Method, device, equipment and medium for detecting causal relationship of battery production indicators

Abstract

The present application relates to the field of battery technology, and in particular to a method, device, equipment and medium for detecting the causal relationship of battery production indicators. In the present application, a causal relationship map is constructed based on the sequence between each process in battery production and the causal effect between each production indicator, and the optimal causal chain from each source node to the target node in the causal relationship map is determined; all optimal causal chains are filtered to obtain a target causal chain, and a target relationship map is obtained based on the target causal chain, so that according to the causal relationship in the target relationship map, the production indicators when problems occur in the battery cell capacity can be quickly located, so as to make timely intervention adjustments, thereby avoiding production fluctuations and quality problems, and ensuring product consistency and high-quality output.

Description

Technical Field

[0001] The present application relates to the field of battery technology, and in particular to a method, device, equipment and medium for detecting the causal relationship of battery production indicators. Background Art

[0002] With the rapid development of the new energy industry, batteries, as core energy storage components, are widely used in new energy vehicles, energy storage equipment, consumer electronics and other fields. Continuous innovation in battery technology has not only promoted industry progress, but also placed higher demands on production efficiency, quality control and product consistency. However, the battery production process is complex, involving multiple variables such as material ratios, process parameters and environmental conditions. These factors influence each other and ultimately determine the capacity of the battery cell. Currently, correlation analysis methods based on machine learning can detect the correlation between indicators and capacity, but correlation analysis methods have difficulty in effectively capturing the interactive relationship between indicators, and therefore cannot accurately locate the root cause of the problem. Therefore, how to determine the interactive relationship between indicators to accurately locate the root cause of abnormal battery cell capacity has become an urgent problem that needs to be solved. Summary of the Invention

[0003] In view of this, the embodiments of the present application provide a method, device, equipment and medium for detecting the causal relationship of battery production indicators to solve the problem of being unable to accurately locate the root cause of abnormal battery cell capacity.

[0004] In a first aspect, an embodiment of the present application provides a method for detecting causal relationships of battery production indicators, the method comprising:

[0005] Determine the sequence of each process in battery production and the production indicators that affect the cell capacity in each process;

[0006] Determine an initial causal relationship between the production indicators according to the sequence, and construct a causal relationship graph based on the initial causal relationship, with the production indicators as source nodes, the cell capacity as target nodes, and the results of the causal relationship as a guide, wherein the causal relationship graph includes a node set and an edge set;

[0007] Obtaining N sets of data, and calculating, based on the N sets of data, a causal effect of each edge in the causal relationship graph to obtain a causal effect value of the corresponding edge, wherein each set of data includes an indicator value of a production indicator and a capacity value of a battery cell capacity in the causal relationship graph, and N is an integer greater than zero;

[0008] Determine the optimal causal chain from each source node to the target node based on the causal effect value of each edge;

[0009] All optimal causal chains are filtered to obtain a target causal chain, and a target relationship graph is obtained based on the target causal chain.

[0010] In a second aspect, an embodiment of the present application provides a device for detecting causal relationships of battery production indicators, the device comprising:

[0011] The first determination module is used to determine the sequence between each process in battery production and the production indicators that affect the battery cell capacity in each process;

[0012] a construction module, configured to determine an initial causal relationship between the production indicators according to the sequence, and construct a causal relationship graph based on the initial causal relationship, with the production indicators as source nodes, the cell capacity as target nodes, and the results of the causal relationship as a guide, wherein the causal relationship graph includes a node set and an edge set;

[0013] a calculation module, configured to obtain N sets of data, and calculate, based on the N sets of data, a causal effect of each edge in the causal relationship graph to obtain a causal effect value of the corresponding edge, wherein each set of data includes an indicator value of a production indicator and a capacity value of a battery cell capacity in the causal relationship graph, and N is an integer greater than zero;

[0014] A second determining module is configured to determine an optimal causal chain from each node in the node set to the target node based on the causal effect value of each edge;

[0015] The filtering module is used to filter all optimal causal chains to obtain a target causal chain, and obtain a target relationship graph based on the target causal chain.

[0016] In a third aspect, an embodiment of the present application provides a computer device, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the detection method described in the first aspect when executing the computer program.

[0017] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the detection method as described in the first aspect is implemented.

[0018] Compared with the prior art, the present invention has the following beneficial effects:

[0019] In this application, a causal relationship map is constructed based on the sequence between each process in battery production and the causal effect between each production indicator, and the optimal causal chain from each source node to the target node in the causal relationship map is determined; all optimal causal chains are filtered to obtain the target causal chain, and based on the target causal chain, a target relationship map is obtained, so that according to the causal relationship in the target relationship map, the production indicators when problems occur in the battery cell capacity can be quickly located, so as to make timely intervention and adjustments, thereby avoiding production fluctuations and quality problems, and ensuring product consistency and high-quality output. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0021] Figure 1 This is an application environment of a method for detecting causal relationships of battery production indicators provided in one embodiment of the present application;

[0022] Figure 2 This is a flow chart of a method for detecting causal relationships of battery production indicators provided in one embodiment of the present application;

[0023] Figure 3 This is a schematic diagram of a cause-effect relationship diagram provided by an embodiment of the present application;

[0024] Figure 4 This is a schematic structural diagram of a device for detecting causal relationships of battery production indicators provided by one embodiment of the present application;

[0025] Figure 5 It is a structural diagram of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0027] In the following description, specific details such as specific system structures and techniques are provided for purposes of illustration rather than limitation to facilitate a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obscuring the description of the present application with unnecessary detail.

[0028] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0029] It will also be understood that the term "and / or" used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0030] As used in this specification and the appended claims, the term "if" can be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.

[0031] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0032] References to "one embodiment" or "some embodiments" in this specification mean that a particular feature, structure, or characteristic described in conjunction with that embodiment is included in one or more embodiments of the present application. Thus, phrases such as "in one embodiment," "in some embodiments," "in other embodiments," and "in other embodiments" appearing in various places in this specification do not necessarily refer to the same embodiment, but rather mean "one or more but not all embodiments," unless otherwise specifically emphasized. The terms "including," "comprising," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0033] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, techniques, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results.

[0034] Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interaction systems, and mechatronics. AI software technologies primarily encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0035] It should be understood that the size of the serial numbers of the steps in the following embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0036] In order to illustrate the technical solution of the present application, specific embodiments are provided below.

[0037] An embodiment of the present application provides a method for detecting the causal relationship of battery production indicators, which can be applied in the following situations: Figure 1 In an application environment, clients communicate with servers. Clients include, but are not limited to, PDAs, desktop computers, laptops, ultra-mobile personal computers (UMPCs), netbooks, cloud computing devices, personal digital assistants (PDAs), and other computer devices. Servers can be standalone servers or cloud servers that provide basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0038] See also Figure 2 , is a flow chart of a method for detecting the causal relationship of battery production indicators provided by an embodiment of the present application. The above-mentioned method for detecting the causal relationship of battery production indicators can be applied to Figure 1 The server in Figure 2 As shown, the method for detecting the causal relationship of battery production indicators may include the following steps.

[0039] S201: Determine the sequence of each process in battery production and the production indicators that affect the battery cell capacity in each process.

[0040] In step S201 , the sequence between each process in battery production is the sequence of the processes performed in sequence in battery production, and the production index is the parameter that affects the capacity of the battery cell.

[0041] In this embodiment, the battery production process involves multiple complex processes, such as coating process, cutting process and capacity conversion process, etc. Each process contains multiple indicators that affect the capacity of the battery cell, such as thickness index, uniformity index, viscosity index, drying temperature index, etc. in the coating process, electrode size index, edge flatness index, etc. in the cutting process, voltage peak index, time-frequency peak index, frequency band power index, etc. in the capacity conversion process. There is not only a correlation between these indicators, but also a causal relationship. For example, the parameter fluctuation of a certain process may directly affect the quality of the subsequent process, and thus ultimately have a significant impact on the capacity of the battery cell. According to the sequence between each process, the sequence of production indicators can be determined, and then the causal relationship of production indicators can be determined according to the sequence of production indicators and the causal theory that the cause occurs before the result.

[0042] It should be noted that when determining the production indicators that affect the capacity of the battery cells in each process, the data in each process may be summary data or curve data, depending on the characteristics of each process. Summary data is discrete data, while curve data is continuous data. Therefore, the production indicators may be the original features in the summary data, such as the thickness index, uniformity index, viscosity index, etc. in the coating process. These production indicators are features directly obtained from the original data. Production indicators may also be high-dimensional features extracted from the curve data, such as the voltage peak index, time-frequency peak index, and frequency band power index in the capacity conversion process. These indicators are features extracted from the curve data, where the curve data is the charging voltage of the corresponding time series.

[0043] In this embodiment, the order of each process in battery production and the production indicators that affect the cell capacity in each process are determined, so that a causal relationship exists between the production indicators in the previous process and the production indicators in the subsequent process. The causal relationship between the production indicators in the previous process and the production indicators in the subsequent process can be determined based on the order of each process, that is, the production indicators in the previous process are the cause and the production indicators in the subsequent process are the result. This preliminarily establishes the possible causal relationship between the production indicators.

[0044] S202: Determine the initial causal relationship between the production indicators according to the sequence, and construct a causal relationship graph based on the initial causal relationship, with the production indicators as the source nodes, the cell capacity as the target node, and the results in the causal relationship as the guide. The causal relationship graph includes a node set and an edge set.

[0045] In step S202, the initial causal relationship between production indicators is determined based on the order of precedence. This means that the production indicators in the previous process are initially determined to be the cause of the production indicators in the subsequent process. Based on this initial causal relationship, a causal relationship graph is constructed, with the production indicators as the source nodes, the cell capacity as the target node, and the results of the causal relationship as the guide. This is a directed graph, with the direction pointing to the target node. The causal relationship graph contains a set of nodes and a set of edges in the graph.

[0046] In this embodiment, the initial causal relationship between production indicators is determined based on the order of precedence, wherein the production indicator in the previous process is determined to be the cause of the production indicator in the subsequent process, and there is no causal relationship between production indicators in the same process. Based on the initial causal relationship, a causal relationship graph is constructed with the production indicator as the source node, the battery cell capacity as the target node, and the result of the causal relationship as the guide. The source node is the corresponding production indicator, and the target node is the battery performance affected by the production indicator, that is, the battery cell capacity. The causal relationship graph is a directed graph, and the direction in the graph points from the cause production indicator to the result production indicator, and finally to the target node.

[0047] See also Figure 3, is a schematic diagram of a cause-effect relationship map provided by an embodiment of the present application. The negative electrode tab welding, mylar-wrapping and shelling and formation processes in the battery production process, wherein the order of the processes is negative electrode tab welding, mylar-wrapping and shelling, formation, the negative electrode tab welding process includes production indicators such as negative electrode tab welding width, negative electrode tab welding length, and negative electrode tab welding area, the mylar-wrapping and shelling process includes production indicators such as mylar-wrapping and shelling time, mylar-wrapping and shelling temperature, and the formation process includes production indicators such as voltage peak, time-frequency peak, and frequency band power. When constructing the cause-effect relationship map, it is considered that the negative electrode tab welding width, negative electrode tab welding length, and negative electrode tab welding area in the negative electrode tab welding process are the same. Production indicators such as the cathode tab welding area are the cause of production indicators in the mylar wrapping and shelling processes and the formation process. The direction of the production indicators in the negative tab welding process, such as the negative tab welding width, negative tab welding length, and negative tab welding area, is directed towards production indicators in the mylar wrapping and shelling processes and the formation process. For example, the negative tab welding width is sequentially directed towards production indicators in the mylar wrapping and shelling processes and the formation process, the negative tab welding length is sequentially directed towards production indicators in the mylar wrapping and shelling processes and the formation process, and the negative tab welding area is sequentially directed towards production indicators in the mylar wrapping and shelling processes and the formation process. Production indicators such as the mylar wrapping and shelling time and mylar wrapping and shelling temperature in the mylar wrapping and shelling process are both the result of production indicators in the negative tab welding process and the cause of production indicators in the formation process. The direction of the production indicators in the mylar wrapping and shelling process, such as the mylar wrapping and shelling time and mylar wrapping and shelling temperature, are directed towards production indicators in the formation process. If the formation process is the last step in the battery production process, the production indicators in the formation process point to the battery cell capacity. The source nodes are the negative electrode tab welding width, negative electrode tab welding length, negative electrode tab welding area, mylar packing-to-shell time, mylar packing-to-shell temperature, voltage peak, time-frequency peak, and frequency band power, and the battery cell capacity is the target node.

[0048] It should be noted that the causal relationship graph is a directed acyclic graph, which is expressed as G = (V, E) where V is the node set and E is the edge set. ,in, is the source node set, i.e., the production index, is the target node, i.e. the cell capacity, , It is the first production indicator in the first process. It is the second production indicator in the first process. is the nth production indicator in the mth process. ,in, , the edge between the sth production index in the i-th process and the tth production index in the j-th process, that is, the edge between the source nodes. i is less than j, , , the edge between the production index and the cell capacity, that is, the edge between the source node and the target node.

[0049] In this embodiment, a cause-and-effect relationship map is constructed based on the sequence between each process, so that the cause-and-effect relationship map meets the causal requirement of cause first and result later, ensuring that each process from raw material input to finished battery cells can be traced and analyzed, so that when the battery cell capacity is abnormal, the root cause of the problem can be quickly deduced based on the cause-and-effect relationship map.

[0050] S203: Obtain N sets of data, and calculate the causal effect of each edge in the causal relationship graph based on the N sets of data to obtain the causal effect value of the corresponding edge, where each set of data includes the index value of the production index and the capacity value of the battery cell capacity in the causal relationship graph, and N is an integer greater than zero.

[0051] In step S203, N groups of data are obtained, and the causal effect of each edge in the causal relationship graph is calculated based on the N groups of data, wherein each group of data includes the index value of the production index and the capacity value of the battery cell capacity in the causal relationship graph, and the causal effect represents the strength of the causal correlation. The larger the causal effect value, the stronger the causal correlation is considered, and the smaller the causal effect value, the weaker the causal correlation is considered.

[0052] In this embodiment, N groups of data are obtained, each group of data including the index value of the production indicator and the capacity value of the battery cell capacity in the causal relationship graph, that is, based on the index value of the production indicator and the capacity value of the battery cell capacity in each group, the causal effect of each edge in the causal relationship graph is calculated to obtain the causal effect value of the corresponding edge. In particular, when calculating the causal effect of each edge in the causal relationship graph, the corresponding mapping relationship between the nodes is determined based on the N groups of data of the production indicator in the corresponding edge, that is, the causal effect of each edge in the causal relationship graph is calculated.

[0053] It should be noted that the causal effect of each edge in the causal relationship graph can be calculated using the instrumental variable method, the double difference method, etc., which is not limited in this embodiment.

[0054] In this embodiment, based on N sets of data, the causal effect of each edge in the causal relationship graph is calculated to obtain the causal effect value of the corresponding edge. This allows the intensity and direction of the causal effect in each edge to be determined based on the magnitude and sign of the causal effect value. This facilitates determining the intensity of adjustments to production indicators when abnormal cell capacity occurs.

[0055] Optionally, after obtaining N sets of data, the following steps are further included:

[0056] Determine causal pairs based on the edge set in the causal graph;

[0057] Calculate the approximate functional dependency in each causal pair based on the index values ​​of N groups of production indicators and the capacity values ​​of battery cells;

[0058] According to the approximate functional dependency, the causal relationship graph is pruned to obtain a pruned causal relationship graph, and the pruned causal relationship graph is used as the final causal relationship graph.

[0059] In this embodiment, it should be noted that after constructing the causal relationship graph, it is pruned, i.e., edges and isolated points that clearly lack causal relationships are removed. Pruning methods are not limited to those based on expert graphs, those based on approximate functional dependencies, and those based on equivalent replacement of equivalence classes. This embodiment does not impose any limitations on this.

[0060] In this embodiment, a causal pair is determined based on the edge set in the causal relationship graph, wherein the causal pair is the two source nodes in each edge in the edge set, that is, the two production indicators connected by the edge in the causal relationship graph constitute a causal pair. According to the indicator values ​​of the N groups of production indicators and the capacity value of the battery cell capacity, the approximate functional dependency in each causal pair is calculated. According to the approximate functional dependency, it is determined whether the two production indicators need to be merged. If the approximate functional dependency is less than the preset threshold, it is considered that the production indicator in the subsequent process in the causal pair does not have an approximate functional dependency on the production indicator in the prior process. If the approximate functional dependency is not less than the preset threshold, it is considered that the production indicator in the subsequent process has an approximate functional dependency on the production indicator in the prior process. The two production indicators in the corresponding causal pair can be merged to obtain a pruned causal relationship graph.

[0061] For example, causal ,Right now , there is an approximate functional dependency, causal pair ,Right now , there is also an approximate functional dependency, which is considered is equivalent to Calculate as the same node. When i≤j, keep .

[0062] In this embodiment, the causal relationship graph is pruned to remove redundant or obviously non-existent causal relationships, so as to reduce the amount of computation required for subsequent causal relationship calculations.

[0063] Optionally, based on N sets of data, the causal effect of each edge in the causal relationship graph is calculated to obtain the causal effect value of the corresponding edge, including:

[0064] For any edge, determine the first node and the second node of the edge;

[0065] Calculate a mapping relationship between the first node and the second node based on the N groups of data of the first node and the second node;

[0066] According to the mapping relationship, the causal effect value of the edge is determined.

[0067] In this embodiment, a regression adjustment method is used to calculate the causal effect, with the first node in the corresponding edge as the independent variable and the second node in the corresponding edge as the dependent variable, where the first node is the cause node and the second node is the effect node. Based on N sets of data from the first and second nodes, a mapping relationship between the first and second nodes is calculated. When calculating this mapping relationship, a mathematical model is constructed between the dependent and independent variables, and other variables that may affect the dependent variable are included in the model as control variables to estimate the independent effect of a specific independent variable on the dependent variable. Based on the N sets of data from the corresponding edge nodes, a regression model is estimated to obtain the regression coefficient of the independent variable on the dependent variable. Model adjustments (such as adding interaction terms and nonlinear terms) are then used to optimize the model fit. Based on the regression coefficients, the direction and strength of the independent variable's influence on the dependent variable, i.e., the causal effect value, is determined.

[0068] It should be noted that the causal effect value can be positive or negative. If it is positive, it is assumed that the effect in the corresponding edge is positively correlated with the cause. If it is negative, it is assumed that the effect in the corresponding edge is negatively correlated with the cause. The absolute value of the causal effect value is the strength of the influence of the cause in the corresponding edge on the effect.

[0069] S204: Determine the optimal causal chain from each source node to the target node based on the causal effect value of each edge.

[0070] In step S204, the causal chain is the path from the source node to the target node in the causal relationship graph. The optimal causal chain is the path with the strongest causal effect among the paths from the source node to the target node.

[0071] In this embodiment, in the causal relationship graph, there are multiple paths from a source node to a target node, such as The path from the start to the destination node is recorded as . Determine one path from multiple paths as the optimal causal chain.

[0072] In this embodiment, the causal effect in each path is calculated, and the path with the maximum causal effect value is determined as the optimal causal chain. The formula for calculating the causal effect value of each path is as follows:

[0073]

[0074] in, From the source node The causal effect value of the path from the start to the target node, for Node Pair The causal effect value of the node, From the source node Start from the node that reaches the target node, path P, path p .

[0075] In this embodiment, according to Size, select The path with the maximum value is the optimal causal chain.

[0076] In this embodiment, the optimal causal chain from each source node to the target node is determined based on the causal effect value of each edge, so as to simplify the causal relationship graph based on the optimal causal chain. This facilitates the subsequent abnormal analysis of the battery cell capacity based on the simplified causal relationship graph, and accurately locates one of the production indicators where the root cause of the problem is located, avoiding the multiple production indicators that may be determined based on multiple causal chains, so that the root cause of the abnormality can be determined in a targeted manner.

[0077] Optionally, based on the causal effect value of each edge, determining the optimal causal chain from each source node to the target node in the node set includes:

[0078] According to the causal effect value of each edge, the weight value of the corresponding edge is calculated;

[0079] According to the weight value of each edge, the optimal causal chain from each source node to the target node is determined.

[0080] In this embodiment, the weight value of the corresponding edge is calculated based on the causal effect value of each edge, where the calculation formula of the weight value is as follows:

[0081]

[0082] in, For path p, Node Pair The weight value of the edge where the causal effect of the node is located, From the source node Start from the node and reach the node included in the path P of the target node. for Node Pair The causal effect value of the node.

[0083] According to the weight value of each edge, the optimal causal chain from each source node to the target node is determined. In this embodiment, the calculation of the optimal causal chain from each source node to the target node is converted into the calculation of the shortest path, and the causal effect value of each path is calculated. Taking the logarithm, we get ,in, From the source node The causal effect value of the path from the start to the target node, for Node Pair The causal effect value of the node, From the source node Start from the node and reach the node included in the path P of the target node. In this way, the problem is transformed into a classic shortest path problem. According to the weight value of each edge, the source node is calculated. The shortest path from the source node to the target node is determined as the optimal causal chain. In this embodiment, the Bellman-Ford Algorithm is used to calculate the source node The shortest path from the node to the destination node can also be calculated using other algorithms, which are not limited in this embodiment.

[0084] In this embodiment, since the causal effect value of the edge includes positive and negative values, the corresponding weight value also includes positive and negative values. The problem of determining the optimal causal chain from each source node to the target node in the node set is transformed into the problem of calculating the shortest path from the source node to the target node. This can avoid the optimal causal chain including the corresponding negative weight edge, and prevent the causal relationship from forming a loop and causing deadlock.

[0085] S205: Filter all optimal causal chains to obtain a target causal chain, and obtain a target relationship graph based on the target causal chain.

[0086] In step S205, after determining the optimal causal chain from each source node to the target node, all optimal causal chains are filtered to filter out contradictory causal chains to obtain the target causal chain, and the target relationship graph is obtained based on the target causal chain.

[0087] In this embodiment, the optimal causal chains with contradictions are detected, and the optimal causal chains with contradictions are filtered to obtain a target causal chain. That is, among the optimal causal chains with contradictions, one of the optimal causal chains is retained as the target causal chain.

[0088] After filtering all the optimal causal chains, the target causal chains form the target relationship graph. In the target relationship graph, each causal chain from the source node to the target node contains one, and the direction of the causal effect from the same source node to the target node in different causal chains is equal.

[0089] In this embodiment, all optimal causal chains are filtered to obtain a target causal chain. Based on the target causal chain, a target relationship map is obtained, and causal chains with contradictory relationships in the causal relationship map are removed to ensure the accuracy of the target relationship map.

[0090] Optionally, all optimal causal chains are filtered to obtain a target causal chain, including:

[0091] When there is a contradiction between any two optimal causal chains, all optimal causal chains are filtered to obtain filtered causal chains;

[0092] The filtered causal chain is screened according to the filtered causal chain and the causal effect value of each edge in the filtered causal chain to obtain the target causal chain.

[0093] In this embodiment, to determine whether any two optimal causal chains are inconsistent, the first step is to determine whether the two optimal causal chains contain the same two nodes, where a node is a source node and a target node. If the two optimal causal chains contain the same two nodes, the second step is to determine whether the direction of the causal effect of the same node on the other node in the two optimal causal chains is the same. If the direction of the causal effect of the same node on the other node in the two optimal causal chains is the same, then the two optimal causal chains are determined to be inconsistent. If the direction of the causal effect of the same node on the other node in the two optimal causal chains is different, then the two optimal causal chains are determined to be inconsistent. For example, the first optimal causal chain contains source nodes A, B, C, and F; the second optimal causal chain contains source nodes B, D, E, and F; and the third optimal causal chain contains source nodes C, M, N, and F. The first and second optimal causal chains contain the same two nodes, B and F. If the causal effect of source node B on source node F in the first optimal causal chain is positive, and the causal effect of source node B on source node F in the second optimal causal chain is negative, then the first and second optimal causal chains are considered to be inconsistent. If any two optimal causal chains are inconsistent, all optimal causal chains are filtered, eliminating the inconsistent optimal causal chains until the remaining optimal causal chains are consistent.

[0094] It should be noted that when determining the direction of the causal effect of source node B on source node F, the direction of the causal effect of each source node on source node F in the corresponding optimal causal chain can be determined. For example, if in the first optimal causal chain, the causal effect of source node B on source node C is positive, and the direction of the causal effect of source node C on source node F is also positive, then the direction of the causal effect of source node B on source node F is determined to be positive. If in the second optimal causal chain, the direction of the causal effect of source node B on source node D is positive, the direction of the causal effect of source node D on source node E is negative, and the direction of the causal effect of source node E on source node F is positive, then the direction of the causal effect of source node B on source node F is determined to be negative.

[0095] It should be noted that when filtering out the contradictory optimal causal chains, the optimal causal chain with the smallest causal effect value can be filtered out, and the optimal causal chain with the largest causal effect value can be retained.

[0096] The filtered causal chain is screened according to the filtered causal chain and the causal effect value of each edge in the filtered causal chain to obtain the target causal chain.

[0097] In this embodiment, when screening the filtered causal chains based on the causal effect values ​​of the filtered causal chains and each edge in the filtered causal chains, the causal effect of the filtered causal chains is calculated using the formula for calculating the causal effect of a path, thereby obtaining the corresponding filtered causal chain causal effect value. All filtered causal chain causal effect values ​​are sorted, and the top K filtered causal chains with the highest causal effect values ​​are selected as target causal chains.

[0098] In this embodiment, when any two optimal causal chains conflict, all optimal causal chains are filtered to eliminate conflicting ones, ensuring the logical consistency of causal inference and enabling clearer identification of causal paths. The filtered causal chains are then screened based on the causal effect values ​​of the filtered causal chains and each edge in the filtered causal chains, retaining the optimal causal chains with the largest causal effect values. This ensures that the nodes in the target causal chain maintain a strong causal relationship, and the target causal chain reduces a large number of causal links, reducing the amount of subsequent analysis of the root cause of the anomaly.

[0099] Optionally, all optimal causal chains are filtered to obtain filtered causal chains, including:

[0100] For any optimal causal chain, determine an optimal causal chain set that does not conflict with the optimal causal chain;

[0101] Taking each optimal causal chain as a node, construct multiple complete subgraphs based on the optimal causal chain set corresponding to each optimal causal chain;

[0102] The complete subgraph containing the most source nodes in the complete subgraph is selected as the maximum complete subgraph, and the optimal causal chain contained in the maximum complete subgraph is determined as the filtered causal chain.

[0103] In this embodiment, for any optimal causal chain, an optimal causal chain set that does not conflict with the optimal causal chain is determined. A complete subgraph is constructed based on the optimal causal chain set corresponding to each optimal causal chain. When constructing the target complete subgraph, each optimal causal chain is used as a node, and multiple complete subgraphs are constructed based on the optimal causal chain set corresponding to each optimal causal chain. That is, no two optimal causal chains in the complete subgraph are in conflict. The complete subgraph containing the most source nodes is selected as the maximum complete subgraph, and the optimal causal chain contained in the maximum complete subgraph is determined as the filtered causal chain.

[0104] It should be noted that, when constructing a complete subgraph, the Bron-Kerbosch algorithm, enumeration method, etc. may be used, but this embodiment does not limit this.

[0105] In this embodiment, when filtering all optimal causal chains, the corresponding optimal causal chains are retained by constructing a complete subgraph, and the remaining optimal causal chains are filtered out. This can obtain a causal module with strong correlation in the graph to assist causal inference.

[0106] In this application, a causal relationship map is constructed based on the sequence between each process in battery production and the causal effect between each production indicator, and the optimal causal chain from each source node to the target node in the causal relationship map is determined; all optimal causal chains are filtered to obtain the target causal chain, and based on the target causal chain, a target relationship map is obtained, so that according to the causal relationship in the target relationship map, the production indicators when problems occur in the battery cell capacity can be quickly located, so as to make timely intervention and adjustments, thereby avoiding production fluctuations and quality problems, and ensuring product consistency and high-quality output.

[0107] See also Figure 4 , is a schematic diagram of the structure of a battery production index causal relationship detection device provided by an embodiment of the present application. The battery production index causal relationship detection device corresponds to the battery production index causal relationship detection method in the above embodiment. For details, please refer to Figure 2 For the convenience of explanation, only the parts related to this embodiment are shown. Figure 4 The detection device 40 includes: a first determination module 41 , a construction module 42 , a calculation module 43 , a second determination module 44 , and a filtering module 45 .

[0108] The first determining module 41 is used to determine the sequence of each process in battery production and the production index that affects the battery cell capacity in each process.

[0109] Construction module 42 is used to determine the initial causal relationship between production indicators according to the sequence, and to construct a causal relationship graph based on the initial causal relationship, with the production indicators as the source nodes, the battery cell capacity as the target node, and the results in the causal relationship as the guide. The causal relationship graph includes a node set and an edge set.

[0110] The calculation module 43 is used to obtain N groups of data, calculate the causal effect of each edge in the causal relationship graph based on the N groups of data, and obtain the causal effect value of the corresponding edge. Each group of data includes the indicator value of the production indicator and the capacity value of the battery cell capacity in the causal relationship graph, and N is an integer greater than zero.

[0111] The second determining module 44 is configured to determine the optimal causal chain from each node in the node set to the target node according to the causal effect value of each edge.

[0112] The filtering module 45 is used to filter all the optimal causal chains to obtain the target causal chain, and obtain the target relationship graph based on the target causal chain.

[0113] Optionally, the second determining module includes:

[0114] The calculation unit is used to calculate the weight value of the corresponding edge according to the causal effect value of each edge.

[0115] The determination unit is used to determine the optimal causal chain from each node in the node set to the target node according to the causal effect value of each edge and the weight value of the corresponding edge.

[0116] Optionally, the calculation module 43 includes:

[0117] According to the edge set in the causal graph, the causal pairs are determined.

[0118] The second calculation unit is used to calculate the approximate functional dependency in each causal pair according to the indicator values ​​of the N groups of production indicators and the capacity value of the battery cell capacity.

[0119] The pruning unit is used to prune the causal relationship graph according to the approximate functional dependency to obtain a pruned causal relationship graph, and use the pruned causal relationship graph as the final causal relationship graph.

[0120] Optionally, the calculation module 43 includes:

[0121] The second determining unit is configured to determine, for any edge, a first node and a second node of the edge.

[0122] The third calculation unit is configured to calculate a mapping relationship between the first node and the second node according to the N groups of data of the first node and the second node.

[0123] The third determining unit is used to determine the causal effect value of the edge according to the mapping relationship.

[0124] Optionally, the filtering module 45 includes:

[0125] The fourth determining unit is used to determine the causal chain pairs containing the same source node.

[0126] The judgment unit is used to filter the causal chain pair to obtain the target causal chain if the directions of the causal effects of the same source node on the target node are opposite in the causal chain pair.

[0127] Optionally, the judgment unit includes:

[0128] The calculation subunit is used to calculate the causal effect of each optimal causal chain in the causal chain pair according to the causal effect value of each edge, and obtain the chain causal effect value of the corresponding optimal causal chain.

[0129] The subunit is obtained, which is used to filter the optimal causal chain with the smallest chain causal effect value to obtain the target causal chain.

[0130] It should be noted that the information interaction, execution process and other contents between the above modules are based on the same concept as the method embodiment of this application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0131] Figure 5 This is a schematic diagram of the structure of a computer device provided in one embodiment of the present application. Figure 5 As shown, the computer device of this embodiment includes: at least one processor ( Figure 5 Only one is shown), a memory, and a computer program stored in the memory and executable on at least one processor, wherein when the processor executes the computer program, the steps of the above-mentioned method for detecting the causal relationship of any of the battery production indicators are implemented.

[0132] The computer device may include, but is not limited to, a processor and a memory. It will be understood by those skilled in the art that Figure 5 This is merely an example of a computer device and does not constitute a limitation on the computer device. The computer device may include more or fewer components than shown in the figure, or a combination of certain components, or different components, for example, it may also include a network interface, etc.

[0133] The processor may be a CPU, other general-purpose processors, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0134] Memory includes readable storage media, internal memory, and the like. Internal memory can be the internal memory of a computer device, providing an environment for the operation of the operating system and computer-readable instructions stored in the readable storage medium. The readable storage medium can be the computer device's hard drive. In other embodiments, it can also be an external storage device, such as a plug-in hard drive, a Smart Media Card (SMC), a Secure Digital (SD) card, or a flash memory card. Furthermore, memory can include both the computer device's internal storage unit and external storage devices. Memory is used to store the operating system, application programs, boot loaders, data, and other programs, such as the program code of computer programs. Memory can also be used to temporarily store data that has been output or is about to be output.

[0135] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned device can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here. If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the process steps in the above-described method embodiments by instructing the relevant hardware through a computer program. The computer program may be stored in a computer-readable storage medium. When executed by a processor, the computer program implements the steps of the above-described method embodiments. The computer program includes computer program code, which may be in source code form, object code form, executable file, or some intermediate form. Computer-readable media may include at least: any entity or device capable of carrying computer program code, recording media, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunications signals, and software distribution media. Examples include USB flash drives, removable hard drives, magnetic disks, or optical disks. In some jurisdictions, based on legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunications signals.

[0136] The present application implements all or part of the processes in the above-mentioned embodiment method, and can also be completed through a computer program product. When the computer program product runs on a computer device, the computer device can implement the steps in the above-mentioned method embodiment when executing it.

[0137] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0138] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0139] In the embodiments provided in this application, it should be understood that the disclosed apparatus / computer equipment and methods can be implemented in other ways. For example, the apparatus / computer equipment embodiments described above are merely schematic. For example, the division of modules or units is merely a logical function division. 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. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of the apparatus or unit, which can be electrical, mechanical or other forms.

[0140] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0141] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A method for detecting causal relationships of battery production indicators, characterized in that: The detection method comprises: Determine the sequence of each process in battery production and the production indicators that affect the cell capacity in each process; Determine an initial causal relationship between the production indicators according to the sequence, and construct a causal relationship graph based on the initial causal relationship, with the production indicators as source nodes, the cell capacity as target nodes, and the results of the causal relationship as a guide, wherein the causal relationship graph includes a node set and an edge set; Obtaining N sets of data, and calculating, based on the N sets of data, a causal effect of each edge in the causal relationship graph to obtain a causal effect value of the corresponding edge, wherein each set of data includes an indicator value of a production indicator and a capacity value of a battery cell capacity in the causal relationship graph, and N is an integer greater than zero; Determine the optimal causal chain from each source node to the target node based on the causal effect value of each edge; Filter all optimal causal chains to obtain a target causal chain, and obtain a target relationship graph based on the target causal chain; The above method filters all the optimal causal chains to obtain the target causal chain, including: When there is a contradiction between any two optimal causal chains, all optimal causal chains are filtered to obtain filtered causal chains; Filtering the filtered causal chains according to the filtered causal chains and the causal effect value of each edge in the filtered causal chains to obtain the target causal chain; The filtering of all optimal causal chains to obtain filtered causal chains includes: For any optimal causal chain, determine an optimal causal chain set that does not conflict with the optimal causal chain; Taking each optimal causal chain as a node, construct multiple complete subgraphs based on the optimal causal chain set corresponding to each optimal causal chain; A complete subgraph containing the most source nodes in the complete subgraph is selected as the maximum complete subgraph, and the optimal causal chain contained in the maximum complete subgraph is determined as the filtered causal chain.

2. The detection method according to claim 1, wherein Determining the optimal causal chain from each source node to the target node based on the causal effect value of each edge includes: According to the causal effect value of each edge, the weight value of the corresponding edge is calculated; According to the weight value of each edge, the optimal causal chain from each source node to the target node is determined.

3. The detection method according to claim 1, wherein After obtaining N sets of data, the method further includes: Determine causal pairs based on the edge set in the causal graph; Calculate the approximate functional dependency in each causal pair based on the index values ​​of N groups of production indicators and the capacity values ​​of battery cells; The causal relationship graph is pruned according to the approximate functional dependency to obtain a pruned causal relationship graph, and the pruned causal relationship graph is used as the final causal relationship graph.

4. The detection method according to claim 1, wherein The step of calculating the causal effect of each edge in the causal relationship graph based on the N sets of data to obtain the causal effect value of the corresponding edge includes: For any edge, determine the first node and the second node of the edge; Calculate a mapping relationship between the first node and the second node based on the N groups of data of the first node and the second node; According to the mapping relationship, the causal effect value of the edge is determined.

5. A device for detecting causal relationships of battery production indicators, characterized in that: The detection device comprises: The first determination module is used to determine the sequence between each process in battery production and the production indicators that affect the battery cell capacity in each process; a construction module, configured to determine an initial causal relationship between the production indicators according to the sequence, and construct a causal relationship graph based on the initial causal relationship, with the production indicators as source nodes, the cell capacity as target nodes, and the results of the causal relationship as a guide, wherein the causal relationship graph includes a node set and an edge set; a calculation module, configured to obtain N sets of data, and calculate, based on the N sets of data, a causal effect of each edge in the causal relationship graph to obtain a causal effect value of the corresponding edge, wherein each set of data includes an indicator value of a production indicator and a capacity value of a battery cell capacity in the causal relationship graph, and N is an integer greater than zero; A second determining module is configured to determine an optimal causal chain from each node in the node set to the target node based on the causal effect value of each edge; A filtering module is used to filter all optimal causal chains to obtain a target causal chain, and obtain a target relationship graph based on the target causal chain; The above method filters all the optimal causal chains to obtain the target causal chain, including: When there is a contradiction between any two optimal causal chains, all optimal causal chains are filtered to obtain filtered causal chains; Filtering the filtered causal chains according to the filtered causal chains and the causal effect value of each edge in the filtered causal chains to obtain the target causal chain; The filtering of all optimal causal chains to obtain filtered causal chains includes: For any optimal causal chain, determine an optimal causal chain set that does not conflict with the optimal causal chain; Taking each optimal causal chain as a node, construct multiple complete subgraphs based on the optimal causal chain set corresponding to each optimal causal chain; A complete subgraph containing the most source nodes in the complete subgraph is selected as the maximum complete subgraph, and the optimal causal chain contained in the maximum complete subgraph is determined as the filtered causal chain.

6. The detection device according to claim 5, characterized in that The second determining module includes: A calculation unit is used to calculate the weight value of the corresponding edge according to the causal effect value of each edge; A determination unit is used to determine the optimal causal chain from each node in the node set to the target node according to the causal effect value of each edge and the weight value of the corresponding edge.

7. A computer device, characterized in that: The computer device includes a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the detection method according to any one of claims 1 to 4 is implemented.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the detection method according to any one of claims 1 to 4 is implemented.

Images