Grid-connected control method and system of container-type electrochemical energy storage system

Abstract

The application provides a kind of container type electrochemical energy storage system grid-connected control method and system, belong to energy storage and grid control technical field, first obtain the electrochemical characteristic data of container type electrochemical energy storage system (including energy storage unit charge migration rate, electrode reaction activity, electrolyte distribution etc.) And grid dynamic admission data (cover real-time power surplus and deficiency, voltage fluctuation tolerance, frequency regulation response etc.);Then construct deep learning collaborative mapping model to establish the dynamic correlation of electrochemical characteristics and grid admission capacity, generate grid-connected collaborative control strategy;Again grid-connected collaborative control strategy is disassembled into electrochemical parameter adjustment instruction and transmission parameter adjustment instruction;According to the grid-connected feedback control instruction generated after adjusting electrochemical parameter adjustment instruction and transmission parameter adjustment instruction and sent to grid-connected control unit, maintain stable grid-connected state.The application improves the stability, reliability and efficiency of energy storage system grid-connected.

Description

Technical Field

[0001] This invention relates to the field of energy storage and grid control technology, and more specifically, to a grid-connected control method and system for a containerized electrochemical energy storage system. Background Technology

[0002] Containerized electrochemical energy storage systems (CES) play an increasingly crucial role in the power grid as an important energy storage and regulation device. They effectively smooth the volatility of renewable energy generation and improve the stability and reliability of the grid. However, existing grid-connection control methods for CES have several shortcomings. On the one hand, traditional control methods often focus only on the electrochemical characteristics of the energy storage system itself, such as simple charge and discharge control based solely on the charge state of the energy storage units, while ignoring the dynamic acceptance capacity of the grid. This results in the energy storage system being unable to rationally adjust its operating state according to dynamic information such as the actual power surplus / deficit, voltage fluctuations, and frequency regulation requirements of the grid during grid connection, easily leading to mismatches with the grid and affecting grid connection effectiveness. On the other hand, while some methods consider certain grid parameters, they fail to establish a deep dynamic correlation between the electrochemical characteristics of the energy storage system and the grid's acceptance capacity, making it impossible to generate accurate and effective grid-connection coordinated control strategies and achieve efficient and stable coordinated operation between the energy storage system and the grid. Summary of the Invention

[0003] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, the present invention provides a grid-connected control method for a containerized electrochemical energy storage system, the method comprising:

[0004] The electrochemical characteristic data of the containerized electrochemical energy storage system and the dynamic grid acceptance data are obtained. The electrochemical characteristic data of the containerized electrochemical energy storage system includes charge migration rate data of the energy storage unit, electrode reaction activity data of the energy storage unit, and electrolyte distribution data inside the energy storage system. The dynamic grid acceptance data includes real-time power surplus / deficit data of the grid, voltage fluctuation tolerance data of the grid, and frequency regulation response data of the grid.

[0005] Based on the electrochemical characteristic data of the containerized electrochemical energy storage system and the dynamic acceptance data of the power grid, a deep learning collaborative mapping model is constructed. The dynamic correlation between changes in electrochemical characteristics and the grid acceptance capacity is established through the deep learning collaborative mapping model, and a grid-connected collaborative control strategy is generated.

[0006] The grid-connected coordinated control strategy is decomposed into electrochemical parameter adjustment commands for the containerized electrochemical energy storage system and transmission parameter adjustment commands for the grid-connected interface, resulting in two sets of related adjustment commands.

[0007] The internal electrochemical operating state of the containerized electrochemical energy storage system is adjusted according to the electrochemical parameter adjustment command to obtain the adjusted electrochemical operating state data. At the same time, the transmission operating parameters of the grid connection interface are adjusted according to the transmission parameter adjustment command to obtain the adjusted transmission operating parameter data.

[0008] Based on the adjusted electrochemical operating status data and the adjusted transmission operating parameter data, a grid-connected feedback control command is generated and sent to the grid-connected control unit of the containerized electrochemical energy storage system, so that the grid-connected control unit maintains a stable grid-connected state according to the grid-connected feedback control command.

[0009] In another aspect, the present invention also provides a grid-connected control system for a containerized electrochemical energy storage system, comprising: a processor; a machine-readable storage medium for storing machine-executable instructions of the processor; wherein the processor is configured to execute the above-described grid-connected control method for the containerized electrochemical energy storage system by executing the machine-executable instructions.

[0010] In another aspect, the present invention also provides a computer program product, the computer program product including machine-executable instructions, the machine-executable instructions being stored in a computer-readable storage medium, the processor of the grid-connected control system of the containerized electrochemical energy storage system reading the machine-executable instructions from the computer-readable storage medium, the processor executing the machine-executable instructions, causing the grid-connected control system of the containerized electrochemical energy storage system to execute the above-described grid-connected control method of the containerized electrochemical energy storage system.

[0011] Based on the above, by acquiring electrochemical characteristic data of the containerized electrochemical energy storage system and dynamic grid acceptance data, a deep learning collaborative mapping model was constructed based on the above data. This successfully established a dynamic correlation between changes in electrochemical characteristics and grid acceptance capacity, enabling the generation of grid-connected collaborative control strategies in real time and accurately. The grid-connected collaborative control strategy is decomposed into two sets of related adjustment instructions for the electrochemical parameters of the energy storage system and the transmission parameters of the grid connection interface. This achieves refined and collaborative control of the internal operation and grid connection transmission of the energy storage system. After adjustments are made according to the adjustment instructions, grid-connected feedback control instructions are generated based on the adjusted data and sent to the grid connection control unit. This enables the grid connection control unit to adjust in a timely manner based on real-time feedback information, maintaining a stable grid connection state and greatly improving the stability, reliability, and efficiency of the containerized electrochemical energy storage system's grid connection. Attached Figure Description

[0012] Figure 1 This is a schematic diagram of the execution flow of the grid connection control method for a containerized electrochemical energy storage system provided in an embodiment of the present invention. Detailed Implementation

[0013] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1 This is a schematic flowchart of a grid-connected control method for a containerized electrochemical energy storage system provided in one embodiment of the present invention.

[0014] Step S110: Obtain electrochemical characteristic data of the containerized electrochemical energy storage system and dynamic grid acceptance data. The electrochemical characteristic data of the containerized electrochemical energy storage system includes charge migration rate data of the energy storage unit, electrode reaction activity data of the energy storage unit, and electrolyte distribution data inside the energy storage system. The dynamic grid acceptance data includes real-time power surplus / deficit data of the grid, voltage fluctuation tolerance data of the grid, and frequency regulation response data of the grid.

[0015] In this embodiment, a containerized lithium iron phosphate battery energy storage system in an industrial park is used as the application scenario. The energy storage system has a specific capacity and a specific access voltage level, and is used to smooth out the output fluctuations of photovoltaic power stations in the park and participate in grid peak regulation.

[0016] To achieve the above data acquisition, multiple types of sensors are first deployed inside the energy storage system. Among them, charge migration rate data is acquired by connecting a high-frequency response current sensor and a high-precision voltage sensor in series on the positive and negative leads of the battery cells. The sampling frequency is set to a specific Hertz. By measuring the change in electrode interface impedance at different times, the dynamic characteristics of charge migration rate are indirectly reflected. The resulting charge migration rate data is a multi-dimensional data set containing time and space dimensions. For example, at each sampling time, charge migration rate-related parameters are collected for multiple battery modules in the energy storage system, forming a two-dimensional array data structure.

[0017] Electrode reaction activity data are collected through a miniature three-electrode sensor integrated on the electrode surface. The redox reaction potential changes and reaction current density distribution on the electrode surface are monitored in real time. The sampling frequency is a specific Hertz, and the data is also multi-dimensional, including activity parameters at different electrode positions and different reaction stages.

[0018] Electrolyte distribution data is acquired through a distributed array of pressure and concentration sensors installed inside the battery module. The pressure sensors are used to monitor the pressure distribution of the electrolyte in the flow channel, and the concentration sensors are used to monitor the ion concentration of the electrolyte in different regions. The sampling frequency is a specific Hertz. The resulting electrolyte distribution data is a three-dimensional dataset that covers multiple dimensions of information such as spatial location, pressure, and concentration.

[0019] The dynamic acceptance data of the power grid is obtained in real time through the communication interface with the power grid dispatch center. The real-time power surplus and deficit data is the surplus or deficit value of active and reactive power that the power grid can accept at the current moment. It is presented in the form of a multi-time period sequence, with each time period corresponding to a power value, forming time series data. The voltage fluctuation tolerance data includes parameters such as the allowable fluctuation range of the power grid voltage and the fluctuation duration threshold, which are expressed in the form of a combination of interval range and time threshold. The frequency regulation response data includes the current value of the power grid frequency, the allowable deviation range, the regulation response time requirement, etc., which also exist in the form of a combination of multiple parameters.

[0020] During the data acquisition process, for privacy-sensitive data such as the battery status inside the energy storage system, data anonymization technology is used to anonymize the battery number, specific location information, etc. in the original data. At the same time, the data is transmitted from the sensor to the data processing center through an encrypted transmission protocol to ensure that the data is not leaked during the transmission process.

[0021] Step S120: Based on the electrochemical characteristic data of the containerized electrochemical energy storage system and the dynamic acceptance data of the power grid, a deep learning collaborative mapping model is constructed. The dynamic correlation between changes in electrochemical characteristics and the grid acceptance capacity is established through the deep learning collaborative mapping model, and a grid-connected collaborative control strategy is generated.

[0022] Step S121: Determine the model functional hierarchy of the deep learning collaborative mapping model, wherein the model functional hierarchy includes a multi-source data coupling perception layer, a temporal correlation modeling layer, and a collaborative strategy generation layer.

[0023] In the application scenario of the energy storage system in this industrial park, the heterogeneity and complexity of electrochemical characteristic data and grid dynamic acceptance data need to be considered when determining the functional levels of the model. The multi-source data coupling sensing layer is responsible for processing raw data from different data sources, achieving effective extraction of cross-domain features; the temporal correlation modeling layer aims to capture the dynamic patterns of data changes over time and the temporal correlation between the two; the collaborative strategy generation layer generates specific grid-connected control strategies based on the processing results of the previous two layers. These three layers are independent yet closely connected, forming the core framework of the deep learning collaborative mapping model. The input of the multi-source data coupling sensing layer is preprocessed electrochemical characteristic data and grid dynamic acceptance data, and the output is the extracted sensitive features and adaptation features; the input of the temporal correlation modeling layer is the weighted set of coupling features, and the output is the temporal correlation model and the bidirectional influence relationship matrix; the input of the collaborative strategy generation layer is the bidirectional influence relationship matrix, and the output is multiple sets of initial grid-connected control strategies and the final grid-connected collaborative control strategy.

[0024] Step S122: Input the electrochemical characteristic data of the containerized electrochemical energy storage system and the dynamic acceptance data of the power grid into the multi-source data coupling sensing layer, perform cross-domain feature association processing on the electrochemical characteristic data of the containerized electrochemical energy storage system and the dynamic acceptance data of the power grid, and extract the sensitive features related to power grid acceptance in the electrochemical characteristic data and the adaptation features related to electrochemical response in the dynamic acceptance data.

[0025] In this embodiment, after inputting the acquired electrochemical characteristic data and grid dynamic access data into the multi-source data coupling sensing layer, the data is first standardized. Since the electrochemical characteristic data and grid dynamic access data have different dimensions—for example, the unit of charge mobility rate and the unit of power surplus / deficit are specific units—it is necessary to standardize the data to enable the mutual calculation of feature parameters with different dimensions. The standardization process uses the Z-score standardization method, converting each feature data into a standard normal distribution with a mean of 0 and a standard deviation of 1. For example, for a certain feature value in the charge mobility rate data, the standardized data is obtained by converting it using the mean and standard deviation of that feature across all sampling times. After standardization, cross-domain feature association processing is performed to extract sensitive and adaptive features.

[0026] Step S1221: Construct a cross-domain feature association graph in the feature interaction unit, using the feature items of electrochemical characteristic data as the first type of node in the cross-domain feature association graph, and using the feature items of dynamically received data as the second type of node in the cross-domain feature association graph.

[0027] Step S1222: Calculate the association strength between each feature item in the first type of node and each feature item in the second type of node in the cross-domain feature association graph. Add association edges between nodes in the cross-domain feature association graph according to the association strength. When the association strength meets the preset association strength standard, the representation weight of the association edge is set according to the corresponding rule.

[0028] In this application scenario, the correlation strength is calculated using a mutual information algorithm. For a feature item in the first type of node and a feature item in the second type of node, their standardized data sequences within the same time period are used as input to calculate the mutual information value between them. The magnitude of this mutual information value reflects the correlation strength between the two feature items. For example, the mutual information value between the interface impedance feature item in the charge migration rate data and the active power surplus / deficit feature item in the real-time power surplus / deficit data over a specific time period is calculated. The preset correlation strength standard has multiple levels. For example, when the mutual information value is within a specific range, the correlation strength is extremely strong; within a specific range, the correlation strength is strong; within a specific range, the correlation strength is moderate; within a specific range, the correlation strength is weak; and below a specific value, the correlation strength is extremely weak. When the correlation strength is extremely strong or strong, it is considered to meet the preset correlation strength standard. At this time, a correlation edge is added between the corresponding nodes, and the representation weight of the correlation edge is set according to the correlation strength level. For example, the representation weight corresponding to extremely strong correlation strength is a specific value, and the representation weight corresponding to strong correlation strength is a specific value. For node pairs with a correlation strength of moderate or lower, no correlation edge is added or a lower representation weight is set.

[0029] Step S1223: Based on the distribution of associated edges in the cross-domain feature association graph, identify feature item pairs whose associated edge representation weights exceed a preset association threshold. The feature item pairs include feature items from the first type of nodes in the cross-domain feature association graph and feature items from the second type of nodes in the cross-domain feature association graph.

[0030] In this embodiment, the preset association threshold is determined based on actual application requirements and historical data statistics. For example, the preset association threshold can be set to a specific value, which corresponds to the lower limit of a strong association strength. All added association edges in the cross-domain feature association graph are traversed, and the representation weight of each association edge is checked to see if it exceeds the preset association threshold. For association edges whose representation weight exceeds the preset association threshold, the first-type node feature items and the second-type node feature items connected to them form a feature item pair. For example, if the representation weight of the association edge between the interface impedance feature item in the charge mobility rate data and the active power surplus / deficit feature item in the real-time power surplus / deficit data exceeds the preset association threshold, then these two feature items form a feature item pair.

[0031] Step S1224: Perform feature attribute analysis on the identified feature pairs to determine the type of influence of the feature pairs in the first type of nodes of the cross-domain feature association map on the grid acceptance capability. The influence type includes types that promote grid acceptance and types that restrict grid acceptance.

[0032] In this application scenario, when performing characteristic attribute analysis on each pair of features, the relationship between the changing trends of the first type of node features and the changing trends of the second type of node features (representing indicators related to grid acceptance capacity) is first determined. For example, for the feature pair of interface impedance and active power surplus / deficit, when the interface impedance decreases, the active power surplus / deficit shows an increasing trend (i.e., the active power that the grid can accept increases), indicating that the decrease in interface impedance has a promoting effect on grid acceptance capacity, and vice versa. Through the analysis of a large amount of historical data, a causal relationship model between the first type of node features and the second type of node features is established, and the type of influence is determined based on this model. For example, when the value of the first type of node feature increases, if the value of the corresponding second type of node feature also increases (i.e., grid acceptance capacity is enhanced), then the first type of node feature belongs to the type that promotes grid acceptance; if the increase in the value of the first type of node feature leads to the decrease in the value of the second type of node feature (i.e., grid acceptance capacity is weakened), then the first type of node feature belongs to the type that restricts grid acceptance.

[0033] Step S1225: Classify the first type of node feature items in the cross-domain feature association map that belong to the type of promoting grid acceptance and whose association strength meets the requirements as sensitive features related to grid acceptance, and record the degree of influence corresponding to each sensitive feature related to grid acceptance.

[0034] In this embodiment, the required correlation strength means that the weight of the correlation edge of the feature item pair exceeds a preset correlation threshold. For a first-class node feature item determined by feature attribute analysis to promote grid acceptance, and whose correlation strength with a second-class node feature item meets the requirements, it is classified as a sensitive feature. For example, if the reaction current density feature item in the electrode reaction activity data is determined to promote grid acceptance, and the weight of its correlation edge with the voltage fluctuation range feature item in the voltage fluctuation tolerance data exceeds a preset correlation threshold, then the reaction current density feature item is classified as a sensitive feature. The degree of influence is measured by the weight of the correlation edge; the larger the weight, the greater the influence of the sensitive feature on grid acceptance capability. The weight of each sensitive feature is recorded as a quantitative indicator of its influence.

[0035] Step S1226: Perform reverse analysis on the feature pairs to determine the constraint type of the feature pairs in the second type of nodes of the cross-domain feature correlation spectrum on the electrochemical characteristic response. The constraint type includes types that support electrochemical stability and types that interfere with electrochemical stability.

[0036] In this application scenario, reverse analysis refers to analyzing the impact of changes in the second type of node characteristic terms on the stability of the first type of node characteristic terms (representing electrochemical response indicators). For example, for a characteristic term pair consisting of the active power surplus / deficit characteristic term in real-time power surplus / deficit data and the ion concentration characteristic term in electrolyte distribution data, when the active power surplus / deficit increases (i.e., the grid needs more active power), can the ion concentration remain within a stable range? If the ion concentration fluctuates little and remains stable, it indicates that the active power surplus / deficit characteristic term supports the stability of the electrochemical response; if the ion concentration fluctuates beyond the normal range, leading to electrochemical instability, it indicates that the characteristic term interferes with electrochemical stability. By establishing a relationship model between the stability of the second type of node characteristic terms and the first type of node characteristic terms, the type of constraint can be determined.

[0037] Step S1227: Classify the second type of node feature items of the cross-domain feature association spectrum that belong to the type that supports electrochemical stability and whose correlation strength meets the requirements as the adaptation features related to electrochemical response, and record the degree of constraint corresponding to each adaptation feature related to electrochemical response.

[0038] In this embodiment, similar to step S1225, the requirement of meeting the correlation strength requirement also means that the characterization weight of the correlation edge of the feature item pair exceeds a preset correlation threshold. For second-type node feature items that are determined to support the electrochemical stability type and whose correlation strength meets the requirement, they are classified as adaptive features. For example, if the frequency deviation range feature item in the frequency regulation response data is determined to support the electrochemical stability type, and the characterization weight of its correlation edge with the migration current feature item in the charge migration rate data exceeds a preset correlation threshold, then the frequency deviation range feature item is classified as an adaptive feature. The degree of constraint is also measured by the characterization weight of the correlation edge. The larger the characterization weight, the greater the degree of constraint of the adaptive feature on the electrochemical characteristic response. The characterization weight of each adaptive feature is recorded as a quantitative indicator of its degree of constraint.

[0039] Step S1228: Collect all cross-domain feature association graph first-class node feature items classified as sensitive features related to grid acceptance and cross-domain feature association graph second-class node feature items classified as adaptability features related to electrochemical response, forming a sensitive feature list and an adaptability feature list, as the output results of cross-domain feature association processing.

[0040] In this application scenario, the node features classified as sensitive features (Category I) are sorted from highest to lowest impact to form a sensitive feature list. Each element in the list includes the feature name, the category of electrochemical characteristic data to which it belongs, the type of impact, and the degree of impact. Similarly, the node features classified as adaptation features (Category II) are sorted from highest to lowest constraint to form an adaptation feature list. Each element in the list includes the feature name, the category of dynamic grid acceptance data to which it belongs, the type of constraint, and the degree of constraint.

[0041] Step S123: Perform dimensional unification processing on the extracted sensitive features and adaptive features, and assign corresponding weights according to the degree of influence of the sensitive features and adaptive features on grid connection control to obtain a weighted set of coupling features.

[0042] In this industrial park energy storage system application scenario, sensitive features and adaptability features may have different dimensions. For example, the number of feature items in the sensitive feature list may be specific, and the number of feature items in the adaptability feature list may also be specific. Their dimensions need to be unified for subsequent coupling processing. Dimension unification is achieved through feature concatenation, where feature items from the sensitive and adaptability feature lists are concatenated sequentially to form a new high-dimensional feature vector. For instance, the influence level of each feature item in the sensitive feature list is used as one element of the vector, and the constraint level of each feature item in the adaptability feature list is used as another element. The concatenated vector has a length equal to the sum of the number of sensitive and adaptability features. Then, weights are assigned to each feature item based on its influence level and the constraint level of the adaptability features. Feature items with higher influence or constraint levels have higher weights. Weight allocation uses a normalization method, with the sum of the influence or constraint levels of all feature items used as the denominator, and the influence or constraint level of each feature item used as the numerator, to obtain the weight value for each feature item. Finally, the value of each feature term is multiplied by its corresponding weight value to obtain the weighted feature value. All the weighted feature values ​​form a weighted coupled feature set, which contains the weighted information of sensitive features and adaptation features.

[0043] Step S124: Transmit the weighted coupling feature set to the time-series correlation modeling layer, perform time-series dynamic modeling on the weighted coupling feature set, capture the correlation change pattern between electrochemical characteristic data and dynamic acceptance data at different time nodes, and generate a time-series correlation model.

[0044] In this embodiment, the weighted coupling feature set is a time-varying sequence of data, with each time node corresponding to a coupling feature vector. This sequence data is transmitted to the temporal correlation modeling layer, which uses a Long Short-Term Memory (LSTM) network as the core modeling tool to capture dynamic temporal correlations. LSTM networks can effectively handle dependencies in long sequence data and are suitable for analyzing the complex time-varying correlations between electrochemical characteristic data and grid dynamic acceptance data.

[0045] Step S1241: Divide the weighted set of coupling features into multiple time segments according to time order, and each time segment contains coupling feature data at the corresponding time node.

[0046] In this application scenario, based on the operating characteristics of the energy storage system and the time scale of grid dispatch, the time axis is divided into multiple time windows of equal length, each of which is a time series segment. For example, the time window length can be set to a specific minute, and each time series segment contains coupled feature data from all sampling moments within that specific minute. Assuming the sampling frequency is a specific Hertz, each time series segment contains a specific number of coupled feature vectors. In this way, continuous time series data is converted into multiple discrete time series segments, facilitating batch processing by the LSTM network.

[0047] Step S1242: Input the first time segment into the input gate of the long short-term memory unit of the time series correlation modeling layer. Use the input gate of the long short-term memory unit to filter the importance of the coupling feature data in the time segment, retain the feature data that is valuable for time series correlation modeling, and filter out redundant feature data.

[0048] In this embodiment, the input gate of the Long Short-Term Memory (LSTM) unit consists of a sigmoid activation function and a tanh activation function. First, the coupled feature data from the first time segment is input to the sigmoid function of the input gate, which outputs a value between 0 and 1 to determine the importance of each feature data point; the closer the value is to 1, the more important the feature data, and the closer it is to 0, the less important it is. Then, the coupled feature data is input to the tanh function to generate a new candidate value vector. Finally, the output of the sigmoid function and the output of the tanh function are multiplied element-wise to obtain the output of the input gate. This output retains important feature data and filters out redundant feature data. For example, for a certain feature item in the coupled feature set, if the sigmoid function output value is a specific value, then the weight of that feature item in the input gate output is its original value multiplied by this specific value, thus achieving importance filtering.

[0049] Step S1243: The feature association information in the historical time sequence is filtered through the forget gate of the long short-term memory unit, and the current time sequence feature data after filtering is fused with the retained historical association information through the cell state of the long short-term memory unit to generate the cell state update data of the current time node.

[0050] In this application scenario, the forget gate also employs the sigmoid activation function. Its input consists of the cell state from the previous time point and the feature data of the current time segment after processing by the input gate. The forget gate determines which information to discard from the cell state; the closer the sigmoid function output value is to 0, the more the corresponding historical association information needs to be forgotten, while the closer it is to 1, the more it needs to be retained. The output of the forget gate is multiplied element-wise with the cell state from the previous time point to obtain the retained historical association information. Simultaneously, the output of the input gate is added to a candidate cell state processed by the tanh function to obtain the updated cell state data for the current time point. Through this method, the fusion of current feature data and historical association information is achieved, enabling the cell state to reflect long-term temporal relationships.

[0051] Step S1244: The cell state update data is filtered by the output gate of the long short-term memory unit, and the feature information that can characterize the correlation between the electrochemical characteristic data and the dynamic acceptance data at the current time point is extracted as the current time series output.

[0052] In this embodiment, the output gate consists of a sigmoid function and a tanh function. First, the cell state update data of the current time node is input into the tanh function, compressing its value to between -1 and 1. Then, the hidden state of the previous time node and the feature data of the current time segment are input into the sigmoid function to obtain an output gate control value. Finally, the output of the tanh function is multiplied element-wise with the output of the sigmoid function to obtain the current time-series output. This current time-series output contains key feature information about the correlation between the electrochemical characteristic data and the dynamic acceptance data of the current time node.

[0053] Step S1245: Process all time segments sequentially to obtain the time output corresponding to each time node, and arrange all time outputs in chronological order to form a time-related sequence.

[0054] In this application scenario, the second, third, and so on, up to the last time segment, are sequentially input into the Long Short-Term Memory (LSTM) unit according to time order. Steps S1242 to S1244 are repeated to obtain the time-series output corresponding to each time segment. These time-series outputs are then arranged according to their corresponding time nodes to form a complete time-series correlation sequence. Each element in this sequence corresponds to a time node and contains the correlation characteristics between electrochemical characteristic data and dynamic acceptance data at that time node.

[0055] Step S1246: Based on the time-series correlation sequence, analyze the change magnitude of the correlation between adjacent time nodes, and at the same time analyze the change trend of the correlation between time interval nodes to construct a correlation change law model.

[0056] In this embodiment, the magnitude of change in the correlation between adjacent time nodes is measured by calculating the Euclidean distance between two adjacent elements in the time-series correlation sequence; a larger Euclidean distance indicates a greater magnitude of change. For the changing trend of the correlation between time-interval time nodes, a sliding window method is used, setting different time intervals and calculating the correlation coefficient between elements in the time-series correlation sequence at different intervals. A positive correlation coefficient indicates a positive correlation trend, and a negative correlation indicates a negative correlation trend; the absolute value indicates the trend strength. Based on the calculated magnitude and trend of change, a correlation change pattern model is constructed. This correlation change pattern model can be a statistical learning-based model, such as using support vector machines or decision trees, with time nodes, magnitude of change, and trend as input features to predict changes in the correlation at future time nodes.

[0057] Step S1247: Integrate the correlation change law model with the network parameters of the long short-term memory unit to form a temporal correlation model that can capture temporal correlation changes.

[0058] In this application scenario, the network parameters of the Long Short-Term Memory (LSTM) unit include the weight matrices and bias terms of the input gate, forget gate, and output gate. These parameters are optimized during model training using the backpropagation algorithm. Integrating the association change pattern model with these network parameters means treating the association change pattern model as an additional module of the LSTM network, connecting it to the output layer of the LSTM network. This allows the temporal association model to not only output temporal association sequences but also predict future changes in association relationships based on the association change pattern model. The integrated temporal association model can more comprehensively capture the temporal association change patterns between electrochemical characteristic data and dynamic acceptance data.

[0059] Step S125: Analyze the impact trend of changes in electrochemical characteristics on grid acceptance capacity based on the time-series correlation model, and at the same time analyze the constraints of changes in grid acceptance capacity on electrochemical characteristics, and construct a two-way influence relationship matrix.

[0060] In this embodiment, firstly, based on the time-series correlation sequence output by the time-series correlation model, the time-varying curves of sensitive features in the electrochemical characteristic data and the time-varying curves of adaptability features in the grid dynamic acceptance data are extracted. Then, by calculating the cross-correlation coefficient between the sensitive feature change curves and the adaptability feature change curves, the influence trend of electrochemical characteristic changes on grid acceptance capability is analyzed. A positive and large correlation coefficient indicates that the enhancement of electrochemical characteristics will significantly promote the improvement of grid acceptance capability, while a negative and large correlation coefficient indicates that the enhancement of electrochemical characteristics will limit grid acceptance capability. The constraint requirements of grid acceptance capability changes on electrochemical characteristics are determined by analyzing the constraint boundary of the adaptability feature change curve on the sensitive feature change curve. For example, when the adaptability feature is within a certain range, the sensitive feature can remain stable; outside this range, the sensitive feature becomes unstable. Based on these analysis results, a two-dimensional matrix is ​​constructed. The rows of the matrix represent different sensitive features of electrochemical characteristics, and the columns represent different adaptability features of grid acceptance capability. The element values ​​in the matrix represent the degree of bidirectional influence between the corresponding sensitive feature and adaptability feature; positive values ​​indicate a promoting or supporting effect, negative values ​​indicate a limiting or interfering effect, and the absolute value indicates the degree of influence, thus forming a bidirectional influence relationship matrix.

[0061] Step S126: Transmit the bidirectional influence relationship matrix to the collaborative strategy generation layer, and perform strategy mapping processing on the bidirectional influence relationship matrix through the multilayer perceptron unit in the collaborative strategy generation layer to generate multiple sets of initial grid-connected control strategies.

[0062] In this embodiment, the core of the collaborative strategy generation layer is a multilayer perceptron unit, which consists of multiple hidden layers and an output layer. It can map the high-dimensional features of the bidirectional influence relationship matrix into specific grid-connected control strategy parameters.

[0063] For example, in step S1261: the bidirectional influence relationship matrix is ​​transmitted to the collaborative strategy generation layer, and the key correlation dimensions characterizing the relationship between changes in electrochemical characteristics and grid acceptance capability are extracted from the bidirectional influence relationship matrix. The key correlation dimensions include the charge migration rate correlation dimension, the electrode reactivity correlation dimension, and the electrolyte distribution correlation dimension.

[0064] In this application scenario, the rows in the bidirectional influence matrix correspond to sensitive features of electrochemical properties. These sensitive features belong to different categories such as charge migration rate, electrode reactivity, and electrolyte distribution. Therefore, the matrix rows corresponding to sensitive features belonging to the same category can be merged to obtain three key correlation dimensions. For example, by averaging or weighting the row vectors corresponding to all sensitive features related to charge migration rate, a vector representation of the charge migration rate correlation dimension can be obtained; similarly, vector representations of the electrode reactivity correlation dimension and the electrolyte distribution correlation dimension can be obtained. Each key correlation dimension is a vector containing the degree of correlation with all suitable features, which can centrally reflect the correlation between electrochemical properties and grid acceptance capability under that dimension.

[0065] Step S1262: Perform feature enhancement processing on the extracted key correlation dimensions to obtain the feature-enhanced bidirectional influence relationship matrix.

[0066] In this embodiment, the feature enhancement process employs Principal Component Analysis (PCA) to reduce the dimensionality of the vectors in the three key correlation dimensions, removing redundant information and retaining the main features. Simultaneously, nonlinear transformations are applied to the dimensionality-reduced feature vectors, such as using polynomial kernel functions or radial basis functions, to increase the nonlinear expressive power of the features. Furthermore, external knowledge, such as operational experience data of energy storage systems and grid dispatch rules, can be introduced as additional features to the key correlation dimensions, further enriching the feature information. Through these processes, a feature-enhanced bidirectional influence relationship matrix is ​​obtained, which possesses higher information density and stronger expressive power.

[0067] Step S1263: Determine multiple association strength screening thresholds. The association strength screening thresholds are used to distinguish feature data with different degrees of association in key association dimensions. Each association strength screening threshold corresponds to a degree of association judgment criterion.

[0068] In this application scenario, the determination of the association strength screening threshold is based on the distribution of element values ​​in the bidirectional influence relationship matrix. First, the value range of all elements in the matrix is ​​statistically analyzed. Then, according to the requirements for classifying the degree of association in the actual application, the value range is divided into multiple intervals, and the boundary value of each interval is a correlation strength screening threshold. For example, element values ​​can be divided into multiple levels such as extremely strong association, strong association, moderate association, weak association, and extremely weak association. Each level corresponds to a correlation strength screening threshold; for example, the screening threshold for extremely strong association is a specific value, and the screening threshold for strong association is a specific value, etc. Each correlation strength screening threshold corresponds to a correlation degree judgment standard; that is, when the matrix element value is greater than or equal to the threshold, it is judged as having the corresponding level of association degree.

[0069] Step S1264: Select the first association strength screening threshold, apply the association strength screening threshold to the bidirectional influence relationship matrix after feature enhancement, and screen out feature data whose association strength exceeds the association strength screening threshold to form the first screening matrix segment.

[0070] In this embodiment, the first correlation strength screening threshold is typically set to the threshold corresponding to the highest level of correlation, such as the threshold for extremely strong correlation. This threshold is applied to the feature-enhanced bidirectional influence relationship matrix. Each element in the matrix is ​​traversed; if the element value is greater than or equal to the threshold, the element and its corresponding row and column index information are retained; otherwise, the element value is set to 0 or the row / column is deleted. Through this method, a first screening matrix fragment is obtained. This first screening matrix fragment contains only feature data with extremely strong correlation strength, highlighting the most critical correlations.

[0071] Step S1265: Input the first selection matrix fragment into the first hidden layer of the multilayer perceptron unit in the collaborative strategy generation layer, and perform a nonlinear transformation on the first selection matrix fragment through the activation function in the first hidden layer to convert the matrix-form feature data into the first-level feature mapping result in vector form.

[0072] In this application scenario, the first hidden layer of the multilayer perceptron unit contains a specific number of neurons, each connected to all elements of the first selection matrix segment. The activation function is the ReLU function, meaning that for each neuron's input value, if it is greater than 0, the corresponding value is output; otherwise, 0 is output. The first selection matrix segment is first flattened into a one-dimensional vector, serving as the input to the first hidden layer. Then, a linear transformation is performed using the weight matrix and bias term of this layer, followed by a non-linear transformation using the ReLU activation function, yielding the first-level feature mapping result. This first-level feature mapping result is a vector with the same number of neurons as the first hidden layer, containing the feature information after the non-linear transformation.

[0073] Step S1266: Input the first-level feature mapping result into the second hidden layer of the multilayer perceptron unit, and perform interactive operations on the different dimensional features in the first-level feature mapping result through the cross-dimensional interaction unit in the second hidden layer to capture the potential correlation information between different key correlation dimensions and generate the second-level feature mapping result.

[0074] In this embodiment, the cross-dimensional interaction unit of the second hidden layer is implemented using feature concatenation and fully connected layers. First, the first-level feature mapping result is divided according to key correlation dimensions, for example, into three sub-vectors: charge migration rate correlation dimension features, electrode reactivity correlation dimension features, and electrolyte distribution correlation dimension features. Then, these sub-vectors are concatenated pairwise or in groups of three to form new interaction feature vectors. Next, these interaction feature vectors are input into the fully connected neurons of the second hidden layer, where a linear transformation is performed using the weight matrix and bias term, followed by a non-linear transformation using the ReLU activation function to generate the second-level feature mapping result. This second-level feature mapping result not only contains the feature information of each key correlation dimension itself but also the interaction information between different dimensions, thus more comprehensively reflecting the potential correlations between features.

[0075] Step S1267: Input the second-level feature mapping result into the output layer of the multilayer sensor unit, and convert the second-level feature mapping result into a specific grid-connected control parameter combination through the strategy mapping function in the output layer. The grid-connected control parameter combination includes electrochemical parameter adjustment suggestions and transmission parameter adjustment suggestions, forming the first set of initial grid-connected control strategies.

[0076] In this application scenario, the number of neurons in the output layer corresponds to the number of grid-connected control parameters. For example, electrochemical parameter adjustment suggestions include adjustments for charge migration rate, electrode reactivity, and electrolyte distribution; transmission parameter adjustment suggestions include adjustments for active power transmission, reactive power transmission, voltage adjustment, and frequency adjustment. The number of neurons in the output layer is the sum of these parameters. The strategy mapping function uses a linear function, transforming the second-level feature mapping results linearly through the weight matrix and bias term of the output layer to obtain the specific value or adjustment range of each grid-connected control parameter, forming electrochemical parameter adjustment suggestions and transmission parameter adjustment suggestions. These suggestions are combined to form the first set of initial grid-connected control strategies. This first set of initial grid-connected control strategies is generated for feature data with extremely strong correlation and has a high priority.

[0077] Step S1268: Select the second correlation strength screening threshold, repeat the above steps of screening matrix fragment generation, first-level feature mapping, second-level feature mapping and strategy conversion, apply the second correlation strength screening threshold to the bidirectional influence relationship matrix after feature enhancement to obtain the second screening matrix fragment, and generate the second set of initial grid-connected control strategies after processing by the multilayer perceptron unit.

[0078] In this embodiment, the second association strength screening threshold is the threshold corresponding to the second-highest level of association, such as the threshold corresponding to strong association. Following the same process as steps S1264 to S1267, this threshold is applied to the bidirectional influence relationship matrix after feature enhancement to screen out feature data with strong association strength, forming a second screening matrix segment. Then, this segment is input into the first hidden layer of the multilayer perceptron unit for nonlinear transformation to obtain the first-level feature mapping result; then it is input into the second hidden layer for cross-dimensional interaction to obtain the second-level feature mapping result; finally, the policy mapping function of the output layer generates a second set of initial grid-connected control strategies. This second set of initial grid-connected control strategies is generated for feature data with strong association strength, supplementing the association relationships not covered by the first set of strategies.

[0079] Step S1269: The remaining correlation strength screening thresholds are applied sequentially to the bidirectional influence relationship matrix after feature enhancement, and corresponding screening matrix fragments are generated respectively. Each screening matrix fragment is transformed by the first hidden layer nonlinear transformation, the second hidden layer cross-dimensional interaction and the output layer strategy mapping of the multilayer perceptron unit to generate the corresponding initial grid-connected control strategy.

[0080] In this application scenario, the remaining correlation strength screening thresholds include thresholds corresponding to medium correlation and weak correlation. Following the same method as steps S1264 to S1268, the above thresholds are sequentially applied to the feature-enhanced bidirectional influence relationship matrix to generate corresponding screening matrix fragments. These fragments are then processed by a multilayer perceptron unit to generate multiple sets of initial grid-connected control strategies. Each strategy corresponds to feature data with different degrees of correlation, enabling the generated initial grid-connected control strategies to cover various possible correlation relationships, thus improving the diversity and comprehensiveness of the strategies.

[0081] Step S12610: Collect all initial grid connection control strategies generated through different correlation strength screening thresholds, form multiple sets of initial grid connection control strategies, and associate and label the multiple sets of initial grid connection control strategies with the corresponding correlation strength screening thresholds.

[0082] In this embodiment, each group of generated initial grid-connected control strategies is arranged in descending order of correlation strength screening threshold, and a correlation strength tag is added to each group of strategies, indicating its corresponding screening threshold and correlation level. For example, the first group of strategies is marked as "extremely strong correlation," corresponding to the first screening threshold; the second group is marked as "strong correlation," corresponding to the second screening threshold, and so on. The correlation tags facilitate subsequent collaborative analysis and screening of the initial grid-connected control strategies.

[0083] Step S127: Perform a synergy analysis on multiple initial grid connection control strategies, select the initial grid connection control strategy that meets the requirements for stable electrochemical characteristics and is compatible with the grid acceptance capability, use the initial grid connection control strategy as the grid connection collaborative control strategy, and associate and store the grid connection collaborative control strategy with the corresponding two-way influence relationship matrix.

[0084] In this embodiment, the synergy analysis is mainly conducted from two aspects: first, the stability requirements of electrochemical characteristics, and second, the grid acceptance capability adaptation requirements. Regarding the electrochemical characteristic stability requirements, based on the design parameters and operational experience of the energy storage system, stable ranges for electrochemical characteristic parameters are set, such as the allowable fluctuation range of charge migration rate, the minimum threshold of electrode reactivity, and the uniformity index of electrolyte distribution. The electrochemical parameter adjustment suggestions in each initial grid connection control strategy are applied to the digital twin model or simulation platform of the energy storage system to simulate the changes in the electrochemical characteristics of the energy storage system under this strategy, and to check whether the simulation results are within the stable range. Regarding the grid acceptance capability adaptation requirements, the transmission parameter adjustment suggestions in each initial grid connection control strategy are compared with the dynamic grid acceptance data to check whether the active and reactive power transmission amounts are within the real-time power surplus / deficit range of the grid, and whether the voltage and frequency adjustments meet the grid voltage fluctuation tolerance and frequency regulation response requirements. Simultaneously, the synergy between electrochemical parameter adjustments and transmission parameter adjustments is analyzed to ensure that the adjustment directions and magnitudes are consistent, avoiding conflicts. For example, if the electrochemical parameter adjustment recommendations call for increasing the charge migration rate to improve output power, while the transmission parameter adjustment recommendations limit power output, then this set of strategies has poor synergy. Based on the above analysis, an initial grid-connected control strategy that simultaneously meets the requirements of electrochemical stability and grid acceptance capability adaptation, and exhibits good synergy, is selected as the final grid-connected coordinated control strategy. Then, this grid-connected coordinated control strategy and its corresponding bidirectional influence matrix are stored in a database, and an index is established for subsequent querying and tracing.

[0085] Step S130: Decompose the grid-connected collaborative control strategy into electrochemical parameter adjustment instructions for the containerized electrochemical energy storage system and transmission parameter adjustment instructions for the grid-connected interface, resulting in two sets of associated adjustment instructions.

[0086] In this embodiment, the grid-connected coordinated control strategy includes adjustment requirements for the electrochemical state inside the energy storage system and the transmission state of the grid-connected interface. These requirements need to be broken down into specific instructions so that they can be sent to the electrochemical control module of the energy storage system and the transmission control module of the grid-connected interface for execution.

[0087] Step S131: Analyze the strategy elements of the grid-connected collaborative control strategy, and extract the electrochemical control elements and transmission control elements contained in the grid-connected collaborative control strategy. The electrochemical control elements are used to adjust the electrochemical state inside the energy storage system, and the transmission control elements are used to adjust the transmission state of the grid-connected interface.

[0088] In this application scenario, strategy element analysis employs entity recognition and relation extraction techniques from natural language processing to analyze the textual descriptions or structured data of the grid-connected collaborative control strategy. Electrochemical control elements include information such as adjustment parameters, adjustment direction, and adjustment magnitude related to charge migration rate, electrode reactivity, and electrolyte distribution, such as "increasing the charge migration rate by a specific proportion" or "maintaining electrode reactivity within a specific range." Transmission control elements include information such as adjustment parameters, adjustment direction, and adjustment magnitude related to power transmission, voltage adaptation, and frequency synchronization, such as "increasing active power output by a specific value" or "adjusting the grid-connected voltage to a specific range." Through element analysis, the grid-connected collaborative control strategy is decomposed into a set of electrochemical control elements and a set of transmission control elements.

[0089] Step S132: Determine the types of electrochemical parameters that need to be adjusted based on the electrochemical control elements. The types of electrochemical parameters include charge migration rate parameters, electrode reactivity parameters, and electrolyte distribution parameters. At the same time, determine the adjustment direction and adjustment range corresponding to the charge migration rate parameters, the adjustment direction and adjustment range corresponding to the electrode reactivity parameters, and the adjustment direction and adjustment range corresponding to the electrolyte distribution parameters.

[0090] In this embodiment, the types of electrochemical parameters that need to be adjusted are determined based on the specific content of the electrochemical control element set. For example, if the electrochemical control elements include adjustment requirements related to "charge migration rate," then the charge migration rate parameter is determined to be the type that needs adjustment; similarly, electrode reactivity parameters and electrolyte distribution parameters are determined. For each parameter type, the adjustment direction and adjustment magnitude are extracted from the electrochemical control elements. The adjustment direction includes increasing, decreasing, maintaining, etc., and the adjustment magnitude can be a specific numerical change, a percentage change, or a target range.

[0091] Step S1321: Classify the element attributes of electrochemical control elements into charge migration correlation attributes, electrode reaction correlation attributes, and electrolyte distribution correlation attributes.

[0092] In this application scenario, the element attribute of an electrochemical control element refers to the type of electrochemical parameter associated with each control element. For example, the control element "reducing interface impedance by a specific value" is associated with the charge migration rate parameter, so its element attribute is a charge migration-related attribute; "maintaining the electrode surface redox potential within a specific range" is associated with the electrode reactivity parameter, so its element attribute is an electrode reactivity-related attribute; and "improving the uniformity of electrolyte ion concentration distribution by a specific level" is associated with the electrolyte distribution parameter, so its element attribute is an electrolyte distribution-related attribute. By classifying the element attributes of each electrochemical control element, the adjustment requirements for various electrochemical parameters can be more clearly determined.

[0093] Step S1322: Based on the trend of the change of the attribute value of the charge migration associated attribute, determine whether the charge migration rate parameter needs to be adjusted. If the trend of the change of the attribute value of the charge migration associated attribute indicates that the charge migration rate cannot meet the grid acceptance requirements, then it is determined that the charge migration rate parameter needs to be adjusted.

[0094] In this embodiment, the trend of the charge migration-related attribute values ​​is determined by analyzing their time-varying curves. For example, if the interface impedance value in the charge migration-related attribute continues to increase, based on historical data and simulation models, when the interface impedance increases to a specific value, the charge migration rate will decrease to the point where it cannot meet the grid's active power output requirements. In this case, it is determined that the charge migration rate parameter needs to be adjusted. During the determination process, the power surplus / deficit data in the grid's acceptance requirements needs to be considered. If the power output capacity corresponding to the current charge migration rate is less than the grid's active power demand, then the charge migration rate parameter needs to be adjusted to increase power output.

[0095] Step S1323: Based on the mapping relationship between charge migration correlation attributes and charge migration rate parameters, calculate the magnitude of adjustment required for the charge migration rate parameters, and determine the adjustment direction of the charge migration rate parameters according to the changing direction of grid acceptance requirements.

[0096] In this application scenario, the mapping relationship between charge migration correlation attributes and charge migration rate parameters is established through experimental data and theoretical models. For example, there is a negative correlation between interface impedance and charge migration rate, which can be fitted to obtain an empirical formula with interface impedance as input and charge migration rate as output. Based on this mapping relationship, when the attribute value of the charge migration correlation attribute needs to be adjusted to a specific target value, the target value of the corresponding charge migration rate parameter can be calculated. The target value is compared with the current value to obtain the required adjustment range. The adjustment direction is determined according to the changing direction of grid acceptance requirements. If the grid needs to increase active power acceptance, the charge migration rate needs to be increased, and the adjustment direction is to increase; if it needs to reduce active power output, the adjustment direction is to decrease.

[0097] Step S1324: Similarly, based on the trend of the change of the attribute values ​​of the electrode reaction-related attributes, determine whether it is necessary to adjust the electrode reaction activity parameters. If the trend of the change of the attribute values ​​of the electrode reaction-related attributes indicates that the electrode reaction activity cannot maintain the stable operation of the energy storage system, it is determined that the electrode reaction activity parameters need to be adjusted.

[0098] In this embodiment, the method for analyzing the trend of changes in the attribute values ​​of electrode reaction-related attributes is similar to that for charge migration-related attributes. For example, if the reaction current density in the electrode reaction activity parameter continuously decreases, and when it drops to a certain threshold, the charge-discharge efficiency of the electrode will significantly decrease, potentially causing the energy storage system to malfunction. In this case, it is determined that the electrode reaction activity parameter needs to be adjusted. In addition, the side reactions on the electrode surface must also be considered. If the concentration of side reaction products increases over time, it indicates that the electrode reaction activity is suppressed, and the electrode reaction activity parameter also needs to be adjusted.

[0099] Step S1325: Based on the mapping relationship between electrode reaction correlation attributes and electrode reaction activity parameters, calculate the adjustment range of electrode reaction activity parameters, and determine the adjustment direction of electrode reaction activity parameters according to the stable operation requirements of the energy storage system.

[0100] In this application scenario, the mapping relationship between electrode reaction-related attributes and electrode reaction activity parameters is also established based on experimental and theoretical models. For example, there is a specific functional relationship between redox potential and reaction current density; changes in reaction current density can be calculated by measuring changes in redox potential. Based on the stable operation requirements of the energy storage system, such as maintaining charge / discharge efficiency above a certain percentage and achieving a certain number of cycle lives, target values ​​for electrode reaction activity parameters are determined. The difference between the current value and the target value is calculated to obtain the adjustment range. The adjustment direction is determined based on the trend of attribute value changes and stable operation requirements. If the reaction activity decreases, it needs to be increased, and the adjustment direction is to increase; if excessive activity leads to exacerbated side reactions, the adjustment direction is to decrease.

[0101] Step S1326: Based on the trend of the attribute values ​​of the electrolyte distribution associated attributes, determine whether the electrolyte distribution parameters need to be adjusted. If the trend of the attribute values ​​of the electrolyte distribution associated attributes indicates that the uneven distribution of electrolyte affects the electrochemical characteristics, then it is determined that the electrolyte distribution parameters need to be adjusted.

[0102] In this embodiment, the attribute values ​​associated with electrolyte distribution include pressure distribution non-uniformity and ion concentration gradient. If pressure distribution non-uniformity increases over time, it may lead to poor electrolyte flow, excessively low or high ion concentrations in local areas, affecting charge migration and electrode reactions. In this case, it is determined that the electrolyte distribution parameters need to be adjusted. By analyzing the changing trends of the attribute values ​​associated with electrolyte distribution, combined with the changes in parameters such as charge migration rate and electrode reactivity in the electrochemical characteristic data, a comprehensive judgment is made as to whether the electrolyte distribution has an adverse effect on the electrochemical characteristics.

[0103] Step S1327: Based on the mapping relationship between electrolyte distribution correlation attributes and electrolyte distribution parameters, calculate the adjustment range of electrolyte distribution parameters, and determine the adjustment direction of electrolyte distribution parameters according to the requirement of uniform electrolyte distribution.

[0104] In this application scenario, the mapping relationship between electrolyte distribution attributes and electrolyte distribution parameters is established through fluid dynamics simulation and experimental measurements. For example, there is a specific relationship between the pressure difference and flow rate within the electrolyte flow channel; adjusting the pressure difference can change the flow rate, thereby affecting the electrolyte distribution. Based on the requirements for uniform electrolyte distribution, such as the ion concentration gradient needing to be less than a specific value and the pressure distribution non-uniformity needing to be less than a specific value, the target state of the electrolyte distribution parameters is determined, and the difference between the current state and the target state is calculated to obtain the adjustment range. The adjustment direction is determined based on the difference; for example, if the ion concentration in a certain area is too low, it is necessary to increase the electrolyte flow rate in that area, and the adjustment direction is to increase the pressure in that area or change the flow path.

[0105] Step S1328: Summarize the type information of the charge migration rate parameter, electrode reactivity parameter and electrolyte distribution parameter that need to be adjusted, and record the adjustment direction and adjustment range of the charge migration rate parameter, the electrode reactivity parameter and the electrolyte distribution parameter, to form an electrochemical parameter adjustment list.

[0106] In this embodiment, the types, directions, and magnitudes of electrochemical parameters requiring adjustment determined in steps S1322 to S1327 are summarized and categorized according to parameter type to form an electrochemical parameter adjustment list. Each entry in the list includes information such as parameter name, current value, target value, adjustment direction, adjustment magnitude, and adjustment priority. For example, the charge migration rate parameter entry includes adjustment information for sub-parameters such as interfacial impedance and migration current; the electrode reactivity parameter entry includes adjustment information for sub-parameters such as redox potential and reaction current density; and the electrolyte distribution parameter entry includes adjustment information for sub-parameters such as pressure distribution and ion concentration.

[0107] Step S133: Generate independent electrochemical parameter adjustment sub-instructions for the charge migration rate parameter based on the determined charge migration rate parameter type, the corresponding adjustment direction and adjustment magnitude, generate independent electrochemical parameter adjustment sub-instructions for the electrode reactivity parameter based on the determined electrode reactivity parameter type, the corresponding adjustment direction and adjustment magnitude, and generate independent electrochemical parameter adjustment sub-instructions for the electrolyte distribution parameter based on the determined electrolyte distribution parameter type, the corresponding adjustment direction and adjustment magnitude. All independent electrochemical parameter adjustment sub-instructions constitute a set of electrochemical parameter adjustment instructions for the containerized electrochemical energy storage system.

[0108] In this embodiment, the format of the independent electrochemical parameter adjustment sub-instruction for the charge migration rate parameter includes fields such as instruction identifier, parameter type, adjustment direction, adjustment magnitude, target value, and execution time. For example, the instruction identifier is used to uniquely identify the sub-instruction, the parameter type is clearly defined as interfacial impedance in the charge migration rate parameter, the adjustment direction is reduction, the adjustment magnitude is a specific percentage, the target value is a specific value, and the execution time is a specific moment. Based on the information in the electrochemical parameter adjustment list, a corresponding independent electrochemical parameter adjustment sub-instruction is generated for each sub-parameter that needs adjustment. Similarly, independent adjustment sub-instructions are generated for each sub-parameter of the electrode reactivity parameter and the electrolyte distribution parameter. All these independent sub-instructions are sorted according to parameter type and adjustment priority to form an electrochemical parameter adjustment instruction set.

[0109] Step S134: Determine the types of transmission parameters that need to be adjusted based on the transmission control elements. The types of transmission parameters include power transmission parameters, voltage adaptation parameters, and frequency synchronization parameters. At the same time, determine the adjustment direction and adjustment range corresponding to the power transmission parameters, the adjustment direction and adjustment range corresponding to the voltage adaptation parameters, and the adjustment direction and adjustment range corresponding to the frequency synchronization parameters.

[0110] In this embodiment, the transmission control elements include various parameter adjustment requirements related to grid-connected interface transmission. By analyzing these elements, the types of transmission parameters requiring adjustment are determined. Power transmission parameters include active power transmission and reactive power transmission; voltage adaptation parameters include grid-connected voltage amplitude and voltage phase; frequency synchronization parameters include output frequency and frequency synchronization accuracy. For each parameter type, the adjustment direction and adjustment magnitude are extracted from the transmission control elements, such as "increasing the active power transmission by a specific value" or "adjusting the grid-connected voltage amplitude to a specific range." The adjustment direction is determined based on real-time power surplus / deficit, voltage fluctuation tolerance, and frequency regulation response data of the power grid, while the adjustment magnitude is determined comprehensively based on the grid's acceptance capacity and the energy storage system's output capacity.

[0111] Step S135: Generate independent transmission parameter adjustment sub-instructions for power transmission parameters based on the determined power transmission parameter type, the corresponding adjustment direction and adjustment magnitude of the power transmission parameters; generate independent transmission parameter adjustment sub-instructions for voltage adaptation parameters based on the determined voltage adaptation parameter type, the corresponding adjustment direction and adjustment magnitude of the voltage adaptation parameters; generate independent transmission parameter adjustment sub-instructions for frequency synchronization parameters based on the determined frequency synchronization parameter type, the corresponding adjustment direction and adjustment magnitude of the frequency synchronization parameters; all independent transmission parameter adjustment sub-instructions constitute a set of transmission parameter adjustment instructions for the grid-connected interface.

[0112] In this embodiment, the process of generating independent transmission parameter adjustment sub-instructions is similar to that of electrochemical parameter adjustment sub-instructions. Each sub-instruction includes fields such as instruction identifier, parameter type, adjustment direction, adjustment range, target value / range, and execution time.

[0113] Step S1351: For the power transmission parameters, determine the target value of the power transmission parameters according to the adjustment direction and adjustment range corresponding to the power transmission parameters. The target value is the result obtained by adjusting the current power transmission parameter value and adjustment range according to the adjustment direction.

[0114] In this application scenario, the current power transmission parameter values ​​are measured in real time by power sensors at the grid connection interface, for example, the current active power transmission is a specific value. When the adjustment direction is to increase, the target value is the current value plus the adjustment range; when the adjustment direction is to decrease, the target value is the current value minus the adjustment range. The adjustment range is determined based on the real-time power surplus / deficit data of the power grid. If the active power surplus / deficit of the power grid is positive and large, the adjustment range will be increased accordingly to fully utilize the grid's capacity.

[0115] Step S1352: Generate the instruction structure of the power transmission parameter adjustment sub-instruction. The instruction structure includes parameter type identifier, target value information and adjustment execution time information. The parameter type identifier is used to clarify that the instruction is aimed at the power transmission parameter.

[0116] In this embodiment, the parameter type is identified by a preset code, such as "PT" representing power transmission parameters, where "PTA" represents active power transmission and "PTQ" represents reactive power transmission. The target value information includes target active power and target reactive power values, represented in a specific format, such as "PTA: target value, PTQ: target value". The execution time information is adjusted to the start time and duration of the instruction, such as "start time: specific moment, duration: specific minutes". This information is combined into a fixed instruction structure, facilitating parsing and execution by the transmission control module of the grid-connected interface.

[0117] Step S1353: Combine the parameter type identifier, target value information, and adjustment execution time information according to a preset format to form an independent transmission parameter adjustment sub-instruction for power transmission parameters.

[0118] In this application scenario, the preset format can be a string concatenation, such as "instruction identifier|parameter type identifier|target value information|adjustment execution time information". For example, an independent transmission parameter adjustment sub-instruction for active power transmission might be "CMD001|PTA|target value|start time: specific time, duration: specific minutes". This format combination ensures the clarity and uniqueness of the instruction.

[0119] Step S1354: For the voltage adaptation parameters, following the same process as for the power transmission parameters, determine the target value range of the voltage adaptation parameters based on the adjustment direction and adjustment magnitude corresponding to the voltage adaptation parameters.

[0120] In this embodiment, the target value range of the voltage adaptation parameter is determined based on the grid voltage fluctuation tolerance data, for example, the allowable fluctuation range of the grid voltage is a specific range. When the adjustment direction is to increase the voltage, the target value range is the current voltage value plus the adjustment amplitude; when the adjustment direction is to decrease the voltage, it is the current voltage value minus the adjustment amplitude. The adjustment amplitude is determined based on the difference between the current value of the grid voltage and the boundary value of the allowable fluctuation range, ensuring that the adjusted voltage is within the allowable range.

[0121] Step S1355: Generate the instruction structure of the voltage adaptation parameter adjustment sub-instruction. The instruction structure includes parameter type identifier, target value range information and adjustment execution time information. The parameter type identifier is used to clarify that the instruction is aimed at the voltage adaptation parameter.

[0122] In this application scenario, the parameter type is identified as "VA," indicating a voltage adaptation parameter. The target value range information is expressed in the format of "upper limit - lower limit," such as "specific value - specific value." The adjustment execution time information is in the same format as the power transfer parameter adjustment sub-instruction.

[0123] Step S1356: Combine the parameter type identifier, target value range information, and adjustment execution time information according to a preset format to form an independent transmission parameter adjustment sub-instruction for voltage adaptation parameters.

[0124] In this embodiment, the combination format is similar to the power transmission parameter adjustment sub-instruction, such as "CMD002|VA|specific value-specific value|start time: specific time, duration: specific minutes".

[0125] Step S1357: For the frequency synchronization parameters, determine the target synchronization accuracy of the frequency synchronization parameters according to the adjustment direction and adjustment magnitude corresponding to the frequency synchronization parameters.

[0126] In this application scenario, the target synchronization accuracy of the frequency synchronization parameters is determined based on the allowable frequency deviation range and regulation response time requirements in the grid frequency regulation response data. When the adjustment direction is to improve synchronization accuracy, the target synchronization accuracy is the current synchronization accuracy value minus the adjustment magnitude; when the adjustment direction is to decrease synchronization accuracy, it is the current value plus the adjustment magnitude. The adjustment magnitude is determined based on the grid frequency stability requirements and the frequency regulation capability of the energy storage system.

[0127] Step S1358: Generate the instruction structure of the frequency synchronization parameter adjustment sub-instruction. The instruction structure includes parameter type identifier, target synchronization accuracy information and adjustment execution time information. The parameter type identifier is used to clarify that the instruction is aimed at the frequency synchronization parameter.

[0128] In this embodiment, the parameter type is identified as "FS", indicating frequency synchronization parameters. The target synchronization accuracy information is expressed in the format of "allowable deviation range, response time", such as "specific Hertz, specific milliseconds".

[0129] Step S1359: Combine the parameter type identifier, target synchronization accuracy information, and adjustment execution time information according to a preset format to form an independent transmission parameter adjustment sub-instruction for the frequency synchronization parameter.

[0130] In this application scenario, the combination format is, for example, "CMD003|FS|specific Hertz, specific milliseconds|start time: specific moment, duration: specific minutes".

[0131] Step S13510: Collect independent transmission parameter adjustment sub-instructions for power transmission parameters, independent transmission parameter adjustment sub-instructions for voltage adaptation parameters, and independent transmission parameter adjustment sub-instructions for frequency synchronization parameters, arrange them in order of priority of parameter adjustment, and form a set of transmission parameter adjustment instructions for grid-connected interfaces.

[0132] In this embodiment, the priority of parameter adjustments is determined based on the safety and stability requirements of power grid operation. Typically, frequency synchronization parameters have the highest priority, followed by voltage adaptation parameters, and finally power transmission parameters. The generated independent transmission parameter adjustment sub-instructions are sorted according to this priority to form a transmission parameter adjustment instruction set.

[0133] Step S136: Establish the correlation between the set of electrochemical parameter adjustment instructions for the containerized electrochemical energy storage system and the set of transmission parameter adjustment instructions for the grid-connected interface, so that the independent electrochemical parameter adjustment sub-instructions for charge migration rate parameters form correlation pairs with the corresponding independent transmission parameter adjustment sub-instructions, the independent electrochemical parameter adjustment sub-instructions for electrode reactivity parameters form correlation pairs with the corresponding independent transmission parameter adjustment sub-instructions, and the independent electrochemical parameter adjustment sub-instructions for electrolyte distribution parameters form correlation pairs with the corresponding independent transmission parameter adjustment sub-instructions, so that the adjustment directions of the two types of instructions in each correlation pair are consistent.

[0134] In this embodiment, the correlation is established based on the causal relationship between electrochemical parameter adjustments and transmission parameter adjustments. For example, if an independent electrochemical parameter adjustment sub-instruction for the charge migration rate parameter requires increasing the charge migration rate to improve active power output, then the corresponding independent transmission parameter adjustment sub-instruction should require increasing the active power transmission amount. Since the adjustment directions are consistent, they form a correlation pair. By analyzing the impact of electrochemical parameter adjustments on the energy storage system's output power, voltage, frequency, etc., and the requirements of transmission parameter adjustments on these outputs, a one-to-one correlation is established. For each electrochemical parameter adjustment sub-instruction, the transmission parameter adjustment sub-instruction related to its impact is found, forming a correlation pair, and the identification information of the correlation pair is recorded.

[0135] Step S137: Perform synergy verification on the independent electrochemical parameter adjustment sub-instructions and the independent transmission parameter adjustment sub-instructions in each associated pair, so that the adjustment range of the independent electrochemical parameter adjustment sub-instructions in each associated pair matches the adjustment range of the independent transmission parameter adjustment sub-instructions.

[0136] In this embodiment, the synergy verification is performed through simulation or calculation. For example, the electrochemical parameter adjustment sub-instruction in the associated pair is applied to the energy storage system model to calculate the changes in parameters such as power, voltage, and frequency that the energy storage system can output under this adjustment. Simultaneously, the adjustment magnitude of the corresponding transmission parameter adjustment sub-instruction is converted into the expected changes in power, voltage, and frequency. The consistency of these two changes is compared; if they are consistent, the synergy verification passes; if they are inconsistent, the adjustment magnitude of one of the sub-instructions is adjusted until they match. For example, if the active power increase calculated by the electrochemical parameter adjustment sub-instruction is a specific value, and the expected active power increase by the transmission parameter adjustment sub-instruction is also a specific value, then the adjustment magnitude of either the electrochemical parameter adjustment sub-instruction or the transmission parameter adjustment sub-instruction needs to be adjusted to make them equal.

[0137] Step S138: The set of electrochemical parameter adjustment instructions for the containerized electrochemical energy storage system and the set of transmission parameter adjustment instructions for the grid-connected interface, which have passed the synergy verification, are taken as the final two sets of associated adjustment instructions.

[0138] In this embodiment, the electrochemical parameter adjustment instruction set and the transmission parameter adjustment instruction set that have passed the collaboration verification undergo final format and integrity checks to ensure that the instruction format is correct, the parameters are complete, and there are no conflicts. Then, association identifiers are added to these two instruction sets to indicate their correspondence; for example, the sub-instructions in the two instruction sets are associated and indexed using the identifier information of the association pairs. Finally, these two sets of associated adjustment instruction sets are stored in an instruction issuance queue, awaiting dispatch to the corresponding control modules for execution.

[0139] Step S140: Adjust the internal electrochemical operating state of the containerized electrochemical energy storage system according to the electrochemical parameter adjustment command to obtain the adjusted electrochemical operating state data. At the same time, adjust the transmission operating parameters of the grid connection interface according to the transmission parameter adjustment command to obtain the adjusted transmission operating parameter data.

[0140] In this embodiment, after the electrochemical parameter adjustment instruction set and the transmission parameter adjustment instruction set are generated, they are sent to the electrochemical control module of the energy storage system and the transmission control module of the grid-connected interface, respectively, via a communication network. Upon receiving the instructions, the electrochemical control module controls the corresponding actuators to adjust the internal electrochemical operating state of the energy storage system; upon receiving the instructions, the transmission control module controls the power electronic devices of the grid-connected interface to adjust the transmission operating parameters. During the adjustment process, relevant data is collected in real time to form adjusted electrochemical operating state data and transmission operating parameter data.

[0141] Step S141: Transmit the set of electrochemical parameter adjustment instructions to the electrochemical control module of the containerized electrochemical energy storage system, so that the charge migration adjustment unit in the electrochemical control module of the containerized electrochemical energy storage system receives independent electrochemical parameter adjustment sub-instructions for the charge migration rate parameter.

[0142] In this application scenario, the electrochemical control module is a microprocessor system integrating multiple control units, communicating with the data processing center via industrial Ethernet or fieldbus. The set of electrochemical parameter adjustment instructions is transmitted using a specific communication protocol format, such as Modbus or IEC61850. Upon receiving the instruction set, the electrochemical control module parses it and, based on the parameter type identifier in the instruction, distributes independent electrochemical parameter adjustment sub-instructions targeting the charge migration rate parameter to the charge migration adjustment unit. The charge migration adjustment unit is a dedicated functional module within the electrochemical control module, responsible for executing adjustment operations related to the charge migration rate.

[0143] Step S142: Change the charge transport path inside the energy storage unit through the charge migration adjustment unit so that the charge migration rate reaches the target value in the independent electrochemical parameter adjustment sub-instruction for the charge migration rate parameter, and collect charge migration rate data in real time.

[0144] In this embodiment, the charge migration adjustment unit changes the charge transport path by controlling switching elements or variable impedance elements on the charge transport path within the energy storage unit. For example, by controlling the switching state of the auxiliary circuit connected in parallel at the electrode interface, the number of charge migration channels is increased or decreased, thereby changing the charge migration rate. During the adjustment process, the charge migration adjustment unit reads data collected in real time from the high-frequency response current sensor and the high-precision voltage sensor, monitors the real-time value of the charge migration rate by calculating the change in electrode interface impedance, compares it with the target value in the independent electrochemical parameter adjustment sub-instruction, and continuously adjusts the state of the switching elements or variable impedance elements through a feedback control algorithm until the charge migration rate reaches the target value.

[0145] For example, step S1421: Obtain the target value of the charge migration rate parameter in the independent electrochemical parameter adjustment sub-instruction for the charge migration rate parameter, and at the same time obtain the current charge migration rate data of the energy storage unit.

[0146] In this application scenario, the charge migration adjustment unit parses the target value of the charge migration rate parameter from the received independent electrochemical parameter adjustment sub-instructions, such as a specific target value for the interface impedance. Simultaneously, through a communication interface with the sensor, it reads the current charge migration rate data. This data is the result of sensor acquisition and preprocessing, and includes the current values ​​of parameters such as interface impedance and migration current.

[0147] Step S1422: Analyze the difference between the current charge migration rate data of the energy storage unit and the target value of the charge migration rate parameter, and determine the charge transport path segment that needs to be changed. The charge transport path segment includes the transport segment from the positive electrode to the electrolyte, the transport segment inside the electrolyte, and the transport segment from the electrolyte to the negative electrode.

[0148] In this embodiment, the difference analysis is achieved by calculating the difference between the interfacial impedance and the target value in the current charge migration rate data. According to electrochemical theory, impedance changes in the transport segment from the positive electrode to the electrolyte, the transport segment within the electrolyte, and the transport segment from the electrolyte to the negative electrode all affect the overall charge migration rate. By analyzing the impedance contribution ratio of each transport segment, it is determined which transport segment(s) is the main cause of the current difference, thereby identifying the charge transport path segment that needs to be modified. For example, if the impedance of the transport segment from the positive electrode to the electrolyte accounts for the largest proportion of the total impedance, and the difference between the current total impedance and the target value is mainly caused by this transport segment, then it is determined that the transport segment from the positive electrode to the electrolyte needs to be modified.

[0149] Step S1423: Determine the degree of adjustment for each charge transport path segment that needs to be changed based on the magnitude of the difference. When the difference meets the first preset difference standard, the degree of adjustment for the corresponding path segment is set according to the first rule. When the difference meets the second preset difference standard, the degree of adjustment for the corresponding path segment is set according to the second rule.

[0150] In this application scenario, the first preset difference standard is a difference value greater than or equal to a specific threshold, indicating a larger adjustment degree. The first rule is set to reduce the impedance of this path segment by a specific percentage. The second preset difference standard is a difference value less than but greater than or equal to the specific threshold, indicating a medium adjustment degree. The second rule is set to reduce the impedance of this path segment by a specific percentage. For differences less than the specific threshold, the adjustment degree is smaller, and the third rule is set to reduce the impedance of this path segment by a specific percentage. The determination of the adjustment degree also needs to consider the maximum adjustable range of the path segment to ensure that the adjusted impedance does not exceed the safe operating limits.

[0151] Step S1424: Change the conduction state of the corresponding charge transport path segment by changing the path control element in the charge transfer adjustment unit, and increase or decrease the cross-sectional area of ​​the charge transport channel of the path segment to adjust the charge transfer rate.

[0152] In this embodiment, the path control element can be a variable on-resistor or a switch array made of semiconductor material. For path segments where impedance needs to be reduced, the path control element is controlled to increase its on-section area, for example, by closing more switches in the switch array, or by decreasing the resistance value of the variable on-resistor, thereby reducing the impedance of the path segment and increasing the charge migration rate. For path segments where impedance needs to be increased (less common, typically when charge migration rate needs to be limited), the on-section area is decreased or the resistance value is increased. During adjustment, the state changes of the path control element are precisely controlled through pulse width modulation (PWM) signals or digital control signals.

[0153] Step S1425: During the adjustment process, the charge migration rate data of the energy storage unit is collected in real time by the rate acquisition element in the charge migration adjustment unit, and the rate change after each adjustment is recorded.

[0154] In this application scenario, the rate acquisition element is connected to the high-frequency response current sensor and high-precision voltage sensor mentioned in step S141, and receives data collected by the sensors in real time. After each adjustment of the state of the path control element, a specific time is waited for the system to reach a stable state, and then the charge migration rate data at this time is collected. Information such as the data change before and after the adjustment, the adjustment time, and the state of the path control element are recorded to form an adjustment log.

[0155] Step S1426: Compare the real-time collected charge migration rate data of the energy storage unit with the target value of the charge migration rate parameter in the independent electrochemical parameter adjustment sub-instruction for the charge migration rate parameter. If the real-time collected charge migration rate data of the energy storage unit reaches the target value of the charge migration rate parameter, then stop the operation of changing the charge transport path.

[0156] In this embodiment, the comparison process is performed using continuous sampling. At specific time intervals, the real-time collected charge migration rate data is compared with the target value, and the absolute error between the two is calculated. If the absolute error is less than or equal to a preset allowable error range, it is determined that the charge migration rate data has reached the target value. The charge migration adjustment unit issues a stop command to stop the adjustment operation on the charge transport path and maintain the current state of the path control element.

[0157] Step S1427: If the charge migration rate data of the energy storage unit collected in real time does not reach the target value of the charge migration rate parameter, then the charge transport path segment that needs to be adjusted and the degree of adjustment are re-determined based on the current difference, and the above operation of changing the conduction state of the corresponding charge transport path segment is repeated.

[0158] In this application scenario, if the absolute error exceeds the preset allowable error range, steps S1422 to S1424 are repeated to analyze the current difference, determine the new charge transport path segment requiring adjustment and the degree of adjustment, and change the conduction state of the path control element again. This process is repeated until the charge migration rate data reaches the target value or the maximum number of adjustments is reached (to prevent infinite looping). If the target value is not reached even after reaching the maximum number of adjustments, an alarm signal is issued to prompt maintenance personnel to conduct an inspection.

[0159] Step S1428: After stopping the adjustment operation, continuously collect the charge migration rate data of the energy storage unit through the rate acquisition element for a set time period, so that the charge migration rate is stabilized at the target value of the charge migration rate parameter, and the stable charge migration rate data within the set time period is used as the final collected charge migration rate data.

[0160] In this embodiment, the set time period is typically a specific number of minutes to ensure that the charge migration rate stabilizes near the target value after adjustment stops, avoiding the impact of instantaneous fluctuations. Within this time period, charge migration rate data is continuously collected, and the average and standard deviation of the data are calculated. If the standard deviation is less than the stabilization threshold, the charge migration rate is considered stable, and the average data within this time period is used as the final collected charge migration rate data. If the standard deviation is still large, the set time period is extended or fine-tuning is performed again. The final collected charge migration rate data is stored in a local database and uploaded to the data processing center.

[0161] Step S143: The electrode reaction adjustment unit in the electrochemical control module of the containerized electrochemical energy storage system receives an independent electrochemical parameter adjustment sub-instruction for the electrode reaction activity parameter. The electrode reaction adjustment unit adjusts the reaction environment of the energy storage unit electrode so that the electrode reaction activity reaches the target value in the independent electrochemical parameter adjustment sub-instruction for the electrode reaction activity parameter, and at the same time, the electrode reaction activity data is collected in real time.

[0162] In this embodiment, the electrode reaction adjustment unit alters the electrode reactivity by regulating environmental parameters such as electrode temperature, pressure, and electrolyte composition. For example, adjusting the electrode temperature via heating or cooling devices typically accelerates the reaction rate and enhances electrode reactivity; changing the contact pressure between the electrode and electrolyte via pressure regulation optimizes the reaction interface; and adjusting the electrolyte composition by adding specific additives improves ionic conductivity and electrode surface activity. Upon receiving independent electrochemical parameter adjustment sub-instructions, the electrode reaction adjustment unit analyzes target values, such as the target range of redox potential or the target value of reaction current density, and then controls the corresponding adjustment devices to operate. Simultaneously, a micro-three-electrode sensor integrated on the electrode surface collects electrode reactivity data in real time, such as redox potential and reaction current density, and compares it with the target values, achieving precise adjustment through feedback control.

[0163] Step S144: The electrolyte adjustment unit in the electrochemical control module of the containerized electrochemical energy storage system receives independent electrochemical parameter adjustment sub-instructions for electrolyte distribution parameters. The electrolyte adjustment unit changes the flow path and distribution mode of the electrolyte to make the electrolyte distribution reach the target state in the independent electrochemical parameter adjustment sub-instructions for electrolyte distribution parameters, while collecting electrolyte distribution data in real time.

[0164] In this embodiment, the electrolyte adjustment unit alters the electrolyte flow path and distribution by controlling the rotational speed of the electrolyte circulation pump, the valve opening, and the position of the guide plate. For example, increasing the circulation pump speed can improve the electrolyte flow rate and pressure, promoting uniform electrolyte distribution within the battery module; adjusting the valve opening in different areas can change the electrolyte distribution ratio, allowing more electrolyte to flow to areas with lower concentrations; changing the position of the guide plate can optimize the electrolyte flow pattern and reduce dead volume. After receiving independent electrochemical parameter adjustment sub-instructions, the electrolyte adjustment unit analyzes the target state, such as the target value for pressure distribution unevenness or the target value for ion concentration gradient, and then controls the actuator to make adjustments. Simultaneously, electrolyte distribution data is collected in real time through a distributed pressure sensor and concentration sensor array, compared with the target state, and uniform electrolyte distribution is achieved through closed-loop control.

[0165] Step S145: Integrate the real-time collected charge migration rate data, real-time collected electrode reaction activity data, and real-time collected electrolyte distribution data to form the adjusted electrochemical operating status data.

[0166] In this embodiment, the data collected in real time by the charge transfer adjustment unit, electrode reaction adjustment unit, and electrolyte adjustment unit are aligned according to timestamps to ensure data temporal consistency. Then, the data undergoes format conversion and standardization, converting the raw data collected by different sensors into a unified data format, such as JSON or XML, and supplementing metadata such as data identifiers, acquisition time, and sensor numbers. Next, the data undergoes quality checks, removing outliers and missing values, and filling in missing values ​​using interpolation methods. Finally, the processed data is categorized and organized according to electrochemical parameter types, combined into adjusted electrochemical operating state data, stored in a data server, and sent to the data processing center for subsequent feedback control command generation.

[0167] Step S146: The set of transmission parameter adjustment instructions is transmitted to the transmission control module of the grid-connected interface, and the power adjustment unit in the transmission control module of the grid-connected interface receives the independent transmission parameter adjustment sub-instructions for the power transmission parameters.

[0168] In this embodiment, the transmission parameter adjustment command set is transmitted through a communication link with the grid-connected interface transmission control module. This communication link can be fiber optic or wireless communication, employing a highly reliable communication protocol. Upon receiving the command set, the transmission control module parses and distributes the individual transmission parameter adjustment sub-commands for power transmission parameters to the power adjustment unit. The power adjustment unit is the core control unit in the grid-connected interface, typically implemented by a digital signal processor (DSP) or field-programmable gate array (FPGA), responsible for controlling the operating status of power conversion components such as the grid-connected inverter.

[0169] Step S147: Change the state of the power conversion element inside the grid-connected interface through the power adjustment unit so that the power transmission parameters reach the target value in the independent transmission parameter adjustment sub-instruction for the power transmission parameters, and collect power transmission data in real time.

[0170] In this embodiment, the power conversion elements mainly refer to switching devices such as IGBTs (Insulated Gate Bipolar Transistors) in the grid-connected inverter. The power adjustment unit adjusts the target active and reactive power values ​​in the sub-instruction based on the independent transmission parameters. It generates control signals using a space vector pulse width modulation (SVPWM) algorithm to control the IGBT's on and off timing, changing the inverter's output voltage and current, thereby adjusting the amount of active and reactive power transmitted. For example, to increase active power output, the power adjustment unit increases the modulation ratio, increasing the fundamental amplitude of the inverter's output voltage, thus increasing the output current and active power. During the adjustment process, the power adjustment unit collects the output active and reactive power data in real time through current and voltage sensors at the grid interface, compares it with the target values, and performs closed-loop regulation using a proportional-integral-derivative (PID) controller or model predictive control (MPC) algorithm to ensure that the power transmission parameters remain stable at the target values.

[0171] Step S148: The voltage adjustment unit in the transmission control module of the grid-connected interface receives the independent transmission parameter adjustment sub-instruction for the voltage adaptation parameter. The voltage adjustment unit adjusts the voltage conversion element parameters of the grid-connected interface so that the voltage adaptation parameter reaches the target value range in the independent transmission parameter adjustment sub-instruction for the voltage adaptation parameter, and at the same time, the voltage adaptation data is collected in real time.

[0172] In this embodiment, the voltage conversion elements include transformers, reactors, and capacitors. The voltage regulation unit adjusts the grid-connected voltage by controlling the parameters of these elements. For example, for an on-load tap-changing transformer, the voltage regulation unit sends a control signal to change the transformer's tap position, adjusts the turns ratio, and thus changes the output voltage amplitude. For parallel reactors or capacitors, the unit changes their connected capacity through switching control, adjusts the reactive power balance of the system, and thus regulates the voltage. After receiving the independent transmission parameter adjustment sub-instruction, the voltage regulation unit analyzes the target value range, such as the grid-connected voltage amplitude should be within a specific range. Then, based on the deviation between the current voltage measurement value and the target range, it controls the voltage conversion elements to make adjustments. Simultaneously, the voltage sensor collects the amplitude and phase data of the grid-connected voltage in real time, i.e., voltage adaptation data, for feedback control and status monitoring.

[0173] Step S149: The frequency adjustment unit in the transmission control module of the grid-connected interface receives the independent transmission parameter adjustment sub-instruction for the frequency synchronization parameter. The frequency adjustment unit adjusts the state of the frequency synchronization element of the grid-connected interface so that the frequency synchronization parameter reaches the target synchronization accuracy in the independent transmission parameter adjustment sub-instruction for the frequency synchronization parameter, and at the same time, the frequency synchronization data is collected in real time.

[0174] In this embodiment, the frequency synchronization components mainly refer to the phase-locked loop (PLL) circuit and the inertial response control module. The frequency adjustment unit adjusts the target synchronization accuracy in the sub-instructions based on independent transmission parameters, such as the allowable frequency deviation range and adjustment response time. It also adjusts the control parameters of the PLL, such as proportional gain and integral gain, to optimize the dynamic response characteristics of the PLL and improve frequency tracking accuracy and stability. Simultaneously, the inertial response control module simulates the inertial characteristics of a traditional synchronous generator. When the grid frequency fluctuates, it quickly adjusts the active power output of the energy storage system to help stabilize the grid frequency. The frequency adjustment unit collects grid-connected frequency data in real time through a frequency sensor, compares it with the target synchronization accuracy, and ensures that the frequency synchronization parameters meet the requirements through feedback control.

[0175] Step S1410: Integrate the real-time power transmission data, real-time voltage adaptation data, and real-time frequency synchronization data to form the adjusted transmission operation parameter data.

[0176] In this embodiment, similar to the process of integrating the adjusted electrochemical operating status data, the power transmission data, voltage adaptation data, and frequency synchronization data collected by the power adjustment unit, voltage adjustment unit, and frequency adjustment unit are aligned according to timestamps, and then undergo format conversion, standardization processing, and quality checks. Then, they are categorized and organized according to transmission parameter types, combined into adjusted transmission operating parameter data, stored, and uploaded to the data processing center.

[0177] Step S150: Generate a grid-connected feedback control command based on the adjusted electrochemical operating status data and the adjusted transmission operating parameter data, and send the grid-connected feedback control command to the grid-connected control unit of the containerized electrochemical energy storage system so that the grid-connected control unit maintains a stable grid-connected state according to the grid-connected feedback control command.

[0178] In this embodiment, after receiving the adjusted electrochemical operating status data and transmission operating parameter data, the data processing center performs comprehensive analysis on the above data, evaluates the current operating status and grid connection of the energy storage system, and generates grid connection feedback control commands to further optimize and stabilize the grid connection operation of the energy storage system.

[0179] For example, in step S151: perform data correlation analysis on the adjusted electrochemical operating status data and the adjusted transmission operating parameter data to determine the matching relationship between the adjusted electrochemical operating status data and the adjusted transmission operating parameter data. If the matching degree between the adjusted electrochemical operating status data and the adjusted transmission operating parameter data meets the requirements for stable grid connection, then generate basic feedback control commands based on the adjusted electrochemical operating status data and the adjusted transmission operating parameter data.

[0180] In this embodiment, the data association analysis employs correlation analysis and consistency testing methods. Correlation analysis calculates the correlation coefficient between key parameters in the electrochemical operating status data (such as charge migration rate and electrode reactivity) and key parameters in the transmission operating parameter data (such as active power and voltage). A high correlation coefficient indicates a good match between the two. Consistency testing checks whether the output capability after adjusting the electrochemical parameters is consistent with the actual transmission amount after adjusting the transmission parameters. For example, it checks whether the maximum output active power calculated after adjusting the electrochemical parameters is greater than or equal to the active power transmission amount in the transmission parameters. The matching degree is evaluated by combining the correlation coefficient and consistency test results. A matching degree threshold is set; if the evaluation result is greater than or equal to this threshold, it is considered to meet the stable grid connection requirements. At this point, using the current values ​​of the adjusted electrochemical operating status data and the transmission operating parameter data as a benchmark, a basic feedback control command is generated. The command content includes maintaining the stable operation of the current electrochemical parameters and transmission parameters.

[0181] Step S152: If the matching degree between the adjusted electrochemical operating status data and the adjusted transmission operating parameter data does not meet the requirements for stable grid connection, analyze the corresponding mismatch reasons, and correct the electrochemical parameter adjustment instructions corresponding to the adjusted electrochemical operating status data or the transmission parameter adjustment instructions corresponding to the adjusted transmission operating parameter data according to the mismatch reasons to obtain the corrected adjustment instructions.

[0182] In this embodiment, the mismatch may be caused by insufficient adjustment of electrochemical parameters, resulting in insufficient output power to meet transmission requirements; conflict between the adjustment direction of transmission parameters and the adjustment direction of electrochemical parameters; or parameter mismatch caused by external interference (such as sudden changes in grid voltage). The specific cause of the mismatch is determined by comparing and analyzing the changing trends and deviation values ​​of the electrochemical operating status data and the transmission operating parameter data. For example, if the electrochemical operating status data shows that the charge migration rate has reached the target value, but the active power in the transmission operating parameter data has not yet reached the target, it may be due to losses in the transmission path or reduced power conversion efficiency. In this case, the transmission parameter adjustment command needs to be revised to increase the adjustment range of the power transmission amount. If the transmission parameters are forced to adjust due to grid voltage fluctuations, and the electrochemical parameters do not respond in time, the electrochemical parameter adjustment command needs to be revised to adjust the electrode reactivity or electrolyte distribution to adapt to the new transmission requirements. During the revision process, historical amendment examples and expert knowledge bases are referenced to ensure that the revised adjustment command can effectively solve the mismatch problem.

[0183] Step S153: Based on the revised adjustment instructions, readjust the electrochemical operating state or transmission operating parameters to obtain the revised electrochemical operating state data or the revised transmission operating parameter data.

[0184] In this embodiment, the corrected electrochemical parameter adjustment command or transmission parameter adjustment command is sent to the corresponding control module according to the process of step S140, and the adjustment operation is re-executed. For example, if the adjustment command for the charge migration rate parameter is corrected, the charge migration adjustment unit adjusts the charge transmission path according to the new command, re-collects the electrochemical operating status data, and obtains the corrected electrochemical operating status data. If the adjustment command for the power transmission parameter is corrected, the power adjustment unit re-controls the power conversion element, adjusts the power transmission amount, and obtains the corrected transmission operating parameter data.

[0185] Step S154: Perform correlation analysis on the corrected electrochemical operating status data and the corrected transmission operating parameter data until the matching degree between the corrected electrochemical operating status data and the corrected transmission operating parameter data meets the requirements for stable grid connection. Generate basic feedback control commands based on the corrected electrochemical operating status data and the corrected transmission operating parameter data that meet the requirements.

[0186] In this application scenario, the correlation analysis process in step S151 is repeated to check whether the matching degree between the corrected electrochemical operating status data and the transmitted operating parameter data meets the requirements for stable grid connection. If it still does not meet the requirements, steps S152 to S154 are continued until the matching degree meets the requirements. This iterative process ensures that the energy storage system and the grid connection interface can maintain a stable matching relationship in a dynamically changing environment. Once the matching degree meets the requirements, basic feedback control commands are generated based on the corrected parameter data.

[0187] Step S155: Perform instruction enhancement processing on the basic feedback control instruction and add a dynamic adjustment factor. The dynamic adjustment factor is used to enable the basic feedback control instruction to adaptively adjust according to changes in subsequent electrochemical operating state or transmission operating parameters.

[0188] In this embodiment, the dynamic adjustment factor is determined based on historical data and a predictive model. For example, a calculation model for the dynamic adjustment factor is established based on the rate of change of electrochemical operating parameters, the fluctuation amplitude of transmission operating parameters, and the changing trend of grid dynamic reception data over a past period. The dynamic adjustment factor can be a function related to the rate of change, fluctuation amplitude, and trend strength. When a change in electrochemical operating status or transmission operating parameters is detected, the target value or adjustment amplitude of the parameters in the basic feedback control command is automatically adjusted according to the direction and degree of the change. For example, if the grid frequency shows an upward trend, the dynamic adjustment factor will improve the target synchronization accuracy of the frequency synchronization parameter in the basic feedback control command to respond to frequency changes more quickly. The dynamic adjustment factor is added to the basic feedback control command as an additional field to give the command adaptive capabilities.

[0189] Step S156: The basic feedback control command after adding the dynamic adjustment factor is used as the grid-connected feedback control command. The grid-connected feedback control command includes electrochemical state maintenance parameters, transmission state maintenance parameters, and dynamic adjustment rules.

[0190] In this embodiment, the electrochemical state maintenance parameters include the target value range and allowable fluctuation range of parameters such as charge migration rate, electrode reactivity, and electrolyte distribution; the transmission state maintenance parameters include the target value range and allowable deviation of parameters such as power transmission, voltage amplitude, and frequency; the dynamic adjustment rules specify the conditions under which the dynamic adjustment factor is triggered, the calculation method of the adjustment factor, and the limitations on the adjustment amplitude. The grid-connected feedback control command adopts a structured data format, such as JSON, to facilitate parsing and execution by the grid-connected control unit. For example, the electrochemical state maintenance parameter section of the command may contain entries such as "charge migration rate: target range, allowable fluctuation range," the transmission state maintenance parameter section may contain entries such as "active power: target value, allowable deviation," and the dynamic adjustment rules section may contain content such as "when the grid frequency change rate exceeds a specific value, the dynamic adjustment factor is activated; the adjustment factor is calculated using a specific formula."

[0191] Step S157: The grid connection feedback control command is transmitted to the grid connection control unit of the containerized electrochemical energy storage system. After receiving the grid connection feedback control command, the grid connection control unit of the containerized electrochemical energy storage system maintains the internal electrochemical stability of the energy storage system according to the electrochemical state maintenance parameters in the grid connection feedback control command, maintains the transmission stability of the grid connection interface according to the transmission state maintenance parameters in the grid connection feedback control command, and responds to parameter changes according to the dynamic adjustment rules in the grid connection feedback control command.

[0192] In this embodiment, the grid-connected control unit is the central control unit of the energy storage system, and it has communication connections with both the electrochemical control module and the transmission control module. Upon receiving the grid-connected feedback control command, the grid-connected control unit parses the command and extracts the electrochemical state maintenance parameters, transmission state maintenance parameters, and dynamic adjustment rules. Then, it sends the electrochemical state maintenance parameters to the electrochemical control module to guide it in adjusting its control strategy to maintain the electrochemical stability within the energy storage system. For example, when the charge migration rate approaches the upper limit of the allowable fluctuation range, it performs fine-tuning in advance to prevent exceeding the range. The transmission state maintenance parameters are sent to the transmission control module to stabilize the transmission parameters of the grid-connected interface within the target range. Simultaneously, the grid-connected control unit runs its own dynamic adjustment rules, monitoring the electrochemical operating status data and transmission operating parameter data in real time. When a parameter change is detected that meets the triggering conditions specified in the dynamic adjustment rules, it automatically calculates the dynamic adjustment factor and corrects the electrochemical state maintenance parameters or transmission state maintenance parameters according to the adjustment factor, then sends the correction to the corresponding control module for execution.

[0193] Step S158: Continuously receive grid connection status data fed back by the grid connection control unit of the containerized electrochemical energy storage system, and guide the grid connection control unit to adjust the operating parameters based on the continuously received grid connection status data.

[0194] In this embodiment, the grid-connected control unit periodically (e.g., every specific second) uploads grid-connected status data to the data processing center. This data includes current electrochemical operating status data, transmitted operating parameter data, control command execution status, alarm information, etc. The data processing center monitors and analyzes the received grid-connected status data in real time. If it detects that certain parameters deviate from their stable range or exhibit abnormalities, it promptly generates new grid-connected feedback control commands and sends them to the grid-connected control unit to guide it in adjusting operating parameters. For example, if it detects increased grid frequency fluctuations, the data processing center analyzes the data and generates new frequency synchronization parameter adjustment commands, sending them to the grid-connected control unit. The grid-connected control unit then adjusts the control strategy of the frequency adjustment unit accordingly, improving frequency synchronization accuracy and response speed. Through this continuous closed-loop feedback mechanism, the energy storage system is ensured to maintain a stable grid-connected operating state.

Claims

1. A grid-connected control method for a containerized electrochemical energy storage system, characterized in that, The method comprises: obtaining electrochemical characteristic data of a container-type electrochemical energy storage system and dynamic accommodation data of a power grid, wherein the electrochemical characteristic data of the container-type electrochemical energy storage system comprises charge migration rate data of an energy storage unit, electrode reaction activity data of the energy storage unit and electrolyte distribution data inside the energy storage system, and the dynamic accommodation data of the power grid comprises real-time power surplus / deficit data of the power grid, voltage fluctuation tolerance data of the power grid and frequency regulation response data of the power grid; constructing a deep learning collaborative mapping model based on the electrochemical characteristic data of the container-type electrochemical energy storage system and the dynamic accommodation data of the power grid, establishing a dynamic correlation between electrochemical characteristic changes and power grid accommodation capacity through the deep learning collaborative mapping model, and generating a grid-connected collaborative control strategy; decomposing the grid-connected collaborative control strategy into an electrochemical parameter adjustment instruction for the container-type electrochemical energy storage system and a transmission parameter adjustment instruction for a grid-connected interface, to obtain two sets of associated adjustment instruction sets; adjusting the internal electrochemical operating state of the container-type electrochemical energy storage system according to the electrochemical parameter adjustment instruction to obtain adjusted electrochemical operating state data, and adjusting the transmission operating parameter of the grid-connected interface according to the transmission parameter adjustment instruction to obtain adjusted transmission operating parameter data; generating a grid-connected feedback control instruction based on the adjusted electrochemical operating state data and the adjusted transmission operating parameter data, and sending the grid-connected feedback control instruction to a grid-connected control unit of the container-type electrochemical energy storage system, so that the grid-connected control unit maintains a stable grid-connected state according to the grid-connected feedback control instruction.

2. The grid-tie control method of a container-type electrochemical energy storage system according to claim 1, characterized by, The deep learning collaborative mapping model is constructed based on the electrochemical characteristic data of the container-type electrochemical energy storage system and the dynamic accommodation data of the power grid, the dynamic correlation between electrochemical characteristic changes and power grid accommodation capacity is established through the deep learning collaborative mapping model, and the grid-connected collaborative control strategy is generated, which comprises: determining a model function level of the deep learning collaborative mapping model, wherein the model function level comprises a multi-source data coupling perception layer, a time sequence correlation modeling layer and a collaborative strategy generation layer; inputting the electrochemical characteristic data of the container-type electrochemical energy storage system and the dynamic accommodation data of the power grid into the multi-source data coupling perception layer, performing cross-domain feature correlation processing on the electrochemical characteristic data of the container-type electrochemical energy storage system and the dynamic accommodation data of the power grid, extracting sensitive features related to power grid accommodation in the electrochemical characteristic data and adaptive features related to electrochemical response in the dynamic accommodation data; performing dimension unification processing on the extracted sensitive features and adaptive features, and assigning corresponding weights according to the influence degree of the sensitive features and the adaptive features on grid-connected control to obtain a weighted coupled feature set; transmitting the weighted coupled feature set to the time sequence correlation modeling layer, performing time sequence dynamic modeling on the weighted coupled feature set, capturing the correlation change law between the electrochemical characteristic data and the dynamic accommodation data at different time nodes, and generating a time sequence correlation model; Based on the time sequence correlation model, influence trend of the change of the electrochemical characteristics on the power grid accommodation capacity is analyzed, and constraint requirement of the change of the power grid accommodation capacity on the electrochemical characteristics is analyzed, and a bidirectional influence relationship matrix is constructed; The bidirectional influence relationship matrix is transmitted to a collaborative strategy generation layer, and strategy mapping processing is performed on the bidirectional influence relationship matrix by a multilayer perceptron unit in the collaborative strategy generation layer, and a plurality of groups of initial grid-connected control strategies are generated; Collaborative analysis is performed on the plurality of groups of initial grid-connected control strategies, and an initial grid-connected control strategy meeting the electrochemical characteristic stability requirement and adapting to the power grid accommodation capacity is selected, the initial grid-connected control strategy is taken as a grid-connected collaborative control strategy, and the grid-connected collaborative control strategy is stored in association with the corresponding bidirectional influence relationship matrix.

3. The grid-tie control method of a container-type electrochemical energy storage system according to claim 2, characterized by, The electrochemical characteristic data of the container-type electrochemical energy storage system and the dynamic accommodation data of the power grid are processed in cross-domain feature correlation, sensitive features related to the power grid accommodation in the electrochemical characteristic data and adaptive features related to the electrochemical response in the dynamic accommodation data are extracted, including: A cross-domain feature correlation graph is constructed in a feature interaction unit, feature items of the electrochemical characteristic data are taken as first-type nodes of the cross-domain feature correlation graph, and feature items of the dynamic accommodation data are taken as second-type nodes of the cross-domain feature correlation graph; The correlation strength of each feature item in the first-type nodes and each feature item in the second-type nodes in the cross-domain feature correlation graph is calculated, an association edge is added between the nodes in the cross-domain feature correlation graph according to the correlation strength, and the representation weight of the association edge is set according to the corresponding rule when the correlation strength meets a preset correlation strength standard; Based on the distribution of the association edges in the cross-domain feature correlation graph, a feature item pair with a representation weight of the association edge exceeding a preset correlation threshold is identified, the feature item pair includes a feature item from the first-type nodes of the cross-domain feature correlation graph and a feature item from the second-type nodes of the cross-domain feature correlation graph; Feature attribute analysis is performed on the identified feature item pair, and the influence type of the feature item pair in the first-type nodes of the cross-domain feature correlation graph on the power grid accommodation capacity is determined, the influence type includes a type of promoting the power grid accommodation and a type of limiting the power grid accommodation; The first-type nodes of the cross-domain feature correlation graph belonging to the type of promoting the power grid accommodation and meeting the requirement of the correlation strength are classified as sensitive features related to the power grid accommodation, and the influence degree corresponding to each sensitive feature related to the power grid accommodation is recorded; Reverse analysis is performed on the feature item pair, and the constraint type of the feature item pair in the second-type nodes of the cross-domain feature correlation graph on the electrochemical characteristic response is determined, the constraint type includes a type of supporting the electrochemical stability and a type of interfering with the electrochemical stability; The second-type nodes of the cross-domain feature correlation graph belonging to the type of supporting the electrochemical stability and meeting the requirement of the correlation strength are classified as adaptive features related to the electrochemical response, and the constraint degree corresponding to each adaptive feature related to the electrochemical response is recorded. Collecting all cross-domain feature association graph first type node feature items belonging to sensitive features related to grid acceptance and cross-domain feature association graph second type node feature items belonging to adaptive features related to electrochemical response, forming a sensitive feature list and an adaptive feature list as the output result of the cross-domain feature association processing.

4. The grid-tie control method of a container-type electrochemical energy storage system according to claim 2, characterized by, The weighted coupling feature set is transmitted to a time sequence association modeling layer, and the weighted coupling feature set is dynamically modeled in time sequence to capture the association change law between electrochemical characteristic data and dynamic acceptance data at different time nodes, and a time sequence association model is generated, including: The weighted coupling feature set is divided into multiple time sequence segments in chronological order, and each time sequence segment contains coupling feature data at the corresponding time node; The first time sequence segment is input into the input gate of the long short-term memory unit of the time sequence association modeling layer, the coupling feature data in the time sequence segment is screened for importance through the input gate of the long short-term memory unit, and feature data valuable for time sequence association modeling is retained, and redundant feature data is filtered out; The feature association information in the historical time sequence segment is screened through the forget gate of the long short-term memory unit, and the current time sequence segment feature data and the retained historical association information are fused through the cell state of the long short-term memory unit to generate cell state update data at the current time node; The cell state update data is output screened through the output gate of the long short-term memory unit, and feature information representing the association relationship between the electrochemical characteristic data and the dynamic acceptance data at the current time node is extracted as the current time sequence output; All time sequence segments are processed in sequence to obtain the time sequence output corresponding to each time node, and all time sequence outputs are arranged in chronological order to form a time sequence association sequence; Based on the time sequence association sequence, the change amplitude of the association relationship between adjacent time nodes is analyzed, and the change trend of the association relationship between interval time nodes is analyzed, and an association change law model is constructed; The association change law model and the network parameters of the long short-term memory unit are integrated to form a time sequence association model capable of capturing time sequence association changes.

5. The grid-tie control method of a containerized electrochemical energy storage system of claim 1, wherein, The grid-connected collaborative control strategy is decomposed into an electrochemical parameter adjustment instruction for the container-type electrochemical energy storage system and a transmission parameter adjustment instruction for the grid-connected interface, and two sets of associated adjustment instruction sets are obtained, including: The grid-connected collaborative control strategy is analyzed to extract electrochemical control elements and transmission control elements contained in the grid-connected collaborative control strategy, the electrochemical control elements are used to adjust the electrochemical state inside the energy storage system, and the transmission control elements are used to adjust the transmission state of the grid-connected interface; Based on the electrochemical control elements, the types of electrochemical parameters that need to be adjusted are determined, the types of electrochemical parameters include charge transfer rate parameters, electrode reaction activity parameters, and electrolyte distribution parameters, and the adjustment direction and adjustment amplitude of the charge transfer rate parameters, the adjustment direction and adjustment amplitude of the electrode reaction activity parameters, and the adjustment direction and adjustment amplitude of the electrolyte distribution parameters are determined. According to the determined charge migration rate parameter type, the adjustment direction and the adjustment amplitude corresponding to the charge migration rate parameter, an independent electrochemical parameter adjustment sub-instruction for the charge migration rate parameter is generated, according to the determined electrode reaction activity parameter type, the adjustment direction and the adjustment amplitude corresponding to the electrode reaction activity parameter, an independent electrochemical parameter adjustment sub-instruction for the electrode reaction activity parameter is generated, according to the determined electrolyte distribution parameter type, the adjustment direction and the adjustment amplitude corresponding to the electrolyte distribution parameter, an independent electrochemical parameter adjustment sub-instruction for the electrolyte distribution parameter is generated, and all the independent electrochemical parameter adjustment sub-instructions constitute an electrochemical parameter adjustment instruction set for the container type electrochemical energy storage system; Based on the transmission control element, the transmission parameter type that needs to be adjusted is determined, the transmission parameter type includes the power transmission parameter, the voltage adaptation parameter and the frequency synchronization parameter, and the adjustment direction and the adjustment amplitude corresponding to the power transmission parameter, the adjustment direction and the adjustment amplitude corresponding to the voltage adaptation parameter and the adjustment direction and the adjustment amplitude corresponding to the frequency synchronization parameter are determined; According to the determined power transmission parameter type, the adjustment direction and the adjustment amplitude corresponding to the power transmission parameter, an independent transmission parameter adjustment sub-instruction for the power transmission parameter is generated, according to the determined voltage adaptation parameter type, the adjustment direction and the adjustment amplitude corresponding to the voltage adaptation parameter, an independent transmission parameter adjustment sub-instruction for the voltage adaptation parameter is generated, according to the determined frequency synchronization parameter type, the adjustment direction and the adjustment amplitude corresponding to the frequency synchronization parameter, an independent transmission parameter adjustment sub-instruction for the frequency synchronization parameter is generated, and all the independent transmission parameter adjustment sub-instructions constitute a transmission parameter adjustment instruction set for the grid-connected interface; An association relationship between the electrochemical parameter adjustment instruction set for the container type electrochemical energy storage system and the transmission parameter adjustment instruction set for the grid-connected interface is established, so that the independent electrochemical parameter adjustment sub-instruction for the charge migration rate parameter and the corresponding independent transmission parameter adjustment sub-instruction form an associated pair, the independent electrochemical parameter adjustment sub-instruction for the electrode reaction activity parameter and the corresponding independent transmission parameter adjustment sub-instruction form an associated pair, and the independent electrochemical parameter adjustment sub-instruction for the electrolyte distribution parameter and the corresponding independent transmission parameter adjustment sub-instruction form an associated pair, so that the adjustment directions of the two types of instructions in each associated pair are consistent; The independent electrochemical parameter adjustment sub-instruction and the independent transmission parameter adjustment sub-instruction in each associated pair are verified for cooperativity, so that the adjustment amplitude of the independent electrochemical parameter adjustment sub-instruction and the adjustment amplitude of the independent transmission parameter adjustment sub-instruction in each associated pair are matched; The electrochemical parameter adjustment instruction set for the container type electrochemical energy storage system and the transmission parameter adjustment instruction set for the grid-connected interface that pass through the cooperativity verification are used as the final two sets of associated adjustment instruction sets.

6. The grid-tie control method of a containerized electrochemical energy storage system according to claim 5, wherein, The method comprises the following steps of: determining the type of the electrochemical parameter that needs to be adjusted based on the electrochemical control element, wherein the electrochemical parameter type comprises a charge migration rate parameter, an electrode reaction activity parameter and an electrolyte distribution parameter, and the adjustment direction and adjustment amplitude of the charge migration rate parameter, the adjustment direction and adjustment amplitude of the electrode reaction activity parameter and the adjustment direction and adjustment amplitude of the electrolyte distribution parameter are determined; performing element attribute classification on the electrochemical control element, and dividing the element attribute of the electrochemical control element into a charge migration correlation attribute, an electrode reaction correlation attribute and an electrolyte distribution correlation attribute; determining whether the charge migration rate parameter needs to be adjusted according to the attribute value variation trend of the charge migration correlation attribute, and determining that the charge migration rate parameter needs to be adjusted if the attribute value variation trend of the charge migration correlation attribute indicates that the charge migration rate cannot meet the power grid receiving requirement; calculating the amplitude of the charge migration rate parameter that needs to be adjusted based on the mapping relationship between the charge migration correlation attribute and the charge migration rate parameter, and determining the adjustment direction of the charge migration rate parameter according to the variation direction of the power grid receiving requirement; determining whether the electrode reaction activity parameter needs to be adjusted according to the attribute value variation trend of the electrode reaction correlation attribute, and determining that the electrode reaction activity parameter needs to be adjusted if the attribute value variation trend of the electrode reaction correlation attribute indicates that the electrode reaction activity cannot maintain the stable operation of the energy storage system; calculating the amplitude of the electrode reaction activity parameter that needs to be adjusted based on the mapping relationship between the electrode reaction correlation attribute and the electrode reaction activity parameter, and determining the adjustment direction of the electrode reaction activity parameter according to the stable operation requirement of the energy storage system; determining whether the electrolyte distribution parameter needs to be adjusted according to the attribute value variation trend of the electrolyte distribution correlation attribute, and determining that the electrolyte distribution parameter needs to be adjusted if the attribute value variation trend of the electrolyte distribution correlation attribute indicates that the electrolyte distribution is uneven and affects the electrochemical characteristics; calculating the amplitude of the electrolyte distribution parameter that needs to be adjusted based on the mapping relationship between the electrolyte distribution correlation attribute and the electrolyte distribution parameter, and determining the adjustment direction of the electrolyte distribution parameter according to the requirement of the uniform electrolyte distribution; and summarizing the type information of the charge migration rate parameter, the electrode reaction activity parameter and the electrolyte distribution parameter that need to be adjusted, recording the adjustment direction and adjustment amplitude of the charge migration rate parameter, the adjustment direction and adjustment amplitude of the electrode reaction activity parameter and the adjustment direction and adjustment amplitude of the electrolyte distribution parameter, and forming an electrochemical parameter adjustment list. ​ ​ ​ ​ ​ ​ ​ ​ 7. The grid-tie control method of a containerized electrochemical energy storage system according to claim 5, wherein, The independent transmission parameter adjustment sub-instruction for the power transmission parameter is generated according to the determined power transmission parameter type, the adjustment direction and the adjustment amplitude corresponding to the power transmission parameter, the independent transmission parameter adjustment sub-instruction for the voltage adaptation parameter is generated according to the determined voltage adaptation parameter type, the adjustment direction and the adjustment amplitude corresponding to the voltage adaptation parameter, and the independent transmission parameter adjustment sub-instruction for the frequency synchronization parameter is generated according to the determined frequency synchronization parameter type, the adjustment direction and the adjustment amplitude corresponding to the frequency synchronization parameter, and all the independent transmission parameter adjustment sub-instructions constitute a transmission parameter adjustment instruction set for the grid-connected interface, which comprises: For the power transmission parameter, the target value of the power transmission parameter is determined according to the adjustment direction and the adjustment amplitude corresponding to the power transmission parameter, and the target value is the result obtained by the current power transmission parameter value and the adjustment amplitude according to the adjustment direction; An instruction structure of the power transmission parameter adjustment sub-instruction is generated, the instruction structure comprises parameter type identification, target value information and adjustment execution time information, and the parameter type identification is used to explicitly indicate that the instruction is for the power transmission parameter; The parameter type identification, the target value information and the adjustment execution time information are combined according to a preset format to form the independent transmission parameter adjustment sub-instruction for the power transmission parameter; For the voltage adaptation parameter, the target value range of the voltage adaptation parameter is determined according to the adjustment direction and the adjustment amplitude corresponding to the voltage adaptation parameter according to the same process as that of the power transmission parameter; An instruction structure of the voltage adaptation parameter adjustment sub-instruction is generated, the instruction structure comprises parameter type identification, target value range information and adjustment execution time information, and the parameter type identification is used to explicitly indicate that the instruction is for the voltage adaptation parameter; The parameter type identification, the target value range information and the adjustment execution time information are combined according to a preset format to form the independent transmission parameter adjustment sub-instruction for the voltage adaptation parameter; For the frequency synchronization parameter, the target synchronization precision of the frequency synchronization parameter is determined according to the adjustment direction and the adjustment amplitude corresponding to the frequency synchronization parameter; An instruction structure of the frequency synchronization parameter adjustment sub-instruction is generated, the instruction structure comprises parameter type identification, target synchronization precision information and adjustment execution time information, and the parameter type identification is used to explicitly indicate that the instruction is for the frequency synchronization parameter; The parameter type identification, the target synchronization precision information and the adjustment execution time information are combined according to a preset format to form the independent transmission parameter adjustment sub-instruction for the frequency synchronization parameter; The independent transmission parameter adjustment sub-instruction for the power transmission parameter, the independent transmission parameter adjustment sub-instruction for the voltage adaptation parameter and the independent transmission parameter adjustment sub-instruction for the frequency synchronization parameter are collected, arranged according to the priority order of parameter adjustment, and a transmission parameter adjustment instruction set for the grid-connected interface is formed.

8. The grid-tie control method of a containerized electrochemical energy storage system according to claim 7, wherein, The internal electrochemical operation state of the container-type electrochemical energy storage system is adjusted according to the electrochemical parameter adjustment instruction to obtain adjusted electrochemical operation state data, and meanwhile, the transmission operation parameter of the grid-connected interface is adjusted according to the transmission parameter adjustment instruction to obtain adjusted transmission operation parameter data, which comprises: transmit the set of electrochemical parameter adjustment instructions to the electrochemical control module of the container-type electrochemical energy storage system, so that a charge migration adjustment unit in the electrochemical control module of the container-type electrochemical energy storage system receives an independent electrochemical parameter adjustment sub-instruction for the charge migration rate parameter; change the charge transmission path inside the energy storage unit through the charge migration adjustment unit, so that the charge migration rate reaches the target value in the independent electrochemical parameter adjustment sub-instruction for the charge migration rate parameter, while real-time collection of charge migration rate data is performed; an electrode reaction adjustment unit in the electrochemical control module of the container-type electrochemical energy storage system receives an independent electrochemical parameter adjustment sub-instruction for the electrode reaction activity parameter, adjusts the reaction environment of the electrode of the energy storage unit through the electrode reaction adjustment unit, so that the electrode reaction activity reaches the target value in the independent electrochemical parameter adjustment sub-instruction for the electrode reaction activity parameter, while real-time collection of electrode reaction activity data is performed; an electrolyte adjustment unit in the electrochemical control module of the container-type electrochemical energy storage system receives an independent electrochemical parameter adjustment sub-instruction for the electrolyte distribution parameter, changes the flow path and distribution mode of the electrolyte through the electrolyte adjustment unit, so that the electrolyte distribution reaches the target state in the independent electrochemical parameter adjustment sub-instruction for the electrolyte distribution parameter, while real-time collection of electrolyte distribution data is performed; integrate the real-time collected charge migration rate data, the real-time collected electrode reaction activity data, and the real-time collected electrolyte distribution data to form adjusted electrochemical operation state data; transmit the set of transmission parameter adjustment instructions to the transmission control module of the grid-connected interface, and a power adjustment unit in the transmission control module of the grid-connected interface receives an independent transmission parameter adjustment sub-instruction for the power transmission parameter; change the state of the power conversion element inside the grid-connected interface through the power adjustment unit, so that the power transmission parameter reaches the target value in the independent transmission parameter adjustment sub-instruction for the power transmission parameter, while real-time collection of power transmission data is performed; a voltage adjustment unit in the transmission control module of the grid-connected interface receives an independent transmission parameter adjustment sub-instruction for the voltage adaptation parameter, adjusts the voltage conversion element parameter of the grid-connected interface through the voltage adjustment unit, so that the voltage adaptation parameter reaches the target value range in the independent transmission parameter adjustment sub-instruction for the voltage adaptation parameter, while real-time collection of voltage adaptation data is performed; a frequency adjustment unit in the transmission control module of the grid-connected interface receives an independent transmission parameter adjustment sub-instruction for the frequency synchronization parameter, adjusts the frequency synchronization element state of the grid-connected interface through the frequency adjustment unit, so that the frequency synchronization parameter reaches the target synchronization accuracy in the independent transmission parameter adjustment sub-instruction for the frequency synchronization parameter, while real-time collection of frequency synchronization data is performed; integrate the real-time collected power transmission data, the real-time collected voltage adaptation data, and the real-time collected frequency synchronization data to form adjusted transmission operation parameter data.

9. A grid-tie control system for a containerized electrochemical energy storage system, characterized by, comprise: a processor; a machine-readable storage medium for storing machine-executable instructions of the processor; The processor is configured to execute the machine-executable instructions to perform the grid-connected control method of the container-type electrochemical energy storage system according to any one of claims 1 to 8.

10. A computer program product, characterised in that, The computer program product comprises machine-executable instructions stored in a computer-readable storage medium, and a processor of the grid-connected control system of the container-type electrochemical energy storage system reads the machine-executable instructions from the computer-readable storage medium, and the processor executes the machine-executable instructions, so that the grid-connected control system of the container-type electrochemical energy storage system performs the grid-connected control method of the container-type electrochemical energy storage system according to any one of claims 1 to 8.

Images