Energy storage inverter parallel operation stability control method

Abstract

The invention relates to the technical field of energy storage inverters, in particular to an energy storage inverter parallel operation stability control method, which comprises the steps of training a load prediction model based on historical load data and environmental data of a first region, and obtaining a load prediction demand and a total load demand of the first region in a first time; based on the historical power generation data of the renewable energy sources in the first region, total power generation output in the first time is predicted; when the total power generation output is greater than the total load demand, redundant electric quantity is stored through an energy storage inverter; when the total power generation output is smaller than the total load demand, the energy storage inverter discharges to provide supplementary power; the working state of the energy storage inverter is obtained in real time, and when one or more energy storage inverters break down, the fault energy storage inverter is disconnected from the parallel connection group by adopting a fault isolation mechanism; obtaining the health degree of the remaining energy storage inverters, and distributing output power for each energy storage inverter; and the power distribution strategy of the energy storage inverter is adjusted in real time according to the load prediction demand.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage inverters, and particularly to a method for controlling the parallel operation stability of energy storage inverters. Background Art

[0002] In traditional power systems, electricity usually comes from centralized power plants and is dispatched and distributed through the power grid. However, in remote areas, due to the high costs of power grid construction and maintenance, these areas often rely on energy storage systems and distributed energy systems to provide electricity. To ensure the efficient and stable operation of the system, the parallel operation of energy storage inverters has gradually become an important means to solve this problem, especially in the case of large fluctuations in load demand and unstable renewable energy supply.

[0003] The parallel operation of energy storage inverters can improve the reliability and stability of the system. However, in practical applications, many challenges still exist. Especially when facing large load fluctuations, uneven load demands, and fluctuations in renewable energy generation, the existing parallel control methods for energy storage inverters still have deficiencies in terms of system stability and reliability. At the same time, traditional methods for power distribution of energy storage inverters often ignore the health status of the inverters, resulting in some inverters operating overloaded or idling, which affects the stability of the system.

[0004] Therefore, a method for controlling the parallel operation stability of energy storage inverters is proposed. Summary of the Invention

[0005] The purpose of the present invention is to provide a method for controlling the parallel operation stability of energy storage inverters. Based on the historical load data and environmental data of the first region, a load prediction model is trained to obtain the load prediction demand and total load demand in the first region within the first time period; based on the historical power generation data of renewable energy in the first region, the total power generation output within the first time period is predicted; when the total power generation output is greater than the total load demand, the excess power is stored through the energy storage inverter; when the total power generation output is less than the total load demand, the energy storage inverter discharges to provide supplementary power; the working state of the energy storage inverter is obtained in real time. When one and / or more energy storage inverters fail, a fault isolation mechanism is used to disconnect the faulty energy storage inverter from the parallel group; the health status of the remaining energy storage inverters is obtained, and the output power is allocated to each energy storage inverter; and according to the load prediction demand, the power distribution strategy of the energy storage inverter is adjusted in real time.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for controlling the parallel operation stability of energy storage inverters, comprising:

[0008] Train a load prediction model based on the historical load data and environmental data of the first region, and predict the load demand of the first region within the first time period through the load prediction model to obtain the load prediction demands and the total load demand at different time scales within the first time period;

[0009] Generate the total power generation output within the first time period based on the historical power generation data of renewable energy within the first region;

[0010] When the total power generation output is greater than the total load demand, store the excess power through the energy storage inverter; when the total power generation output is less than the total load demand, discharge the energy storage inverter to provide supplementary power;

[0011] The specific process of the energy storage inverter discharging to provide supplementary power is as follows:

[0012] Obtain the working status of the energy storage inverter in real time, and judge whether the energy storage inverter has a fault through the operating parameters and safety thresholds of the energy storage inverter; when one and / or more energy storage inverters have a fault, use the fault isolation mechanism to disconnect the faulty energy storage inverter from the parallel group and obtain the health of the remaining energy storage inverters; when the energy storage inverter has no fault, obtain the health of the energy storage inverter;

[0013] Allocate output power to each parallel energy storage inverter through the health of the energy storage inverter to obtain an output power matrix;

[0014] Dynamically adjust the output power matrix according to the health of the energy storage inverter obtained in real time, and adjust the output power of each energy storage inverter;

[0015] According to the health of the energy storage inverter and the load prediction demand, adjust the power distribution strategy of each energy storage inverter in real time.

[0016] Preferably, the historical load data and environmental data are the load demand, temperature, humidity, wind speed, solar radiation intensity, and rainfall within the same time period;

[0017] The power generation data is the power generation output data within the same time period;

[0018] The historical load data, environmental data, and power generation data include load curves and environmental curves at different time scales;

[0019] The historical load data, environmental data, and power generation data are obtained through the power company and meteorological station in the first region.

[0020] Preferably, the load prediction model includes a historical load data analysis unit, an environmental data analysis unit, a time feature analysis unit, and a load prediction unit;

[0021] The historical load data analysis unit analyzes the historical load data through moving average to obtain load data information; the load data information includes the long-term load trend and the load cycle;

[0022] The environmental data analysis unit obtains the correlation between environmental factors and load demand through correlation analysis;

[0023] The time feature analysis unit generates time tags through time window division;

[0024] The load prediction unit comprehensively analyzes the load data information, the correlation between environmental factors and load demand, and the time tags to generate a load prediction demand, and smooths the load prediction demand to obtain a load demand prediction value.

[0025] Preferably, the load prediction model includes a data preprocessing unit, an LSTM model training unit, and a load prediction unit;

[0026] The data preprocessing unit includes data cleaning, standardization, and feature extraction; extracting time features, lag features, and moving average;

[0027] The LSTM model training unit includes an input layer, a hidden layer, a fully connected layer, and an output layer; the input layer is a multi-dimensional feature vector of historical load data and environmental data; the hidden layer includes n LSTM hidden layers, and each layer contains m LSTM units, which are used to obtain complex time dependencies and non-linear features; the fully connected layer generates the final load prediction value by connecting the output of the LSTM layer to a fully connected layer; the output layer outputs the load demand prediction values for multiple future moments;

[0028] The load prediction unit is used to obtain the total load demand and smooth the total load demand;

[0029] The total power generation output is obtained by predicting the power generation data through an LSTM network.

[0030] Preferably, the specific steps for judging the failure of the energy storage inverter are as follows:

[0031] Convert the collected operating parameters of each energy storage inverter into digital signals and perform preprocessing, data synchronization, and data storage; the operating parameters include voltage, current, frequency, temperature, power factor, and harmonic content;

[0032] Compare the real-time operating parameters with the safety thresholds of the preset operating parameters for anomaly determination; the anomaly determination includes single-parameter anomaly and multi-parameter anomaly;

[0033] Perform parameter correlation analysis on the anomaly determination for anomaly confirmation.

[0034] Preferably, the formula for calculating the health degree of the energy storage inverter is:

[0035]

[0036] where α V is the voltage weight, V is the current output voltage, V nominal is the rated output voltage, V max is the voltage upper limit, V min is the voltage lower limit, α I is the current weight, I is the current output current, I max is the rated output current, α f is the output frequency weight, f is the current output frequency, f nominal is the rated frequency, △f is the allowable frequency deviation range, α T is the temperature weight, HI T is the temperature, α PT is the power factor weight, PT is the current power factor, α HD is the harmonic distortion rate weight, HD is the current harmonic distortion rate, α OT is the running time weight, OT is the cumulative running time, e is the natural constant, λ is the running time decay coefficient, α FC is the failure times weight, FC is the number of failures, μ is the failure times decay coefficient, T is the current internal temperature, T max is the temperature upper limit.

[0037] Preferably, the specific process of allocating output power to each parallel energy storage inverter is as follows:

[0038] S10. Normalize the health degree of each energy storage inverter to obtain the first health degree of the energy storage inverter;

[0039] S20. Calculate the first output power allocated to each energy storage inverter according to the first health degree;

[0040] S30. Determine whether the first output power exceeds the rated power of the energy storage inverter;

[0041] S40. If an output power does not exceed the rated power, execute according to the allocated power;

[0042] S50. If the first output power exceeds the rated power, obtain the second output power that exceeds the range; obtain the first energy storage inverter whose output power does not exceed the rated power, and allocate output power to the first energy storage inverter according to the second output power;

[0043] S60. Obtain the output powers of all energy storage inverters to obtain an output power matrix.

[0044] Preferably, the process of obtaining the output power allocated to each parallel energy storage inverter also includes:

[0045] Obtain the health of each energy storage inverter in real time, and dynamically adjust the output power matrix according to the health;

[0046] Adjusting the output power of the energy storage inverter according to the output power matrix;

[0047] According to the load forecasting requirements of different time scales within the first time output by the load forecasting model, the power allocation strategy is adjusted in advance; the adjustment strategy includes:

[0048] S100. Based on the load curves of different time scales of the load forecast demand, a higher proportion of output power is preferentially allocated to energy storage inverters with higher health;

[0049] S200. Dynamically optimize the power allocation ratio of each energy storage inverter according to the real-time health change of the energy storage inverter;

[0050] S300. When predicting an increase in future load demand, increase the output power of the energy storage inverter with high health in advance;

[0051] S400. When it is predicted that the future load demand will decrease, the output power of some energy storage inverters is reduced and energy storage charging operations are performed.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] 1. The present invention obtains historical load data and environmental data of the first region and adopts a long-short-term memory network load prediction model to predict load demand with high accuracy. The accuracy of the prediction is improved by obtaining the long-short-term dependency and nonlinear characteristics in the load data. Based on the load prediction capability, the energy storage inverter can more reasonably allocate the output power, optimize the utilization efficiency of power resources, and improve the overall operation stability of the system.

[0054] 2. The present invention provides multi-parameter comprehensive monitoring and health assessment, and identifies the operating status and potential faults of the inverter by real-time monitoring of key operating parameters such as voltage, current, frequency, and temperature of the energy storage inverter. Multi-parameter analysis improves the accuracy and response speed of fault detection, and reduces the incidence of false alarms and missed alarms. When an energy storage inverter fails, the fault isolation mechanism is automatically triggered to quickly disconnect the faulty inverter from the parallel group, ensuring that other healthy inverters can adjust power output in a timely manner, thereby improving the stability of continuous operation.

[0055] 3. Through the dynamic evaluation of the health of the energy storage inverter, the present invention realizes the load balancing and optimal allocation among energy storage inverters; the output power is allocated according to the health of each inverter, thereby extending the service life of the inverter and improving the overall performance of the energy storage inverter. In addition, combined with the load prediction results, the power distribution strategy is adjusted in advance to optimize the response time and resource utilization rate, further improving the operation efficiency and stability of the energy storage system in a complex power environment. Description of the Drawings

[0056] Figure 1 It is a schematic flow chart of a method for controlling the parallel operation stability of an energy storage inverter provided by the present invention;

[0057] Figure 2 It is a flow chart of the parallel operation stability of the energy storage inverter provided by the embodiment of the present invention;

[0058] Figure 3 It is a flow chart for allocating output power to the energy storage inverter provided by the embodiment of the present invention. Detailed Embodiments

[0059] 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 of 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.

[0060] Embodiment 1

[0061] As an implementation manner of the present invention, referring to Figure 1 and Figure 2 , the present invention provides a method for controlling the parallel operation stability of an energy storage inverter, and the technical solution is as follows:

[0062] Based on the historical load data and environmental data of the first region, a load prediction model is trained, and the load demand in the first region within the first time is predicted through the load prediction model to obtain the load prediction demand and the total load demand within the first time; wherein the first time is consistent with the time range of the load prediction demand;

[0063] Based on the historical power generation data of renewable energy in the first region, the total power generation output within the first time is predicted and obtained;

[0064] Furthermore, the historical load data and environmental data are the load demand, temperature, humidity, wind speed, solar radiation intensity, and rainfall within the same time;

[0065] The power generation data is the power generation output data within the same time;

[0066] The historical load data, environmental data, and power generation data include load curves and environmental curves at different time scales; the load curves at different time scales include load curves for hours, days, and months.

[0067] The historical load data, environmental data, and power generation data are obtained through the power company and meteorological station in the first region.

[0068] In this embodiment, the present invention captures both short-term fluctuations and long-term trends by fusing data at multiple time scales of hours, days, and months, ensuring the comprehensiveness of the prediction results. And through the historical operation and meteorological data obtained by the power company and meteorological station, the data is accurate, laying a foundation for model training and the health state assessment of the energy storage inverter.

[0069] Furthermore, the load prediction model includes a historical load data analysis unit, an environmental data analysis unit, a time feature analysis unit, and a load prediction unit.

[0070] The historical load data analysis unit analyzes the historical load data through moving average to obtain load data information; the load data information includes the long-term trend of the load and the load cycle.

[0071] The environmental data analysis unit obtains the correlation between environmental factors and load demand through correlation analysis.

[0072] The time feature analysis unit generates time tags through time window partitioning.

[0073] The load prediction unit generates a load prediction demand through comprehensive analysis of the load data information, the correlation between environmental factors and load demand, and the time tags, and performs smoothing processing on the load prediction demand to obtain a load demand prediction value.

[0074] In the present invention, the load prediction model divides data analysis into different functional units, facilitating system integration; through the integration of multiple data sources such as historical load, environmental factors, and time features, it improves the processing ability of the prediction model for multi-dimensional information; and performs smoothing processing on the prediction results, effectively reducing the noise interference caused by short-term fluctuations, making subsequent scheduling more stable and reliable.

[0075] Furthermore, the load prediction model includes a data preprocessing unit, an LSTM model training unit, and a load prediction unit.

[0076] The data preprocessing unit includes data cleaning, standardization, and feature extraction; extracting time features, lag features, and moving average.

[0077] The LSTM model training unit includes an input layer, a hidden layer, a fully connected layer, and an output layer; the input layer inputs multi-dimensional feature vectors of historical load data and environmental data; the hidden layer includes two layers of LSTM hidden layers, each layer containing 50 LSTM units, which are used to obtain complex time-dependent relationships and non-linear features; the fully connected layer generates the final load prediction value by connecting the output of the LSTM layer to a fully connected layer; the output layer outputs the predicted values of the load demands at multiple future moments; in this embodiment, n is 2 and m is 50;

[0078] The load prediction unit is used to obtain the total load demand and smooth the total load demand;

[0079] The total power generation output is obtained by predicting the power generation data through an LSTM network.

[0080] To further illustrate the advantages of the above LSTM model in load prediction, Table 1 below gives the comparison results of the present invention and other traditional prediction methods in terms of prediction accuracy and operation performance;

[0081] Table 1 Load prediction accuracy

[0082] Method MAPE (%) Prediction time (s) The present invention 2.1 0.15 Linear regression 7.9 0.06 Traditional RNN 5.6 0.08

[0083] As can be seen from Table 1, under the same dataset and hardware test environment, the LSTM model adopted by the present invention has a significant improvement in prediction accuracy, and at the same time, the prediction time remains within an acceptable range.

[0084] Judge whether the total power generation output is greater than the total load demand; when the total power generation output is greater than the total load demand, store the excess power through the energy storage inverter; when the total power generation output is less than the total load demand, discharge the energy storage inverter to provide supplementary power;

[0085] The specific process of the energy storage inverter discharging to provide supplementary power is as follows:

[0086] Obtain the working state of the energy storage inverter in real time and judge whether the energy storage inverter has a fault; when one and / or more energy storage inverters have a fault, use the fault isolation mechanism to disconnect the faulty energy storage inverter from the parallel group and obtain the health of the remaining energy storage inverters; when the energy storage inverter has no fault, obtain the health of the energy storage inverter.

[0087] Further, the specific steps for judging whether the energy storage inverter has a fault are as follows:

[0088] Convert the collected operation parameters of each energy storage inverter into digital signals and perform preprocessing, data synchronization, and data storage; the operation parameters include voltage, current, frequency, temperature, power factor, and harmonic content;

[0089] Anomaly determination is performed by comparing real-time operating parameters with the safety thresholds of preset operating parameters; the anomaly determination includes single-parameter anomalies and multi-parameter anomalies;

[0090] Parameter correlation analysis is performed on the anomaly determination for anomaly confirmation.

[0091] The safety thresholds refer to the technical manuals of energy storage inverter manufacturers and industry standards, and are also statistically optimized based on the long-term operating data of this system in remote areas. First, the upper and lower limit ranges of each parameter are initially determined according to standards such as rated voltage, temperature upper limit, and harmonic content; subsequently, by combining a large amount of historical data collected during actual operation, the fluctuation ranges of each parameter under normal and faulty states are compared and analyzed to calibrate the accuracy of the thresholds. When necessary, multi-level warning and alarm intervals can be set for key parameters such as voltage, current, and temperature. For example, a mild alarm is triggered when the voltage deviates from the rated value by 5%, and a serious anomaly is determined when it exceeds 10%. For occasions with harsh operating environments or large load fluctuations, the safety thresholds can also be updated regularly through an online learning mechanism to ensure that potential faults can still be accurately identified and timely warned when the inverter gradually ages or the external environment changes.

[0092] Compared with the traditional method of only monitoring voltage or current, the present invention monitors multi-dimensional information such as voltage, current, frequency, temperature, power factor, and harmonic content, and can identify faults more accurately; and through single-parameter or multi-parameter anomaly determination, faults are confirmed through correlation analysis, effectively reducing false alarms and missed alarms, and enhancing the system's security and fault self-healing ability.

[0093] To further illustrate the advantages of the multi-parameter fault detection method in terms of accuracy rate and false alarm rate, Table 2 below gives the comparison data between the present invention and different single-parameter or double-parameter detection methods.

[0094] Table 2 Fault Detection Sensitivity

[0095] Fault detection method Detection rate (%) False alarm rate (%) The present invention (multi-parameter fault detection) 98.0 1.0 Only monitor voltage 81.3 3.9 Monitor voltage and current 88.2 2.4 Only monitor output power 85.7 3.1

[0096] As can be seen from Table 2, the multi-parameter monitoring of the present invention has better performance than single-parameter or double-parameter monitoring in terms of detection rate and false alarm rate, greatly improving the accuracy and timeliness of fault identification.

[0097] Furthermore, the formula for calculating the health degree of the energy storage inverter is:

[0098]

[0099] where α V is the voltage weight, V is the current output voltage, V nominal is the rated output voltage, V max is the voltage upper limit, V min is the voltage lower limit, αI is the current weight, I is the current output current, I max is the rated output current, α f is the output frequency weight, f is the current output frequency, f nominal is the rated frequency, △f is the allowable frequency deviation range, α T is the temperature weight, HI T is the temperature, α PT is the power factor weight, PT is the current power factor, α HD is the harmonic distortion rate weight, HD is the current harmonic distortion rate, α OT is the running time weight, OT is the cumulative running time, e is the natural constant, λ is the running time decay coefficient, α FC is the number of faults weight, FC is the number of faults, μ is the number of faults decay coefficient, T is the current internal temperature, T max is the temperature upper limit.

[0100] The formula for calculating the health degree of the energy storage inverter provided by the present invention provides a scientific basis for subsequent dynamic power distribution by quantifying the health status of each energy storage inverter; by training and dynamically adjusting the weights of key parameters for different scenarios, the health degree evaluation process has higher adaptability.

[0101] Based on the health degree of the energy storage inverter, the output power is allocated to each parallel energy storage inverter; and according to the predicted load demand, the power distribution strategy of the energy storage inverter is adjusted in real time.

[0102] Further, for the specific process of allocating the output power to each parallel energy storage inverter, refer to Figure 3 ;

[0103] S10. Normalize the health degree of each energy storage inverter to obtain the first health degree of the energy storage inverter;

[0104] S20. Calculate the first output power allocated to each energy storage inverter according to the first health degree;

[0105] S30. Determine whether the first output power exceeds the rated power of the energy storage inverter;

[0106] S40. If an output power does not exceed the rated power, execute according to the allocated power;

[0107] S50. If the first output power exceeds the rated power, obtain the second output power exceeding the range; obtain the first energy storage inverter whose output power does not exceed the rated power, and allocate the output power to the first energy storage inverter according to the second output power;

[0108] S60. Obtain the output powers of all energy storage inverters to obtain an output power matrix.

[0109] In the present invention, by allocating greater power to inverters with high health and reducing the load or standby for inverters with low health, the overall efficiency of the system can be effectively improved and the equipment lifespan can be extended. When a new energy storage inverter is added or a faulty unit exits, the system can automatically complete reallocation after normalization.

[0110] Furthermore, the process of obtaining the output power allocated to each parallel energy storage inverter further includes:

[0111] Obtaining the health of each energy storage inverter in real time and dynamically adjusting the output power matrix based on the health;

[0112] Adjusting the output power of the energy storage inverter according to the output power matrix;

[0113] Based on the load prediction demands at different time scales within the first time output by the load prediction model, adjusting the power distribution strategy in advance; the adjustment strategies include:

[0114] S100. According to the load curves at different time scales of the load prediction demands, preferentially allocating a higher proportion of the output power to the energy storage inverters with higher health;

[0115] S200. Dynamically optimizing the power distribution ratio of each energy storage inverter according to the real-time health changes of the energy storage inverter to ensure the high efficiency and stability of the overall system operation;

[0116] S300. When it is predicted that the future load demand will increase significantly, increasing the output power of the energy storage inverters with high health in advance to meet the growth of the load demand;

[0117] S400. When it is predicted that the future load demand will decrease, appropriately reducing the output power of some energy storage inverters or performing energy storage charging operations to optimize the energy utilization efficiency.

[0118] In this embodiment, by combining the real-time health and the prediction model, the output power matrix can be automatically corrected during the system operation, improving the ability to adapt to emergencies and demand fluctuations; based on the prediction of the load and power generation, the system can perform prior scheduling to avoid energy waste or load collapse caused by delayed response.

[0119] The present invention accurately predicts the load demand and renewable energy power generation by combining long short-term memory networks, and introduces a multi-parameter fault detection and inverter health assessment mechanism into the parallel operation control of energy storage inverters. First, since the LSTM model can capture the non-linearity and long-term and short-term dependencies of time series data, the accuracy of load and power generation prediction is improved. Second, for the diverse fault risks that inverters often face in remote areas, the present invention monitors multi-dimensional parameters such as voltage, current, frequency, temperature, and harmonics and conducts correlation analysis to identify single-parameter or multi-parameter anomalies, and quickly isolates the faulty inverter after a fault occurs, minimizing the impact on the system. In addition, the overall operating state of the inverter is quantified through a health calculation formula, and a flexible power distribution strategy is implemented based on the health of each inverter, allocating smaller loads or backups to inverters with low health, thereby improving system efficiency and extending the service life of the equipment. Finally, by dynamically adjusting the power in real time by obtaining the health and combining the load prediction requirements, the present invention can also achieve the high efficiency and safety of the energy storage system in remote areas with limited resources and complex operating environments.

[0120] Embodiment 2

[0121] Train a load prediction model based on the historical load data and environmental data of the first region, and predict the load demand of the first region within the first time through the load prediction model to obtain the load prediction demand and the total load demand within the first time;

[0122] Based on the historical power generation data of renewable energy within the first region, predict and obtain the total power generation output within the first time;

[0123] Judge whether the total power generation output is greater than the total load demand; when the total power generation output is greater than the total load demand, store the excess power through the energy storage inverter; when the total power generation output is less than the total load demand, discharge the energy storage inverter to provide supplementary power;

[0124] The specific process of the energy storage inverter discharging to provide supplementary power is as follows:

[0125] Obtain the working state of the energy storage inverter in real time and judge whether the energy storage inverter has a fault; when one and / or more energy storage inverters have a fault, use a fault isolation mechanism to disconnect the faulty energy storage inverter from the parallel group and obtain the health of the remaining energy storage inverters; when the energy storage inverter has no fault, obtain the health of the energy storage inverter.

[0126] Further, the specific steps for judging whether the energy storage inverter has a fault are as follows:

[0127] Convert the collected operating parameters of each energy storage inverter into digital signals and perform preprocessing, data synchronization, and data storage; the operating parameters include voltage, current, frequency, temperature, power factor, and harmonic content;

[0128] Anomaly determination is performed by comparing real-time operating parameters with the safety thresholds of preset operating parameters; the anomaly determination includes single-parameter anomalies and multi-parameter anomalies;

[0129] Parameter correlation analysis is performed on the anomaly determination for anomaly confirmation.

[0130] Furthermore, the formula for calculating the health degree of the energy storage inverter is:

[0131]

[0132] where α V is the voltage weight, V is the current output voltage, V nominal is the rated output voltage, V max is the voltage upper limit, V min is the voltage lower limit, α I is the current weight, I is the current output current, I max is the rated output current, α f is the output frequency weight, f is the current output frequency, f nominal is the rated frequency, △f is the allowable frequency deviation range, α T is the temperature weight, HI T is the temperature, α PT is the power factor weight, PT is the current power factor, α HD is the harmonic distortion rate weight, HD is the current harmonic distortion rate, α OT is the operating time weight, OT is the cumulative operating time, λ is the operating time decay coefficient, α FC is the failure times weight, FC is the number of failures, μ is the failure times decay coefficient, T is the current internal temperature, T max is the temperature upper limit.

[0133] In this embodiment, by acquiring and normalizing the data of 5 energy storage inverters, the corresponding health degrees are obtained, and the specific data can be referred to Table 3;

[0134] Table 3 Health Degrees of Energy Storage Inverters

[0135] Sample number Voltage Current Frequency Temperature Harmonic Health degree 1 0.95 0.88 0.98 0.90 0.94 0.93 2 0.96 0.85 0.97 0.87 0.95 0.91 3 0.94 0.90 0.99 0.88 0.92 0.92 4 0.92 0.87 0.96 0.85 0.90 0.89 5 0.93 0.91 0.97 0.86 0.93 0.91

[0136] Output power is allocated to each parallel energy storage inverter according to the health degree of the energy storage inverter to obtain an output power matrix;

[0137] According to the health degree of the energy storage inverter obtained in real time, the output power matrix is dynamically adjusted, and the output power of each energy storage inverter is adjusted; according to the health degree of the energy storage inverter and the predicted load demand, the power distribution strategy of each energy storage inverter is adjusted in real time.

[0138] Furthermore, the specific process of allocating output power to each parallel energy storage inverter is as follows:

[0139] S10. Normalize the health degree of each energy storage inverter to obtain the first health degree of the energy storage inverter;

[0140] S20. Calculate the first output power allocated to each energy storage inverter according to the first health degree;

[0141] The formula for obtaining the first output power is:

[0142]

[0143] where P i is the first output power corresponding to the i-th energy storage inverter, HI i is the health degree of the i-th energy storage inverter, N is the total number of energy storage inverters, and P total is the total output power;

[0144] S30. Determine whether the first output power exceeds the rated power of the energy storage inverter;

[0145] S40. If an output power does not exceed the rated power, execute according to the allocated power;

[0146] S50. If the first output power exceeds the rated power, obtain the second output power exceeding the range; obtain the first energy storage inverter whose output power does not exceed the rated power, and allocate the output power to the first energy storage inverter according to the second output power;

[0147] S60. Obtain the output powers of all energy storage inverters to obtain an output power matrix.

[0148] Based on the 5 energy storage inverters provided in Table 3, power is allocated to them, and the data is referred to Table 4;

[0149] As can be seen from Table 4, the inverters with high health degrees undertake larger power allocations and have almost no overload; the inverters with relatively low health degrees undertake relatively less power or are prone to overload when the system pressure increases, thus highlighting the rationality and safety of the health degree-driven allocation strategy.

[0150] Table 4 Comparison table of power allocation efficiency driven by health degree

[0151] Sample number Health degree Allocated power Average load rate Overload times System efficiency 1 0.93 15.2 80.1 0 95.2 2 0.91 14.7 78.3 0 94.8 3 0.92 15.0 79.5 0 95.0 4 0.89 13.8 75.2 1 92.7 5 0.91 14.5 77.9 0 94.3

[0152] Furthermore, the process of obtaining the output power allocated to each parallel energy storage inverter further includes:

[0153] Obtain the health degree of each energy storage inverter in real time, and dynamically adjust the output power matrix for adjusting the output power of the energy storage inverter;

[0154] Adjust the power distribution strategy in advance according to the predicted load demand value output by the load prediction model.

[0155] The present invention improves the operation stability and efficiency of the energy storage inverter parallel system through multi-parameter health calculation, fault detection and LSTM load prediction. A high-precision prediction model is constructed using historical load and environmental data to achieve accurate prediction of multi-time scale demands; power is distributed according to the health of the inverter to avoid overload or expansion of faults; if a fault occurs, it is automatically isolated and redistributed to ensure continuous power supply. It comprehensively uses threshold self-adaptation and multi-parameter correlation analysis to effectively reduce false alarms and missed alarms, enhance the fault self-healing ability and the utilization rate of renewable energy, and is applicable to the power supply scenarios in remote areas.

[0156] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for controlling the stability of parallel operation of energy storage inverters, characterized in that: include: Training a load forecasting model based on historical load data and environmental data of the first region, forecasting the load demand of the first region within a first period of time by using the load forecasting model, and obtaining load forecasting demands and total load demands at different time scales within the first period of time; generating a total power generation output in a first time period based on historical power generation data of renewable energy sources in a first region; When the total power generation output is greater than the total load demand, the excess power is stored through the energy storage inverter; when the total power generation output is less than the total load demand, the energy storage inverter is discharged to provide supplementary power; The specific process of energy storage inverter discharging to provide supplementary power is: Obtain the working status of the energy storage inverter in real time, and judge the fault of the energy storage inverter through the operating parameters and safety thresholds of the energy storage inverter; when one or more energy storage inverters fail, use the fault isolation mechanism to disconnect the faulty energy storage inverter from the parallel group and obtain the health of the remaining energy storage inverters; when the energy storage inverter does not fail, obtain the health of the energy storage inverter; According to the health of the energy storage inverter, the output power is allocated to each parallel energy storage inverter to obtain the output power matrix; According to the real-time acquired health status of the energy storage inverter, the output power matrix is ​​dynamically adjusted, and the output power of each energy storage inverter is adjusted; According to the health of the energy storage inverter and the load forecast demand, the power allocation strategy of each energy storage inverter is adjusted in real time.

2. A method for controlling the stability of parallel operation of energy storage inverters according to claim 1, characterized in that: The historical load data and environmental data are load demand, temperature, humidity, wind speed, solar radiation intensity and rainfall within the same period of time; The power generation data is the power generation output data within the same period of time; The historical load data, environmental data and power generation data include load curves and environmental curves of different time scales; The historical load data, environmental data and power generation data are obtained through the power company and the weather station in the first area.

3. A method for controlling the stability of parallel operation of energy storage inverters according to claim 1, characterized in that: The load prediction model includes a historical load data analysis unit, an environmental data analysis unit, a time characteristic analysis unit and a load prediction unit; The historical load data analysis unit analyzes the historical load data by moving average to obtain load data information; the load data information includes a long-term load trend and a load cycle; The environmental data analysis unit obtains the correlation between the environmental factors and the load demand through correlation analysis; The time feature analysis unit generates a time tag by dividing the time window; The load prediction unit generates a load prediction demand by comprehensively analyzing the load data information, the correlation between the environmental factors and the load demand, and the time tag, and performs smoothing processing on the load prediction demand to obtain a load demand prediction value.

4. A method for controlling the stability of parallel operation of energy storage inverters according to claim 1, characterized in that: The load prediction model includes a data preprocessing unit, an LSTM model training unit and a load prediction unit; The data preprocessing unit includes data cleaning, standardization and feature extraction; extracting time features, lag features and moving average; The LSTM model training unit includes an input layer, a hidden layer, a fully connected layer and an output layer; the input layer includes a multi-dimensional feature vector of historical load data and environmental data; the hidden layer includes n layers of LSTM hidden layers, each layer includes m LSTM units; the fully connected layer generates the final load forecast value by connecting the output of the LSTM layer to a fully connected layer; the output layer outputs the load demand forecast value at multiple moments in the future; The load prediction unit is used to obtain the total load demand and perform smoothing on the total load demand; The total power generation output is obtained by predicting the power generation data through the LSTM network.

5. The method for controlling the stability of parallel operation of energy storage inverters according to claim 1, characterized in that: The specific steps to determine if the energy storage inverter has a fault are as follows: Convert the collected operating parameters of each energy storage inverter into digital signals and perform preprocessing, data synchronization and data storage; the operating parameters include voltage, current, frequency, temperature, power factor and harmonic content; By comparing the real-time operating parameters with the safety thresholds of the preset operating parameters, abnormality determination is performed; the abnormality determination includes single parameter abnormality and multi-parameter abnormality; The abnormality judgment is confirmed by performing parameter correlation analysis.

6. A method for controlling the stability of parallel operation of energy storage inverters according to claim 1, characterized in that: The calculation formula for the health of the energy storage inverter is: Among them, α V is the voltage weight, V is the current output voltage, V nominal is the rated output voltage, V max is the voltage upper limit, V min is the voltage lower limit, α I is the current weight, I is the current output current, I max is the rated output current, α f is the output frequency weight, f is the current output frequency, f nominal is the rated frequency, △f is the allowable frequency deviation range, α T is the temperature weight, HI T is temperature, α PT is the power factor weight, PT is the current power factor, α HD is the harmonic distortion weight, HD is the current harmonic distortion, α OT is the running time weight, OT is the cumulative running time, e is a natural constant, λ is the running time attenuation coefficient, α FC is the fault number weight, FC is the fault number, μ is the fault number attenuation coefficient, T is the current internal temperature, T max The upper temperature limit.

7. A method for controlling the stability of parallel operation of energy storage inverters according to claim 1, characterized in that: The specific process of allocating output power to each parallel energy storage inverter is as follows: S10. Normalizing the health of each energy storage inverter to obtain a first health of the energy storage inverter; S20. Calculate the first output power allocated to each energy storage inverter according to the first health level; S30. Determine whether the first output power exceeds the rated power of the energy storage inverter; S40. If an output power does not exceed the rated power, execute according to the allocated power; S50. If the first output power exceeds the rated power, obtain a second output power that exceeds the range; obtain a first energy storage inverter that does not exceed the rated power, and allocate output power to the first energy storage inverter according to the second output power; S60. Obtain the output power of all energy storage inverters and obtain an output power matrix.

8. A method for controlling the stability of parallel operation of energy storage inverters according to claim 7, characterized in that: The process of obtaining the output power distribution of each parallel energy storage inverter also includes: Obtain the health of each energy storage inverter in real time, and dynamically adjust the output power matrix according to the health; Adjusting the output power of the energy storage inverter according to the output power matrix; According to the load forecasting requirements of different time scales within the first time output by the load forecasting model, the power allocation strategy is adjusted in advance; the adjustment strategy includes: S100. Based on the load curves of different time scales of the load forecast demand, a higher proportion of output power is preferentially allocated to energy storage inverters with higher health; S200. Dynamically optimize the power allocation ratio of each energy storage inverter according to the real-time health change of the energy storage inverter; S300. When predicting an increase in future load demand, increase the output power of the energy storage inverter with high health in advance; S400. When it is predicted that the future load demand will decrease, the output power of some energy storage inverters is reduced and energy storage charging operations are performed.

Images