power generation system, the optimization goal is to minimize the weighted sum of output voltage error, frequency error and phase adjustment error. At the same time, an error threshold is set as a constraint in the optimization process.

[0179] Based on the optimization objective, a fitness function is constructed. This function should be able to evaluate the performance of the grid-connected model under given parameters, that is, calculate the error between the model output and the actual data, and calculate the fitness value based on the error. The lower the fitness value, the better the model performance.

[0180] Determine the parameters of the genetic algorithm, including population size (i.e. the number of initial solutions), number of iterations (i.e. the total number of rounds of the optimization process), crossover probability (i.e. the probability that two parent individuals cross to generate offspring), and mutation probability (i.e. the probability that individual genes undergo random changes).

[0181] Generate a set of random grid-connected model parameters as the initial solution to form the initial population. These parameters may include various coefficients and thresholds in the grid-connected model.

[0182] Use the fitness function to evaluate the fitness value of each individual in the current population. Substitute the individual's parameters into the grid-connected model, calculate the error between the model output and the actual data, and calculate the fitness value based on the error.

[0183] Select excellent individuals as parents based on their fitness values. Generally, individuals with lower fitness values ​​are more likely to be selected.

[0184] Perform a crossover operation on the selected parent individuals to generate new offspring individuals. The crossover operation can be achieved by exchanging some genes of the parent individuals.

[0185] Perform mutation operations on offspring individuals, that is, randomly change the values ​​of some of their genes. Mutation operations help increase the diversity of the population and prevent the optimization process from falling into a local optimal solution.

[0186] The selection and crossover mutation process is repeated until the predetermined number of iterations is reached. In each iteration, a new population is generated and its fitness value is evaluated. Through continuous iterations, the individuals in the population gradually approach the optimal solution.

[0187] After the iteration, the individual with the lowest fitness value is selected from the population as the optimal solution. This optimal solution is the optimized grid-connected model parameters.

[0188] Output the optimized grid-connected model parameters, and use these parameters in the subsequent grid-connected control of the photovoltaic synchronous machine pile power generation system. The accuracy and stability of the grid-connected control can be further improved through the optimized parameters.

[0189] Through the implementation of step S105, the present invention can optimize the grid-connected model parameters using a genetic algorithm, thereby reducing the output voltage error, frequency error and phase adjustment error, and improving the accuracy and stability of the grid-connected photovoltaic synchronous machine stack power generation system.

[0190] The photovoltaic synchronous machine stack power generation system grid-connected control method of the present invention effectively solves the problem of unstable output voltage of the generator during the grid-connected process of the existing photovoltaic synchronous machine stack power generation system, and the problem that the output voltage, frequency and phase of the generator do not match the voltage, frequency and phase of the grid by means of the following technical means:

[0191] Acquire PV system data (including control commands, PV module data, inverter data, meteorological data, power generation data, and grid-connected control data) and PV power generation task data, and perform preprocessing.

[0192] The random forest model is trained using the preprocessed data to obtain the grid-connected model of the photovoltaic synchronous machine stack power generation system. The grid-connected model of the photovoltaic synchronous machine stack power generation system can accurately receive and process new photovoltaic system data and photovoltaic power generation task data, and output accurate grid-connected control parameters.

[0193] Collect the photovoltaic array operation data and natural environment data, substitute them into the preset photovoltaic array power generation prediction model, and obtain the photovoltaic array power generation prediction data. Through the photovoltaic array power generation prediction model, the power generation capacity of the photovoltaic array can be understood in advance, providing an important reference for subsequent grid-connected control.

[0194] Obtain real-time PV array operation data and compare it with the PV array power generation forecast data. If the real-time PV array operation data is higher than the forecast data, generate early warning information, which helps to detect and handle abnormal situations in time to prevent system overload or damage.

[0195] If the real-time PV array operation data is lower than the predicted data, the real-time PV array operation data is substituted into the grid-connected model of the PV synchronous machine stack power generation system, and the grid-connected control parameters are output. The grid-connected control parameters are compared with the real-time system parameters, and the parameter comparison errors (including output voltage error, frequency error and phase adjustment error) are calculated. The genetic algorithm is used to optimize the grid-connected model, and the model parameters are continuously adjusted through historical data and parameter comparison errors to improve the accuracy and stability of the grid-connected control.

[0196] The optimized grid-connected model is used to process the real-time photovoltaic array operation data and output the optimized real-time parameters. The photovoltaic synchronous machine stack power generation system executes the optimized real-time parameters, achieves synchronization with the grid, converts the generated electrical energy into AC power that meets the grid standards, and achieves connection with the grid.

[0197] Through these steps, the method of the present invention can monitor and adjust the output of the photovoltaic synchronous machine stack power generation system in real time, ensuring that the output voltage, frequency and phase of the generator are highly matched with the power grid, thereby solving the problems of unstable output voltage and parameter mismatch in the prior art, and improving the stability and grid-connected efficiency of the photovoltaic power generation system.

Claims

1. A photovoltaic synchronous machine stack power generation system grid-connected control method, characterized in that: include: Step S101, obtaining photovoltaic system data and photovoltaic power generation task data, the photovoltaic system data including control commands, photovoltaic component data, inverter data, meteorological data, power generation data and grid-connected control data, preprocessing the photovoltaic system data and photovoltaic power generation task data to obtain preprocessed photovoltaic system data and photovoltaic power generation task data, training the preprocessed photovoltaic system data and photovoltaic power generation task data on a random forest model to obtain a photovoltaic synchronous machine stack power generation system grid-connected model; Step S102, collecting photovoltaic array operation data and natural environment data, substituting the collected photovoltaic array operation data and natural environment data into a preset photovoltaic array power generation prediction model to obtain photovoltaic array power generation prediction data; Step S103, obtaining real-time photovoltaic array operation data, substituting the real-time photovoltaic array operation data into the photovoltaic array power generation prediction data for comparison, and generating photovoltaic array power generation warning information if the real-time photovoltaic array operation data is higher than the photovoltaic array power generation prediction data; Step S104, if the real-time photovoltaic array operation data is lower than the photovoltaic array power generation prediction data, the real-time photovoltaic array operation data is substituted into the photovoltaic synchronous machine stack power generation system grid-connected model, and the photovoltaic synchronous machine stack power generation system grid-connected control parameters are output; Step S105, compare the grid-connected control parameters of the photovoltaic synchronous machine stack power generation system with the real-time parameters of the photovoltaic synchronous machine stack power generation system to obtain parameter comparison errors, the parameter comparison errors include output voltage errors, frequency errors and phase adjustment errors, obtain historical data of photovoltaic array operation data, use the historical data of photovoltaic array operation data and parameter comparison errors to optimize the grid-connected model of the photovoltaic synchronous machine stack power generation system using a genetic algorithm, use the optimized grid-connected model of the photovoltaic synchronous machine stack power generation system to process the real-time photovoltaic array operation data, output the optimized real-time parameters of the photovoltaic synchronous machine stack power generation system, the photovoltaic synchronous machine stack power generation system executes the optimized real-time parameters of the photovoltaic synchronous machine stack power generation system to achieve synchronization with the power grid, convert the electric energy generated by the photovoltaic synchronous machine stack power generation system into alternating current that meets the power grid standards, and achieve connection with the power grid.

2. The photovoltaic synchronous machine stack power generation system grid-connected control method according to claim 1, characterized in that: The step S101 includes: Volt component data including power, efficiency, and temperature; Inverter data includes inverter conversion efficiency and inverter working status; Meteorological data include solar irradiance, temperature, and wind speed; Grid-connected control data include grid-connected voltage, grid-connected current and grid-connected frequency; Photovoltaic power generation mission data includes expected power generation, power generation period, and power regulation requirements; The preprocessed photovoltaic system data and photovoltaic power generation task data are used as input to train the random forest model. After the training is completed, the random forest model becomes a grid-connected model of the photovoltaic synchronous machine stack power generation system, which is used to receive new photovoltaic system data and photovoltaic power generation task data and output grid-connected control parameters.

3. The photovoltaic synchronous machine stack power generation system grid-connected control method according to claim 1, characterized in that: The step S102 includes: The operation data of the photovoltaic array includes the voltage of the photovoltaic module, the current of the photovoltaic module, the power of the photovoltaic module, and the real-time parameters of the temperature of the photovoltaic module; The preprocessed photovoltaic array operation data and natural environment data are used as input and substituted into the preset photovoltaic array power generation prediction model. The preset photovoltaic array power generation prediction model is started, the input data is processed, and the power generation prediction data of the photovoltaic array is calculated through the preset photovoltaic array power generation prediction model. The power generation prediction data of the photovoltaic array includes predicted power generation, predicted power generation efficiency and predicted power output.

4. The photovoltaic synchronous machine stack power generation system grid-connected control method according to claim 1, characterized in that: The step S103 includes: Obtain the prediction data of the corresponding time period from the photovoltaic array power generation prediction model, and align the real-time photovoltaic array operation data with the photovoltaic array power generation prediction data in time, that is, the real-time photovoltaic array operation data and the photovoltaic array power generation prediction data correspond to the data of the same time point or time period; Compare the real-time PV array operation data with the PV array power generation prediction data item by item to check whether the parameters of the real-time PV array are higher than the predicted values; When the power output of the real-time photovoltaic array is higher than the predicted power by, for example, 10% for 5 consecutive minutes, an early warning is triggered, which includes the time, parameters and level of the triggering warning.

5. The photovoltaic synchronous machine stack power generation system grid-connected control method according to claim 1, characterized in that: The step S104 includes: Determine whether the real-time photovoltaic array operation data is lower than the photovoltaic array power generation prediction data, compare the real-time photovoltaic array operation data with the photovoltaic array power generation prediction data in terms of voltage, current and power; if the real-time photovoltaic array operation data is lower than the prediction data, use the voltage, current and power of the real-time photovoltaic array operation data as input information; Determine the photovoltaic synchronous machine stack power generation system grid-connected model to be used. The photovoltaic synchronous machine stack power generation system grid-connected model is established based on historical data, system characteristics and control objectives. The photovoltaic synchronous machine stack power generation system grid-connected model is used to output grid-connected control parameters; Substitute the real-time photovoltaic array operation data into the photovoltaic synchronous machine pile power generation system grid-connected model, and correspond the voltage, current and timestamp of the data to the input variables of the model; The photovoltaic synchronous machine stack power generation system grid-connected model is started, and the photovoltaic synchronous machine stack power generation system grid-connected model processes the input real-time photovoltaic array operation data, and the photovoltaic synchronous machine stack power generation system grid-connected model calculates the grid-connected control parameters; The output grid-connected control parameters are obtained from the grid-connected model of the photovoltaic synchronous machine stack power generation system. The grid-connected control parameters include output voltage, frequency and phase adjustment. The grid-connected control parameters are used to adjust the output of the photovoltaic synchronous machine stack power generation system so that the photovoltaic synchronous machine stack power generation system is synchronized with the power grid.

6. The photovoltaic synchronous machine stack power generation system grid-connected control method according to claim 1, characterized in that: The step S105 includes: Obtaining the photovoltaic synchronous machine stack power generation system grid-connected control parameters outputted by step S104, the photovoltaic synchronous machine stack power generation system grid-connected control parameters are calculated based on the photovoltaic synchronous machine stack power generation system grid-connected model, and are intended to synchronize the system output with the power grid; Real-time acquisition of actual operating parameters of the photovoltaic synchronous machine stack power generation system, including the current output voltage, frequency and phase; The grid-connected control parameters and the real-time system parameters are aligned in time, that is, the grid control parameters and the real-time parameters of the photovoltaic synchronous machine stack power generation system correspond to the same moment; Compare the target output voltage, target frequency and target phase in the grid-connected control parameters with the actual output voltage, actual frequency and actual phase in the real-time parameters of the photovoltaic synchronous machine stack power generation system item by item, and calculate the error value of each parameter, that is, the difference between the real-time value of the photovoltaic synchronous machine stack power generation system and the target value in the grid-connected control parameters; For the output voltage, the difference between the actual output voltage and the target output voltage is calculated to obtain the output voltage error; For frequency, the difference between the actual frequency and the target frequency is calculated to obtain the frequency error; For the phase, the difference between the actual phase and the target phase is calculated to obtain the phase adjustment error.

7. The photovoltaic synchronous machine stack power generation system grid-connected control method according to claim 6, characterized in that: The step S105 includes: According to the grid connection requirements and control objectives of the photovoltaic synchronous machine stack power generation system, the optimization objectives are determined, and the optimization objectives include minimizing the weighted sum of the output voltage error, frequency error and phase adjustment error and the error threshold; Based on the optimization objective, a fitness function is constructed, which is used to evaluate the grid-connected model under given parameters; Determine the parameters of the genetic algorithm, such as population size, number of iterations, crossover probability, and mutation probability; Initialize the population, that is, generate a set of random grid-connected model parameters as the initial solution; Use the fitness function to evaluate the fitness value of each individual in the current population, substitute the individual's parameters into the grid-connected model, calculate the error between the model output and the actual data, and calculate the fitness value based on the error; Select excellent individuals as parents according to their fitness values, and perform crossover operations to generate new offspring; Perform mutation operations on offspring individuals, that is, randomly change the values ​​of some of their genes, and repeat the selection and crossover mutation process until the predetermined number of iterations is reached; After the iteration, the individual with the lowest fitness value is selected from the population as the optimal solution, that is, the optimized grid-connected model parameters; The optimized grid-connected model parameters are output and used in the subsequent grid-connected control of the photovoltaic synchronous machine stack power generation system.

Images