CN117192373A - Power battery fault analysis method, device, computer equipment and storage medium - Google Patents

Abstract

The application relates to a power battery fault analysis method, a power battery fault analysis device, computer equipment and a storage medium. The method comprises the following steps: acquiring power battery fault alarm data based on a power battery fault alarm database, determining a power battery fault knowledge graph based on a power battery fault knowledge base, determining a fault knowledge aggregation abnormal pattern based on the power battery fault knowledge graph, determining a fault alarm abnormal pattern based on the power battery fault alarm data, determining a power battery fault abnormal pattern based on the fault knowledge aggregation abnormal pattern and the fault alarm abnormal pattern, determining a power battery fault evolution path and a fault item importance degree based on the power battery fault abnormal pattern, and determining a fault item analysis report based on the power battery fault evolution path and the fault item importance degree. The association relation of the power battery faults can be accurately and effectively analyzed, and meanwhile, the reliability of the power battery fault analysis is improved.

Description

Power battery failure analysis method, device, computer equipment and storage medium

Technical field

The present application relates to the field of fault analysis technology, and in particular to a power battery fault analysis method, device, computer equipment and storage medium.

Background technique

As more and more electric vehicles are put into use in the market, power batteries serve as the power source of the entire vehicle and provide power sources for electric vehicles. The maintenance of power batteries is very important. Routine maintenance of power batteries usually includes fault analysis and troubleshooting. In traditional technology, fault analysis of power batteries usually only analyzes a single fault, and cannot discover the correlation between faults and potential dangerous factors. At the same time, power battery failure analysis mainly relies on manual experience analysis. Maintenance personnel need to master comprehensive and systematic professional knowledge and business experience. There are technical problems such as insufficient knowledge and experience reserves that make manual analysis of power battery failures difficult and unstable in reliability.

Therefore, there is an urgent need in the related technology for a method that can analyze the relationship between power battery faults and improve the reliability of power battery fault analysis.

Contents of the invention

Based on this, it is necessary to address the above technical problems and provide a power battery fault analysis method, device, computer equipment and computer-readable storage medium that can analyze the correlation between power battery faults and improve the reliability of power battery fault analysis.

In the first aspect, this application provides a power battery failure analysis method. The methods include:

Obtain power battery fault alarm data based on the power battery fault alarm database;

The power battery fault knowledge graph is determined based on the power battery fault knowledge base, and the fault knowledge aggregation heterogeneous graph is determined based on the power battery fault knowledge graph. The fault knowledge aggregation heterogeneous graph includes entity nodes and topological relationships between nodes, so The entity nodes are each fault item, and the attributes of the entity node include theoretical characteristics of the fault item;

A fault alarm heterogeneous graph is determined based on the power battery fault alarm data. The fault alarm heterogeneous graph includes entity nodes and topological relationships between each node. The entity nodes are fault items, and the attributes of the entity nodes include Power battery status characteristics, the topological relationship includes fault item interactive alarm characteristics;

The power battery fault heterogeneous graph is determined based on the fault knowledge aggregation heterogeneous graph and the fault alarm heterogeneous graph. The power battery fault heterogeneous graph includes entity nodes and topological relationships between each node. The entity nodes are faults. items, the attributes of the entity nodes include theoretical characteristics of fault items and power battery status characteristics, and the topological relationships include interactive alarm characteristics of fault items;

Determine the power battery fault evolution path and the importance of fault items based on the power battery fault heterogeneous graph;

A fault item analysis report is determined based on the power battery fault evolution path and the importance of the fault item.

Optionally, in one embodiment of the present application, determining the fault knowledge aggregation heterogeneous graph based on the power battery fault knowledge graph includes:

Determine source node characteristics, edge characteristics, and target node characteristics based on the power battery fault knowledge graph;

The fault knowledge aggregation heterogeneous graph is determined based on the source node characteristics, edge characteristics and target node characteristics. The target node includes fault items, and the source node includes fault phenomena, fault causes, fault mechanisms, fault consequences, and maintenance plans. At least two items, and the side characteristics include at least two items among phenomena, causes, mechanisms, consequences, and solutions.

Optionally, in one embodiment of the present application, determining the fault knowledge aggregation heterogeneous graph based on the source node characteristics, edge characteristics and target node characteristics includes:

Aggregate the source node characteristics and edge characteristics to obtain the theoretical characteristics of the fault item;

The target node characteristics are updated based on the theoretical characteristics of the fault item, and the battery fault knowledge aggregation heterogeneous graph is determined based on the target node characteristics.

Optionally, in one embodiment of the present application, the power battery fault alarm data includes cloud fault alarm data and vehicle-side fault alarm data, and the determination of the fault alarm heterogeneous graph based on the power battery fault alarm data includes:

Determine the characteristics of cloud interactive alarm and vehicle-side interactive alarm based on cloud fault alarm data and vehicle-side fault alarm data;

The total alarm characteristics are determined based on the cloud interactive alarm characteristics and the vehicle interactive alarm characteristics.

Optionally, in one embodiment of the present application, the power battery fault alarm data includes cloud power battery indicator data and vehicle-side power battery indicator data, and the fault alarm heterogeneous graph is determined based on the power battery fault alarm data. Also includes:

Determine the overall state characteristics of the power battery based on the cloud power battery indicator data and the vehicle end power battery indicator data;

A fault alarm heterogeneous graph is determined based on the total state characteristics and total alarm characteristics of the power battery.

Optionally, in one embodiment of the present application, determining the cloud interactive alarm characteristics and the vehicle interactive alarm characteristics based on the cloud fault alarm data and the vehicle fault alarm data includes:

Determine the cloud fault interactive alarm confidence and the cloud fault alarm correlation based on the cloud fault alarm data, and determine the cloud interactive alarm characteristics based on the cloud fault interactive alarm confidence and the cloud fault alarm correlation;

The vehicle terminal fault interactive alarm confidence and the vehicle terminal fault alarm correlation are determined based on the vehicle terminal fault alarm data, and the vehicle terminal interactive alarm characteristics are determined based on the vehicle terminal fault interactive alarm confidence and the vehicle terminal fault alarm correlation.

Optionally, in one embodiment of the present application, determining the power battery fault evolution path and fault item importance based on the power battery fault heterogeneous graph includes:

Based on the power battery fault heterogeneous graph, clustering and merging are performed to determine the power battery fault cluster subgraph;

Determine the shortest path between fault items based on the power battery fault group subgraph, and determine the power battery fault evolution path;

The importance of the fault item is determined based on the power battery fault group subgraph.

In a second aspect, this application also provides a power battery failure analysis device. The device includes:

The data acquisition module is used to obtain power battery fault alarm data based on the power battery fault alarm database;

A fault knowledge aggregation heterogeneous graph determination module is used to determine a power battery fault knowledge graph based on a power battery fault knowledge base, and determine a fault knowledge aggregation heterogeneous graph based on the power battery fault knowledge graph. The fault knowledge aggregation heterogeneous graph includes entities. Nodes and topological relationships between nodes, the entity nodes include fault items, and the attributes of the entity nodes include theoretical characteristics of the fault items;

A fault alarm heterogeneous graph determination module is used to determine a fault alarm heterogeneous graph based on the power battery fault alarm data. The fault alarm heterogeneous graph includes entity nodes and topological relationships between nodes, and the entity nodes include faults. item, the attributes of the entity node include power battery status characteristics, and the topological relationship includes fault item interactive alarm characteristics;

The power battery fault heterogeneous graph determination module is used to determine the power battery fault heterogeneous graph based on the fault knowledge aggregation heterogeneous graph and the fault alarm heterogeneous graph. The power battery fault heterogeneous graph includes entity nodes and the connections between each node. The topological relationship, the entity node includes a fault item, the attributes of the entity node include the theoretical characteristics of the fault item and the power battery status characteristics, the topological relationship includes the interactive alarm characteristics of the fault item;

A power battery fault analysis module, used to determine the power battery fault evolution path and the importance of fault items based on the power battery fault heterogeneous graph;

A fault item analysis report determination module is used to determine a fault item analysis report based on the power battery fault evolution path and the importance of the fault item.

In a third aspect, this application also provides a computer device. The computer device includes a memory and a processor, the memory stores a computer program, and the processor executes the steps of the methods described in the above embodiments.

In a fourth aspect, this application also provides a computer-readable storage medium. The computer-readable storage medium has a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method described in each of the above embodiments are implemented.

The above-mentioned power battery fault analysis method, device, computer equipment and storage medium firstly obtain the power battery fault alarm data based on the power battery fault alarm database, and then determine the power battery fault knowledge map based on the power battery fault knowledge base. The fault knowledge graph determines a fault knowledge aggregation heterogeneous graph. The fault knowledge aggregation heterogeneous graph includes entity nodes and topological relationships between nodes. The entity nodes are fault items, and the attributes of the entity nodes include fault item theory. Features, and then determine a fault alarm heterogeneous graph based on the power battery fault alarm data. The fault alarm heterogeneous graph includes entity nodes and topological relationships between each node. The entity nodes are each fault item, and the entity The attributes of the nodes include power battery status characteristics, and the topological relationship includes fault item interactive alarm characteristics. Afterwards, the power battery fault heterogeneous graph is determined based on the fault knowledge aggregation heterogeneous graph and the fault alarm heterogeneous graph. The power battery fault is The heterogeneous graph includes entity nodes and topological relationships between each node. The entity nodes are fault items. The attributes of the entity nodes include theoretical characteristics of the fault items and power battery status characteristics. The topological relationships include interactive alarms of the fault items. Features, then, determine the power battery fault evolution path and the fault item importance based on the power battery fault heterogeneous graph, and finally determine the fault item analysis report based on the power battery fault evolution path and the fault item importance. That is to say, by comprehensively considering the theoretical text knowledge of power battery faults, the possibility of interactive alarms of each fault item of the power battery, and the power battery status characteristics when the fault item alarms, the propagation routes and needs between the fault items are determined. Focusing on fault items can accurately and effectively analyze the correlation between power battery faults, while improving the reliability of power battery fault analysis.

Description of the drawings

Figure 1 is an application environment diagram of the power battery failure analysis method in one embodiment;

Figure 2 is a schematic flow chart of a power battery failure analysis method in one embodiment;

Figure 3 is a schematic diagram of a power battery fault knowledge graph in one embodiment;

Figure 4 is a schematic flowchart of determining a fault knowledge aggregation heterogeneous graph based on a power battery fault knowledge graph in one embodiment;

Figure 5 is a schematic flowchart of determining the shortest path between fault items in one embodiment;

Figure 6 is a schematic flowchart of specific steps of a power battery failure analysis method in one embodiment;

Figure 7 is a schematic diagram of a fault alarm heterogeneous graph in an embodiment;

Figure 8 is a structural block diagram of a power battery failure analysis device in one embodiment;

Figure 9 is an internal structure diagram of a computer device in one embodiment.

Detailed ways

In order to make the purpose, technical solutions and advantages of the present application more clear, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present application and are not used to limit the present application.

The power battery failure analysis method provided by the embodiment of the present application can be applied in the application environment as shown in Figure 1. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system may store data that server 104 needs to process. The data storage system can be integrated on the server 104, or placed on the cloud or other network servers. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablets, Internet of Things devices and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart vehicle-mounted devices, etc. . Portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The server 104 can be implemented as an independent server or a server cluster composed of multiple servers.

In one embodiment, as shown in Figure 2, a power battery failure analysis method is provided. This method is explained by taking the method applied to the server in Figure 1 as an example, and includes the following steps:

S201: Obtain power battery fault alarm data based on the power battery fault alarm database.

In the embodiment of this application, first, the power battery failure alarm data is obtained based on the power battery failure alarm database. The power battery failure alarm data is the alarm when the power battery fails obtained from the cloud power battery failure alarm database and the vehicle-end battery failure alarm database. Data includes historical power battery failure alarm data and real-time power battery failure alarm data, such as abnormal voltage, abnormal temperature rise rate, etc.

S203: Determine a power battery fault knowledge graph based on the power battery fault knowledge base, and determine a fault knowledge aggregation heterogeneous graph based on the power battery fault knowledge graph. The fault knowledge aggregation heterogeneous graph includes entity nodes and topological relationships between nodes. , the entity nodes are each fault item, and the attributes of the entity node include theoretical characteristics of the fault item.

In the embodiment of this application, the power battery fault knowledge graph refers to a knowledge graph created using the relevant theoretical knowledge of power battery faults. Specifically, the relevant theoretical knowledge of power battery faults is obtained through the power battery fault knowledge base, and the power battery fault knowledge graph is created, such as As shown in Figure 3, the power battery fault knowledge graph takes each fault item as the central node and connects characteristic nodes such as fault phenomenon, fault cause, fault mechanism, fault consequence, maintenance plan, etc., corresponding to phenomena, causes, mechanisms, consequences, maintenance, etc. Scheme and other related information are used as attributes of the feature node. After that, based on the power battery fault knowledge graph, feature aggregation of each fault information is performed to obtain a fault knowledge aggregation heterogeneous graph. The fault knowledge aggregation heterogeneous graph includes entity nodes and the topological relationships between each node. The entity nodes are each fault. Item, the attribute of the entity node is the theoretical characteristic of the fault item obtained after feature aggregation.

S205: Determine a fault alarm heterogeneous graph based on the power battery fault alarm data. The fault alarm heterogeneous graph includes entity nodes and topological relationships between nodes. The entity nodes are fault items, and the entity nodes are The attributes include power battery status characteristics, and the topological relationship includes fault item interactive alarm characteristics.

In the embodiment of this application, after the power battery fault alarm data is obtained, calculations are performed based on the power battery fault alarm data to determine the power battery status characteristics and the fault item interactive alarm characteristics. The power battery status characteristics refer to the corresponding power battery when each fault item alarms. The real-time status is mainly manifested as changes in power battery indicator data. The interactive alarm feature of fault items refers to the correlation of interactive alarms between fault items, that is, when one fault item alarms, the possibility of another fault item alarming and the two Afterwards, each fault item is used as an entity node, the power battery status characteristics are used as attributes of the entity node, and the interactive alarm characteristics of the fault items are used as edges to obtain a fault alarm heterogeneous graph, that is, a fault alarm heterogeneous graph. It includes entity nodes and topological relationships between each node. The entity nodes are fault items. The attributes of the entity nodes include power battery status characteristics. The topological relationships include interactive alarm characteristics of fault items.

S207: Determine a power battery fault heterogeneous graph based on the fault knowledge aggregation heterogeneous graph and the fault alarm heterogeneous graph. The power battery fault heterogeneous graph includes entity nodes and topological relationships between each node. The entity nodes are For each fault item, the attributes of the entity node include theoretical characteristics of the fault item and power battery status characteristics, and the topological relationship includes interactive alarm characteristics of the fault item.

In the embodiment of this application, after determining the fault knowledge aggregation heterogeneous graph and the fault alarm heterogeneous graph, the nodes and edges are merged correspondingly to obtain the power battery fault heterogeneous graph. The power battery fault heterogeneous graph includes entity nodes and various The topological relationship between nodes. The entity nodes are the corresponding fault items in the fault knowledge aggregation heterogeneous graph and the fault alarm heterogeneous graph. The attributes of the entity nodes include the theoretical characteristics of the fault items in the fault knowledge aggregation heterogeneous graph and the fault alarm anomalies. The power battery status characteristics and topological relationships in the composition include the interactive alarm characteristics of fault items in the fault alarm heterogeneous graph.

S209: Determine the power battery fault evolution path and the fault item importance based on the power battery fault heterogeneous graph.

In the embodiment of this application, after determining the power battery fault heterogeneous diagram, by analyzing the theoretical characteristics of the fault items, the power battery status characteristics, and the interactive alarm characteristics of the fault items, the correlation between the fault items is obtained, and the power battery fault is determined. Evolution path and fault item importance. Among them, the power battery fault evolution path refers to the propagation route between fault items that are most likely to cause an alarm when a certain fault item alarms. The fault item importance refers to the core fault item.

S211: Determine a fault item analysis report based on the power battery fault evolution path and fault item importance.

In the embodiment of this application, after determining the power battery fault evolution path and the fault item importance based on the power battery fault heterogeneous graph, a fault item analysis report is determined based on the power battery fault evolution path and the fault item importance. The fault item analysis report may include The relationship between each fault item, the fault items that need to be detected, and the fault item repair methods, etc.

In the above power battery fault analysis method, first, the power battery fault alarm data is obtained based on the power battery fault alarm database, and then, the power battery fault knowledge graph is determined based on the power battery fault knowledge base, and the fault knowledge aggregation is determined based on the power battery fault knowledge graph. Heterogeneous graph. The fault knowledge aggregation heterogeneous graph includes entity nodes and topological relationships between nodes. The entity nodes include fault items. The attributes of the entity nodes include theoretical characteristics of the fault items. Then, based on the power The battery fault alarm data determines a fault alarm heterogeneous graph. The fault alarm heterogeneous graph includes entity nodes and topological relationships between nodes. The entity nodes include fault items. The attributes of the entity nodes include power battery status characteristics. The topological relationship includes fault item interactive alarm features. Afterwards, the power battery fault heterogeneous graph is determined based on the fault knowledge aggregation heterogeneous graph and the fault alarm heterogeneous graph. The power battery fault heterogeneous graph includes entity nodes and each node. The topological relationship between the entity node includes a fault item, the attributes of the entity node include the theoretical characteristics of the fault item and the power battery status characteristics, the topological relationship includes the interactive alarm characteristics of the fault item, and then, based on the power battery failure The heterogeneous graph determines the power battery fault evolution path and the fault item importance. Finally, the fault item analysis report is determined based on the power battery fault evolution path and the fault item importance. That is to say, by comprehensively considering the theoretical text knowledge of power battery faults, the possibility of interactive alarms of each fault item of the power battery, and the power battery status characteristics when the fault item alarms, the propagation routes and needs between the fault items are determined. Focusing on fault items can accurately and effectively analyze the correlation between power battery faults, while improving the reliability of power battery fault analysis.

In one embodiment of the present application, determining the fault knowledge aggregation heterogeneous graph based on the power battery fault knowledge graph includes:

S301: Determine source node characteristics, edge characteristics, and target node characteristics based on the power battery fault knowledge graph.

S303: Determine the fault knowledge aggregation heterogeneous graph based on the source node characteristics, edge characteristics and target node characteristics. The target node includes fault items, and the source node includes fault phenomena, fault causes, fault mechanisms, fault consequences, and maintenance plans. At least two of the side characteristics include at least two of phenomena, causes, mechanisms, consequences, and solutions.

In one embodiment of the present application, as shown in Figure 4, first a word embedding operation is performed based on the text knowledge in the power battery fault knowledge graph, the corresponding feature tensor is obtained, and the source node features, edge features and target node features are obtained. Construct a power battery fault knowledge heterogeneous graph. After that, the source node characteristics, edge characteristics and target node characteristics of each edge in the power battery fault knowledge heterogeneous graph are obtained through the message generation function. Among them, the target node includes the fault item, and the source node includes the fault phenomenon, fault cause, fault mechanism, At least two of the consequences of the failure and the maintenance plan, and the side characteristics include at least two of the phenomena, causes, mechanisms, consequences, and plans. Afterwards, with each target node, that is, the fault item, as the center, messages are disseminated to the source nodes connected to the target node, and new features are aggregated and assigned to the target node to form a fault knowledge aggregation heterogeneous graph.

In this embodiment, the source node characteristics, edge characteristics, and target node characteristics are determined through the power battery fault knowledge graph, and the fault knowledge aggregation heterogeneous graph is determined based on the source node characteristics, edge characteristics, and target node characteristics, which can make the information of the fault item node more detailed. Rich.

In one embodiment of the present application, determining the fault knowledge aggregation heterogeneous graph based on the source node characteristics, edge characteristics and target node characteristics includes:

S401: Aggregate the source node characteristics and edge characteristics to obtain theoretical characteristics of the fault item.

S403: Update the target node characteristics based on the theoretical characteristics of the fault item, and determine the battery fault knowledge aggregation heterogeneous graph based on the target node characteristics.

In one embodiment of the present application, as shown in Figure 4, after obtaining the source node characteristics, edge characteristics, and target node characteristics of each edge in the power battery fault knowledge heterogeneous graph through the message generation function, they are respectively named u_feat, e_feat, v_feat. After that, the feature tensor of the source node and the feature tensor of the edge are summed, recorded as u_e_feat. With the target node, that is, the fault item as the center, the sum of all source node features and edge features flowing into the target node is aggregated through the aggregation function u_e_feat, and the fault is obtained. The theoretical feature is denoted as total_v_feat. After that, through the node update function, the target node characteristics are updated based on the fault item theoretical feature total_v_feat, and the final battery fault knowledge aggregation heterogeneous graph is determined based on the target node characteristics.

In this embodiment, by aggregating the source node features and edge features, the fault item theoretical features are obtained, the target node features are updated based on the fault item theoretical features, and the battery fault knowledge aggregation heterogeneous graph is determined based on the target node features, which can Unifying the information of the fault item node into a target node can more effectively merge it with the fault alarm heterogeneous graph.

In one embodiment of the present application, the power battery fault alarm data includes cloud fault alarm data and vehicle fault alarm data, and the determination of the fault alarm heterogeneous graph based on the power battery fault alarm data includes:

S501: Determine the cloud interactive alarm characteristics and the vehicle interactive alarm characteristics based on the cloud fault alarm data and the vehicle terminal fault alarm data.

S503: Determine the total alarm characteristics based on the cloud interactive alarm characteristics and the vehicle interactive alarm characteristics.

In one embodiment of the present application, the power battery fault alarm data includes cloud fault alarm data and vehicle-side fault alarm data, where the vehicle-side fault alarm data refers to unseen data stored in the vehicle-side battery fault alarm database and obtained directly from the vehicle side. Processed fault alarm data. Cloud fault alarm data refers to the processed fault alarm data obtained from the vehicle stored in the cloud battery fault alarm database. The fault alarm data includes abnormal voltage, abnormal temperature rise rate, and abnormal battery state of charge display. wait. After obtaining the power battery fault alarm data, the cloud interactive alarm characteristics and the vehicle interactive alarm characteristics are determined based on the cloud fault alarm data and the vehicle fault alarm data respectively, that is, the cloud interactive alarm coefficient and the vehicle interactive alarm coefficient, which are recorded as Alarm_Coeff _cloud and Alarm_Coeff _car , where the interactive alarm feature refers to the correlation between each fault item, such as confidence, correlation, etc. After that, the total alarm characteristics are determined based on the cloud interactive alarm characteristics and the vehicle interactive alarm characteristics. Specifically, the weights are determined according to the importance of the cloud and the vehicle, which are recorded as α and 1-α respectively. The total alarm is obtained after the weighted sum of the two. The characteristic, that is, the total alarm coefficient, is recorded as Alarm_Coeff _total . The specific calculation method is as shown in the following formula.

Alarm_Coeff _total =α×Alarm_Coeff _cloud +(1-α)×Alarm_Coeff _car

In this embodiment, the cloud interactive alarm feature and the vehicle interactive alarm feature are determined based on the cloud fault alarm data and the vehicle fault alarm data, and the total alarm feature is determined based on the cloud interactive alarm feature and the vehicle interactive alarm feature, and the total alarm feature is determined based on the degree of importance. Assign different weights to the cloud and the car to make the fault alarm heterogeneous graph more reasonable.

In one embodiment of the present application, the power battery fault alarm data includes cloud power battery indicator data and vehicle-side power battery indicator data, and determining the fault alarm heterogeneous graph based on the power battery fault alarm data also includes:

S601: Determine the overall state characteristics of the power battery based on the cloud power battery indicator data and the vehicle end power battery indicator data.

S603: Determine a fault alarm heterogeneous graph based on the total state characteristics and total alarm characteristics of the power battery.

In one embodiment of the present application, the power battery fault alarm data includes cloud power battery indicator data and vehicle-side power battery indicator data. It is the corresponding power battery indicator data obtained when each fault item of the power battery is alarmed. The power battery indicator data refers to Various indicator data of the battery when the vehicle reports a certain fault, such as power battery state of charge SOC, power battery health status SOH, power battery remaining energy SOE, battery pack discharge depth DOD, battery voltage difference consistency, battery temperature, etc. . After that, based on the cloud power battery indicator data and the vehicle end power battery indicator data, the mode is calculated to obtain the cloud power battery status characteristics and the vehicle end power battery status characteristics, which are recorded as Feature _cloud and Feature _car respectively, and according to the cloud and The importance of the vehicle end determines the weights, which are recorded as α and 1-α respectively. After weighting and summing the two, _{the total} state characteristics of the power battery are obtained. The specific calculation method is as shown in the following formula.

Feature _total =α×Feature _cloud +(1-α)×Feature _car

After that, the entity nodes are each fault item, and the total state characteristics of the power battery corresponding to each fault item are used as the entity node characteristics of the fault alarm heterogeneous graph, and the total alarm characteristics are used as edge features to obtain the fault alarm heterogeneous graph.

In this embodiment, the overall state characteristics of the power battery are determined through the power battery indicator data when the power battery fails to alarm, and the fault alarm heterogeneous graph is determined based on the overall state characteristics of the power battery and the total alarm characteristics, making the power battery failure analysis more reliable. .

In one embodiment of the present application, determining the cloud interactive alarm characteristics and the vehicle interactive alarm characteristics based on the cloud fault alarm data and the vehicle fault alarm data includes:

S701: Determine the cloud fault interactive alarm confidence and the cloud fault alarm correlation based on the cloud fault alarm data, and determine the cloud interactive alarm characteristics based on the cloud fault interactive alarm confidence and the cloud fault alarm correlation.

S703: Determine the vehicle-side fault interactive alarm confidence and the vehicle-side fault alarm correlation based on the vehicle-side fault alarm data, and determine the vehicle-side interactive alarm characteristics based on the vehicle-side fault interactive alarm confidence and the vehicle-side fault alarm correlation.

In one embodiment of the present application, the cloud fault interactive alarm confidence and the cloud fault alarm correlation are first determined based on the cloud fault alarm data, where the cloud fault interactive alarm confidence means that when a certain fault item alarms, another fault item The frequency of alarms, for example, calculating the frequency of alarms of fault item A when fault item B alarms, is recorded as Confidence(B→A), and the calculation method is as shown in the following formula.

Among them, P(B) is the frequency of alarm of fault B, and P(A∩B) is the frequency of interactive alarm of fault A and fault B.

Cloud fault alarm correlation refers to the correlation between each fault item, for example, calculating the correlation between fault A and fault B, denoted as The calculation method is as shown in the following formula.

Among them, cov(A,B) is the covariance of fault A and fault B, and σ _A is the standard deviation of fault A.

After that, based on the confidence of the vehicle-side fault interactive alarm and the correlation of the vehicle-side fault alarm, the characteristics of the vehicle-side interactive alarm, that is, the vehicle-side interactive alarm coefficient, are determined. For example, calculate the interactive alarm coefficient of fault A after fault B alarm, which is recorded as Alarm_Coeff(B→A). The calculation method is as follows.

Using the same calculation method, the vehicle-side fault interactive alarm confidence and the vehicle-side fault alarm correlation are determined based on the vehicle-side fault alarm data, and the vehicle-side interactive alarm characteristics are determined based on the vehicle-side fault interactive alarm confidence and the vehicle-side fault alarm correlation. .

In this embodiment, the cloud fault interactive alarm confidence and the cloud fault alarm correlation are determined based on the cloud fault alarm data. The cloud interactive alarm characteristics are determined based on the cloud fault interactive alarm confidence and the cloud fault alarm correlation. Based on the vehicle fault, The alarm data determines the vehicle-side fault interactive alarm confidence and the vehicle-side fault alarm correlation. Based on the vehicle-side fault interactive alarm confidence and the vehicle-side fault alarm correlation, the vehicle-side interactive alarm characteristics are determined, comprehensively considering the relationship between each fault item. Correlation and fault interactive alarm probability can more comprehensively consider the possibility of fault interactive alarm.

In one embodiment of the present application, determining the power battery fault evolution path and fault item importance based on the power battery fault heterogeneous graph includes:

S801: Based on the power battery fault heterogeneous graph, cluster and merge to determine the power battery fault cluster subgraph.

S803: Determine the shortest path between fault items based on the power battery fault group subgraph, and determine the power battery fault evolution path.

S805: Determine the importance of fault items based on the power battery fault group subgraph.

In one embodiment of the present application, after the power fault heterogeneous graph is determined, based on the power battery fault heterogeneous graph, each node in the power battery fault heterogeneous graph is clustered and merged, and the power battery fault heterogeneous graph is divided into Multiple power battery faults are divided into cluster subgraphs. Among them, clustering and merging can use graph convolution algorithm, Louvain community detection algorithm, etc. Afterwards, the shortest path between fault items is determined based on the power battery fault group subgraph, and the power battery fault evolution path is determined, that is, the power battery fault potential evolution path. Among them, methods such as Floyd algorithm, breadth-first search algorithm, and depth-first search algorithm can be used to determine the shortest path between fault items. Specifically, taking the Floyd algorithm as an example, as shown in Figure 5, first, extract the adjacency matrix in each power battery fault group subgraph, and use the adjacency matrix as the initialization path matrix P ₀ , indicating that the current subgraph has not gone through any The shortest distance between each node in the case of a transit node, where the value in row A and column B represents the distance from node A to node B under the current situation. After that, the number of faulty nodes is counted as L, and i is assigned a value of 1. After that, using the i fault node as a transit node, traverse all fault nodes and calculate the distance between any two nodes. If the distance is less than the distance in the P _i-1 matrix, the smaller distance is updated in the path matrix. , and use the updated path matrix as the new path matrix _Pi . After that, L nodes are traversed in sequence, and different nodes are used as transit nodes in sequence to update the path matrix. After that, after L nodes are traversed, the final updated path matrix is The shortest path matrix of the clique graph is the potential evolution path of the fault. After that, the importance of the fault item is determined based on the power battery fault group subgraph, that is, based on the theoretical characteristics of the entity node attribute fault item and the power battery status characteristics, as well as the interactive alarm characteristics of the topological relationship fault item, the feature vector centrality or betweenness centrality is used to determine the fault Item importance, specifically, taking eigenvector centrality as an example, first of all, the eigenvector centrality of a node is related to the eigenvector centrality of its surrounding nodes, and is the average of the eigenvector centralities of the other adjacent nodes. It can be expressed in matrix form as:

λc=Ac

Among them, c is a vector where elements /> For the eigenvector centrality of node i, the eigenvector centrality of each node is obtained by solving the eigenvector of the adjacency matrix A;

After that, solve for the eigenvalue λ of the adjacency matrix A, and take the largest eigenvalue λ _max . Then, solve for the eigenvector c _max corresponding to the largest eigenvalue λ _max . The elements in the vector are the eigenvector centralities of each node in the graph. , after that, the elements in the feature vector c _max are sorted, and the sorting order is the order of importance of the corresponding labeled fault items.

In this embodiment, the power battery fault clustering subgraph is determined by clustering and merging based on the power battery fault heterogeneous graph, and the power battery fault evolution path and fault item importance are determined based on the power battery fault clustering subgraph. Path evolution analysis and fault item importance analysis in related cluster subgraphs can make the power battery fault evolution analysis more reasonable and efficient, identify the fault items that need to be monitored, and make the power battery fault analysis more reliable.

The following uses a specific embodiment to illustrate the specific implementation steps of the power battery fault analysis method of the present application. As shown in Figure 6, first, S901, obtains power battery fault alarm data based on the power battery fault alarm database. After that, S903, determine the power battery fault knowledge graph based on the power battery fault knowledge base, and determine the fault knowledge aggregation heterogeneous graph based on the power battery fault knowledge graph. The fault knowledge aggregation heterogeneous graph includes entity nodes and links between each node. Topological relationship, the entity node is each fault item, and the attributes of the entity node include the theoretical characteristics of the fault item. Specifically, S905-S909, determine the source node characteristics, edge characteristics, and target node characteristics based on the power battery fault knowledge graph, and aggregate The source node characteristics and edge characteristics are used to obtain the theoretical characteristics of the fault item, the target node characteristics are updated based on the theoretical characteristics of the fault item, and the battery fault knowledge aggregation heterogeneous graph is determined based on the target node characteristics.

Afterwards, S911, determine a fault alarm heterogeneous graph based on the power battery fault alarm data. The fault alarm heterogeneous graph includes entity nodes and topological relationships between nodes. The entity nodes are fault items, and the entities are The attributes of the nodes include power battery status characteristics, and the topological relationships include fault item interactive alarm characteristics. Specifically, S913-S921 determines the cloud fault interactive alarm confidence and the cloud fault alarm correlation based on the cloud fault alarm data, determines the cloud interactive alarm characteristics based on the cloud fault interactive alarm confidence and the cloud fault alarm correlation, based on the vehicle terminal The fault alarm data determines the vehicle terminal fault interactive alarm confidence and the vehicle terminal fault alarm correlation, determines the vehicle terminal interactive alarm characteristics based on the vehicle terminal fault interactive alarm confidence and the vehicle terminal fault alarm correlation, and determines the vehicle terminal interactive alarm characteristics based on the cloud terminal interactive alarm characteristics. Interact with the vehicle-side alarm characteristics to determine the total alarm characteristics, determine the overall status characteristics of the power battery based on the cloud power battery indicator data and the vehicle-side power battery indicator data, and determine the fault alarm heterogeneity based on the total status characteristics of the power battery and total alarm characteristics. picture. As shown in Figure 7, it is a schematic diagram of a fault alarm heterogeneous graph. The entity nodes of the fault alarm heterogeneous graph are each fault item, and the relationship edges of the fault alarm heterogeneous graph are the total alarm characteristics. For example, the fault item B node to The relationship edge of the fault item A node is Alarm_Coeff _total (B→A), the relationship edge of the fault item A node to the fault item B node is Alarm_Coeff _total (A→B), and the node attribute of the fault alarm heterogeneous graph is the total status of the power battery. Features, for example, the attribute of the fault item A node is Feature _total (A).

After that, S923, determine the power battery fault heterogeneous graph based on the fault knowledge aggregation heterogeneous graph and the fault alarm heterogeneous graph. The power battery fault heterogeneous graph includes entity nodes and topological relationships between nodes. The entities The nodes are each fault item, the attributes of the entity node include the theoretical characteristics of the fault item and the power battery status characteristics, and the topological relationship includes the interactive alarm characteristics of the fault item. After that, S925-S929, cluster and merge based on the power battery fault heterogeneous graph, determine the power battery fault cluster subgraph, determine the shortest path between fault items based on the power battery fault cluster subgraph, and determine the power battery fault Evolution path, determine the importance of fault items based on the power battery fault group subgraph.

Finally, S933, determine the fault item analysis report based on the power battery fault evolution path and the fault item importance.

It should be understood that although the steps in the flowcharts involved in the above-mentioned embodiments are shown in sequence as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated in this article, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least some of the steps in the flowcharts involved in the above embodiments may include multiple steps or stages. These steps or stages are not necessarily executed at the same time, but may be completed at different times. The execution order of these steps or stages is not necessarily sequential, but may be performed in turn or alternately with other steps or at least part of the steps or stages in other steps.

Based on the same inventive concept, embodiments of the present application also provide a power battery fault analysis device for implementing the above-mentioned power battery fault analysis method. The solution to the problem provided by this device is similar to the solution recorded in the above method. Therefore, the specific limitations in the embodiments of one or more power battery fault analysis devices provided below can be found in the above article on power battery fault analysis. The limitations of the method will not be repeated here.

In one embodiment, as shown in Figure 8, a power battery fault analysis device 800 is provided, including: a data acquisition module 801, a fault knowledge aggregation heterogeneous graph determination module 803, a fault alarm heterogeneous graph determination module 805, a power battery fault analysis device 800, and a power battery fault analysis device 800. Battery fault heterogeneous map determination module 807, power battery fault analysis module 809 and fault item analysis report determination module 811, among which:

The data acquisition module 801 is used to acquire power battery fault alarm data based on the power battery fault alarm database.

The fault knowledge aggregation heterogeneous graph determination module 803 is used to determine the power battery fault knowledge graph based on the power battery fault knowledge base, and determine the fault knowledge aggregation heterogeneous graph based on the power battery fault knowledge graph. The fault knowledge aggregation heterogeneous graph includes The entity node and the topological relationship between each node, the entity node includes the fault item, and the attributes of the entity node include the theoretical characteristics of the fault item.

The fault alarm heterogeneous graph determination module 805 is used to determine the fault alarm heterogeneous graph based on the power battery fault alarm data. The fault alarm heterogeneous graph includes entity nodes and topological relationships between each node. The entity nodes include Fault item, the attributes of the entity node include power battery status characteristics, and the topological relationship includes interactive alarm characteristics of the fault item.

The power battery fault heterogeneous graph determination module 807 is used to determine the power battery fault heterogeneous graph based on the fault knowledge aggregation heterogeneous graph and the fault alarm heterogeneous graph. The power battery fault heterogeneous graph includes entity nodes and the relationship between each node. The topological relationship between the entities includes a fault item, the attributes of the entity node include theoretical characteristics of the fault item and power battery status characteristics, and the topological relationship includes interactive alarm characteristics of the fault item.

The power battery fault analysis module 809 is used to determine the power battery fault evolution path and the importance of fault items based on the power battery fault heterogeneous graph.

The fault item analysis report determination module 811 is used to determine a fault item analysis report based on the power battery fault evolution path and the fault item importance.

In one embodiment of the present application, the fault knowledge aggregation heterogeneous graph determination module is also used to:

Determine source node characteristics, edge characteristics, and target node characteristics based on the power battery fault knowledge graph;

The fault knowledge aggregation heterogeneous graph is determined based on the source node characteristics, edge characteristics and target node characteristics. The target node includes fault items, and the source node includes fault phenomena, fault causes, fault mechanisms, fault consequences, and maintenance plans. At least two items, and the side characteristics include at least two items among phenomena, causes, mechanisms, consequences, and solutions.

In one embodiment of the present application, the fault knowledge aggregation heterogeneous graph determination module is also used to:

Aggregate the source node characteristics and edge characteristics to obtain the theoretical characteristics of the fault item;

The target node characteristics are updated based on the theoretical characteristics of the fault item, and the battery fault knowledge aggregation heterogeneous graph is determined based on the target node characteristics.

In one embodiment of the present application, the power battery fault alarm data includes cloud fault alarm data and vehicle-end fault alarm data, and the fault alarm heterogeneous graph determination module is also used to:

Determine the characteristics of cloud interactive alarm and vehicle-side interactive alarm based on cloud fault alarm data and vehicle-side fault alarm data;

The total alarm characteristics are determined based on the cloud interactive alarm characteristics and the vehicle interactive alarm characteristics.

In one embodiment of the present application, the power battery fault alarm data includes cloud power battery indicator data and vehicle-side power battery indicator data. The fault alarm heterogeneous graph determination module is also used to:

Determine the overall state characteristics of the power battery based on the cloud power battery indicator data and the vehicle end power battery indicator data;

A fault alarm heterogeneous graph is determined based on the total state characteristics and total alarm characteristics of the power battery.

In one embodiment of this application, the fault alarm heterogeneous graph determination module is also used to:

Determine the cloud fault interactive alarm confidence and the cloud fault alarm correlation based on the cloud fault alarm data, and determine the cloud interactive alarm characteristics based on the cloud fault interactive alarm confidence and the cloud fault alarm correlation;

The vehicle terminal fault interactive alarm confidence and the vehicle terminal fault alarm correlation are determined based on the vehicle terminal fault alarm data, and the vehicle terminal interactive alarm characteristics are determined based on the vehicle terminal fault interactive alarm confidence and the vehicle terminal fault alarm correlation.

In one embodiment of the present application, the power battery fault analysis module is also used to:

Based on the power battery fault heterogeneous graph, clustering and merging are performed to determine the power battery fault cluster subgraph;

Determine the shortest path between fault items based on the power battery fault group subgraph, and determine the power battery fault evolution path;

The importance of the fault item is determined based on the power battery fault group subgraph.

Each module in the above-mentioned power battery fault analysis device can be realized in whole or in part by software, hardware and their combination. Each of the above modules may be embedded in or independent of the processor of the computer device in the form of hardware, or may be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. 9 . The computer device includes a processor, memory, communication interface, display screen and input device connected through a system bus. Wherein, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes non-volatile storage media and internal memory. The non-volatile storage medium stores operating systems and computer programs. This internal memory provides an environment for the execution of operating systems and computer programs in non-volatile storage media. The communication interface of the computer device is used for wired or wireless communication with external terminals. The wireless mode can be implemented through WIFI, mobile cellular network, NFC (Near Field Communication) or other technologies. The computer program implements a power battery failure analysis method when executed by a processor. The display screen of the computer device may be a liquid crystal display or an electronic ink display. The input device of the computer device may be a touch layer covered on the display screen, or may be a button, trackball or touch pad provided on the computer device shell. , it can also be an external keyboard, trackpad or mouse, etc.

Those skilled in the art can understand that the structure shown in Figure 9 is only a block diagram of a partial structure related to the solution of the present application, and does not constitute a limitation on the computer equipment to which the solution of the present application is applied. Specific computer equipment can May include more or fewer parts than shown, or combine certain parts, or have a different arrangement of parts.

In one embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory. When the processor executes the computer program, it implements the steps in the above method embodiments.

In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.

Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage. In the media, when executed, the computer program may include the processes of the above method embodiments. Any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetic variable memory (Magnetoresistive Random) Access Memory (MRAM), Ferroelectric Random Access Memory (FRAM), Phase Change Memory (PCM), graphene memory, etc. Volatile memory may include random access memory (Random Access Memory, RAM) or external cache memory. By way of illustration but not limitation, RAM can be in various forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM). The databases involved in the various embodiments provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include blockchain-based distributed databases, etc., but are not limited thereto. The processors involved in the various embodiments provided in this application may be general-purpose processors, central processing units, graphics processors, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to this.

The technical features of the above embodiments can be combined in any way. To simplify the description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, all possible combinations should be used. It is considered to be within the scope of this manual.

The above-described embodiments only express several implementation modes of the present application, and their descriptions are relatively specific and detailed, but should not be construed as limiting the patent scope of the present application. It should be noted that, for those of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all fall within the protection scope of the present application. Therefore, the scope of protection of this application should be determined by the appended claims.

Claims (10)

1. A power battery fault analysis method, characterized in that the method includes: Obtain power battery fault alarm data based on the power battery fault alarm database; The power battery fault knowledge graph is determined based on the power battery fault knowledge base, and the fault knowledge aggregation heterogeneous graph is determined based on the power battery fault knowledge graph. The fault knowledge aggregation heterogeneous graph includes entity nodes and topological relationships between nodes, so The entity nodes are each fault item, and the attributes of the entity node include theoretical characteristics of the fault item; A fault alarm heterogeneous graph is determined based on the power battery fault alarm data. The fault alarm heterogeneous graph includes entity nodes and topological relationships between each node. The entity nodes are fault items, and the attributes of the entity nodes include Power battery status characteristics, the topological relationship includes fault item interactive alarm characteristics; The power battery fault heterogeneous graph is determined based on the fault knowledge aggregation heterogeneous graph and the fault alarm heterogeneous graph. The power battery fault heterogeneous graph includes entity nodes and topological relationships between each node. The entity nodes are faults. items, the attributes of the entity nodes include theoretical characteristics of fault items and power battery status characteristics, and the topological relationships include interactive alarm characteristics of fault items; Determine the power battery fault evolution path and the importance of fault items based on the power battery fault heterogeneous graph; A fault item analysis report is determined based on the power battery fault evolution path and the importance of the fault item. 2. The method according to claim 1, wherein determining the fault knowledge aggregation heterogeneous graph based on the power battery fault knowledge graph includes: Determine source node characteristics, edge characteristics, and target node characteristics based on the power battery fault knowledge graph; The fault knowledge aggregation heterogeneous graph is determined based on the source node characteristics, edge characteristics and target node characteristics. The target node includes fault items, and the source node includes fault phenomena, fault causes, fault mechanisms, fault consequences, and maintenance plans. At least two items, and the side characteristics include at least two items among phenomena, causes, mechanisms, consequences, and solutions. 3. The method according to claim 2, wherein determining the fault knowledge aggregation heterogeneous graph based on the source node characteristics, edge characteristics and target node characteristics includes: Aggregate the source node characteristics and edge characteristics to obtain the theoretical characteristics of the fault item; The target node characteristics are updated based on the theoretical characteristics of the fault item, and the battery fault knowledge aggregation heterogeneous graph is determined based on the target node characteristics. 4. The method according to claim 1, wherein the power battery fault alarm data includes cloud fault alarm data and vehicle fault alarm data, and the fault alarm heterogeneous graph is determined based on the power battery fault alarm data. include: Determine the characteristics of cloud interactive alarm and vehicle-side interactive alarm based on cloud fault alarm data and vehicle-side fault alarm data; The total alarm characteristics are determined based on the cloud interactive alarm characteristics and the vehicle interactive alarm characteristics. 5. The method according to claim 4, wherein the power battery fault alarm data includes cloud power battery indicator data and vehicle end power battery indicator data, and the fault alarm abnormality is determined based on the power battery fault alarm data. The composition also includes: Determine the overall state characteristics of the power battery based on the cloud power battery indicator data and the vehicle end power battery indicator data; A fault alarm heterogeneous graph is determined based on the total state characteristics and total alarm characteristics of the power battery. 6. The method according to claim 4, wherein determining the cloud interactive alarm characteristics and the vehicle interactive alarm characteristics based on the cloud fault alarm data and the vehicle fault alarm data includes: Determine the cloud fault interactive alarm confidence and the cloud fault alarm correlation based on the cloud fault alarm data, and determine the cloud interactive alarm characteristics based on the cloud fault interactive alarm confidence and the cloud fault alarm correlation; The vehicle terminal fault interactive alarm confidence and the vehicle terminal fault alarm correlation are determined based on the vehicle terminal fault alarm data, and the vehicle terminal interactive alarm characteristics are determined based on the vehicle terminal fault interactive alarm confidence and the vehicle terminal fault alarm correlation. 7. The method of claim 1, wherein determining the power battery fault evolution path and fault item importance based on the power battery fault heterogeneous graph includes: Based on the power battery fault heterogeneous graph, clustering and merging are performed to determine the power battery fault cluster subgraph; Determine the shortest path between fault items based on the power battery fault group subgraph, and determine the power battery fault evolution path; The importance of the fault item is determined based on the power battery fault group subgraph. 8. A power battery failure analysis device, characterized in that the device includes: The data acquisition module is used to obtain power battery fault alarm data based on the power battery fault alarm database; A fault knowledge aggregation heterogeneous graph determination module is used to determine a power battery fault knowledge graph based on a power battery fault knowledge base, and determine a fault knowledge aggregation heterogeneous graph based on the power battery fault knowledge graph. The fault knowledge aggregation heterogeneous graph includes entities. Nodes and topological relationships between nodes, the entity nodes include fault items, and the attributes of the entity nodes include theoretical characteristics of the fault items; A fault alarm heterogeneous graph determination module is used to determine a fault alarm heterogeneous graph based on the power battery fault alarm data. The fault alarm heterogeneous graph includes entity nodes and topological relationships between nodes, and the entity nodes include faults. item, the attributes of the entity node include power battery status characteristics, and the topological relationship includes fault item interactive alarm characteristics; The power battery fault heterogeneous graph determination module is used to determine the power battery fault heterogeneous graph based on the fault knowledge aggregation heterogeneous graph and the fault alarm heterogeneous graph. The power battery fault heterogeneous graph includes entity nodes and the connections between each node. The topological relationship, the entity node includes a fault item, the attributes of the entity node include the theoretical characteristics of the fault item and the power battery status characteristics, the topological relationship includes the interactive alarm characteristics of the fault item; A power battery fault analysis module, used to determine the power battery fault evolution path and the importance of fault items based on the power battery fault heterogeneous graph; A fault item analysis report determination module is used to determine a fault item analysis report based on the power battery fault evolution path and the importance of the fault item. 9. A computer device, comprising a memory and a processor, the memory stores a computer program, characterized in that when the processor executes the computer program, the method of any one of claims 1 to 7 is implemented. step. 10. A computer-readable storage medium with a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.