A photovoltaic grid low-voltage ride-through coordination method and system

Abstract

The present invention discloses a photovoltaic grid low-voltage ride-through coordination method and system, which relates to the technical field of photovoltaic grids and includes the following steps: real-time monitoring of voltage fluctuations in the photovoltaic system, obtaining a voltage steady-state coefficient based on the voltage fluctuations, and sending a voltage dip signal if the voltage is lower than the voltage steady-state coefficient; obtaining multiple segments of voltage dip values, and obtaining performance impact data of the photovoltaic inverter and grid anomaly impact data under each segment of voltage dip value; constructing a photovoltaic system voltage dip state evaluation model based on machine learning according to the performance impact data and grid anomaly impact data; performing primary screening and secondary screening on the output of the photovoltaic system voltage dip state evaluation model to obtain a preset voltage dip value; and judging according to the preset voltage dip value and controlling a corresponding instruction set for the photovoltaic system to perform a low-voltage ride-through operation. The present invention solves the problem of abnormal low-voltage ride-through caused by unreasonable threshold setting in the photovoltaic grid through reasonable coordination of low-voltage ride-through.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic power grids, and more specifically, to a method and system for coordinating low-voltage ride-through of a photovoltaic power grid. Background Art

[0002] In recent years, with the progress of technology and policy support, photovoltaic power generation has been widely used. The photovoltaic system converts direct current into alternating current through an inverter and connects to the power grid to transmit power to the power grid. Low-voltage ride-through refers to the ability of a photovoltaic power generation system to continue operating or automatically recover and transmit power to the power grid when the power grid experiences short-term voltage fluctuations or faults, rather than immediately disconnecting. This is a requirement for the inverter of the photovoltaic system under abnormal power grid conditions, ensuring that the photovoltaic power station can maintain stable operation when the power grid voltage temporarily drops below the specified value, avoiding exacerbating the burden on the power grid due to the disconnection of large-scale photovoltaic power generation, and promoting the recovery and stability of the power grid.

[0003] For example, a method for optimizing the low-voltage ride-through scheduling of a high-penetration photovoltaic distribution network disclosed in the invention patent announcement with the publication number: CN116264400A includes: real-time monitoring of the power grid voltage value to judge the power grid state: judging whether the power grid is in the low-voltage ride-through process, and controlling the active power and reactive power of the power grid during the low-voltage ride-through process of the power grid; judging whether the power grid is in the low-voltage ride-through recovery process, and controlling the active power and reactive power of the power grid during the low-voltage ride-through recovery process of the power grid; judging whether the power grid parameters have recovered to the level before low-voltage ride-through, and switching the power grid after low-voltage ride-through recovery back to the original control mode. When the AC power grid experiences disturbances such as short-circuit faults or impacts, resulting in a large voltage drop in the AC power grid voltage, a low-voltage ride-through phenomenon occurs in the power grid voltage. The dual-power optimization strategy provided in this application controls the rise of the active current and the decrease of the reactive current simultaneously, thereby regulating the stability of the grid connection frequency of the power grid and maximizing the power consumption capacity of the power system for photovoltaic output.

[0004] For example, a method, device, and storage medium for regulating overcurrent during voltage ride-through of a photovoltaic inverter disclosed in the invention patent announcement with the publication number: CN117134417B include: monitoring the AC-side voltage of the photovoltaic inverter in a photovoltaic power station; when the AC-side voltage drops and is lower than the low-voltage ride-through threshold, low-voltage ride-through control is performed; when the AC-side voltage rises and is higher than the high-voltage ride-through threshold, high-voltage ride-through control is performed; when the AC-side voltage is between the high-voltage ride-through threshold and the low-voltage ride-through threshold, conventional control is executed. The technical solution adopted in the present invention actively supports the stability of the power grid voltage during continuous low-voltage ride-through or high-voltage ride-through in the process of DC commutation failure; when DC is blocked, the voltage support is withdrawn to reduce the overvoltage at the moment of DC blocking, support the safe and stable operation of the power grid, and improve the acceptance capacity of photovoltaic power generation at the DC sending end.

[0005] Among the above-disclosed technical solutions, at least the following technical problems exist:

[0006] When the voltage of the photovoltaic grid system drops due to a fault or short circuit and drops to the set threshold, low-voltage ride-through is performed to maintain the stability of the grid. However, if the set threshold is too large, the low-voltage ride-through operation will be too late, resulting in the equipment being in a low-voltage state for a long time, increasing the risk of equipment damage, and reducing the stability of the system. If the set threshold is too small, it may lead to frequent protection actions and affect the normal operation of the equipment.

[0007] In view of the above problems, the present invention proposes a solution. Summary of the Invention

[0008] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a method and system for coordinating low-voltage ride-through of a photovoltaic grid, which can solve the problem of abnormal low-voltage ride-through caused by unreasonable threshold setting in the photovoltaic grid through reasonable coordination of low-voltage ride-through.

[0009] To achieve the above object, the present invention provides the following technical solutions:

[0010] A method and system for coordinating low-voltage ride-through of a photovoltaic grid, comprising the following steps: continuously monitoring voltage fluctuations in the photovoltaic system, obtaining a voltage steady-state coefficient based on the voltage fluctuations, and if the voltage is lower than the voltage steady-state coefficient, sending a voltage drop signal to the photovoltaic system; based on the voltage drop signal, obtaining multiple segments of voltage drop values, and obtaining performance impact data of the photovoltaic inverter and grid anomaly impact data under each segment of voltage drop value; constructing a photovoltaic system voltage drop state evaluation model based on machine learning according to the performance impact data and grid anomaly impact data; performing preliminary screening and secondary screening on the output of the photovoltaic system voltage drop state evaluation model to obtain a voltage drop preset value; judging according to the voltage drop preset value and controlling a corresponding instruction set for the photovoltaic system to perform low-voltage ride-through operation.

[0011] In a preferred embodiment, the performance impact data of the photovoltaic inverter includes a photovoltaic inverter sensitivity coefficient, and the specific method for obtaining the photovoltaic inverter sensitivity coefficient is as follows: obtaining the input signal of the photovoltaic inverter, as well as the light radiation and temperature data around the photovoltaic inverter under the current voltage drop; performing data analysis and feature extraction on the light radiation and temperature data around the photovoltaic inverter to obtain a light radiation impact factor and a temperature impact factor that affect the propagation of the voltage signal; obtaining the output signal waveform after the action of the photovoltaic inverter based on the input signal of the photovoltaic inverter; analyzing the output signal waveform to obtain the response time interval between the time when the photovoltaic inverter receives the input signal and then outputs the signal; calculating the photovoltaic inverter sensitivity coefficient based on the response time interval, in combination with the light radiation impact factor and the temperature impact factor, according to a preset photovoltaic inverter sensitivity formula.

[0012] In a preferred embodiment, the steps for analyzing the light radiation and temperature data around the photovoltaic inverter to obtain the light radiation influence factor and temperature influence factor that affect the voltage signal propagation are as follows: collect the light radiation and temperature data for feature extraction to obtain the weight coefficients of the light radiation and temperature data, and remove the random fluctuations in the data based on a smoothing algorithm; plot a scatter diagram of the light radiation and temperature data against the signal influence on the photovoltaic inverter for the processed light radiation and temperature data; obtain the change trend of the light radiation and temperature data on the signal propagation of the photovoltaic inverter according to the scatter diagram of the signal influence; perform a quantitative process on the change trend based on regression analysis to obtain the light radiation influence factor and temperature influence factor respectively.

[0013] In a preferred embodiment, the method for obtaining the stability coefficient of the photovoltaic inverter is as follows: obtain the waveform diagram of the output power of the photovoltaic inverter; analyze the change trend of the output power before and after the voltage dip based on the waveform diagram of the output power according to a preset standard output power to obtain the efficiency of the power conversion unit; obtain the working state of the historical protection device in the photovoltaic inverter under the voltage dip; perform regression analysis on the working state data of the historical protection device to obtain the influencing factors of the working state of the protection device, where the influencing factors of the working state include the fault repair data of the protection device and the aging rate of the protection device; calculate the sensitivity coefficient of the protection device based on the fault repair data and the aging rate of the protection device according to a preset sensitivity formula of the protection device; calculate the stability coefficient of the photovoltaic inverter based on the efficiency of the power conversion unit and the sensitivity coefficient of the protection device according to a preset stability formula of the photovoltaic inverter.

[0014] In a preferred embodiment, the method for obtaining the abnormal grid influence data is as follows: obtain the state data of grid equipment, and obtain the load data of grid equipment based on feature analysis and extraction; perform a differential analysis on the load data of grid equipment based on a preset stable load data to obtain the load difference value of grid equipment; monitor the synchronism of different regions of the grid system to obtain the synchronization rate of the grid system; calculate the abnormal grid influence data based on the load difference value of grid equipment and the synchronization rate of the grid system according to a preset formula for abnormal grid influence data.

[0015] In a preferred embodiment, the steps of preliminarily screening and secondarily screening the output of the photovoltaic system voltage sag state evaluation model to obtain the voltage sag preset value are as follows: Obtain the voltage sag state evaluation coefficient according to the output of the photovoltaic system voltage sag state evaluation model; Monitor the state of the protection device in the photovoltaic system to obtain the voltage sag value that causes the photovoltaic system to frequently perform protection actions due to too small voltage sag; Use the voltage sag state evaluation coefficient corresponding to the voltage sag value of the frequently performed protection actions as the limit minimum value; Obtain the voltage sag value that causes the inverter to disconnect due to too large voltage sag in the photovoltaic system; Use the voltage sag state evaluation coefficient corresponding to the voltage sag value that causes the inverter to disconnect as the limit maximum value; Obtain all voltage sag state evaluation coefficients and perform preliminary screening based on the limit minimum value and the limit maximum value; Exclude the voltage sag state evaluation coefficients that are less than the limit minimum value and greater than the limit maximum value to obtain a preliminarily screened voltage sag state evaluation data set; Calculate the average value of the voltage sag state evaluation data set and set the average value as the voltage sag state evaluation stability value; Based on the voltage sag state evaluation stability value, perform secondary screening on the voltage sag state evaluation data set through comparison, and based on the gradient descent method, obtain the optimal voltage sag state evaluation coefficient;

[0016] Set the voltage sag value corresponding to the optimal voltage sag state evaluation coefficient as the voltage sag preset value.

[0017] In a preferred embodiment, the steps of the corresponding instruction set for judging according to the voltage sag preset value and controlling the photovoltaic system to perform low-voltage ride-through operations are as follows: If the voltage sag value is less than the voltage sag preset value, the photovoltaic system does not perform low-voltage ride-through operations, and the system monitors in real time; If the voltage sag value is greater than or equal to the voltage sag preset value, the system automatically schedules to perform low-voltage ride-through operations on the photovoltaic system.

[0018] In a preferred embodiment, the specific calculation formulas for the light radiation influence factor and the temperature influence factor are as follows: The specific calculation formula for the photovoltaic inverter sensitivity coefficient is as follows: In the formula, S is the photovoltaic inverter sensitivity coefficient, is the response time interval, is the light radiation influence factor, is the temperature influence factor, is the standard data of the photovoltaic inverter sensitivity, is an adjustable constant, is the light radiation, is the light radiation weight coefficient, is the temperature data, is the temperature data weight coefficient.

[0019] In a preferred embodiment, the voltage sag state evaluation model has the following specific calculation formula: In the formula, is the voltage sag state evaluation coefficient, is the power grid anomaly influence data, S is the photovoltaic inverter sensitivity coefficient, is the photovoltaic inverter stability coefficient, is the power grid anomaly influence data weight coefficient, is the photovoltaic inverter sensitivity coefficient weight coefficient, is the photovoltaic inverter stability coefficient weight coefficient.

[0020] In a preferred embodiment, a system for a photovoltaic power grid low-voltage ride-through coordination method is characterized by comprising a signal transmission module, a data acquisition module, a model construction module, a data screening module, and an instruction generation module: The signal transmission module is used to monitor the voltage fluctuation in the photovoltaic system in real time, obtain the voltage steady-state coefficient based on the voltage fluctuation, and send a voltage sag signal to the photovoltaic system if the voltage is lower than the voltage steady-state coefficient; The data acquisition module is used to obtain multiple segments of voltage sag values based on the voltage sag signal, and obtain the performance influence data and the power grid anomaly influence data of the photovoltaic inverter under each segment of voltage sag value; The model construction module is used to construct a voltage sag state evaluation model of the photovoltaic system based on the performance influence data and the power grid anomaly influence data by machine learning; The data screening module is used to perform primary screening and secondary screening on the output of the voltage sag state evaluation model of the photovoltaic system to obtain a preset voltage sag value; The instruction generation module is used to judge according to the preset voltage sag value and control the corresponding instruction set for the photovoltaic system to perform low-voltage ride-through operations.

[0021] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0022] 1. By comprehensively analyzing the voltage sag state evaluation coefficient of the photovoltaic system, combining the state of the protection device and the inverter disconnection condition, a multi-level screening and optimization process is established, thereby accurately determining the preset voltage sag value. This method can effectively eliminate abnormal data, reduce the negative impacts caused by frequent protection actions and inverter disconnections, and improve the stability and reliability of the system. At the same time, the secondary screening and optimization process based on the gradient descent method helps to accurately adjust the preset voltage sag value to ensure the stable operation of the photovoltaic system under various voltage sag conditions.

[0023] 2. By comprehensively analyzing the input signals of the photovoltaic inverter, the surrounding light radiation and temperature data, and combining the response time intervals of the output signal waveforms, the sensitivity coefficient of the photovoltaic inverter is accurately calculated. By extracting features, the influence factors of light radiation and temperature on the voltage signal transmission are obtained, which can quantify the influence of external environmental factors on the inverter performance, and optimize the system response in combination with the sensitivity formula of the photovoltaic inverter. This method helps to improve the adaptability and stability of the photovoltaic inverter under different environmental conditions, optimize its output signal, and thus enhance the overall efficiency and reliability of the photovoltaic system. Description of the Drawings

[0024] Figure 1 It is a schematic structural diagram of a low-voltage ride-through coordination method for a photovoltaic power grid provided by an embodiment of the present application.

[0025] Figure 2 It is a schematic structural diagram of a low-voltage ride-through coordination system for a photovoltaic power grid provided by an embodiment of the present application. Detailed Embodiments

[0026] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0027] Embodiment 1 Figure 1 It is a schematic structural diagram of a low-voltage ride-through coordination method for a photovoltaic power grid provided by an embodiment of the present application, including the following steps:

[0028] S1. Monitor the voltage fluctuations in the photovoltaic system in real time, obtain the voltage steady-state coefficient based on the voltage fluctuations, and send a voltage dip signal to the photovoltaic system if the voltage is lower than the voltage steady-state coefficient.

[0029] In this example, voltage dips are usually caused by load changes, equipment failures, or grid instantaneous fluctuations in the power system. By obtaining voltage fluctuation data in real time, the system can quickly identify voltage dip phenomena and issue alarms or activate emergency response mechanisms in a timely manner. Among them, the voltage steady-state coefficient refers to the standard voltage that maintains the normal and stable operation of the system in the photovoltaic system. The specific method for obtaining the voltage steady-state coefficient is as follows:

[0030] Monitor and obtain the voltages in the photovoltaic system at multiple time intervals in real time, integrate the voltages at multiple time intervals into a data set and calculate the standard deviation of the data set, and set the standard deviation as the voltage steady-state coefficient. By calculating the voltage steady-state coefficient, the photovoltaic system can evaluate in real time whether the voltage is within the normal operating range. When the voltage monitored in real time in the photovoltaic system is lower than the voltage steady-state coefficient within a preset time, the monitoring device will send a voltage drop signal to the photovoltaic system.

[0031] S2. Based on the voltage drop signal, obtain multiple voltage drop values, and obtain the performance impact data of the photovoltaic inverter and the grid anomaly impact data under each voltage drop value.

[0032] In this example, the performance impact data of the photovoltaic inverter refers to the impact on the performance of the photovoltaic inverter when the voltage drops. Among them, the performance impact data of the photovoltaic inverter includes the photovoltaic inverter sensitivity coefficient and the photovoltaic inverter stability coefficient.

[0033] The photovoltaic inverter sensitivity coefficient refers to that when a voltage drop occurs, it will affect the sensitivity of the photovoltaic inverter at this time, resulting in a delay problem when the photovoltaic inverter receives the voltage signal of the photovoltaic system and outputs to maintain the stability of the photovoltaic system voltage. Obtaining the photovoltaic inverter sensitivity coefficient has the following advantages for analyzing the performance of the photovoltaic inverter during voltage drop:

[0034] Improve the response prediction ability of the inverter to voltage changes: The performance of the photovoltaic inverter may be greatly affected when the grid voltage changes, especially during voltage drops (such as voltage sags or voltage defects). By determining the voltage sensitivity coefficient of the inverter, its response during voltage drop can be accurately predicted. For example, the inverter may enter the islanding protection mode when the voltage is lower than a certain threshold, and the sensitivity coefficient helps to understand the degree of this change.

[0035] Optimize the anti-interference ability and stability of the inverter: By obtaining the sensitivity coefficient, the stability of the inverter during voltage drop can be effectively evaluated, and it can be judged whether it is prone to frequent shutdowns or restarts, thereby improving the stability and reliability of the system. Especially in areas with unstable grid quality, understanding the voltage sensitivity of the inverter can help designers select appropriate inverter configurations and improve the anti-interference ability of the system.

[0036] Support the performance optimization and intelligent control of the inverter: Obtaining the voltage sensitivity coefficient of the inverter helps to optimize the intelligent control strategy of the inverter. For example, the inverter can adaptively adjust its operating mode according to the characteristics of voltage changes, thereby reducing the impact of voltage drops on system performance. The role of the sensitivity coefficient is particularly prominent in aspects such as dynamic voltage control, frequency regulation, and fault recovery, which helps to achieve the intelligent operation and high-efficiency performance of the inverter.

[0037] The specific method for obtaining the sensitivity coefficient of the photovoltaic inverter is as follows:

[0038] Obtain the input signal of the photovoltaic inverter, as well as the light radiation and temperature data around the photovoltaic inverter, under the current voltage dip;

[0039] Perform data analysis and feature extraction on the light radiation and temperature data around the photovoltaic inverter to obtain the light radiation influence factor and temperature influence factor that affect the voltage signal propagation;

[0040] Obtain the output signal waveform after the action of the photovoltaic inverter based on the input signal of the photovoltaic inverter;

[0041] Analyze the output signal waveform to obtain the response time interval between the time when the photovoltaic inverter receives the input signal and then outputs the signal;

[0042] Calculate the sensitivity coefficient of the photovoltaic inverter based on the response time interval, combined with the light radiation influence factor and temperature influence factor, according to a preset photovoltaic inverter sensitivity formula.

[0043] Among them, the specific steps for performing data analysis and feature extraction on the light radiation and temperature data around the photovoltaic inverter to obtain the light radiation influence factor and temperature influence factor that affect the voltage signal propagation are as follows:

[0044] Collect light radiation and temperature data for feature extraction to obtain the weight coefficients of the light radiation and temperature data, and remove the random fluctuations in the data based on a smoothing algorithm;

[0045] Plot a scatter diagram of the processed light radiation and temperature data showing the signal influence of the light radiation and temperature data on the photovoltaic inverter;

[0046] Obtain the change trend of the signal propagation influence of the light radiation and temperature data on the photovoltaic inverter according to the scatter diagram of the signal influence;

[0047] Perform quantitative processing on the change trend based on regression analysis to obtain the light radiation influence factor and temperature influence factor respectively.

[0048] Among them, generally, the stronger the light radiation, the higher the output voltage. The influence of temperature on the photovoltaic system is usually negative. The higher the temperature, the lower the efficiency of the photovoltaic module, resulting in a voltage drop.

[0049] The specific calculation formulas for the light radiation influence factor and temperature influence factor are as follows:

[0050]

[0051]

[0052] The sensitivity coefficient of the photovoltaic inverter is specifically calculated as follows:

[0053]

[0054] In the formula, S is the sensitivity coefficient of the photovoltaic inverter, is the response time interval, is the light radiation influence factor, is the temperature influence factor, is the standard data of the sensitivity of the photovoltaic inverter, is an adjustable constant, is the light radiation, is the light radiation weight coefficient, is the temperature data, is the temperature data weight coefficient.

[0055] It should be noted that the adjustable constant reflects the sensitivity of the photovoltaic inverter to voltage changes. When analyzing and extracting the characteristics of light radiation and temperature data around the photovoltaic inverter, the purpose is to identify and quantify the impact of these environmental factors on the output voltage and performance of the photovoltaic system. By analyzing the response time interval between when the photovoltaic inverter receives the input signal and then outputs the signal, the sensitivity of the photovoltaic inverter can be better reflected. The longer the response time, the greater the impact of the photovoltaic inverter on voltage sag at this time, and the lower the sensitivity.

[0056] The stability coefficient of the photovoltaic inverter refers to that when there is a voltage sag, it will affect the output efficiency of the inverter and its own adaptability, resulting in instability of the photovoltaic inverter and a decline in the low-voltage ride-through operation performance of the photovoltaic system.

[0057] The method for obtaining the stability coefficient of the photovoltaic inverter is as follows:

[0058] Obtain the waveform diagram of the output power of the photovoltaic inverter;

[0059] Based on the waveform diagram of the output power, analyze the change trend of the output power before and after the voltage sag according to the preset standard output power to obtain the efficiency of the power conversion unit;

[0060] Obtain the working state of the historical protection device in the photovoltaic inverter under voltage sag;

[0061] Based on the working state data of the historical protection device, perform regression analysis to obtain the influencing factors of the working state of the protection device. The influencing factors of the working state include the fault repair data of the protection device and the aging rate of the protection device;

[0062] Calculate the sensitivity coefficient of the protection device based on the fault repair data and the aging rate of the protection device according to the preset sensitivity formula of the protection device;

[0063] The stability coefficient of the photovoltaic inverter is calculated based on a preset photovoltaic inverter stability formula for the efficiency of the power conversion unit and the sensitivity coefficient of the protection device.

[0064] The specific calculation formula for the sensitivity coefficient of the protection device is as follows:

[0065]

[0066] The specific calculation formula for the stability coefficient of the photovoltaic inverter is as follows:

[0067]

[0068] In the formula, is the stability coefficient of the photovoltaic inverter, is the efficiency of the power conversion unit, is the sensitivity coefficient of the protection device, is the number of faults of the photovoltaic inverter, is the maximum fault tolerance of the photovoltaic inverter, is the fault repair data, is the aging rate of the protection device, is the weight coefficient of the fault repair data, is the weight coefficient of the aging rate of the protection device.

[0069] It should be noted that voltage dips may affect the maximum power point tracking of the power conversion unit of the inverter for photovoltaic modules, resulting in the inability to adjust to the optimal working state in a timely manner, thereby reducing power output and affecting the stability of the photovoltaic inverter. When a voltage dip occurs in the power grid, if the sensitivity of the protection setting of the inverter decreases, it may cause the system to be disconnected erroneously, affecting the stability of the photovoltaic system and the power supply. Therefore, analyzing the efficiency of the power conversion unit of the photovoltaic module and the sensitivity coefficient of the protection device plays an important role in analyzing the stability of the protection device under voltage dips.

[0070] Grid anomaly impact data refers to data used to evaluate the impact of voltage dips on grid stability. When the voltage dip depth is relatively large, it will cause load fluctuations, a sharp change in power demand, resulting in grid load imbalance, thereby causing problems such as overloaded grid equipment not working properly and grid system instability.

[0071] Analyzing the grid anomaly impact data has the following advantages for evaluating the state of the photovoltaic system under voltage dips:

[0072] Adaptive Control Optimization for Photovoltaic Systems: Inverters in photovoltaic systems usually automatically adjust the output power according to the changes in grid voltage to ensure synchronization with the grid. By analyzing the data on the impact of grid anomalies, it is possible to help design more intelligent adaptive control strategies, enabling the inverter to take more appropriate response measures when sudden situations such as voltage dips occur. This optimization can not only improve the adaptability of the system in an unstable grid environment but also reduce the negative impact of the photovoltaic system on the grid.

[0073] Improving Power Quality and Grid Coordination: Analyzing grid anomaly data helps in the coordination between the photovoltaic system and the grid. By real-time monitoring of the voltage fluctuations in the grid, the photovoltaic power generation system can perform more precise load scheduling with the grid to mitigate the impact of voltage dips on photovoltaic power generation. This not only improves the power generation efficiency of the photovoltaic system but also helps to improve the overall power quality of the grid and reduce the impact of voltage dips on other users.

[0074] Enhancing System Stability: Grid anomaly events can have a direct impact on the performance of the photovoltaic power generation system. The photovoltaic inverter may experience power outages or performance degradation during voltage dips. By analyzing the grid anomaly data, it is possible to identify in advance the potential fluctuation trends in the grid and adjust the operation strategy of the photovoltaic system, thereby reducing system downtime or power reduction caused by grid problems and enhancing the overall stability of the system.

[0075] The data on the impact of grid anomalies is obtained specifically as follows:

[0076] Obtain the status data of grid equipment and get the load data of grid equipment based on feature analysis and extraction;

[0077] Perform differential analysis on the load data of grid equipment based on the preset stable load data to obtain the load difference value of grid equipment;

[0078] Monitor the synchronization of different regions of the grid system to obtain the synchronization rate of the grid system;

[0079] Calculate the data on the impact of grid anomalies based on the load difference value of grid equipment and the synchronization rate of the grid system using the preset formula for the data on the impact of grid anomalies.

[0080] The specific calculation formula for the data on the impact of grid anomalies is as follows:

[0081]

[0082] In the formula, is the data on the impact of grid anomalies, is the load difference value of grid equipment, is the stable load of the th device, where, = 1, 2, 3, ..., R, where R is an integer and N is the total number of devices, is the load difference of the th device, and is the synchronization rate of the power grid system.

[0083] It should be noted that the differential analysis of the load data of power grid devices based on the preset stable load data can be understood as follows: when the voltage drops, the load stability of power grid devices is affected, and the load difference value of the power grid devices is obtained. The larger the load difference value of the power grid devices, the more unstable the power grid stability at this time. The synchronization rate of the power grid system can be understood as follows: when the synchronization rate of the power grid system is larger, it means that the transmission synchronization rate between the power grids is consistent at this time, and any deviation will affect the instability of the power grid.

[0084] S3. Based on the performance impact data and the power grid anomaly impact data, construct a photovoltaic system voltage drop state evaluation model using machine learning.

[0085] Obtain the performance impact data and the power grid anomaly impact data, train the performance impact data and the power grid anomaly impact data using machine learning to obtain the proportion weights of the performance impact data and the power grid anomaly impact data on the entire photovoltaic system under voltage drop, and automatically adjust and update the weights to construct a photovoltaic system voltage drop state evaluation model. It should be noted that machine learning here is an existing technology and will not be elaborated too much.

[0086] The specific calculation formula of the voltage drop state evaluation model is as follows:

[0087]

[0088] In the formula, is the voltage drop state evaluation coefficient, is the power grid anomaly impact data, S is the photovoltaic inverter sensitivity coefficient, is the photovoltaic inverter stability coefficient, is the power grid anomaly impact data weight coefficient, is the photovoltaic inverter sensitivity coefficient weight coefficient, is the photovoltaic inverter stability coefficient weight coefficient.

[0089] It should be noted that when the power grid anomaly impact data becomes smaller and smaller, the photovoltaic inverter sensitivity coefficient becomes larger and larger, and the photovoltaic inverter stability coefficient becomes larger and larger, the voltage drop state evaluation coefficient will also become larger and larger.

[0090] S4. Perform preliminary screening and secondary screening on the output of the photovoltaic system voltage drop state evaluation model to obtain the voltage drop preset value. The specific steps are as follows:

[0091] Obtain the voltage sag state evaluation coefficient according to the output of the voltage sag state evaluation model of the photovoltaic system;

[0092] Monitor the state of the protection device in the photovoltaic system, and obtain the voltage sag value that causes the photovoltaic system to frequently perform protection actions due to too small voltage sag;

[0093] Take the voltage sag state evaluation coefficient corresponding to the voltage sag value with frequent protection actions as the limit minimum value;

[0094] Obtain the voltage sag value that causes the inverter to disconnect due to too large voltage sag in the photovoltaic system;

[0095] Take the voltage sag state evaluation coefficient corresponding to the voltage sag value that causes the inverter to disconnect as the limit maximum value;

[0096] Obtain all voltage sag state evaluation coefficients and perform preliminary screening based on the limit minimum value and the limit maximum value;

[0097] Exclude the voltage sag state evaluation coefficients that are less than the limit minimum value and greater than the limit maximum value to obtain a preliminarily screened voltage sag state evaluation data set;

[0098] Calculate the average value of the voltage sag state evaluation data set, and set the average value as the voltage sag state evaluation stability value;

[0099] Based on the voltage sag state evaluation stability value, perform secondary screening on the voltage sag state evaluation data set through comparison, and based on the gradient descent method, obtain the best voltage sag state evaluation coefficient;

[0100] Set the voltage sag value corresponding to the best voltage sag state evaluation coefficient as the voltage sag preset value.

[0101] S5. According to the voltage sag preset value, judge and control the corresponding instruction set for the photovoltaic system to perform low-voltage ride-through operation. The specific steps are as follows:

[0102] If the voltage sag value is less than the voltage sag preset value, the photovoltaic system does not need to perform low-voltage ride-through operation, and the system monitors in real time;

[0103] If the voltage sag value is greater than or equal to the voltage sag preset value, the system automatically schedules to perform low-voltage ride-through operation on the photovoltaic system. The inverter maintains the grid voltage level through reactive power compensation, reduces the risk of further voltage drop, and stops the low-voltage ride-through operation of the photovoltaic system when the voltage returns to the standard operating voltage, and the photovoltaic inverter resumes normal operation.

[0104] Embodiment 2 Figure 2Schematic diagram of a low-voltage ride-through coordination system for a photovoltaic power grid provided by an embodiment of the present application, including a signal transmission module, a data acquisition module, a model construction module, a data screening module, and an instruction generation module:

[0105] The signal transmission module is used to monitor the voltage fluctuation in the photovoltaic system in real time, obtain a voltage steady-state coefficient based on the voltage fluctuation, and send a voltage dip signal to the photovoltaic system if the voltage is lower than the voltage steady-state coefficient;

[0106] The data acquisition module is used to obtain multiple segments of voltage dip values based on the voltage dip signal, and obtain the performance impact data of the photovoltaic inverter and the grid anomaly impact data under each segment of voltage dip value;

[0107] The model construction module is used to construct an evaluation model for the voltage dip state of the photovoltaic system based on the performance impact data and the grid anomaly impact data by machine learning;

[0108] The data screening module is used to perform a preliminary screening and a secondary screening on the output of the evaluation model for the voltage dip state of the photovoltaic system to obtain a preset voltage dip value;

[0109] The instruction generation module is used to judge according to the preset voltage dip value and control the corresponding instruction set for the photovoltaic system to perform a low-voltage ride-through operation.

[0110] The above formulas are all dimensionless and take their numerical values for calculation. The formula is obtained by collecting a large amount of data and performing software simulation to obtain a formula that is closest to the actual situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0111] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product.

[0112] Those of ordinary skill in the art can realize that the modules and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0113] In addition, the functional modules in each embodiment of the present application can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0114] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims described above.

[0115] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A photovoltaic power grid low voltage ride through coordination method, characterized in that: The steps include: Real-time monitoring of voltage fluctuations in the photovoltaic system, obtaining a voltage steady-state coefficient based on the voltage fluctuations, and sending a voltage drop signal to the photovoltaic system if the voltage is lower than the voltage steady-state coefficient; Based on the voltage drop signal, multiple voltage drop values ​​are obtained, and the performance impact data of the photovoltaic inverter and the grid abnormality impact data under each voltage drop value are obtained; According to the performance impact data and grid abnormality impact data, a photovoltaic system voltage drop status assessment model is constructed based on machine learning, specifically: In the formula, is the voltage drop status assessment coefficient, is the grid abnormality impact data, S is the photovoltaic inverter sensitivity coefficient, is the photovoltaic inverter stability factor, is the weight coefficient of power grid abnormality impact data, is the sensitivity coefficient of the photovoltaic inverter, is the weight coefficient of the stability factor of the photovoltaic inverter; Obtain the voltage drop value that causes the photovoltaic system to frequently perform protection actions, and use the corresponding voltage drop state assessment coefficient as the limit minimum value; Obtain the voltage drop value that causes the inverter to disconnect, and use the corresponding voltage drop state assessment coefficient as the maximum limit value; The voltage drop state assessment coefficients that are less than the limit minimum value and greater than the limit maximum value are excluded to obtain a preliminary screened voltage drop state assessment data set; The data set is averaged to obtain a voltage drop state assessment stable value, and secondary screening is performed based on a gradient downward method and compared with the voltage drop state assessment data set to obtain an optimal voltage drop state assessment coefficient, and the corresponding voltage drop value is set as a voltage drop preset value; The corresponding instruction set is used to judge and control the photovoltaic system to perform low voltage ride-through operation according to the preset voltage drop value.

2. A photovoltaic power grid low voltage ride through coordination method according to claim 1, characterized in that: The performance influencing data of the photovoltaic inverter includes the photovoltaic inverter sensitivity coefficient. The specific method for obtaining the photovoltaic inverter sensitivity coefficient is as follows: Obtain the input signal of the photovoltaic inverter and the light radiation and temperature data around the photovoltaic inverter under the current voltage drop; Performing data analysis and feature extraction on the light radiation and temperature data around the photovoltaic inverter to obtain a light radiation influencing factor and a temperature influencing factor that affect the propagation of the voltage signal; Acquiring an output signal waveform of the photovoltaic inverter based on the photovoltaic inverter input signal; The output signal waveform is analyzed to obtain the response time interval between when the photovoltaic inverter receives the input signal and when it outputs the signal; The photovoltaic inverter sensitivity coefficient is calculated based on the response time interval combined with the light radiation influence factor and the temperature influence factor based on the preset photovoltaic inverter sensitivity formula.

3. A photovoltaic power grid low voltage ride through coordination method according to claim 2, characterized in that: The specific steps of performing data analysis and feature extraction on the light radiation and temperature data around the photovoltaic inverter to obtain the light radiation influencing factor and the temperature influencing factor that affect the propagation of the voltage signal are as follows: Collect light radiation and temperature data for feature extraction, obtain weight coefficients of light radiation and temperature data, and remove random fluctuations in the data based on a smoothing algorithm; The processed light radiation and temperature data are used to draw a scatter diagram of the influence of light radiation and temperature data on the signal of the photovoltaic inverter; Obtain the changing trend of the influence of light radiation and temperature data on the signal propagation of photovoltaic inverters based on the scatter plot of signal influence; The change trend is quantified based on regression analysis to obtain the light radiation influence factor and the temperature influence factor respectively.

4. A photovoltaic power grid low voltage ride through coordination method according to claim 1, characterized in that: The performance influencing data of the photovoltaic inverter includes a photovoltaic inverter stability coefficient. The photovoltaic inverter stability coefficient is specifically obtained in the following manner: Get the waveform of the photovoltaic inverter output power; Based on the output power waveform, the efficiency of the power conversion unit is obtained by analyzing the change trend of the output power before and after the voltage drop according to the preset standard output power; Obtain the working status of the historical protection device in the photovoltaic inverter during voltage drop; Performing regression analysis based on historical working status data of the protection device to obtain working status influencing factors of the protection device, wherein the working status influencing factors include fault repair data of the protection device and an aging rate of the protection device; The fault repair data and the aging rate of the protection device are calculated based on the preset protection device sensitivity formula to obtain the protection device sensitivity coefficient; The efficiency of the power conversion unit and the sensitivity coefficient of the protection device are calculated based on a preset photovoltaic inverter stability formula to obtain a photovoltaic inverter stability coefficient.

5. A photovoltaic power grid low voltage ride through coordination method according to claim 1, characterized in that: The specific method for obtaining the power grid abnormality impact data is as follows: Obtain the status data of power grid equipment, and obtain the load data of power grid equipment based on feature analysis and extraction; Performing differential analysis on the load data of the power grid equipment based on preset stable load data to obtain a load differential value of the power grid equipment; Monitor the synchronization of different areas of the power grid system and obtain the synchronization rate of the power grid system; The grid abnormality impact data is calculated based on the preset grid abnormality impact data formula according to the load difference value of the grid equipment and the synchronization rate of the grid system.

6. A photovoltaic power grid low voltage ride through coordination method according to claim 1, characterized in that: The specific steps of the corresponding instruction set for judging and controlling the photovoltaic system to perform low voltage ride-through operation according to the voltage drop preset value are as follows: If the voltage drop value is less than the voltage drop preset value, the photovoltaic system does not perform low voltage ride-through operation, and the system performs real-time monitoring; If the voltage drop value is greater than or equal to the voltage drop preset value, the system automatically schedules the photovoltaic system to perform low voltage ride-through operation.

7. A photovoltaic power grid low voltage ride through coordination method according to claim 2, characterized in that: The specific calculation formulas of the light radiation influence factor and the temperature influence factor are as follows: The specific calculation formula of the photovoltaic inverter sensitivity coefficient is as follows: In the formula, S is the sensitivity coefficient of the photovoltaic inverter, is the response time interval, is the light radiation influencing factor, is the temperature influence factor, is the standard data of photovoltaic inverter sensitivity, is an adjustable constant, is the light radiation, is the light radiation weight coefficient, is the temperature data, is the temperature data weight coefficient.

8. A system using a photovoltaic power grid low voltage ride through coordination method according to any one of claims 1 to 7, characterized in that: It includes signal transmission module, data acquisition module, model building module, data screening module and instruction generation module: A signal transmission module is used to monitor voltage fluctuations in the photovoltaic system in real time, obtain a voltage steady-state coefficient based on the voltage fluctuations, and send a voltage drop signal to the photovoltaic system if the voltage is lower than the voltage steady-state coefficient; A data acquisition module is used to obtain multiple voltage drop values ​​based on the voltage drop signal, and obtain performance impact data of the photovoltaic inverter and power grid abnormality impact data under each voltage drop value; A model building module is used to build a photovoltaic system voltage drop state assessment model based on machine learning according to performance impact data and grid abnormality impact data; A data screening module, used for performing preliminary screening and secondary screening on the output of the photovoltaic system voltage drop state assessment model to obtain a voltage drop preset value; The instruction generation module is used to judge and control the corresponding instruction set of the photovoltaic system to perform low voltage ride-through operation according to the preset value of voltage drop.

Images