Energy storage inverter control method based on adaptive algorithm

Abstract

The invention provides an energy storage inverter control method based on an adaptive algorithm, and relates to the technical field of energy storage inverter control, and the method comprises the steps: monitoring the operation states of a power grid and a battery in real time, and obtaining the operation data of the battery and the operation data of the power grid; predicting a battery state based on the battery operation data to obtain a battery prediction result; evaluating a current power grid disturbance condition based on the power grid operation data to obtain a power grid disturbance result; and combining and analyzing the battery prediction result and the power grid evaluation result with an adaptive algorithm, and adjusting the output of an inverter. The method comprises the following steps: monitoring the operation states of a power grid and a battery in real time, obtaining battery operation data and power grid operation data, predicting the battery state and performing power grid disturbance analysis to obtain a battery prediction result and a power grid disturbance result; a battery prediction result, a power grid evaluation result and an adaptive algorithm are combined and analyzed to adjust output of the energy storage inverter, energy utilization and distribution can be optimized, system safety and stability are improved, and the service life of the battery is prolonged.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage inverter control, and in particular to an energy storage inverter control method based on an adaptive algorithm. Background Art

[0002] As a key interface device between the energy storage system and the power grid, the performance of the energy storage inverter directly affects the overall efficiency of the energy storage system and the stability of the power grid. However, traditional energy storage inverter control methods often use fixed control parameters and strategies, which are difficult to adapt to the complex and changeable power system environment. How to flexibly control the output of the energy storage inverter has become one of the current research focuses.

[0003] Therefore, the present invention provides an energy storage inverter control method based on an adaptive algorithm. Summary of the invention

[0004] The present invention provides an energy storage inverter control method based on an adaptive algorithm, which is used to monitor the operating status of a power grid and a battery in real time, obtain battery operating data and power grid operating data, predict battery status and power grid disturbance analysis, and obtain battery prediction results and power grid disturbance results; the battery prediction results and power grid evaluation results are combined with an adaptive algorithm for analysis to adjust the output of the energy storage inverter, which can optimize energy utilization and distribution, improve system safety and stability, and extend battery life.

[0005] The present invention provides an energy storage inverter control method based on an adaptive algorithm, comprising:

[0006] Step 1: Monitor the operating status of the power grid and battery in real time, and obtain battery operating data and power grid operating data;

[0007] Step 2: predicting a battery state based on the battery operation data to obtain a battery prediction result;

[0008] Step 3: Evaluate the current grid disturbance situation based on the grid operation data to obtain a grid disturbance result;

[0009] Step 4: Combine the battery prediction result with the grid assessment result and the adaptive algorithm for analysis, and adjust the inverter output.

[0010] Preferably, real-time monitoring of the operating status of the power grid and the battery, and acquisition of battery operating data and power grid operating data include:

[0011] Using preset sensors to monitor the operating status of the battery and the power grid in real time, and obtain first data and second data respectively;

[0012] After data cleaning and filtering are performed on the first data and the second data, they are output as battery operation data and power grid operation data respectively.

[0013] Preferably, predicting the battery state based on the battery operation data to obtain a battery prediction result includes:

[0014] According to the current battery type and application requirements, select and set the battery model and set the initial parameters;

[0015] A Kalman filter algorithm is introduced to perform model training on the set battery model based on historical operation data of a preset amount of the current battery extracted from a battery database to obtain a first prediction model;

[0016] Using preset test data to test the first prediction model to obtain a test result;

[0017] The test results are evaluated using a set effect evaluation index to obtain a prediction effect coefficient;

[0018] When the prediction effect coefficient is less than a set prediction threshold, adjusting the first key parameter in the first prediction model by using a set parameter optimization algorithm to obtain a battery prediction model;

[0019] When the prediction effect coefficient is not less than a set prediction threshold, using the first prediction model as a battery prediction model;

[0020] Inputting battery operation data into the battery prediction model and outputting battery prediction data;

[0021] Corresponding prediction data of performance-related parameters are extracted from the battery prediction data and output as a battery prediction result.

[0022] Preferably, the first key parameters refer to the state transfer matrix, the observation matrix and the initial state estimate.

[0023] Preferably, evaluating the current grid disturbance situation based on the grid operation data to obtain a grid disturbance result includes:

[0024] Extracting and integrating features of the power grid operation data using signal processing technology to obtain a first operation feature set;

[0025] Selecting a feature associated with a power grid disturbance from the first operation feature set and considering it as a first disturbance-associated feature;

[0026] Extracting and aggregating disturbance-related features of a preset amount of historical power grid operation data extracted from a power grid operation database to obtain a first feature training set;

[0027] Using the first feature training set as input, training the neural network to obtain a disturbance recognition model;

[0028] Inputting the first disturbance association feature into a disturbance identification model, and outputting a disturbance identification result;

[0029] According to the disturbance identification result, if there is a disturbance identification type, the disturbance is located based on the topological structure of the current power grid to determine the disturbance time and the disturbance impact range;

[0030] The disturbance identification type, disturbance time and disturbance impact range are output as power grid disturbance results.

[0031] Preferably, the features associated with the power grid disturbance are selected from the first operation feature set and regarded as the first disturbance-associated features, including:

[0032] Obtaining historical feature vectors of first operation features of the power grid before and after various disturbance types occur within a preset time period;

[0033] Based on the historical feature vectors, the historical key parameters of the first operating characteristics before and after each disturbance type occurs are calculated;

[0034] Determining a disturbance sensitivity coefficient of a current first operating characteristic based on historical key parameters;

[0035] The calculation formula of the disturbance sensitivity coefficient is as follows:

[0036] Where P represents the disturbance sensitivity coefficient of the current first operating characteristic; n represents the total number of disturbance types; s1 represents the occurrence coefficient of the i-th disturbance type within the preset time period; s0 represents the total number of disturbances occurring within the preset time period; x ij1 It is represented as the historical key parameters of the current first operating feature before the jth disturbance under the i-th disturbance type; x ij0 It is represented as the historical key parameter of the current first operating feature after the jth disturbance under the i-th disturbance type; ε j It is expressed as the contribution weight of historical key parameters to the disturbance;

[0037] Marking the first operating feature whose disturbance sensitivity coefficient exceeds the set sensitivity threshold as a possible feature;

[0038] Using Euclidean distance to calculate type difference values ​​of the possible features under different disturbance types;

[0039] Possible features whose type difference values ​​exceed the set difference threshold are regarded as the first disturbance-associated features.

[0040] Preferably, the battery prediction result and the grid assessment result are combined with an adaptive algorithm for analysis to adjust the inverter output, including:

[0041] Determine the first adjustment target of the PID control algorithm based on the battery prediction results and the grid disturbance assessment results;

[0042] Based on the PID control principle, combined with the first adjustment target and the disturbance type, determine the current PID parameter adjustment strategy to adjust the parameters of the PID control algorithm;

[0043] Using the battery operation data and the grid operation data as input, the target power is calculated using the PID control algorithm after adjusting the parameters;

[0044] According to the target power, the output of the energy storage inverter is adjusted.

[0045] Preferably, adjusting the output of the energy storage inverter according to the target power includes:

[0046] Identify the target power type and determine whether the target power is discharge power or charge power;

[0047] When the target power is the discharge power, the working mode of the energy storage inverter is set to the discharge mode;

[0048] The energy storage inverter increases the output current and voltage, and monitors in real time until the output power of the energy storage inverter is consistent with the target discharge power. The energy storage inverter currently maintains a stable output and transmits electric energy to the power grid;

[0049] When the target power is the charging power, the working mode of the energy storage inverter is set to the charging mode;

[0050] The energy storage inverter increases the input current and voltage, and monitors in real time until the input power of the inverter is consistent with the target discharge power. The current energy storage inverter maintains stable input current and voltage to provide stable charging power for the battery.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] By real-time monitoring of the operating status of the power grid and battery, obtaining battery operating data and power grid operating data, predicting battery status and power grid disturbance analysis, and obtaining battery prediction results and power grid disturbance results; combining the battery prediction results with the power grid assessment results and the adaptive algorithm for analysis to adjust the output of the energy storage inverter, it can optimize energy utilization and distribution, improve system safety and stability, and extend battery life.

[0053] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0054] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0056] Figure 1 The present invention is a flowchart of a method for controlling an energy storage inverter based on an adaptive algorithm in an embodiment of the present invention. DETAILED DESCRIPTION

[0057] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0058] The embodiment of the present invention provides a control method of an energy storage inverter based on an adaptive algorithm, such as Figure 1 As shown, including:

[0059] Step 1: Monitor the operating status of the power grid and battery in real time, and obtain battery operating data and power grid operating data;

[0060] Step 2: predicting a battery state based on the battery operation data to obtain a battery prediction result;

[0061] Step 3: Evaluate the current grid disturbance situation based on the grid operation data to obtain a grid disturbance result;

[0062] Step 4: Based on the battery prediction results and the grid assessment results, an adaptive algorithm is used to dynamically adjust the charging and discharging power and the inverter output.

[0063] In this embodiment, the energy storage inverter (PCS) can realize AC / DC conversion between the battery and the power grid, and complete the two-way energy flow between the two; the battery operation data is the data obtained after data cleaning and filtering processing of the first data; the power grid operation data is the data obtained after data cleaning and filtering processing of the second data; the first data refers to the battery operation data collected by real-time monitoring of the battery status using preset sensors, including voltage, current, temperature, etc.; the second data refers to the power grid operation data collected by real-time monitoring of the power grid status using preset sensors, including voltage, frequency, phase, etc.

[0064] In this embodiment, the battery prediction result refers to the corresponding prediction data of the performance-related parameters extracted from the battery prediction data, wherein the performance-related parameters are parameters closely related to the battery performance determined by the technical data of the current battery, including remaining power, power output, cycle life, internal resistance and self-discharge rate; the grid disturbance result is composed of the disturbance identification type, disturbance time and disturbance impact range.

[0065] The beneficial effects of the above technical solution are: by real-time monitoring of the operating status of the power grid and the battery, obtaining the battery operating data and the power grid operating data, predicting the battery status and power grid disturbance analysis, and obtaining the battery prediction results and the power grid disturbance results; combining the battery prediction results and the power grid assessment results with the adaptive algorithm for analysis to adjust the output of the energy storage inverter, it can optimize energy utilization and distribution, improve system safety and stability, and extend battery life.

[0066] The embodiment of the present invention provides a method for controlling an energy storage inverter based on an adaptive algorithm, which monitors the operating status of a power grid and a battery in real time, and obtains battery operating data and power grid operating data, including:

[0067] Using preset sensors to monitor the operating status of the battery and the power grid in real time, and obtain first data and second data respectively;

[0068] After data cleaning and filtering are performed on the first data and the second data, they are output as battery operation data and power grid operation data respectively.

[0069] In this embodiment, the preset sensors are used to monitor the operating status of the battery and the power grid in real time and collect operating data of the battery and the power grid, including voltage sensors, current sensors, temperature sensors, frequency sensors, etc.; the first data refers to the battery operating data collected by monitoring the battery status in real time using the preset sensors, including voltage, current, temperature, etc.; the second data refers to the power grid operating data collected by monitoring the power grid status in real time using the preset sensors, including voltage, frequency, phase, etc.; the battery operating data is the data obtained after data cleaning and filtering processing is performed on the first data; the power grid operating data is the data obtained after data cleaning and filtering processing is performed on the second data.

[0070] The beneficial effects of the above technical solution are as follows: by cleaning and filtering the operating data obtained by real-time monitoring of the operating status of the battery and the power grid, the timeliness and accuracy of the data can be ensured, providing strong support for subsequent decision-making.

[0071] The embodiment of the present invention provides a method for controlling an energy storage inverter based on an adaptive algorithm, which predicts a battery state based on the battery operation data to obtain a battery prediction result, including:

[0072] According to the current battery type and application requirements, select and set the battery model and set the initial parameters;

[0073] A Kalman filter algorithm is introduced to perform model training on the set battery model based on historical operation data of a preset amount of the current battery extracted from a battery database to obtain a first prediction model;

[0074] Using preset test data to test the first prediction model to obtain a test result;

[0075] The test results are evaluated using a set effect evaluation index to obtain a prediction effect coefficient;

[0076] When the prediction effect coefficient is less than a set prediction threshold, adjusting the first key parameter in the first prediction model by using a set parameter optimization algorithm to obtain a battery prediction model;

[0077] When the prediction effect coefficient is not less than a set prediction threshold, using the first prediction model as a battery prediction model;

[0078] Inputting battery operation data into the battery prediction model and outputting battery prediction data;

[0079] Corresponding prediction data of performance-related parameters are extracted from the battery prediction data and output as a battery prediction result.

[0080] In this embodiment, a battery refers to a device that can directly convert chemical energy, light energy, thermal energy, nuclear energy, and other forms of energy into electrical energy, wherein battery types include lithium-ion batteries, nickel-metal hydride batteries, and nickel-cadmium batteries, etc.; application requirements refer to the usage scenarios of the batteries, and different usage scenarios have different requirements for batteries and different choices of battery models; setting a battery model is based on the current battery type and application requirements, and is a model selected from a preset battery model matching table, such as a Randles model, a Sundmeier model, etc., wherein the preset battery model matching table is composed of battery type, application requirements, and battery model.

[0081] In this embodiment, the initial parameters include the capacity, internal resistance and open circuit voltage of the battery, etc.; the Kalman filter algorithm is an effective mathematical method for predicting and estimating the state of a linear dynamic system; the battery database is composed of the operating data of the battery, such as voltage, current, and temperature; the preset quantity is predetermined; the preset test data is the historical operating data of the battery randomly selected from the battery database in advance; the first prediction model is obtained by training the selected battery model using the historical operating data in combination with the Kalman filter algorithm, and is used to predict the future operating state of the battery; the first key parameter refers to the state transfer matrix, the observation matrix, and the initial state estimate; the parameter optimization algorithm is set to refer to the random search algorithm.

[0082] In this embodiment, the test result refers to the predicted test result output after the preset test data is input into the first prediction model; the set effect evaluation index refers to the accuracy, mean square error and mean absolute error; the prediction effect coefficient is the weighted average of the index evaluation obtained by evaluating the test result using the set effect evaluation index; the set prediction threshold is predetermined; the performance-related parameters are parameters closely related to the battery performance determined by the technical data of the current battery, including remaining power, power output, cycle life, internal resistance and self-discharge rate; the battery prediction result refers to the corresponding prediction data of the performance-related parameters extracted from the battery prediction data.

[0083] The beneficial effects of the above technical solution are: by introducing the Kalman filter algorithm to train the selected battery model, and then obtaining the battery prediction model after test optimization, the adaptability and generalization ability of the model are effectively enhanced; the battery prediction model is used to realize effective prediction of the future operation of the battery based on real-time operation data.

[0084] The embodiment of the present invention provides a method for controlling an energy storage inverter based on an adaptive algorithm, which evaluates the current grid disturbance situation based on the grid operation data to obtain a grid disturbance result, including:

[0085] Extracting and integrating features of the power grid operation data using signal processing technology to obtain a first operation feature set;

[0086] Selecting a feature associated with a power grid disturbance from the first operation feature set and considering it as a first disturbance-associated feature;

[0087] Extracting and aggregating disturbance-related features of a preset amount of historical power grid operation data extracted from a power grid operation database to obtain a first feature training set;

[0088] Using the first feature training set as input, training the neural network to obtain a disturbance recognition model;

[0089] Inputting the first disturbance association feature into a disturbance identification model, and outputting a disturbance identification result;

[0090] According to the disturbance identification result, if there is a disturbance identification type, the disturbance is located based on the topological structure of the current power grid to determine the disturbance time and the disturbance impact range;

[0091] The disturbance identification type, disturbance time and disturbance impact range are output as power grid disturbance results.

[0092] In this embodiment, the grid operation data refers to the data obtained after data cleaning and filtering processing is performed on the second data, wherein the second data refers to the grid operation data collected by real-time monitoring of the grid status using preset sensors, including voltage, frequency, phase, etc.; signal processing technology is used to extract features from the grid operation data, including time-frequency analysis, wavelet transform, etc.; the first operation feature set is a set of features obtained after feature extraction of the grid operation data using signal processing technology; the first disturbance-associated feature refers to the first operation feature associated with the grid disturbance, including amplitude, frequency, phase, harmonic content, etc.

[0093] In this embodiment, the power grid operation database is composed of the operation data of the power grid; the preset quantity is predetermined; the first feature training set is a feature set obtained by extracting disturbance-related features from the extracted historical power grid operation data using signal processing technology; the disturbance identification model is a model obtained by training a neural network using the first feature training set as training data, and is used to identify the type of power grid disturbance, wherein the disturbance types include short circuit faults, load mutations, voltage sags, voltage swells, voltage interruptions, harmonics, oscillations, etc.; the power grid disturbance result is composed of the disturbance identification type, disturbance time, and disturbance influence range, wherein the disturbance influence range is obtained by mapping the abnormal change points identified by the abnormal monitoring algorithm to the power grid topology structure and propagation path tracking through the feature vector obtained by combining the current disturbance-related features.

[0094] The beneficial effect of the above technical solution is: by training a neural network to build a disturbance recognition model, the current power grid disturbance is identified based on the features obtained by feature extraction of power grid operation data, which can help improve the efficiency and accuracy of disturbance recognition.

[0095] An embodiment of the present invention provides a method for controlling an energy storage inverter based on an adaptive algorithm, wherein a feature associated with a power grid disturbance is selected from the first operating feature set and regarded as a first disturbance-associated feature, including:

[0096] Obtaining historical feature vectors of first operation features of the power grid before and after various disturbance types occur within a preset time period;

[0097] Based on the historical feature vectors, the historical key parameters of the first operating characteristics before and after each disturbance type occurs are calculated;

[0098] Determining a disturbance sensitivity coefficient of a current first operating characteristic based on historical key parameters;

[0099] The calculation formula of the disturbance sensitivity coefficient is as follows:

[0100] Where P represents the disturbance sensitivity coefficient of the current first operating characteristic; n represents the total number of disturbance types; s1 represents the occurrence coefficient of the i-th disturbance type within the preset time period; s0 represents the total number of disturbances occurring within the preset time period; x ij1 It is represented as the historical key parameters of the current first operating feature before the jth disturbance under the i-th disturbance type; x ij0 It is represented as the historical key parameter of the current first operating feature after the jth disturbance under the i-th disturbance type; ε j It is expressed as the contribution weight of historical key parameters to the disturbance;

[0101] Marking the first operating feature whose disturbance sensitivity coefficient exceeds the set sensitivity threshold as a possible feature;

[0102] Using Euclidean distance to calculate type difference values ​​of the possible features under different disturbance types;

[0103] Possible features whose type difference values ​​exceed the set difference threshold are regarded as the first disturbance-associated features.

[0104] In this embodiment, the preset time period is predetermined; the disturbance types include short circuit fault, load mutation, voltage sag, voltage swell, voltage interruption, harmonics, oscillation, etc.; the historical key parameters include mean, variance, etc.; the disturbance sensitivity coefficient is used to quantify the sensitivity of the first operating feature to the disturbance; the set sensitivity threshold is pre-set; the possible feature refers to the first operating feature whose disturbance sensitivity coefficient exceeds the set sensitivity threshold; the first disturbance associated feature refers to the possible feature whose type difference value exceeds the set difference threshold.

[0105] The beneficial effects of the above technical solution are: obtaining the disturbance sensitivity coefficient by performing sensitivity and difference analysis on the first operating characteristics under different disturbance types; screening out the first disturbance-related characteristics based on the disturbance sensitivity coefficient, and providing strong data support for power grid disturbance identification.

[0106] The embodiment of the present invention provides a method for controlling an energy storage inverter based on an adaptive algorithm, combining the battery prediction result and the grid assessment result with the adaptive algorithm for analysis, and adjusting the inverter output, including:

[0107] Determine the first adjustment target of the current PID control algorithm based on the battery prediction results and the grid disturbance assessment results;

[0108] Based on the PID control principle, combined with the first adjustment target and the disturbance type, a PID parameter adjustment strategy is determined to adjust the parameters of the PID control algorithm;

[0109] Using the battery operation data and the grid operation data as input, the target power is calculated using the PID control algorithm after adjusting the parameters;

[0110] According to the target power, the output of the energy storage inverter is adjusted.

[0111] In this embodiment, the battery prediction result refers to the corresponding prediction data of the performance-related parameters extracted from the battery prediction data, wherein the performance-related parameters are parameters closely related to the battery performance determined by the technical data of the current battery, including remaining power, power output, cycle life, internal resistance and self-discharge rate; the first adjustment target is determined by combining the analysis of the battery prediction results and the disturbance type in the power grid disturbance assessment results, including improving the system's response speed, voltage stability, power balance and normal frequency, etc.

[0112] In this embodiment, the PID control principle refers to the effects of proportional gain kp, integral time Ti, and differential time Td; disturbance types include short circuit faults, load mutations, voltage sags, voltage swells, voltage interruptions, harmonics, oscillations, etc.; the PID parameter adjustment strategy is a strategy for adjusting the PID parameter proportional gain kp, integral time Ti, or differential time Td selected from the set parameter adjustment strategy table based on the first adjustment target and the disturbance type, and multiple parameters cannot be adjusted at the same time. For example, if the first adjustment target is to improve the response speed of the system, the corresponding PID parameter adjustment strategy is to gradually increase the proportional gain kp to the corresponding update amount, and then increase the integral time Ti to the corresponding update amount. New quantity, where the update quantity is determined by introducing a fuzzy algorithm based on the difference between the actual output power of the current energy storage system and the target power; the parameter adjustment strategy table is a list established by analyzing the parameter adjustment rules in the historical parameter adjustment data under different battery states and grid disturbance types within a preset time period, and is composed of battery state range, disturbance type and parameter adjustment strategy, where the battery state range includes the remaining power range, cycle life range, etc.; the target power refers to the optimal charge and discharge power of the battery output by the PID control algorithm after adjusting the parameters; the energy storage inverter (PCS) can realize the AC / DC conversion between the battery and the grid, and complete the two-way energy flow between the two.

[0113] The beneficial effect of the above technical solution is: by determining the adjustment target of the PID control algorithm according to the battery prediction results and the grid disturbance assessment results, after adjusting the PID parameters based on the PID control principle, the battery target charging and discharging power is determined, providing data support for adjusting the inverter output.

[0114] The embodiment of the present invention provides a method for controlling an energy storage inverter based on an adaptive algorithm, which adjusts the output of the energy storage inverter according to the target power, including:

[0115] Identify the target power type and determine whether the target power is discharge power or charge power;

[0116] When the target power is the discharge power, the working mode of the energy storage inverter is set to the discharge mode;

[0117] The energy storage inverter increases the output current and voltage, and monitors in real time until the output power of the energy storage inverter is consistent with the target discharge power. The energy storage inverter currently maintains a stable output and transmits electric energy to the power grid;

[0118] When the target power is the charging power, the working mode of the energy storage inverter is set to the charging mode;

[0119] The energy storage inverter increases the input current and voltage, and monitors in real time until the input power of the inverter is consistent with the target discharge power. The current energy storage inverter maintains stable input current and voltage to provide stable charging power for the battery.

[0120] The beneficial effect of the above technical solution is that the energy storage inverter can dynamically adjust the output according to the acquired target power to accurately control the charging and discharging behavior of the battery to meet the needs and constraints of the power grid, thereby helping to improve the safety of the battery system and the battery cycle life.

[0121] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A control method for energy storage inverter based on adaptive algorithm, characterized in that: include: Step 1: Monitor the operating status of the power grid and battery in real time, and obtain battery operating data and power grid operating data; Step 2: predicting a battery state based on the battery operation data to obtain a battery prediction result; Step 3: Evaluate the current grid disturbance situation based on the grid operation data to obtain a grid disturbance result; Step 4: Combine the battery prediction result with the grid assessment result and the adaptive algorithm for analysis, and adjust the inverter output.

2. The energy storage inverter control method based on the adaptive algorithm according to claim 1 is characterized in that: Monitor the operating status of the power grid and batteries in real time, and obtain battery operating data and power grid operating data, including: Using preset sensors to monitor the operating status of the battery and the power grid in real time, and obtain first data and second data respectively; After data cleaning and filtering are performed on the first data and the second data, they are output as battery operation data and power grid operation data respectively.

3. The energy storage inverter control method based on the adaptive algorithm according to claim 1 is characterized in that: Predicting a battery state based on the battery operation data to obtain a battery prediction result includes: According to the current battery type and application requirements, select and set the battery model and set the initial parameters; A Kalman filter algorithm is introduced to perform model training on the set battery model based on historical operation data of a preset amount of the current battery extracted from a battery database to obtain a first prediction model; Using preset test data to test the first prediction model to obtain a test result; The test results are evaluated using a set effect evaluation index to obtain a prediction effect coefficient; When the prediction effect coefficient is less than a set prediction threshold, adjusting the first key parameter in the first prediction model by using a set parameter optimization algorithm to obtain a battery prediction model; When the prediction effect coefficient is not less than a set prediction threshold, using the first prediction model as a battery prediction model; Inputting battery operation data into the battery prediction model and outputting battery prediction data; Corresponding prediction data of performance-related parameters are extracted from the battery prediction data and output as a battery prediction result.

4. The energy storage inverter control method based on the adaptive algorithm according to claim 3 is characterized in that: The first key parameters refer to the state transfer matrix, the observation matrix and the initial state estimate.

5. The energy storage inverter control method based on adaptive algorithm according to claim 1, characterized in that: Evaluating the current grid disturbance situation based on the grid operation data to obtain a grid disturbance result includes: Extracting and integrating features of the power grid operation data using signal processing technology to obtain a first operation feature set; Selecting a feature associated with a power grid disturbance from the first operation feature set and considering it as a first disturbance-associated feature; Extracting and aggregating disturbance-related features of a preset amount of historical power grid operation data extracted from a power grid operation database to obtain a first feature training set; Using the first feature training set as input, training the neural network to obtain a disturbance recognition model; Inputting the first disturbance association feature into a disturbance identification model, and outputting a disturbance identification result; According to the disturbance identification result, if there is a disturbance identification type, the disturbance is located based on the topological structure of the current power grid to determine the disturbance time and the disturbance impact range; The disturbance identification type, disturbance time and disturbance impact range are output as the power grid disturbance results.

6. The energy storage inverter control method based on adaptive algorithm according to claim 5 is characterized in that: Screening out features associated with power grid disturbance from the first operation feature set and considering them as first disturbance-associated features includes: Obtaining historical feature vectors of first operation features of the power grid before and after various disturbance types occur within a preset time period; Based on the historical feature vectors, the historical key parameters of the first operating characteristics before and after each disturbance type occurs are calculated; Determining a disturbance sensitivity coefficient of a current first operating characteristic based on historical key parameters; The calculation formula of the disturbance sensitivity coefficient is as follows: Where, P represents the disturbance sensitivity coefficient of the current first operating characteristic; n represents the total number of disturbance types; si1 represents the occurrence coefficient of the ith disturbance type within the preset time period; s0 represents the total number of disturbances occurring within the preset time period; x ij1 It is represented as the historical key parameters of the current first operating feature before the jth disturbance under the i-th disturbance type; x ij0 It is represented as the historical key parameter of the current first operating feature after the jth disturbance under the i-th disturbance type; ε j It is expressed as the contribution weight of historical key parameters to the disturbance; Marking the first operating feature whose disturbance sensitivity coefficient exceeds the set sensitivity threshold as a possible feature; Using Euclidean distance to calculate type difference values ​​of the possible features under different disturbance types; Possible features whose type difference values ​​exceed the set difference threshold are regarded as the first disturbance-associated features.

7. The energy storage inverter control method based on adaptive algorithm according to claim 1, characterized in that: The battery prediction result is combined with the grid assessment result and the adaptive algorithm for analysis to adjust the inverter output, including: Determine the first adjustment target of the current PID control algorithm based on the battery prediction results and the grid disturbance assessment results; Based on the PID control principle, combined with the first adjustment target and the disturbance type, a PID parameter adjustment strategy is determined to adjust the parameters of the PID control algorithm; Using the battery operation data and the grid operation data as input, the target power is calculated using the PID control algorithm after adjusting the parameters; According to the target power, the output of the energy storage inverter is adjusted.

8. The energy storage inverter control method based on the adaptive algorithm according to claim 7 is characterized in that: According to the target power, adjusting the output of the energy storage inverter includes: Identify the target power type and determine whether the target power is discharge power or charge power; When the target power is the discharge power, the working mode of the energy storage inverter is set to the discharge mode; The energy storage inverter increases the output current and voltage, and monitors in real time until the output power of the energy storage inverter is consistent with the target discharge power. The energy storage inverter currently maintains a stable output and transmits electric energy to the power grid; When the target power is charging power, the working mode of the energy storage inverter is set to charging mode; The energy storage inverter increases the input current and voltage, and monitors in real time until the input power of the inverter is consistent with the target discharge power. The current energy storage inverter maintains stable input current and voltage to provide stable charging power for the battery.

Images