Power adaptive emergency power supply control method for battery swap cabinet based on power grid state

Abstract

The application discloses a power self-adaptive emergency power supply control method for a battery replacement cabinet based on a power grid state, and relates to the technical field of power supply control methods. The method comprises the following steps: real-time monitoring of a target battery replacement cabinet, acquisition of operation data of the target battery replacement cabinet, construction of an abnormality detection model, and identification of abnormal operation information. A power grid health evaluation system is established, a power grid health index is calculated, and an emergency event is determined, and emergency event analysis information is generated. Meanwhile, residual battery state characteristics are extracted from battery replacement cabinet monitoring data, an emergency event strategy library is constructed, and emergency power supply management and control are implemented. For a battery replacement cabinet cluster in a target area, each battery replacement cabinet is defined as a network node, abnormality influence analysis is carried out when an abnormality is monitored, a power self-adaptive compensation strategy is formulated in combination with cluster operation and maintenance states, and power compensation regulation and control are carried out on the cluster. The safety and intelligence of the battery replacement cabinet in an emergency event are improved, and the operation stability of the battery replacement cabinet is ensured.

Description

Technical Field

[0001] This invention relates to the field of emergency control technology for battery swapping cabinets, and in particular to an adaptive emergency power supply control method for battery swapping cabinets based on grid conditions. Background Technology

[0002] With the increasing popularity of battery swapping for two-wheeled electric vehicles, battery swapping cabinets, as a crucial infrastructure, are becoming more and more numerous. However, in existing battery swapping cabinets, some are equipped with lead-acid batteries to power the main control system after a power outage. This poses a safety hazard: the cabinet door may accidentally open when the main control system is interrupted due to a power outage. Furthermore, a power failure in the main control system will cause the real-time monitoring function of the backend management platform to fail. The battery swapping cabinet relies on a battery management system (BMS) for anomaly monitoring and alarm push notifications. If the main control system loses power, the linkage response of these safety protection functions will be interrupted, and fault alarms cannot be sent to maintenance personnel in a timely manner. Simultaneously, under insufficient grid load, effective regulation cannot be performed to ensure the safety of the swapping cabinet. Therefore, how to solve these problems to improve the safety and intelligence of battery swapping cabinets in emergency situations is an urgent issue to be addressed. Summary of the Invention

[0003] This invention overcomes the shortcomings of the prior art and provides a power adaptive emergency power supply control method for battery swapping cabinets based on grid conditions.

[0004] To achieve the above objectives, the first aspect of this invention provides a power adaptive emergency power supply control method for a battery swapping cabinet based on grid conditions, comprising:

[0005] The operation monitoring of the target battery swapping cabinet is carried out to obtain the operation monitoring information of the battery swapping cabinet, and an anomaly detection model is constructed to detect anomalies based on the operation monitoring information of the battery swapping cabinet to obtain anomaly detection information;

[0006] Establish a power grid health evaluation system, calculate the power grid health index by combining the operation monitoring information and abnormal operation detection information of the aforementioned battery swapping cabinet, determine emergency events based on the power grid health index, and obtain emergency event analysis information;

[0007] An emergency event strategy database is constructed. Based on the battery swapping cabinet operation monitoring information, the remaining battery status characteristics of the target battery swapping cabinet are extracted, and emergency power supply management is carried out in combination with the emergency event analysis information.

[0008] Obtain information on the battery swapping cabinet cluster within the target area, define each battery swapping cabinet as a single node to construct a battery swapping cabinet cluster network, and perform an impact analysis on the battery swapping cabinet anomaly when an anomaly is detected to obtain anomaly impact analysis information.

[0009] Obtain the operation and maintenance status information of the target battery swapping cabinet cluster, and formulate a power adaptive compensation strategy based on the abnormal impact analysis information to perform power compensation and regulation on the target battery swapping cabinet cluster.

[0010] In this solution, the step of monitoring the operation of the target battery swapping cabinet to obtain its operation monitoring information, and constructing an anomaly detection model to detect anomalies based on the battery swapping cabinet's operation monitoring information, specifically includes:

[0011] A sensor array is set in the target battery swapping cabinet. The sensor array is used to monitor the cabinet's operating information and the battery status information inside the cabinet. The operating monitoring information of the battery swapping cabinet is obtained through the sensor array.

[0012] The standard deviation of the target battery swapping cabinet operation monitoring data is calculated based on the battery swapping cabinet operation monitoring information. Data that is less than the preset standard deviation threshold is marked as outliers and removed. Data is then interpolated and supplemented using linear interpolation.

[0013] An anomaly detection model is constructed based on generative adversarial networks. The pre-processed battery swapping cabinet operation monitoring information is processed to obtain a time series sequence, which is then input into the anomaly detection model for analysis.

[0014] The input time series sequence is used as the original sequence. The generator performs sequence reconstruction based on the original sequence to obtain the reconstructed sequence. The reconstructed sequence is then output to the discriminator to determine whether to accept the current reconstructed sequence.

[0015] If the discriminator determines that the reconstruction sequence is accepted, the corresponding reconstruction sequence is output; if the discriminator determines that the reconstruction sequence is not accepted, the generator continues to generate the next reconstruction sequence for judgment and outputs the final reconstruction sequence.

[0016] A preset reconstruction error threshold is set, and the reconstruction error between the final reconstructed sequence and the original sequence is calculated. If the error is greater than the reconstruction error threshold, the final reconstructed sequence is compared with the original sequence to extract the difference features and obtain abnormal operation detection information.

[0017] In this plan, the establishment of a power grid health evaluation system specifically includes:

[0018] Based on data retrieval, different historical emergency events are obtained. Feature extraction is performed on each historical emergency event to obtain its features. The K-means clustering algorithm is introduced to classify the historical emergency events according to their features, thus obtaining a historical emergency event dataset.

[0019] Principal component analysis is used to reduce the dimensionality of each category set, and the principal component scores of each feature in the historical emergency event dataset are calculated. The calculated principal component scores are then compared with a preset threshold.

[0020] Define the historical emergency event features corresponding to the principal component scores that are greater than a preset threshold as principal component factors, and use the principal component factors to perform principal component direction projection to obtain a projection scatter plot;

[0021] A preset selection range is used to select corresponding historical emergency event features in the projected scatter plot, generating a first dataset. Based on the first dataset, power grid health evaluation indicators are selected to obtain power grid health evaluation indicators.

[0022] Based on the selected power grid health evaluation indicators, the information entropy of each power grid health evaluation indicator is calculated using the entropy method. The difference coefficient of each power grid health evaluation indicator is obtained by calculating the information entropy.

[0023] The selected power grid health evaluation indicators are weighted using the calculated difference coefficients, and a power grid health evaluation system is constructed by combining the power grid health evaluation indicators and the weight allocation results.

[0024] In this solution, the calculation of the power grid health index is performed by combining the operation monitoring information and abnormal operation detection information of the battery swapping cabinet. Based on the power grid health index, emergency events are determined to obtain emergency event analysis information, specifically including:

[0025] Obtain the operation monitoring information and abnormal operation detection information of the battery swapping cabinet, and extract the operation features for power grid health evaluation indicators from the operation monitoring information and abnormal operation detection information of the battery swapping cabinet according to the established power grid health evaluation system to obtain the first feature information;

[0026] A power grid health analysis model is built based on the constructed power grid health evaluation system. The first feature information is input into the power grid health analysis model to calculate the power grid health index and obtain the power grid health index information.

[0027] Obtain a historical emergency event dataset, and analyze the historical power grid health index of each historical emergency event in the dataset using the power grid health analysis model to obtain historical power grid health index information;

[0028] Based on the historical emergency event dataset and the historical power grid health index information, an emergency event knowledge graph is constructed with historical emergency event types, historical power grid health indices, and historical emergency event characteristics as meta-paths.

[0029] The battery swapping cabinet operation monitoring information, abnormal operation detection information, and power grid health index information are used as input features and imported into the emergency event knowledge graph to perform emergency event judgment and analysis on the abnormal status of the target battery swapping cabinet, thereby obtaining emergency event analysis information.

[0030] In this solution, the construction of an emergency event strategy database, which involves extracting the remaining battery status characteristics of the target battery swapping cabinet based on the battery swapping cabinet operation monitoring information and combining this with the emergency event analysis information for emergency power supply management, specifically includes:

[0031] Based on the big data network, historical power supply management and control instances of various battery swapping cabinet emergency events are obtained to form an instance dataset. The management and control features of each historical power supply management and control instance are extracted from the instance dataset, including management and control strategy features and management and control event features.

[0032] The control strategies and control features are associated, and the membership function is preset to calculate the membership value between each control strategy and control feature, so as to obtain the historical power supply control instance feature dataset.

[0033] The Apriority algorithm is introduced to perform frequent itemset mining on the historical power supply management instance feature dataset. By searching layer by layer, several frequent item subsets are obtained to generate association rules for several management strategies.

[0034] The confidence level of the association rules for each control strategy is calculated. The association rules that meet the preset confidence threshold are filtered by the calculated confidence level and defined as the triggering rules of the corresponding control strategy to build an emergency event strategy library.

[0035] Obtain battery swapping cabinet operation monitoring information and emergency event analysis information, and extract the status characteristics of the remaining storage batteries in the target battery swapping cabinet through the battery swapping cabinet operation monitoring information to obtain the remaining battery status characteristic information;

[0036] The emergency event analysis information is matched and analyzed with the emergency event strategy database to determine the emergency event control strategy that matches the current emergency event of the target battery swapping cabinet, and the matching control strategy information is obtained.

[0037] Based on the remaining battery status information, the remaining battery status in the target battery swapping cabinet is sorted, and the battery with the best status is selected as the power supply battery. An emergency event management strategy is generated in combination with the matching strategy information to control the emergency power supply of the target battery swapping cabinet.

[0038] In this solution, the process of acquiring battery swapping cabinet cluster information within the target area involves defining each battery swapping cabinet as a single node to construct a battery swapping cabinet cluster network. When an anomaly is detected in a battery swapping cabinet, an impact analysis of the anomaly is performed to obtain anomaly impact analysis information, specifically including:

[0039] Obtain information on the battery swapping cabinet cluster within the target area, including the location, type, scale, and power grid connection information of each battery swapping cabinet within the target area;

[0040] Each battery swapping cabinet is defined as a graph node. Based on the battery swapping cabinet cluster information in the target area, the graph nodes of the battery swapping cabinets are assigned auxiliary attributes. The connection features between each battery swapping cabinet and the power grid are extracted through the battery swapping cabinet cluster information in the target area.

[0041] Based on the connection characteristics between each battery swapping cabinet and the power grid, the relationship between each battery swapping cabinet is defined. If two battery swapping cabinets are connected to the same power grid branch, a directed edge is established for the corresponding graph node and the edge weight is set. The edge weight is set according to the electrical distance.

[0042] A battery swapping cabinet cluster network for the target area is constructed using graph nodes, directed edges, and edge weights. The battery swapping cabinet cluster network is then converted into an adjacency matrix for storage. An anomaly impact analysis model is constructed based on the SIR model, and the model is trained and its parameters are adjusted.

[0043] Obtain emergency event analysis information, update the node status of the top-changing cabinet cluster network based on the emergency event analysis information, mark and update the abnormal node status and abnormal events, and generate a new adjacency matrix;

[0044] The new adjacency matrix is ​​input into the anomaly impact analysis model. By iteratively traversing and calculating the infection probability of neighboring nodes, the anomaly impact of the current emergency event on the battery swapping cabinet cluster is analyzed, and anomaly impact analysis information is obtained.

[0045] In this solution, the step of obtaining the operation and maintenance status information of the target battery swapping cabinet cluster, and combining it with the abnormal impact analysis information to formulate a power adaptive compensation strategy, and to perform power compensation regulation on the target battery swapping cabinet cluster, specifically includes:

[0046] Obtain the operation and maintenance status information and anomaly impact analysis information of the target battery swapping cabinet cluster. The operation and maintenance status information of the target battery swapping cabinet cluster includes the power status, battery status and communication status of each battery swapping cabinet in the target area during operation.

[0047] The objective functions are set based on minimizing power deviation and SOC loss. Constraints are set according to grid capacity. An improved NSGA-II algorithm is introduced to combine the target battery swapping cabinet cluster operation and maintenance status information and abnormal impact analysis information to conduct adaptive compensation strategy analysis and generate an initial population.

[0048] The objective function value is calculated for each individual in the initial population. All individuals in the population are sorted in non-dominated order according to the objective function value and divided into several Pareto fronts. The crowding distance is calculated for individuals in each non-dominated layer.

[0049] The tournament selection method is used to select elite individuals by using non-dominated ranking hierarchy and crowding distance. The selected elite individuals are then subjected to crossover and mutation operations, and the parent and offspring generations are merged. The non-dominated ranking and crowding distance are recalculated for all candidate schemes.

[0050] Based on the Pareto frontier and crowding distance, a preset number of elite individuals are selected to form the next generation population for iterative evolution. When the objective function of the population meets the convergence criterion or the preset number of iterations, an adaptive power compensation scheme is output to perform power compensation regulation on the target top-changing cabinet cluster.

[0051] A second aspect of the present invention provides a power adaptive emergency power supply control system for a battery swapping cabinet based on grid conditions. The system includes a memory and a processor. The memory contains a program for a power adaptive emergency power supply control method for the battery swapping cabinet based on grid conditions. When executed by the processor, the power adaptive emergency power supply control method program for the battery swapping cabinet based on grid conditions implements the following steps:

[0052] Establish a power grid health evaluation system, calculate the power grid health index by combining the operation monitoring information and abnormal operation detection information of the aforementioned battery swapping cabinet, determine emergency events based on the power grid health index, and obtain emergency event analysis information;

[0053] An emergency event strategy database is constructed. Based on the battery swapping cabinet operation monitoring information, the remaining battery status characteristics of the target battery swapping cabinet are extracted, and emergency power supply management is carried out in combination with the emergency event analysis information.

[0054] Obtain information on the battery swapping cabinet cluster within the target area, define each battery swapping cabinet as a single node to construct a battery swapping cabinet cluster network, and perform an impact analysis on the battery swapping cabinet anomaly when an anomaly is detected to obtain anomaly impact analysis information.

[0055] Obtain the operation and maintenance status information of the target battery swapping cabinet cluster, and formulate a power adaptive compensation strategy based on the abnormal impact analysis information to perform power compensation and regulation on the target battery swapping cabinet cluster.

[0056] This invention discloses a power adaptive emergency power supply control method for battery swapping cabinets based on grid status. The method includes: real-time monitoring of the target battery swapping cabinet, acquiring its operational data, constructing an anomaly detection model, and identifying abnormal operational information. A grid health evaluation system is established, and a grid health index is calculated to determine emergency events and generate emergency event analysis information. Simultaneously, remaining battery status characteristics are extracted from the battery swapping cabinet monitoring data to construct an emergency event strategy library and implement emergency power supply management. For the battery swapping cabinet cluster within the target area, each battery swapping cabinet is defined as a network node. When anomalies are detected, an anomaly impact analysis is conducted, and a power adaptive compensation strategy is formulated based on the cluster's operation and maintenance status to perform power compensation regulation on the cluster. This improves the safety and intelligence of the battery swapping cabinets during emergency events and ensures the operational stability of the battery swapping cabinets. Attached Figure Description

[0057] To more clearly illustrate the technical solutions in the embodiments or examples of the present invention, the drawings used in the embodiments or examples will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained according to these drawings without creative effort.

[0058] Figure 1 A flowchart of an adaptive emergency power supply control method for a battery swapping cabinet based on grid conditions is provided in an embodiment of the present invention.

[0059] Figure 2 This is a flowchart of an embodiment of the present invention providing a power adaptive emergency power compensation method for a battery swapping cabinet based on power grid conditions;

[0060] Figure 3 A block diagram of an adaptive emergency power supply control system for a battery swapping cabinet based on grid conditions is provided in one embodiment of the present invention.

[0061] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0062] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0063] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0064] Figure 1 An embodiment of the present invention provides an adaptive emergency power supply control for a battery swapping cabinet based on grid conditions;

[0065] like Figure 1 As shown, this invention provides an adaptive emergency power supply control for a battery swapping cabinet based on grid conditions, including:

[0066] S102, monitor the operation of the target battery swapping cabinet to obtain battery swapping cabinet operation monitoring information, construct an anomaly detection model to perform anomaly detection based on the battery swapping cabinet operation monitoring information, and obtain anomaly detection information.

[0067] S104. Establish a power grid health evaluation system, calculate the power grid health index by combining the operation monitoring information and abnormal operation detection information of the battery swapping cabinet, determine emergency events based on the power grid health index, and obtain emergency event analysis information.

[0068] S106, Construct an emergency event strategy database, extract the remaining battery status characteristics of the target battery swapping cabinet based on the battery swapping cabinet operation monitoring information, and combine the emergency event analysis information to carry out emergency power supply management and control.

[0069] S106: Obtain information on the battery swapping cabinet cluster within the target area, define each battery swapping cabinet as a single node to build a battery swapping cabinet cluster network, and perform an abnormality impact analysis when an abnormality is detected in the battery swapping cabinet to obtain abnormality impact analysis information.

[0070] S110: Obtain the operation and maintenance status information of the target battery swapping cabinet cluster, and formulate a power adaptive compensation strategy based on the abnormal impact analysis information to perform power compensation and regulation on the target battery swapping cabinet cluster.

[0071] Furthermore, in a preferred embodiment of the present invention, the step of monitoring the operation of the target battery swapping cabinet to obtain battery swapping cabinet operation monitoring information, and constructing an anomaly detection model to perform anomaly detection based on the battery swapping cabinet operation monitoring information to obtain anomaly detection information, specifically includes:

[0072] A sensor array is set in the target battery swapping cabinet. The sensor array is used to monitor the cabinet's operating information and the battery status information inside the cabinet. The operating monitoring information of the battery swapping cabinet is obtained through the sensor array.

[0073] The standard deviation of the target battery swapping cabinet operation monitoring data is calculated based on the battery swapping cabinet operation monitoring information. Data that is less than the preset standard deviation threshold is marked as outliers and removed. Data is then interpolated and supplemented using linear interpolation.

[0074] An anomaly detection model is constructed based on generative adversarial networks. The pre-processed battery swapping cabinet operation monitoring information is processed to obtain a time series sequence, which is then input into the anomaly detection model for analysis.

[0075] The input time series sequence is used as the original sequence. The generator performs sequence reconstruction based on the original sequence to obtain the reconstructed sequence. The reconstructed sequence is then output to the discriminator to determine whether to accept the current reconstructed sequence.

[0076] If the discriminator determines that the reconstruction sequence is accepted, the corresponding reconstruction sequence is output; if the discriminator determines that the reconstruction sequence is not accepted, the generator continues to generate the next reconstruction sequence for judgment and outputs the final reconstruction sequence.

[0077] A preset reconstruction error threshold is set, and the reconstruction error between the final reconstructed sequence and the original sequence is calculated. If the error is greater than the reconstruction error threshold, the final reconstructed sequence is compared with the original sequence to extract the difference features and obtain abnormal operation detection information.

[0078] It should be noted that a sensor array is deployed in the target battery swapping cabinet to monitor the cabinet's operating status and internal battery condition in real time, and to acquire the cabinet's operational monitoring information through this array. Based on the collected data, its standard deviation is calculated, and data with a standard deviation less than a preset threshold are marked as outliers and removed to avoid the impact of abnormal data on the analysis results. Subsequently, linear interpolation is used to supplement the missing or removed data to ensure data integrity and continuity. An anomaly detection model is constructed based on a Generative Adversarial Network (GAN) to perform time-series processing on the battery swapping cabinet's operational monitoring information, obtain time-series data, and input it into the anomaly detection model for analysis. This model consists of a generator and a discriminator, and is trained in a semi-supervised manner, enabling the model to reconstruct the input abnormal sequence into a normal sequence, thereby completing the anomaly detection task. Specifically, the generator reconstructs the sequence based on the original input sequence, generates a new data sequence, and passes it to the discriminator for evaluation. The discriminator evaluates the reconstructed sequence. If it deems the sequence conforms to a normal data distribution, it accepts and outputs the corresponding reconstructed sequence. If the discriminator determines that the reconstructed sequence cannot pass verification, the generator continues to optimize and generate new reconstructed sequences until they pass the discriminator's judgment, thus finally outputting an optimized reconstructed sequence. The reconstruction error between the final reconstructed sequence and the original sequence is calculated. If the reconstruction error exceeds a set threshold, it indicates that the input data may contain abnormal patterns. In this case, the final reconstructed sequence and the original sequence are compared, and their difference features are extracted to obtain abnormal operation detection information.

[0079] Furthermore, in a preferred embodiment of the present invention, the establishment of the power grid health evaluation system specifically includes:

[0080] Based on data retrieval, different historical emergency events are obtained. Feature extraction is performed on each historical emergency event to obtain its features. The K-means clustering algorithm is introduced to classify the historical emergency events according to their features, thus obtaining a historical emergency event dataset.

[0081] Principal component analysis is used to reduce the dimensionality of each category set, and the principal component scores of each feature in the historical emergency event dataset are calculated. The calculated principal component scores are then compared with a preset threshold.

[0082] Define the historical emergency event features corresponding to the principal component scores that are greater than a preset threshold as principal component factors, and use the principal component factors to perform principal component direction projection to obtain a projection scatter plot;

[0083] A preset selection range is used to select corresponding historical emergency event features in the projected scatter plot, generating a first dataset. Based on the first dataset, power grid health evaluation indicators are selected to obtain power grid health evaluation indicators.

[0084] Based on the selected power grid health evaluation indicators, the information entropy of each power grid health evaluation indicator is calculated using the entropy method. The difference coefficient of each power grid health evaluation indicator is obtained by calculating the information entropy.

[0085] The selected power grid health evaluation indicators are weighted using the calculated difference coefficients, and a power grid health evaluation system is constructed by combining the power grid health evaluation indicators and the weight allocation results.

[0086] It should be noted that different historical emergency events are obtained through data retrieval, and features are extracted from these historical data to form a historical emergency event feature set. The K-means clustering algorithm is introduced to classify events according to their features, ultimately constructing a historical emergency event dataset. For the segmented dataset, Principal Component Analysis (PCA) is used for dimensionality reduction, calculating the principal component scores of each feature in the dataset. These scores are compared with preset thresholds to select the most representative principal component factors. Historical emergency event features corresponding to principal component scores above the preset thresholds are selected as key principal component factors. Principal component direction projection is performed based on these factors to generate a projected scatter plot, visually displaying the distribution of event features in the data space. From the projected scatter plot, historical emergency event features meeting the requirements are selected according to the selection range to form the first dataset. Based on this dataset, power grid health evaluation indicators are selected. The selection of power grid health evaluation indicators mainly focuses on key factors such as power grid operational stability, load balancing, power supply reliability, and the status of battery swapping cabinets. For example, grid operation stability can be measured by indicators such as voltage fluctuation amplitude and frequency deviation; load balance can be referenced by parameters such as load volatility and regional power distribution balance; power supply reliability can be assessed by data such as power outage duration, grid failure rate, and recovery time; and the status of battery swapping cabinets can be evaluated by indicators such as battery remaining capacity, charging and discharging efficiency, and temperature monitoring data. After selecting the evaluation indicators, the information entropy of each grid health evaluation indicator is calculated using the entropy method to quantify the effective information content of each indicator. Using the calculated difference coefficients, weights are allocated to each grid health evaluation indicator to ensure a reasonable weight distribution that effectively reflects the contribution of each indicator to the grid health status. Finally, combining the grid health evaluation indicators and their weight allocation results, a complete grid health evaluation system is constructed, providing accurate references for grid status monitoring and emergency power supply strategy formulation.

[0087] Furthermore, in a preferred embodiment of the present invention, the step of calculating the power grid health index by combining the operation monitoring information and abnormal operation detection information of the battery swapping cabinet, and determining emergency events based on the power grid health index to obtain emergency event analysis information specifically includes:

[0088] Obtain the operation monitoring information and abnormal operation detection information of the battery swapping cabinet, and extract the operation features for power grid health evaluation indicators from the operation monitoring information and abnormal operation detection information of the battery swapping cabinet according to the established power grid health evaluation system to obtain the first feature information;

[0089] A power grid health analysis model is built based on the constructed power grid health evaluation system. The first feature information is input into the power grid health analysis model to calculate the power grid health index and obtain the power grid health index information.

[0090] Obtain a historical emergency event dataset, and analyze the historical power grid health index of each historical emergency event in the dataset using the power grid health analysis model to obtain historical power grid health index information;

[0091] Based on the historical emergency event dataset and the historical power grid health index information, an emergency event knowledge graph is constructed with historical emergency event types, historical power grid health indices, and historical emergency event characteristics as meta-paths.

[0092] The battery swapping cabinet operation monitoring information, abnormal operation detection information, and power grid health index information are used as input features and imported into the emergency event knowledge graph to perform emergency event judgment and analysis on the abnormal status of the target battery swapping cabinet, thereby obtaining emergency event analysis information.

[0093] It should be noted that, firstly, by collecting operational monitoring information and abnormal operation detection information of the battery swapping cabinets, and combining this with the established power grid health evaluation system, key operational characteristics related to the power grid health status are extracted from this data to form the first feature information. This feature information covers various aspects of the battery swapping cabinets, including battery status, power output, operational stability, charging and discharging modes, and abnormal operation conditions, providing comprehensive input data for subsequent health assessments. Based on this, a power grid health analysis model is built, and the first feature information is input into the model to calculate the power grid health index. This index is used to quantify the health status of the power grid and provide an assessment basis for the overall operation of the power grid. Simultaneously, a historical emergency event dataset is acquired, and the power grid health analysis model is used to perform backtracking analysis on this historical data, calculating the power grid health index at the time of each historical emergency event, forming historical power grid health index information. Based on the historical emergency event dataset and historical power grid health index information, an emergency event knowledge graph is constructed. In this graph, historical emergency event types, historical power grid health indices, and historical emergency event characteristics are used as meta-paths to establish the relationships between different emergency events. Knowledge graphs can integrate and represent historical data in a structured way, clarifying the relationships between different types of emergency events, their corresponding health index changes, and key characteristics, thus providing a more intuitive reference for emergency analysis. Finally, the current operation monitoring information, abnormal operation detection information, and calculated power grid health index information of the battery swapping cabinet are used as input features and imported into the emergency event knowledge graph for abnormal status analysis of the target battery swapping cabinet. Through the associative reasoning capabilities of the knowledge graph, the degree of matching between the current power grid state and historical emergency events is determined, thereby identifying potential emergency event types, generating emergency event analysis information, and ensuring that abnormal situations in the battery swapping cabinet can be identified in a timely manner and corresponding emergency power supply control measures can be taken.

[0094] Furthermore, in a preferred embodiment of the present invention, the step of constructing an emergency event strategy library, extracting the remaining battery status characteristics of the target battery swapping cabinet based on the battery swapping cabinet operation monitoring information, and combining the emergency event analysis information for emergency power supply management, specifically includes:

[0095] Based on the big data network, historical power supply management and control instances of various battery swapping cabinet emergency events are obtained to form an instance dataset. The management and control features of each historical power supply management and control instance are extracted from the instance dataset, including management and control strategy features and management and control event features.

[0096] The control strategies and control features are associated, and the membership function is preset to calculate the membership value between each control strategy and control feature, so as to obtain the historical power supply control instance feature dataset.

[0097] The Apriority algorithm is introduced to perform frequent itemset mining on the historical power supply management instance feature dataset. By searching layer by layer, several frequent item subsets are obtained to generate association rules for several management strategies.

[0098] The confidence level of the association rules for each control strategy is calculated. The association rules that meet the preset confidence threshold are filtered by the calculated confidence level and defined as the triggering rules of the corresponding control strategy to build an emergency event strategy library.

[0099] Obtain battery swapping cabinet operation monitoring information and emergency event analysis information, and extract the status characteristics of the remaining storage batteries in the target battery swapping cabinet through the battery swapping cabinet operation monitoring information to obtain the remaining battery status characteristic information;

[0100] The emergency event analysis information is matched and analyzed with the emergency event strategy database to determine the emergency event control strategy that matches the current emergency event of the target battery swapping cabinet, and the matching control strategy information is obtained.

[0101] Based on the remaining battery status information, the remaining battery status in the target battery swapping cabinet is sorted, and the battery with the best status is selected as the power supply battery. An emergency event management strategy is generated in combination with the matching strategy information to control the emergency power supply of the target battery swapping cabinet.

[0102] It should be noted that, firstly, historical power supply management instances from various battery swapping cabinet emergency events are collected through a big data network to form an instance dataset. This dataset covers specific management measures for various historical emergency events, including management strategy characteristics and management event characteristics. Management strategy characteristics describe the control methods used under different circumstances, such as power allocation methods and battery scheduling strategies, while management event characteristics reflect the operational status of different battery swapping cabinets under emergency conditions, such as battery loss rate and changes in grid health status. A correlation is established between management strategies and management features, and a membership function is introduced to calculate the membership value between them. The membership value measures the degree to which a certain management strategy is applicable to a certain management event characteristic, thus forming a historical power supply management instance feature dataset. The Apriori algorithm is introduced to perform frequent itemset mining on the historical power supply management instance feature dataset. The Apriori algorithm extracts frequently occurring subsets of items with high co-occurrence from the dataset through a layer-by-layer search, and further generates association rules for several management strategies. These rules reflect the applicability of different management strategies in historical emergency events and their correlation with specific operating states. In practical applications, operational monitoring information and emergency event analysis information of the battery swapping cabinets are acquired. The remaining battery status characteristics of the target battery swapping cabinet are extracted from the monitoring data to form remaining battery status characteristic information. Subsequently, the emergency event analysis information is matched with an emergency event strategy database to determine whether the current emergency event at the battery swapping cabinet conforms to existing control strategies in the database, thus selecting the control strategy that best matches the current event. Finally, based on the remaining battery status information of the battery swapping cabinet, available batteries are sorted by status, and the battery in the best condition is selected as the power supply battery. Combined with the matched control strategy information, a specific emergency event control strategy is generated to ensure that the battery swapping cabinet can efficiently control power supply in emergency situations. This improves the accuracy and reliability of emergency response.

[0103] Furthermore, in a preferred embodiment of the present invention, the step of acquiring battery swapping cabinet cluster information within the target area, defining each battery swapping cabinet as a single node to construct a battery swapping cabinet cluster network, and performing an anomaly impact analysis when an anomaly is detected in a battery swapping cabinet to obtain anomaly impact analysis information specifically includes:

[0104] Obtain information on the battery swapping cabinet cluster within the target area, including the location, type, scale, and power grid connection information of each battery swapping cabinet within the target area;

[0105] Each battery swapping cabinet is defined as a graph node. Based on the battery swapping cabinet cluster information in the target area, the graph nodes of the battery swapping cabinets are assigned auxiliary attributes. The connection features between each battery swapping cabinet and the power grid are extracted through the battery swapping cabinet cluster information in the target area.

[0106] Based on the connection characteristics between each battery swapping cabinet and the power grid, the relationship between each battery swapping cabinet is defined. If two battery swapping cabinets are connected to the same power grid branch, a directed edge is established for the corresponding graph node and the edge weight is set. The edge weight is set according to the electrical distance.

[0107] A battery swapping cabinet cluster network for the target area is constructed using graph nodes, directed edges, and edge weights. The battery swapping cabinet cluster network is then converted into an adjacency matrix for storage. An anomaly impact analysis model is constructed based on the SIR model, and the model is trained and its parameters are adjusted.

[0108] Obtain emergency event analysis information, update the node status of the top-changing cabinet cluster network based on the emergency event analysis information, mark and update the abnormal node status and abnormal events, and generate a new adjacency matrix;

[0109] The new adjacency matrix is ​​input into the anomaly impact analysis model. By iteratively traversing and calculating the infection probability of neighboring nodes, the anomaly impact of the current emergency event on the battery swapping cabinet cluster is analyzed, and anomaly impact analysis information is obtained.

[0110] It should be noted that, firstly, information on the battery swapping cabinet cluster within the target area is acquired. This information includes the location, type, scale, and connection status of each battery swapping cabinet within the target area to the power grid. Each battery swapping cabinet is defined as a node in the graph, and based on the collected cluster information, associated attributes are assigned to these nodes, such as capacity, battery status, and power supply capability. Simultaneously, the connection characteristics between the battery swapping cabinets and the power grid are analyzed to extract their topological relationship. If two battery swapping cabinets are connected to the same power grid branch, they are considered to be associated, and directed edges are established in the graph structure to represent this relationship. Furthermore, to reflect the degree of mutual influence between different battery swapping cabinets, edge weights are set based on electrical distance. Electrical distance can be calculated based on the impedance, voltage level, and power exchange capability between the battery swapping cabinet and the power grid node, ensuring that the edge weights accurately represent the physical and functional connections between the battery swapping cabinets. After completing the graph structure construction, a battery swapping cabinet cluster network within the target area is established using graph nodes, directed edges, and edge weights, and this network is converted into an adjacency matrix for storage. Subsequently, an anomaly impact analysis model was constructed based on the SIR model, and the model was trained and its parameters were adjusted. The SIR model is a classic propagation model widely used to describe the propagation process of information, faults, or diseases in networks. In this method, the SIR model is used to analyze how anomalies in battery swapping cabinets propagate throughout the entire battery swapping cabinet cluster network to predict the potential scope and severity of the impact of the anomaly. After obtaining emergency event analysis information, the node states in the battery swapping cabinet cluster network are updated based on this information. If an anomaly is detected in a battery swapping cabinet, it is marked as an anomalous node in the network, and its state is updated in the adjacency matrix. Finally, the updated adjacency matrix is ​​input into the anomaly impact analysis model, and the infection probability of each neighboring node is calculated through iterative traversal to simulate the propagation process of the current emergency event's impact on the battery swapping cabinet cluster. During this process, the SIR model calculates the infection probability of nodes in different states and predicts which battery swapping cabinets the anomaly may propagate to and the extent of its impact. Ultimately, the anomaly impact analysis information is obtained through this calculation process, providing a basis for subsequent fault response and emergency power supply control.

[0111] It should be noted that when analyzing the impact of emergency events using the anomaly impact analysis model, it is necessary to set an impact propagation equation, as shown below:

[0112]

[0113] Among them, the The probability that a node is affected is given by β, where β is the propagation rate, determined by the electrical distance and negatively correlated with it. ij S represents the edge weight between nodes i and j. j For the state of node j, I iLet i be the state of node i.

[0114] Furthermore, in a preferred embodiment of the present invention, the step of obtaining the operation and maintenance status information of the target battery swapping cabinet cluster, and formulating a power adaptive compensation strategy in conjunction with the anomaly impact analysis information to perform power compensation regulation on the target battery swapping cabinet cluster specifically includes:

[0115] Obtain the operation and maintenance status information and anomaly impact analysis information of the target battery swapping cabinet cluster. The operation and maintenance status information of the target battery swapping cabinet cluster includes the power status, battery status and communication status of each battery swapping cabinet in the target area during operation.

[0116] The objective functions are set based on minimizing power deviation and SOC loss. Constraints are set according to grid capacity. An improved NSGA-II algorithm is introduced to combine the target battery swapping cabinet cluster operation and maintenance status information and abnormal impact analysis information to conduct adaptive compensation strategy analysis and generate an initial population.

[0117] The objective function value is calculated for each individual in the initial population. All individuals in the population are sorted in non-dominated order according to the objective function value and divided into several Pareto fronts. The crowding distance is calculated for individuals in each non-dominated layer.

[0118] The tournament selection method is used to select elite individuals by using non-dominated ranking hierarchy and crowding distance. The selected elite individuals are then subjected to crossover and mutation operations, and the parent and offspring generations are merged. The non-dominated ranking and crowding distance are recalculated for all candidate schemes.

[0119] Based on the Pareto frontier and crowding distance, a preset number of elite individuals are selected to form the next generation population for iterative evolution. When the objective function of the population meets the convergence criterion or the preset number of iterations, an adaptive power compensation scheme is output to perform power compensation regulation on the target top-changing cabinet cluster.

[0120] It should be noted that an adaptive power compensation strategy is formulated based on the operation and maintenance status information and anomaly impact analysis results of the target battery swapping cabinet cluster, using an improved NSGA-II algorithm. First, the power status, battery status (such as SOC, SOH, and temperature) and communication status of each battery swapping cabinet are integrated. Combined with the affected nodes identified by the anomaly propagation model and their recovery times, a multi-objective optimization problem is constructed: minimizing power deviation and SOC loss rate are the core objectives, with single-node power limits, SOC protection thresholds, and grid capacity constraints set. The improved NSGA-II algorithm generates an initial population through dynamic parameter adjustment, where each individual represents a power adjustment scheme, and its chromosome encodes the power adjustment amount for each battery swapping cabinet. Subsequently, the population is non-dominated, Pareto fronts are defined, and crowding distance calculations ensure solution diversity. A tournament selection method is used, combining non-dominated levels and crowding distance to select elite individuals. Simulated binary crossover and polynomial mutation operations are performed to generate offspring, and the parent and offspring generations are merged and reordered iteratively. The algorithm terminates when it meets the convergence condition (e.g., Pareto front rate of change < 1%) or reaches the preset number of iterations, outputting the Pareto optimal solution set to obtain the adaptive power compensation scheme. Finally, by dynamically issuing power adjustment commands, it minimizes battery SOC loss while meeting the power grid's supply requirements, improves the overall energy efficiency of the battery swapping cabinet, optimizes the adaptability of the compensation strategy, and enhances the stability and power supply reliability of the entire battery swapping cabinet network.

[0121] Figure 2 This is a flowchart of an embodiment of the present invention providing a power adaptive emergency power compensation method for a battery swapping cabinet based on power grid conditions;

[0122] like Figure 2 As shown, the present invention provides a flowchart of a power adaptive emergency power compensation method for battery swapping cabinets based on power grid conditions, including:

[0123] S202, monitor the operation of the target battery swapping cabinet to obtain battery swapping cabinet operation monitoring information, detect abnormal operating conditions and calculate the power grid health index, and make emergency event judgment based on the power grid health index and the detected abnormal conditions to obtain emergency event analysis information;

[0124] S204, Update the node status of the top-changing cabinet cluster network according to the emergency event analysis information, mark and update the abnormal node status and abnormal events, and generate a new adjacency matrix.

[0125] S206. Input the new adjacency matrix into the abnormal impact analysis model, and analyze the abnormal impact of the current emergency event on the battery swapping cabinet cluster by iteratively traversing and calculating the infection probability of neighboring nodes to obtain abnormal impact analysis information.

[0126] S208: Obtain the operation and maintenance status information of the target battery swapping cabinet cluster, and formulate a power adaptive compensation strategy based on the abnormal impact analysis information to perform power compensation and regulation on the target battery swapping cabinet cluster.

[0127] It should be noted that abnormal battery swapping stations may experience reduced power supply capacity due to battery failure, communication interruption, or abnormal power output, thus affecting the overall power grid balance. Power compensation control can dynamically adjust the output power of other battery swapping stations, preventing fluctuations in the local power grid due to power shortages, while optimizing energy dispatch and improving power supply efficiency. Furthermore, reasonable compensation control can balance discharge among different battery swapping stations, reducing SOC imbalance losses, extending battery life, and improving the long-term operating efficiency of the system. Combining anomaly impact analysis information and optimized control algorithms, compensation control can enhance the adaptive adjustment capability of the battery swapping station cluster, enabling the system to operate stably under abnormal conditions, improving robustness and the ability to respond to emergencies. Simultaneously, this measure can also ensure the continuity of battery swapping services for electric vehicle users, avoiding service interruptions due to battery swapping station malfunctions, reducing user waiting time, and improving system availability. To address the issue of traditional battery swapping cabinets lacking the ability to participate in grid demand response and having insufficient functionality, under the condition of large-scale deployment, intelligent and safe battery swapping cabinets can receive remote control commands from virtual power plant aggregation platforms. Through converters, they can invert the DC power from the battery storage batteries in the battery swapping cabinets into AC power and feed it into the grid to supplement the grid load when it is insufficient. This reduces safety hazards such as accidental opening of the cabinet doors and communication interruptions, and improves the stability, power supply reliability and operation and maintenance safety of the entire battery swapping cabinet network.

[0128] Figure 3 An embodiment of the present invention provides a power adaptive emergency power supply control system 3 for a battery swapping cabinet based on power grid status. The system includes a memory 31 and a processor 32. The memory 31 contains a program for a power adaptive emergency power supply control method for the battery swapping cabinet based on power grid status. When the processor 32 executes the program, the power adaptive emergency power supply control method for the battery swapping cabinet based on power grid status performs the following steps:

[0129] The operation monitoring of the target battery swapping cabinet is carried out to obtain the operation monitoring information of the battery swapping cabinet, and an anomaly detection model is constructed to detect anomalies based on the operation monitoring information of the battery swapping cabinet to obtain anomaly detection information;

[0130] Establish a power grid health evaluation system, calculate the power grid health index by combining the operation monitoring information and abnormal operation detection information of the aforementioned battery swapping cabinet, determine emergency events based on the power grid health index, and obtain emergency event analysis information;

[0131] An emergency event strategy database is constructed. Based on the battery swapping cabinet operation monitoring information, the remaining battery status characteristics of the target battery swapping cabinet are extracted, and emergency power supply management is carried out in combination with the emergency event analysis information.

[0132] Obtain information on the battery swapping cabinet cluster within the target area, define each battery swapping cabinet as a single node to construct a battery swapping cabinet cluster network, and perform an impact analysis on the battery swapping cabinet anomaly when an anomaly is detected to obtain anomaly impact analysis information.

[0133] Obtain the operation and maintenance status information of the target battery swapping cabinet cluster, and formulate a power adaptive compensation strategy based on the abnormal impact analysis information to perform power compensation and regulation on the target battery swapping cabinet cluster.

[0134] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0135] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0136] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0137] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0138] Alternatively, if the integrated units of the present invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks. The above descriptions are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A power adaptive emergency power supply control method for a battery swapping cabinet based on grid conditions, characterized in that, include: The operation monitoring of the target battery swapping cabinet is carried out to obtain the operation monitoring information of the battery swapping cabinet, and an anomaly detection model is constructed to detect anomalies based on the operation monitoring information of the battery swapping cabinet to obtain anomaly detection information; Establish a power grid health evaluation system, calculate the power grid health index by combining the operation monitoring information and abnormal operation detection information of the aforementioned battery swapping cabinet, determine emergency events based on the power grid health index, and obtain emergency event analysis information; An emergency event strategy database is constructed. Based on the battery swapping cabinet operation monitoring information, the remaining battery status characteristics of the target battery swapping cabinet are extracted, and emergency power supply management is carried out in combination with the emergency event analysis information. Obtain information on the battery swapping cabinet cluster within the target area, define each battery swapping cabinet as a single node to construct a battery swapping cabinet cluster network, and perform an impact analysis on the battery swapping cabinet anomaly when an anomaly is detected to obtain anomaly impact analysis information. Obtain the operation and maintenance status information of the target battery swapping cabinet cluster, and formulate a power adaptive compensation strategy based on the abnormal impact analysis information to perform power compensation and regulation on the target battery swapping cabinet cluster. Specifically, the process of acquiring battery swapping cabinet cluster information within the target area involves defining each battery swapping cabinet as a single node to construct a battery swapping cabinet cluster network. When an anomaly is detected in a battery swapping cabinet, an anomaly impact analysis is performed to obtain anomaly impact analysis information. This includes: Obtain information on the battery swapping cabinet cluster within the target area, including the location, type, scale, and power grid connection information of each battery swapping cabinet within the target area; Each battery swapping cabinet is defined as a graph node. Based on the battery swapping cabinet cluster information in the target area, the graph nodes of the battery swapping cabinets are assigned auxiliary attributes. The connection features between each battery swapping cabinet and the power grid are extracted through the battery swapping cabinet cluster information in the target area. Based on the connection characteristics between each battery swapping cabinet and the power grid, the relationship between each battery swapping cabinet is defined. If two battery swapping cabinets are connected to the same power grid branch, a directed edge is established for the corresponding graph node and the edge weight is set. The edge weight is set according to the electrical distance. A battery swapping cabinet cluster network for the target area is constructed using graph nodes, directed edges, and edge weights. The battery swapping cabinet cluster network is then converted into an adjacency matrix for storage. An anomaly impact analysis model is constructed based on the SIR model, and the model is trained and its parameters are adjusted. Obtain emergency event analysis information, update the node status of the battery swapping cabinet cluster network based on the emergency event analysis information, mark and update the abnormal node status and abnormal events, and generate a new adjacency matrix; The new adjacency matrix is ​​input into the anomaly impact analysis model. By iteratively traversing and calculating the infection probability of neighboring nodes, the anomaly impact of the current emergency event on the battery swapping cabinet cluster is analyzed, and anomaly impact analysis information is obtained.

2. The method for adaptive emergency power supply control of a battery swapping cabinet based on grid conditions according to claim 1, characterized in that, The process of monitoring the operation of the target battery swapping cabinet to obtain its operation monitoring information, constructing an anomaly detection model, and performing anomaly detection based on the battery swapping cabinet's operation monitoring information to obtain anomaly detection information specifically includes: A sensor array is set in the target battery swapping cabinet. The sensor array is used to monitor the cabinet's operating information and the battery status information inside the cabinet. The operating monitoring information of the battery swapping cabinet is obtained through the sensor array. The standard deviation of the target battery swapping cabinet operation monitoring data is calculated based on the battery swapping cabinet operation monitoring information. Data that is less than the preset standard deviation threshold is marked as outliers and removed. Data is then interpolated and supplemented using linear interpolation. An anomaly detection model is constructed based on generative adversarial networks. The pre-processed battery swapping cabinet operation monitoring information is processed to obtain a time series sequence, which is then input into the anomaly detection model for analysis. The input time series sequence is used as the original sequence. The generator performs sequence reconstruction based on the original sequence to obtain the reconstructed sequence. The reconstructed sequence is then output to the discriminator to determine whether to accept the current reconstructed sequence. If the discriminator determines that the reconstruction sequence is accepted, the corresponding reconstruction sequence is output; if the discriminator determines that the reconstruction sequence is not accepted, the generator continues to generate the next reconstruction sequence for judgment and outputs the final reconstruction sequence. A preset reconstruction error threshold is set, and the reconstruction error between the final reconstructed sequence and the original sequence is calculated. If the error is greater than the reconstruction error threshold, the final reconstructed sequence is compared with the original sequence to extract the difference features and obtain abnormal operation detection information.

3. The method for adaptive emergency power supply control of a battery swapping cabinet based on grid conditions according to claim 1, characterized in that, The establishment of the power grid health assessment system specifically includes: Based on data retrieval, different historical emergency events are obtained. Feature extraction is performed on each historical emergency event to obtain its features. The K-means clustering algorithm is introduced to classify the historical emergency events according to their features, thus obtaining a historical emergency event dataset. Principal component analysis is used to reduce the dimensionality of each category set, and the principal component scores of each feature in the historical emergency event dataset are calculated. The calculated principal component scores are then compared with a preset threshold. Define the historical emergency event features corresponding to the principal component scores that are greater than a preset threshold as principal component factors, and use the principal component factors to perform principal component direction projection to obtain a projection scatter plot; A preset selection range is used to select corresponding historical emergency event features in the projected scatter plot, generating a first dataset. Based on the first dataset, power grid health evaluation indicators are selected to obtain power grid health evaluation indicators. Based on the selected power grid health evaluation indicators, the information entropy of each power grid health evaluation indicator is calculated using the entropy method. The difference coefficient of each power grid health evaluation indicator is obtained by calculating the information entropy. The selected power grid health evaluation indicators are weighted using the calculated difference coefficients, and a power grid health evaluation system is constructed by combining the power grid health evaluation indicators and the weight allocation results.

4. The method for adaptive emergency power supply control of a battery swapping cabinet based on grid conditions according to claim 1, characterized in that, The process involves combining the battery swapping cabinet operation monitoring information and abnormal operation detection information to calculate the power grid health index, and then using this index to determine emergency events, resulting in emergency event analysis information, specifically including: Obtain the operation monitoring information and abnormal operation detection information of the battery swapping cabinet, and extract the operation features for power grid health evaluation indicators from the operation monitoring information and abnormal operation detection information of the battery swapping cabinet according to the established power grid health evaluation system to obtain the first feature information; A power grid health analysis model is built based on the constructed power grid health evaluation system. The first feature information is input into the power grid health analysis model to calculate the power grid health index and obtain the power grid health index information. Obtain a historical emergency event dataset, and analyze the historical power grid health index of each historical emergency event in the dataset using the power grid health analysis model to obtain historical power grid health index information; Based on the historical emergency event dataset and the historical power grid health index information, an emergency event knowledge graph is constructed with historical emergency event types, historical power grid health indices, and historical emergency event characteristics as meta-paths. The battery swapping cabinet operation monitoring information, abnormal operation detection information, and power grid health index information are used as input features and imported into the emergency event knowledge graph to perform emergency event judgment and analysis on the abnormal status of the target battery swapping cabinet, thereby obtaining emergency event analysis information.

5. The method for adaptive emergency power supply control of a battery swapping cabinet based on grid conditions according to claim 1, characterized in that, The construction of the emergency event strategy database, based on the battery swapping cabinet operation monitoring information, extracts the remaining battery status characteristics of the target battery swapping cabinet, and combines the emergency event analysis information to carry out emergency power supply management, specifically including: Based on the big data network, historical power supply management and control instances of various battery swapping cabinet emergency events are obtained to form an instance dataset. The management and control features of each historical power supply management and control instance are extracted from the instance dataset, including management and control strategy features and management and control event features. The control strategies and control features are associated, and the membership function is preset to calculate the membership value between each control strategy and control feature, so as to obtain the historical power supply control instance feature dataset. The Apriority algorithm is introduced to perform frequent itemset mining on the historical power supply management instance feature dataset. By searching layer by layer, several frequent item subsets are obtained to generate association rules for several management strategies. The confidence level of the association rules for each control strategy is calculated. The association rules that meet the preset confidence threshold are filtered by the calculated confidence level and defined as the triggering rules of the corresponding control strategy to build an emergency event strategy library. Obtain battery swapping cabinet operation monitoring information and emergency event analysis information, and extract the status characteristics of the remaining storage batteries in the target battery swapping cabinet through the battery swapping cabinet operation monitoring information to obtain the remaining battery status characteristic information; The emergency event analysis information is matched and analyzed with the emergency event strategy database to determine the emergency event control strategy that matches the current emergency event of the target battery swapping cabinet, and the matching control strategy information is obtained. Based on the remaining battery status information, the remaining battery status in the target battery swapping cabinet is sorted, and the battery with the best status is selected as the power supply battery. An emergency event management strategy is generated by combining the matching management strategy information to control the emergency power supply of the target battery swapping cabinet.

6. The method for adaptive emergency power supply control of a battery swapping cabinet based on grid conditions according to claim 1, characterized in that, The process of obtaining the operation and maintenance status information of the target battery swapping cabinet cluster, and formulating a power adaptive compensation strategy based on the anomaly impact analysis information, to perform power compensation regulation on the target battery swapping cabinet cluster, specifically includes: Obtain the operation and maintenance status information and anomaly impact analysis information of the target battery swapping cabinet cluster. The operation and maintenance status information of the target battery swapping cabinet cluster includes the power status, battery status and communication status of each battery swapping cabinet in the target area during operation. The objective functions are set based on minimizing power deviation and SOC loss. Constraints are set according to grid capacity. An improved NSGA-II algorithm is introduced to combine the target battery swapping cabinet cluster operation and maintenance status information and abnormal impact analysis information to conduct adaptive compensation strategy analysis and generate an initial population. The objective function value is calculated for each individual in the initial population. All individuals in the population are sorted in non-dominated order according to the objective function value and divided into several Pareto fronts. The crowding distance is calculated for individuals in each non-dominated layer. The tournament selection method is used to select elite individuals by using non-dominated ranking hierarchy and crowding distance. The selected elite individuals are then subjected to crossover and mutation operations, and the parent and offspring generations are merged. The non-dominated ranking and crowding distance are recalculated for all candidate schemes. Based on the Pareto frontier and crowding distance, a preset number of elite individuals are selected to form the next generation of the population for iterative evolution. When the objective function of the population changes to meet the convergence criterion or the preset number of iterations, an adaptive power compensation scheme is output to perform power compensation regulation on the target battery swapping cabinet cluster.

7. A power adaptive emergency power supply control system for a battery swapping cabinet based on grid conditions, characterized in that, The system includes a memory and a processor. The memory contains a program for an adaptive emergency power supply control method for a battery swapping cabinet based on grid conditions. When the program for adaptive emergency power supply control method for a battery swapping cabinet based on grid conditions is executed by the processor, it implements the steps of an adaptive emergency power supply control method for a battery swapping cabinet based on grid conditions as described in any one of claims 1-6.

Images