A safety assessment method and device for electrochemical energy storage system

Abstract

The present invention provides a method and device for safety assessment of an electrochemical energy storage system, the method comprising: obtaining data corresponding to each factor node in a pre-established safety status influencing factor structure tree during the operation of the electrochemical energy storage system, the factor nodes including subjective factor nodes and objective factor nodes; obtaining evaluation results corresponding to each subjective factor node; determining the evaluation results corresponding to each objective factor node according to the threshold interval to which the data corresponding to each objective factor node belongs; establishing a fuzzy evaluation membership matrix in accordance with a comprehensive fuzzy evaluation method, combining the safety status influencing factor structure tree, the evaluation results of each factor node, and the weight of each factor node; calculating the fuzzy evaluation membership matrix to obtain the safety assessment result of the electrochemical energy storage system. The present invention can evaluate the safety of the electrochemical energy storage system in different operation stages in combination with the overall characteristics of the overall electrochemical energy storage system, thereby improving the credibility of the evaluation results.

Description

Technical Field

[0001] The present invention relates to the technical field of electrochemical energy storage, and in particular to a safety assessment method and device for an electrochemical energy storage system. Background Art

[0002] Effectively evaluating the overall safety status of an electrochemical energy storage system is an important technical means for the smooth operation of an electrochemical energy storage system and for preventing major safety accidents in the electrochemical energy storage system in advance. At present, the methods and relevant technical standards for evaluating the safety of electrochemical energy storage systems are not yet sound. Most of the reference methods are based on the characteristics of the battery pack itself and the battery management system (BMS) corresponding to the battery pack to evaluate the safety status of a single battery cell or battery pack, which cannot characterize the overall safety status of the entire system, or require other difficult-to-obtain data, and then perform some summary calculations based on statistical data to obtain the safety status. However, the data required for this method is indeed difficult to obtain and count, and it does not have the conditions for adaptive expansion and modification for different electrochemical energy storage systems. At the same time, most of the methods that can be consulted are based on the essential characteristics of the battery, and some statistical calculations are performed based on the basic data of the battery. The degree of integration with some monitoring data in the actual operation of the electrochemical energy storage system itself is not very high, and the measured data of the electrochemical energy storage system is not well utilized for dynamic evaluation. Moreover, even for the same system, the focus is different when evaluating the safety of the system at different stages of operation. The previous methods tend to be fixed in terms of the way the system is evaluated for safety and the values ​​of the parameters required in the evaluation process, which cannot well meet the actual conditions of the electrochemical energy storage system at different stages of operation over a long period of time. Therefore, how to fully consider the characteristics of the electrochemical energy storage system and the multi-faceted influencing factors at all levels of the whole system, and to comprehensively and dynamically evaluate the safety status of the entire system based on the electrochemical energy storage system itself and the actual operation conditions, is the key to determining whether the evaluation results are valuable and referenceable. Summary of the invention

[0003] Therefore, the technical problem to be solved by the present invention is to overcome the defect in the prior art that the safety status of the entire system cannot be comprehensively and dynamically evaluated based on the electrochemical energy storage system itself and the actual operating conditions, thereby providing a method and device for safety assessment of an electrochemical energy storage system.

[0004] The first aspect of the present invention provides a safety assessment method for an electrochemical energy storage system, comprising: during the operation of the electrochemical energy storage system, obtaining data corresponding to each factor node in a pre-established safety status influencing factor structure tree, the factor nodes including subjective factor nodes and objective factor nodes, the safety status influencing factor structure tree being established based on factors related to the safety status of the electrochemical energy storage system; obtaining evaluation results corresponding to each subjective factor node; determining the evaluation results corresponding to each objective factor node according to the threshold interval to which the data corresponding to each objective factor node belongs and the evaluation results corresponding to the threshold interval; each objective factor node corresponds to multiple threshold intervals, different threshold intervals correspond to evaluation results of different levels, and the upper limit and lower limit of each threshold interval are continuously tuned and updated through a first machine learning model; according to a comprehensive fuzzy evaluation method, a fuzzy evaluation membership matrix is ​​established in combination with the safety status influencing factor structure tree, the evaluation results of each factor node, and the weights of each factor node; in the safety status influencing factor structure tree, the sum of the weights between factor nodes belonging to the same level is 1; the fuzzy evaluation membership matrix is ​​calculated to obtain the safety assessment result of the electrochemical energy storage system.

[0005] Optionally, in the electrochemical energy storage system safety assessment method provided by the present invention, the step of obtaining the evaluation results corresponding to each subjective factor node includes: if the prediction performance of the current second machine learning model does not meet the first preset condition, receiving the evaluation results corresponding to the subjective factor nodes input by an external device; adding the data corresponding to the subjective factor nodes and the evaluation results corresponding to the subjective factor nodes to the training set for training the second machine learning model.

[0006] Optionally, in the electrochemical energy storage system safety assessment method provided by the present invention, if the prediction performance of the current second machine learning model meets the first preset condition, the data corresponding to the subjective factor node is input into the second machine learning model, and the evaluation result corresponding to the subjective factor node is obtained through the second machine learning model.

[0007] Optionally, in the electrochemical energy storage system safety assessment method provided by the present invention, the step of optimizing and updating the upper limit and lower limit of each threshold interval by the first machine learning model includes: obtaining multiple groups of sample data, each group of sample data containing data corresponding to each objective factor node; inputting the sample data into the first machine learning model to obtain evaluation results corresponding to each group of sample data; determining the proportion of multiple data corresponding to the objective factor node in the current threshold intervals in the evaluation results of different levels; optimizing and updating the upper limit and lower limit of the threshold interval corresponding to the evaluation results of each level according to the proportion of multiple data corresponding to the objective factor node in the current threshold intervals in the evaluation results of different levels to obtain optimized threshold intervals.

[0008] Optionally, the electrochemical energy storage system safety assessment method provided by the present invention also includes: obtaining a training set, the training set containing data corresponding to each objective factor node and the safety level of the electrochemical energy storage system; training a first machine learning model according to the training set; if the training process of the first machine learning model meets the second preset condition, the upper limit and lower limit of each threshold interval are tuned and updated through the first machine learning model; updating the training set, and returning to the step of training the first machine learning model according to the training set.

[0009] The present invention provides a safety assessment device for an electrochemical energy storage system, comprising: a data acquisition module, used to obtain data corresponding to each factor node in a pre-established safety status influencing factor structure tree during the operation of the electrochemical energy storage system, wherein the factor nodes include subjective factor nodes and objective factor nodes, and the safety status influencing factor structure tree is established based on factors related to the safety status of the electrochemical energy storage system; a first evaluation result acquisition module, used to obtain evaluation results corresponding to each subjective factor node; a second evaluation result acquisition module, used to determine the safety status of each objective factor node according to the threshold interval to which the data corresponding to each objective factor node belongs, and the evaluation results corresponding to the threshold interval. Corresponding evaluation results; each objective factor node corresponds to multiple threshold intervals, different threshold intervals correspond to evaluation results of different levels, and the upper and lower limits of each threshold interval are continuously tuned and updated through the first machine learning model; a fuzzy evaluation membership matrix construction module is used to establish a fuzzy evaluation membership matrix in accordance with a comprehensive fuzzy evaluation method, combining the safety status influencing factor structure tree, the evaluation results of each factor node, and the weight of each factor node; in the safety status influencing factor structure tree, the sum of the weights between factor nodes belonging to the same level is 1; a third evaluation result acquisition module is used to calculate the fuzzy evaluation membership matrix to obtain a safety assessment result of the electrochemical energy storage system.

[0010] Optionally, in the electrochemical energy storage system safety assessment device provided by the present invention, the first evaluation result acquisition module includes: a data receiving submodule, if the current prediction performance of the second machine learning model does not meet the first preset condition, the data receiving submodule is used to receive the evaluation result corresponding to the subjective factor node input by an external device; a model training submodule, used to add the data corresponding to the subjective factor node and the evaluation result corresponding to the subjective factor node to the training set, for training the second machine learning model; an evaluation result calculation submodule, if the current prediction performance of the second machine learning model meets the first preset condition, the evaluation result calculation submodule is used to input the data corresponding to the subjective factor node into the second machine learning model, and obtain the evaluation result corresponding to the subjective factor node through the second machine learning model.

[0011] Optionally, the electrochemical energy storage system safety assessment device provided by the present invention also includes: a sample acquisition module, which is used to obtain multiple groups of sample data, each group of sample data contains data corresponding to each objective factor node; a sample calculation module, which is used to input the sample data into the first machine learning model to obtain evaluation results corresponding to each group of sample data; a sample calculation result analysis module, which is used to determine the proportion of multiple data corresponding to the objective factor nodes in the current threshold intervals in the evaluation results of different levels; a threshold interval optimization module, which is used to adjust and update the upper and lower limits of the threshold intervals corresponding to the evaluation results of each level according to the proportion of multiple data corresponding to the objective factor nodes in the current threshold intervals in the evaluation results of different levels, so as to obtain optimized threshold intervals.

[0012] The third aspect of the present invention provides a computer device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor, thereby executing the electrochemical energy storage system safety assessment method provided in the first aspect of the present invention.

[0013] A fourth aspect of the present invention provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the electrochemical energy storage system safety assessment method provided in the first aspect of the present invention.

[0014] The technical solution of the present invention has the following advantages:

[0015] The electrochemical energy storage system safety assessment method and device provided by the present invention first establish a safety status influencing factor structure tree based on the overall electrochemical energy storage system. When the safety of the electrochemical energy storage system is subsequently assessed, it is based on the structure of the safety status influencing factor structure tree and the factor nodes contained therein. Since the safety status influencing factor structure tree is established based on the overall electrochemical energy storage system, and different electrochemical energy storage systems have different safety status influencing factor structure trees, the electrochemical energy storage system can be evaluated as a whole based on the safety status influencing factor structure tree, and targeted safety status influencing factor structure trees are established for electrochemical energy storage systems with different structural forms to assess the safety status. The present invention has high applicability and good scalability.

[0016] In addition, when the comprehensive fuzzy evaluation method is used to evaluate the safety of the electrochemical energy storage system, the data during the actual operation of the electrochemical energy storage system is used as the main data basis to determine the evaluation results of the subjective factor nodes and the objective factor nodes, so as to obtain the safety evaluation results of the electrochemical energy storage system according to the comprehensive fuzzy evaluation method. The evaluation results of the objective factor nodes are determined according to the threshold interval to which the data corresponding to the objective factor nodes belong. The upper and lower limits of the threshold interval will be continuously tuned and updated through the first machine learning model. In the continuous operation of the electrochemical energy storage system, the upper and lower limits of the threshold interval are also continuously adjusted. Therefore, in different operation stages of the electrochemical energy storage system, its safety can be dynamically evaluated according to the state of the electrochemical energy storage system, which improves the credibility of the safety status evaluation results of the electrochemical energy storage system to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 A flowchart of a specific example of a method for safety assessment of an electrochemical energy storage system in an embodiment of the present invention;

[0019] Figure 2 A schematic diagram of a structure tree of security status influencing factors pre-established in an embodiment of the present invention;

[0020] Figure 3 is a principle block diagram of a specific example of a safety assessment device for an electrochemical energy storage system in an embodiment of the present invention;

[0021] Figure 4 It is a principle block diagram of a specific example of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0023] In the description of the present invention, it should be noted that the terms “first” and “second” are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0024] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0025] The embodiment of the present invention provides a method for safety assessment of an electrochemical energy storage system. Figure 1 As shown, including:

[0026] Step S11: During the operation of the electrochemical energy storage system, data corresponding to each factor node in a pre-established safety status influencing factor structure tree is obtained, wherein the factor nodes include subjective factor nodes and objective factor nodes, and the safety status influencing factor structure tree is established based on factors related to the safety status of the electrochemical energy storage system.

[0027] In an optional embodiment, since different electrochemical energy storage systems have different hardware topologies, operating environments, operating slots, etc., the structures of the safety status influencing factor structure trees of different electrochemical systems are not completely the same.

[0028] In the embodiment of the present invention, the factors affecting the electrochemical energy storage system are divided into different levels, forming a tree structure of safety status influencing factors.

[0029] In an optional embodiment, the factor corresponding to the subjective factor node cannot obtain a fixed value, such as the safety of the fire protection system.

[0030] In an optional embodiment, the factor corresponding to the objective factor node can directly obtain a fixed value, or can be calculated through the collected data to obtain a fixed value, such as a cell voltage value, a current value, a temperature value, etc.

[0031] Step S12: Obtain the evaluation result corresponding to each subjective factor node.

[0032] In an optional embodiment, the evaluation result corresponding to the subjective factor node is obtained based on expert experience or a second machine learning model.

[0033] Step S13: Determine the evaluation results corresponding to each objective factor node according to the threshold interval to which the data corresponding to each objective factor node belongs and the evaluation results corresponding to the threshold interval; each objective factor node corresponds to multiple threshold intervals, different threshold intervals correspond to different levels of evaluation results, and the upper limit value and lower limit value of each threshold interval are continuously optimized and updated through the first machine learning model.

[0034] In an optional embodiment, each objective factor node corresponds to a plurality of threshold intervals, the number of threshold intervals of each objective factor node is the same, and the number of threshold intervals of the objective factor node is the same as the number of levels of the evaluation result. For example, if the evaluation result is divided into 5 levels, the number of threshold intervals corresponding to each objective factor node is 5.

[0035] Step S14: According to the comprehensive fuzzy evaluation method, a fuzzy evaluation membership matrix is ​​established by combining the safety status influencing factor structure tree, the evaluation results of each factor node, and the weight of each factor node; in the safety status influencing factor structure tree, the sum of the weights between factor nodes belonging to the same level is 1.

[0036] In an optional embodiment, when setting the weight of each factor node, the weight coefficients of different experts for factor nodes at different levels in the safety status influencing factor structure tree are weighted averaged to obtain the weight distribution result of each influencing factor node in the structure tree.

[0037] Step S15: Calculate the fuzzy evaluation membership matrix to obtain the safety assessment result of the electrochemical energy storage system.

[0038] In an optional embodiment, the number of levels of the safety assessment result of the electrochemical energy storage system, the evaluation result of the subjective factor node, and the evaluation result of the objective factor node is the same. Exemplarily, the safety assessment result and the evaluation result include five levels: "safe", "relatively safe", "risky", "highly risky", and "unsafe".

[0039] The electrochemical energy storage system safety assessment method provided in the embodiment of the present invention first establishes a safety status influencing factor structure tree based on the overall electrochemical energy storage system. When the safety of the electrochemical energy storage system is subsequently assessed, it is based on the structure of the safety status influencing factor structure tree and the factor nodes contained therein. Since the safety status influencing factor structure tree is established based on the overall electrochemical energy storage system, and the safety status influencing factor structure trees of different electrochemical energy storage systems are different, the electrochemical energy storage system can be evaluated as a whole based on the safety status influencing factor structure tree, and targeted safety status influencing factor structure trees are established for electrochemical energy storage systems with different structural forms to assess the safety status. The present method has high applicability and good scalability.

[0040] In addition, when the comprehensive fuzzy evaluation method is used to evaluate the safety of the electrochemical energy storage system, the data during the actual operation of the electrochemical energy storage system is used as the main data basis to determine the evaluation results of the subjective factor nodes and the objective factor nodes, so as to obtain the safety evaluation results of the electrochemical energy storage system according to the comprehensive fuzzy evaluation method. The evaluation results of the objective factor nodes are determined according to the threshold interval to which the data corresponding to the objective factor nodes belong. The upper and lower limits of the threshold interval will be continuously tuned and updated through the first machine learning model. In the continuous operation of the electrochemical energy storage system, the upper and lower limits of the threshold interval are also continuously adjusted. Therefore, in different operation stages of the electrochemical energy storage system, its safety can be dynamically evaluated according to the state of the electrochemical energy storage system, which improves the credibility of the safety status evaluation results of the electrochemical energy storage system to a certain extent.

[0041] In an optional embodiment, the electrochemical energy storage system safety assessment method provided by the present invention is implemented to construct a safety status influencing factor structure tree of the electrochemical energy storage system through the following steps:

[0042] First, the topological structure of each hardware device and battery pack of the electrochemical energy storage system was sorted out, and relevant experts were invited to discuss and clarify several major factors that affect the safety status of the electrochemical energy storage system in the energy storage container.

[0043] Then, according to the BMS system in the electrochemical energy storage system, including the on-site EMS system that may be contained in the container, the fire protection system, the temperature control system, and the management and association methods of each battery cluster, module, and battery cell in the electrochemical energy storage system, the influencing factors related to the safety status in the electrochemical energy storage system are expanded.

[0044] Subsequently, a safety status influencing factor structure is established, and each node category is further refined layer by layer, and each node is split into quantifiable sub-nodes that need further evaluation, forming a multi-level and multi-node safety status influencing factor structure tree for the electrochemical energy storage system. For example, the safety status influencing factor structure tree constructed is as follows: Figure 2 shown.

[0045] In an optional embodiment, the above step S12 specifically includes:

[0046] If the prediction performance of the current second machine learning model does not meet the first preset condition, the evaluation result is obtained through the following steps:

[0047] First, an evaluation result corresponding to a subjective factor node inputted from an external device is received.

[0048] Then, the data corresponding to the subjective factor node and the evaluation results corresponding to the subjective factor node are added to the training set for training the second machine learning model.

[0049] If the prediction performance of the current second machine learning model does not meet the first preset condition, the data corresponding to the subjective factor node is input into the second machine learning model, and the evaluation result corresponding to the subjective factor node is obtained through the second machine learning model.

[0050] In an embodiment of the present invention, the second machine learning model can predict the evaluation results obtained by experts based on the data of the subjective factor node. When the prediction performance of the second machine learning model does not meet the preset conditions, it means that the second machine learning model cannot accurately predict the expert's evaluation results at this time. Therefore, it is necessary to receive the evaluation results input by the external device, and the evaluation results are obtained by the expert scoring. After obtaining the evaluation results input by the external device, the data corresponding to the subjective factor node and the evaluation results corresponding to the subjective factor node are added to the training set for training the second machine learning model until the prediction performance of the second machine learning model meets the first preset conditions. After obtaining the data corresponding to the subjective factor node, the data is input into the second machine learning model to obtain the evaluation results through the second machine learning model.

[0051] Exemplarily, the prediction performance of the second machine learning model may be the accuracy of the second machine learning model. When the accuracy of the second machine learning model is greater than a preset value, it indicates that the prediction performance of the second machine learning model meets the first preset condition.

[0052] The electrochemical energy storage system safety assessment method provided in an embodiment of the present invention continuously trains the second machine learning model. When the prediction performance of the second machine learning model meets the first preset condition, the evaluation results of the subjective factor nodes are no longer obtained through expert subjective scoring. By gradually weakening and replacing the expert subjective evaluation through this method, the obtained safety status assessment results are closer to the actual situation of the electrochemical energy storage system, and the efficiency of safety assessment is higher.

[0053] In an optional embodiment, the step of optimizing and updating the upper limit and lower limit of each threshold interval by the first machine learning model includes:

[0054] First, multiple groups of sample data are obtained, each group of sample data contains data corresponding to each objective factor node.

[0055] Secondly, the sample data is input into the first machine learning model to obtain the evaluation results corresponding to each group of sample data.

[0056] Then, in the evaluation results at different levels, the proportions of the multiple data corresponding to the objective factor nodes within the current threshold intervals are determined.

[0057] Finally, according to the proportion of multiple data corresponding to the objective factor nodes in the current threshold intervals in the evaluation results of different levels, the upper and lower limits of the threshold intervals corresponding to the evaluation results of each level are adjusted and updated to obtain the optimized threshold intervals.

[0058] Taking the evaluation results of one level as an example, multiple sample data will correspond to the same evaluation result, that is, one evaluation result corresponds to multiple groups of sample data, and each group of sample data contains data corresponding to multiple objective factors. This means that one evaluation result will correspond to multiple data of an objective factor node, and different data of the objective factor node may be in different threshold intervals. In an embodiment of the present invention, the upper and lower limits of the threshold interval are optimized according to the proportion of the data of the objective factor node in different threshold intervals.

[0059] Exemplarily, for the evaluation result of "safety", the value of the objective factor node "battery cell voltage value" may fall within different threshold intervals, and the threshold intervals correspond to "safe", "relatively safe", "risky", "highly risky", and "unsafe", respectively. If the proportion falling within "relatively safe" is the largest, then at least the lower limit value of the threshold interval corresponding to "safety" and the upper limit value of the threshold interval corresponding to "relatively safe" need to be adjusted, so as to increase the proportion of the "battery cell voltage value" falling within the threshold interval corresponding to "safety", so that the result is closer to the actual situation.

[0060] In an optional embodiment, the electrochemical energy storage system safety assessment method provided by the embodiment of the present invention further includes:

[0061] First, a training set is obtained, which contains data corresponding to each objective factor node and the safety level of the electrochemical energy storage system.

[0062] Then, the first machine learning model is trained according to the training set. When the training process of the first machine learning model meets the second preset condition, the following steps are performed:

[0063] The upper limit and lower limit of each threshold interval are tuned and updated through the first machine learning model, the training set is updated, and the step of training the first machine learning model according to the training set is returned.

[0064] In an embodiment of the present invention, during the continuous operation of the electrochemical energy storage system and during the continuous use of the first machine learning model, the first machine learning model will still be continuously trained. After completing one stage of training, the upper and lower limits of each threshold interval are tuned by the first machine learning model. In different operating stages of the electrochemical energy storage system, the emphasis is different when evaluating its safety, and some parameters of the electrochemical energy storage system will change. If the first machine learning model obtained by training according to the initial training set is blindly used to tune the upper and lower limits of each interval threshold, the obtained tuning result does not match the current situation. In an embodiment of the present invention, the first machine learning model is continuously trained and optimized, and the upper and lower limits of each threshold interval are continuously tuned according to the optimized first machine learning model, so that a more accurate safety assessment result can be obtained through the optimized interval threshold.

[0065] In an optional embodiment, the training progress of the first machine learning model can be measured according to parameters such as training duration and number of trainings. For example, when the training progress of the first machine learning model is measured by training duration, if the training duration of the first machine learning model in this round of training reaches a preset duration, then the training progress of the first machine learning model is determined to meet the second preset condition; when the training progress of the first machine learning model is measured by the number of trainings, if the number of trainings of the first machine learning model in this round of training reaches a preset value, then the training progress of the first machine learning model is determined to meet the second preset condition.

[0066] In an optional embodiment, since the safety assessment result of the electrochemical energy storage system is based on a safety status influencing factor structure tree, and the safety status influencing factor structure tree is constructed according to the structure of the electrochemical energy storage system, when a battery pack in the electrochemical energy storage system is replaced, the following adjustments need to be made:

[0067] Step 1: Update the structure tree of factors affecting the safety status of the electrochemical energy storage system:

[0068] First, after a group of batteries in the electrochemical energy storage system is replaced, it is necessary to update each node of the branch related to this group of batteries based on the original safety status influencing factor structure tree, including the topological structure, mechanism characteristics of the newly replaced battery group, etc.

[0069] Then, the relevant sub-nodes under the updated battery pack branch are expanded or deleted, the parameters that are not closely related to the safety status of the replaced battery pack are removed, and the nodes affecting the safety status of the old battery pack are replaced according to the characteristics of the battery pack.

[0070] Finally, the relevant influencing factors of the replaced battery pack are broken down into different levels and refined, including those that need to be evaluated based on a threshold value range and those that require subjective scoring and judgment by experts.

[0071] Step 2: Update relevant evaluation criteria:

[0072] First, for the updated relevant rated parameters of the battery pack, the threshold intervals of the objective factor nodes corresponding to the original system evaluation standard are updated.

[0073] Then, the newly added branch nodes under the replaced battery pack are divided into evaluation levels consistent with the original evaluation standards.

[0074] Finally, based on the evaluation levels of the newly added nodes, an evaluation description document is prepared for the factors for subjective scoring by experts.

[0075] Step 3: Update the importance weight distribution of each influencing factor

[0076] First, relevant experts are invited to re-evaluate the structure tree of influencing factors after replacing the battery pack, and to redefine the importance weights of the sub-nodes of the battery pack and other nodes associated with it.

[0077] Then, the weighted average of the final weight delineation results of each expert is taken as the new weight factor of each node in the updated influencing factor structure tree.

[0078] Step 4: Update machine learning model and system access:

[0079] First, update the relevant characteristic parameters of the battery pack in the first machine learning model, such as rated current, rated voltage, charge and discharge times, etc.

[0080] Then, the training data of the original battery pack is cleared, and the measured data of the new battery pack is re-connected to train the first machine learning model.

[0081] Finally, the above revisions were updated to the background Python calculation script, connected to the electrochemical energy storage system, and the calculation and collection data of the system after the battery pack was replaced, so as to restore the original dynamic assessment status of the safety status assessment of the electrochemical energy storage system.

[0082] It can be seen that when the electrochemical energy storage system is safety assessed by the electrochemical energy storage system safety assessment method provided by the embodiment of the present invention, even if the structure of the electrochemical energy storage system has changed, only the changed part needs to be adjusted, and the evaluation of the updated electrochemical energy storage system can be quickly achieved.

[0083] The embodiment of the present invention provides a safety assessment device for an electrochemical energy storage system, such as Figure 3 As shown, including:

[0084] The data acquisition module 21 is used to obtain data corresponding to each factor node in a pre-established safety status influencing factor structure tree during the operation of the electrochemical energy storage system. The factor nodes include subjective factor nodes and objective factor nodes. The safety status influencing factor structure tree is established based on factors related to the safety status of the electrochemical energy storage system. For details, please refer to the records of the above embodiment and will not be repeated here.

[0085] The first evaluation result acquisition module 22 is used to acquire the evaluation result corresponding to each subjective factor node. For details, please refer to the description of the above embodiment and will not be repeated here.

[0086] The second evaluation result acquisition module 23 is used to determine the evaluation results corresponding to each objective factor node according to the threshold interval to which the data corresponding to each objective factor node belongs, and the evaluation results corresponding to the threshold interval; each objective factor node corresponds to multiple threshold intervals, different threshold intervals correspond to evaluation results of different levels, and the upper limit value and lower limit value of each threshold interval are continuously optimized and updated through the first machine learning model. For details, please refer to the records of the above embodiment, which will not be repeated here.

[0087] The fuzzy evaluation membership matrix construction module 24 is used to establish a fuzzy evaluation membership matrix in accordance with the comprehensive fuzzy evaluation method, combining the safety status influencing factor structure tree, the evaluation results of each factor node, and the weight of each factor node; in the safety status influencing factor structure tree, the sum of the weights between factor nodes belonging to the same level is 1. For details, please refer to the records of the above embodiment and will not be repeated here.

[0088] The third evaluation result acquisition module 25 is used to calculate the fuzzy evaluation membership matrix to obtain the safety evaluation result of the electrochemical energy storage system. For details, please refer to the description of the above embodiment and will not be repeated here.

[0089] In an optional embodiment, the first evaluation result acquisition module 22 includes:

[0090] The data receiving submodule is used to receive the evaluation result corresponding to the subjective factor node input by an external device if the prediction performance of the current second machine learning model does not meet the first preset condition. For details, please refer to the records of the above embodiment and will not be repeated here.

[0091] The model training submodule is used to add the data corresponding to the subjective factor node and the evaluation results corresponding to the subjective factor node to the training set, so as to train the second machine learning model. For details, please refer to the records of the above embodiment and will not be repeated here.

[0092] The evaluation result calculation submodule is used to input the data corresponding to the subjective factor node into the second machine learning model if the prediction performance of the current second machine learning model meets the first preset condition, and obtain the evaluation result corresponding to the subjective factor node through the second machine learning model. For details, please refer to the records of the above embodiment and will not be repeated here.

[0093] In an optional embodiment, the electrochemical energy storage system safety assessment device provided by the embodiment of the present invention further includes:

[0094] The sample acquisition module is used to acquire multiple groups of sample data, each group of sample data contains data corresponding to each objective factor node. For details, please refer to the records of the above embodiment and will not be repeated here.

[0095] The sample calculation module is used to input the sample data into the first machine learning model to obtain the evaluation results corresponding to each group of sample data. For details, please refer to the records of the above embodiments and will not be repeated here.

[0096] The sample calculation result analysis module is used to determine the proportion of multiple data corresponding to the objective factor node within the current threshold range in the evaluation results of different levels. For details, please refer to the records of the above embodiment and will not be repeated here.

[0097] The threshold interval optimization module is used to adjust and update the upper and lower limits of the threshold intervals corresponding to the evaluation results of each level according to the proportion of multiple data corresponding to the objective factor nodes in the current threshold intervals in the evaluation results of different levels, so as to obtain the optimized threshold intervals. For details, please refer to the records of the above embodiments and will not be repeated here.

[0098] An embodiment of the present invention provides a computer device, such as Figure 4 As shown, the computer device mainly includes one or more processors 31 and a memory 32. Figure 4 A processor 31 is taken as an example.

[0099] The computer device may further include: an input device 33 and an output device 34 .

[0100] The processor 31, the memory 32, the input device 33 and the output device 34 may be connected via a bus or other means. Figure 4 The example of connecting through bus is taken in the following.

[0101] The processor 31 may be a central processing unit (CPU). The processor 31 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, or a combination of the above chips. The general-purpose processor may be a microprocessor or the processor may be any conventional processor, etc. The memory 32 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the electrochemical energy storage system safety assessment device, etc. In addition, the memory 32 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory 32 may optionally include a memory remotely arranged relative to the processor 31, and these remote memories may be connected to the electrochemical energy storage system safety assessment device via a network. The input device 33 can receive a calculation request (or other digital or character information) input by a user, and generate a key signal input related to the electrochemical energy storage system safety assessment device. The output device 34 can include a display device such as a display screen to output the calculation result.

[0102] The embodiment of the present invention provides a computer-readable storage medium, which stores computer instructions, and the computer storage medium stores computer executable instructions, which can execute the electrochemical energy storage system safety assessment method in any of the above method embodiments. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory (Flash Memory), a hard disk (Hard Disk Drive, abbreviated: HDD) or a solid-state drive (SSD), etc.; the storage medium can also include a combination of the above types of memory.

[0103] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. A safety assessment method for an electrochemical energy storage system, characterized in that: include: During the operation of the electrochemical energy storage system, data corresponding to each factor node in a pre-established safety status influencing factor structure tree is obtained, wherein the factor node includes a subjective factor node and an objective factor node, and the safety status influencing factor structure tree is established based on factors related to the safety status of the electrochemical energy storage system; Obtaining evaluation results corresponding to each of the subjective factor nodes; Determine the evaluation results corresponding to each objective factor node according to the threshold interval to which the data corresponding to each objective factor node belongs and the evaluation results corresponding to the threshold interval; each objective factor node corresponds to multiple threshold intervals, different threshold intervals correspond to different levels of evaluation results, and the upper limit and lower limit of each threshold interval are continuously tuned and updated through the first machine learning model; According to the comprehensive fuzzy evaluation method, a fuzzy evaluation membership matrix is ​​established by combining the safety status influencing factor structure tree, the evaluation results of each factor node, and the weight of each factor node; In the safety status influencing factor structure tree, the sum of the weights of the factor nodes belonging to the same level is 1; Calculating the fuzzy evaluation membership matrix to obtain a safety assessment result of the electrochemical energy storage system; The step of tuning and updating the upper limit and lower limit of each threshold interval by the first machine learning model includes: Acquire multiple groups of sample data, each group of sample data contains data corresponding to each objective factor node; Inputting the sample data into the first machine learning model to obtain evaluation results corresponding to each group of sample data; In the evaluation results of different levels, determining the proportion of the multiple data corresponding to the objective factor node within each current threshold interval; According to the proportion of the multiple data corresponding to the objective factor nodes in the current threshold intervals in the evaluation results of different levels, the upper limit and lower limit of the threshold interval corresponding to the evaluation results of each level are adjusted and updated to obtain the optimized threshold interval.

2. The electrochemical energy storage system safety assessment method according to claim 1, characterized in that: The obtaining of the evaluation results corresponding to each of the subjective factor nodes comprises: If the prediction performance of the current second machine learning model does not meet the first preset condition, Receiving the evaluation result corresponding to the subjective factor node input by an external device; Adding the data corresponding to the subjective factor node and the evaluation result corresponding to the subjective factor node to a training set for training the second machine learning model; The second machine learning model is used to predict the evaluation results obtained by experts based on the data of the subjective factor node; when the prediction performance of the second machine learning model does not meet the first preset condition, the evaluation result input by the external device is received, and the evaluation result is obtained by the expert scoring; after obtaining the evaluation result input by the external device, the data corresponding to the subjective factor node and the evaluation result corresponding to the subjective factor node are added to the training set, which is used to train the second machine learning model until the prediction performance of the second machine learning model meets the first preset condition. After obtaining the data corresponding to the subjective factor node, the data is input into the second machine learning model to obtain the evaluation result through the second machine learning model; The prediction performance of the second machine learning model is the accuracy of the second machine learning model. When the accuracy of the second machine learning model is greater than the preset value, it indicates that the prediction performance of the second machine learning model meets the first preset condition.

3. The electrochemical energy storage system safety assessment method according to claim 2, characterized in that: If the prediction performance of the current second machine learning model meets the first preset condition, The data corresponding to the subjective factor node is input into the second machine learning model, and the evaluation result corresponding to the subjective factor node is obtained through the second machine learning model.

4. The electrochemical energy storage system safety assessment method according to claim 1, characterized in that: Also includes: Obtaining a training set, wherein the training set contains data corresponding to each objective factor node and a safety level of the electrochemical energy storage system; Training the first machine learning model according to the training set; If the training process of the first machine learning model meets the second preset condition, Tuning and updating the upper limit and lower limit of each threshold interval through the first machine learning model; Update the training set and return to the step of training the first machine learning model according to the training set.

5. A safety assessment device for an electrochemical energy storage system, characterized in that: include: A data acquisition module, used to obtain data corresponding to each factor node in a pre-established safety status influencing factor structure tree during the operation of the electrochemical energy storage system, wherein the factor node includes a subjective factor node and an objective factor node, and the safety status influencing factor structure tree is established based on factors related to the safety status of the electrochemical energy storage system; A first evaluation result acquisition module, used to acquire the evaluation result corresponding to each subjective factor node; A second evaluation result acquisition module is used to determine the evaluation results corresponding to each objective factor node according to the threshold interval to which the data corresponding to each objective factor node belongs and the evaluation results corresponding to the threshold interval; each objective factor node corresponds to multiple threshold intervals, different threshold intervals correspond to different levels of evaluation results, and the upper limit value and the lower limit value of each threshold interval are continuously tuned and updated through the first machine learning model; A fuzzy evaluation membership matrix construction module is used to establish a fuzzy evaluation membership matrix in accordance with a comprehensive fuzzy evaluation method, combining the safety status influencing factor structure tree, the evaluation results of each factor node, and the weight of each factor node; In the safety status influencing factor structure tree, the sum of the weights of the factor nodes belonging to the same level is 1; A third evaluation result acquisition module is used to calculate the fuzzy evaluation membership matrix to obtain a safety assessment result of the electrochemical energy storage system; The step of tuning and updating the upper limit and lower limit of each threshold interval by the first machine learning model includes: Acquire multiple groups of sample data, each group of sample data contains data corresponding to each objective factor node; Inputting the sample data into the first machine learning model to obtain evaluation results corresponding to each group of sample data; In the evaluation results of different levels, determining the proportion of the multiple data corresponding to the objective factor node within each current threshold interval; According to the proportion of the multiple data corresponding to the objective factor nodes in the current threshold intervals in the evaluation results of different levels, the upper limit and lower limit of the threshold interval corresponding to the evaluation results of each level are adjusted and updated to obtain the optimized threshold interval.

6. The electrochemical energy storage system safety assessment device according to claim 5, characterized in that: The first evaluation result acquisition module includes: A data receiving submodule, which is used to receive an evaluation result corresponding to a subjective factor node input by an external device if the prediction performance of the current second machine learning model does not meet the first preset condition; A model training submodule, used for adding the data corresponding to the subjective factor node and the evaluation result corresponding to the subjective factor node to a training set, so as to train the second machine learning model; An evaluation result calculation submodule, which is used to input the data corresponding to the subjective factor node into the second machine learning model if the prediction performance of the current second machine learning model meets the first preset condition, and obtain the evaluation result corresponding to the subjective factor node through the second machine learning model; The second machine learning model is used to predict the evaluation results obtained by experts based on the data of the subjective factor node; when the prediction performance of the second machine learning model does not meet the first preset condition, the evaluation result input by the external device is received, and the evaluation result is obtained by the expert scoring; after obtaining the evaluation result input by the external device, the data corresponding to the subjective factor node and the evaluation result corresponding to the subjective factor node are added to the training set, which is used to train the second machine learning model until the prediction performance of the second machine learning model meets the first preset condition. After obtaining the data corresponding to the subjective factor node, the data is input into the second machine learning model to obtain the evaluation result through the second machine learning model; The prediction performance of the second machine learning model is the accuracy of the second machine learning model. When the accuracy of the second machine learning model is greater than the preset value, it indicates that the prediction performance of the second machine learning model meets the first preset condition.

7. The electrochemical energy storage system safety assessment device according to claim 5, characterized in that: Also includes: The sample acquisition module is used to acquire multiple groups of sample data, each group of sample data contains data corresponding to each objective factor node; A sample calculation module, used to input the sample data into the first machine learning model to obtain evaluation results corresponding to each group of sample data; A sample calculation result analysis module is used to determine the proportion of multiple data corresponding to the objective factor node within each current threshold interval in the evaluation results of different levels; The threshold interval optimization module is used to adjust and update the upper and lower limits of the threshold intervals corresponding to the evaluation results of each level according to the proportion of multiple data corresponding to the objective factor nodes in the current threshold intervals in the evaluation results of different levels, so as to obtain the optimized threshold intervals.

8. A computer device, characterized in that: include: at least one processor; And a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to execute the electrochemical energy storage system safety assessment method as described in any one of claims 1-4.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the electrochemical energy storage system safety assessment method according to any one of claims 1 to 4.

Images