Lithium battery safety assessment method and device, electronic equipment, storage medium and program product

Abstract

The invention relates to the technical field of lithium batteries, and discloses a lithium battery safety assessment method and device, electronic equipment, a storage medium and a program product, and the method comprises the steps: inputting to-be-assessed parameter information into a safety assessment network model, and obtaining an early warning fault type of a lithium battery; in a mapping relation library, matching the early warning fault type with a target confidence interval; wherein a known fault type and a confidence interval corresponding to the known fault type are stored in the mapping relation library, and the confidence interval comprises a parameter value interval of each evaluation parameter during abnormal operation; judging whether to-be-evaluated parameter values in the to-be-evaluated parameter information fall into a target confidence interval corresponding to the early warning fault type or not; and determining that the fault of the early-warning fault type occurs when all the faults fall into the early-warning fault type. According to the invention, the safety assessment capability of the lithium battery is improved.

Description

Technical Field

[0001] The present invention relates to the field of lithium battery technology, and in particular to a lithium battery safety assessment method, device, electronic equipment, storage medium and program product. Background Art

[0002] In the field of energy storage power stations, the widespread use of lithium batteries provides efficient and reliable solutions for energy storage and conversion.

[0003] However, lithium batteries are affected by many factors during operation and are prone to failures such as overcharging, over-discharging, internal and external short circuits, and overheating. These failures will not only seriously affect the performance and life of the battery, but may also cause safety accidents. Summary of the Invention

[0004] The purpose of the present invention is to improve the safety assessment capability of lithium batteries.

[0005] In a first aspect, an embodiment of the present invention provides a lithium battery safety assessment method, the method comprising:

[0006] Obtaining parameter information of the lithium battery to be evaluated during the evaluation period;

[0007] Inputting the parameter information to be evaluated into a safety assessment network model to obtain a warning fault type of the lithium battery;

[0008] In a mapping relationship library, the warning fault type is matched with a target confidence interval; wherein the mapping relationship library stores known fault types and confidence intervals corresponding to the known fault types, and the confidence intervals include parameter value intervals of each evaluation parameter during abnormal operation;

[0009] Determining whether all the values of the parameters to be evaluated in the parameter information to be evaluated fall within the target confidence interval corresponding to the warning fault type;

[0010] If it is determined that the values of the parameters to be evaluated all fall within their respective target confidence intervals, it is determined that a fault of the warning fault type has occurred.

[0011] Furthermore, after determining whether all the values of the parameters to be evaluated in the parameter information to be evaluated fall within their respective target confidence intervals, the method further includes:

[0012] If it is determined that the values of the parameter to be evaluated do not all fall within the target confidence interval corresponding to the warning fault type, comparing the actual change trend of the parameter to be evaluated with the standard change trend of the parameter to be evaluated under the warning fault type;

[0013] When the actual change trend of the parameter to be evaluated is consistent with the standard change trend, it is determined that a fault of the warning fault type has occurred.

[0014] Furthermore, when the parameters do not fall into the above categories, after comparing the actual change trend of the parameter to be evaluated with the standard change trend of the parameter to be evaluated under the early warning fault type, the method further includes:

[0015] When the actual change trend of the parameter to be evaluated is inconsistent with the standard change trend, a suspected fault is determined to have occurred, and the target confidence interval, the value of the parameter to be evaluated, the change trend comparison information of the parameter to be evaluated, and the warning fault type are sent to the client for fault confirmation.

[0016] Furthermore, before inputting the parameter information to be evaluated into the security assessment network model, the method further includes:

[0017] The parameter information to be evaluated is preprocessed, and the preprocessing includes at least one of data deduplication, missing value filling, and outlier detection.

[0018] Furthermore, the preprocessing of the parameter information to be evaluated includes:

[0019] Removing duplicate information from the parameter information to be evaluated based on a hash algorithm or a timestamp method;

[0020] and / or, using multiple interpolation methods in combination with time series analysis techniques to infer missing values based on known data in the parameter information to be evaluated, and filling missing values in the parameter information to be evaluated, wherein the multiple interpolation methods include linear interpolation, spline interpolation, and weighted average based on adjacent data;

[0021] And / or, based on a set threshold, performing outlier detection on the parameter information to be evaluated, and removing or replacing the outlier.

[0022] In a second aspect, an embodiment of the present invention further provides a model building method, the method comprising:

[0023] Acquire sample evaluation parameter information of a lithium battery during abnormal operation and a sample fault type of the sample evaluation parameter information;

[0024] The sample evaluation parameter information is used as sample input, and the sample fault type is used as a sample label to train the initial neural network to obtain a safety evaluation network model for implementing any of the lithium battery safety evaluation methods described in the embodiments of the present invention.

[0025] Furthermore, the initial neural network adopts a network with long-short-term memory function to process the long-term dependency of the time series data in the sample evaluation parameter information.

[0026] Furthermore, before the sample evaluation parameter information is used as sample input and the sample fault type is used as a sample label to train the initial neural network to obtain the safety evaluation network model, the method further includes:

[0027] Determining the number of neurons in the input layer according to the number of evaluation parameter types in the sample evaluation parameter information;

[0028] Determining the number of units in the hidden layer based on the data complexity and time series length of the sample evaluation parameter information;

[0029] Determining the number of neurons in the output layer according to the number of fault types in the sample fault types;

[0030] The initial neural network is determined according to the number of neurons in the input layer, the number of units in the hidden layer, and the number of neurons in the output layer.

[0031] In a third aspect, an embodiment of the present invention provides a method for determining a mapping relationship, the method comprising:

[0032] Based on the known evaluation parameter information and known fault types in the fault records, statistics are collected on the fault records within a first preset period, and a mapping relationship between the known fault types and a set of confidence intervals, as well as a mapping relationship between the known fault types and a single confidence interval, is established. The mapping relationships are stored in a mapping relationship library for implementing any of the lithium battery safety assessment methods described in any embodiment of the present invention, wherein the confidence interval includes a parameter value range for each evaluation parameter during abnormal operation.

[0033] Furthermore, the method of performing statistics on the fault records within the first preset period based on the known evaluation parameter information and the known fault types in the fault records, and establishing a mapping relationship between the known fault types and the confidence interval set, and a mapping relationship between the known fault types and the single confidence interval, includes:

[0034] According to each fault record within the first preset period, the probability of occurrence of different fault types under the confidence intervals corresponding to different evaluation parameters is calculated;

[0035] Sort the fault types in each confidence interval according to the probability value;

[0036] Extracting the top preset number of known fault types for each confidence interval;

[0037] A set of confidence intervals with the same known fault type is determined, and a mapping relationship between the known fault type and the set of confidence intervals, as well as a mapping relationship between the known fault type and a single confidence interval, is established.

[0038] Furthermore, the method of calculating the probability of occurrence of different fault types under confidence intervals corresponding to different evaluation parameters based on each fault record within the first preset period includes:

[0039] Extracting each fault record within a first preset period from the fault feature library, each fault record including evaluation parameter information and fault type corresponding to the fault;

[0040] According to each fault record, the fault type count is performed for the confidence interval of each evaluation parameter;

[0041] Count the total number of occurrences of all fault types within each confidence interval;

[0042] The probability of a single fault type occurring in a single confidence interval is obtained by calculating the ratio between the number of occurrences of a single fault type in a single confidence interval and the total number of occurrences of all fault types in a single confidence interval.

[0043] Furthermore, counting the fault types for the confidence interval of each evaluation parameter according to each fault record includes:

[0044] Determine whether each evaluation parameter value falls within its respective confidence interval in a single fault record;

[0045] When falling into the confidence interval, the number of occurrences of the fault type in a single fault record in the confidence space is accumulated.

[0046] In a fourth aspect, an embodiment of the present invention provides a lithium battery safety assessment device, comprising:

[0047] An information acquisition module is used to obtain parameter information of the lithium battery to be evaluated during the evaluation period;

[0048] A fault prediction module is used to input the parameter information to be evaluated into a safety assessment network model to obtain a warning fault type of the lithium battery;

[0049] An interval determination module is configured to match the warning fault type with a target confidence interval in a mapping relationship library; wherein the mapping relationship library stores known fault types and confidence intervals corresponding to the known fault types, and the confidence intervals include parameter value intervals for each evaluation parameter during abnormal operation;

[0050] An interval matching module is used to determine whether the values of the parameters to be evaluated in the parameter information to be evaluated all fall within the target confidence interval corresponding to the warning fault type;

[0051] The fault determination module is used to determine that a fault of the warning fault type has occurred if it is determined that the values of the parameters to be evaluated all fall within their respective target confidence intervals.

[0052] In a fifth aspect, an embodiment of the present invention provides a model building device, the device comprising:

[0053] A sample acquisition module, used to acquire sample evaluation parameter information and sample fault types of the sample evaluation parameter information when the lithium battery is operating abnormally;

[0054] The model training module is used to use the sample evaluation parameter information as sample input and the sample fault type as sample label to train the initial neural network to obtain a safety evaluation network model for implementing any lithium battery safety evaluation method described in the embodiments of the present invention.

[0055] In a sixth aspect, an embodiment of the present invention provides a device for determining a fault mapping relationship, the device comprising:

[0056] A fault mapping module is used to collect statistics on fault records within a first preset period based on known evaluation parameter information and known fault types in the fault records, establish a mapping relationship between known fault types and a set of confidence intervals, and a mapping relationship between known fault types and a single confidence interval, and store the mapping relationship in a mapping relationship library for implementing any of the lithium battery safety assessment methods described in the embodiments of the present invention, wherein the confidence interval includes a parameter value range for each evaluation parameter during abnormal operation.

[0057] In a seventh aspect, an embodiment of the present invention provides an electronic device, comprising:

[0058] one or more processors; and

[0059] A memory storing computer program instructions, wherein the computer program instructions, when executed, cause the processor to perform the steps of the method as described in any one of the embodiments of the present invention.

[0060] In an eighth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the storage medium stores a computer program, and the computer program is used to execute the steps of any one of the methods described in the embodiments of the present invention.

[0061] In a ninth aspect, an embodiment of the present invention provides a computer program product, comprising a computer program / instruction, which, when executed by a processor, implements the steps of any one of the methods described in the embodiments of the present invention.

[0062] The embodiment of the present invention realizes the fusion analysis of multi-source heterogeneous parameter data by integrating the warning fault types and the confidence intervals corresponding to the warning fault types output by the safety assessment network model, thereby improving the accuracy of lithium battery safety assessment and providing support for the subsequent construction of a more comprehensive assessment model.

[0063] By setting confidence intervals for each evaluation parameter for energy storage power stations, we can discover the data distribution characteristics of lithium batteries under different warning fault states, as well as the data distribution characteristics of different lithium batteries under the same warning fault state. By regularly analyzing the correlation between the confidence interval of each evaluation parameter and the fault types in the battery fault feature library, and calculating the probability of occurrence of each type of fault under different confidence intervals, we can clarify the close connection between the evaluation parameters of each confidence interval and the fault type. This allows us to accurately locate the relationship between the probability of fault occurrence and the relevant parameter intervals, providing a quantitative basis for subsequent fault warnings.

[0064] In addition, even if the operating environment of the lithium battery changes, resulting in changes in the combination of evaluation parameter factors and the warning fault type, since the close relationship within a period of time can be regularly mined, the close relationship after the change can also be mined in a timely manner.

[0065] Based on this, a set of confidence intervals that are strongly correlated with the lithium battery warning fault type in a single energy storage power station was excavated, thereby obtaining a multi-factor combination that leads to the warning fault type. Compared with a single parameter, it can more accurately locate the fault type, stage and degree, reduce false alarms and missed alarms, and improve the accuracy of fault judgment.

[0066] Finally, the confidence interval assessment information is linked with the output of the machine learning model to form a feedback mechanism. This allows for dynamic assessment of the battery safety status and timely updates of model parameters based on this linked information. Furthermore, during this linkage process, the warning fault type output by the safety assessment network model is matched and verified with the confidence interval or set of confidence intervals for that warning fault type, complementing each other. When conflicting assessment results are found, further verification is performed by comparing the actual and standard trends of the target assessment parameters, improving the accuracy of the warning and avoiding the limitations of a single assessment method.

[0067] By using a network with long short-term memory functions (such as LSTM) suitable for processing time series data to build a safety assessment network model, it is possible to effectively capture the long-term dependencies in the assessment parameters (such as changes in battery temperature and voltage over time), thereby better predicting lithium battery failures.

[0068] After multiple rounds of verification, an early warning notification message can be generated when a fault is confirmed, helping operation and maintenance personnel to take timely measures to prevent the fault from expanding and ensure the safe and stable operation of the energy storage power station.

[0069] When trend comparisons are inconsistent, relevant information is fed back to the client, allowing users to judge the actual situation of the energy storage power station and promptly update the safety assessment network model to adapt to changes in battery operating conditions, continuously improving model accuracy and early warning effects.

[0070] In summary, by adopting the technical solutions of the embodiments of the present invention, the present application can achieve efficient and accurate evaluation and prediction of the state of electrochemical energy storage batteries, solving technical problems such as inaccurate battery state evaluation, insufficient fault prediction capabilities, and low data processing efficiency in the prior art. It can also provide more accurate state evaluation and prediction in complex electrochemical energy storage battery operating environments, so that the safety assessment network model can adapt to battery state changes and environmental fluctuations, thereby achieving more accurate state evaluation and prediction. In order to adapt to the performance of the battery under different working conditions, it ensures that high prediction accuracy and safety can be maintained in various dynamic environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 This is a flow chart of a lithium battery safety assessment method provided by an embodiment of the present invention;

[0072] Figure 2 is a flow chart of a lithium battery safety assessment method provided by another embodiment of the present invention;

[0073] Figure 3 is a flow chart of a lithium battery safety assessment method provided by another embodiment of the present invention;

[0074] Figure 4 is a flow chart of a lithium battery safety assessment method provided by another embodiment of the present invention;

[0075] Figure 5 is a flow chart of a model building method provided by an embodiment of the present invention;

[0076] Figure 6 is a flowchart of a method for determining a mapping relationship provided by an embodiment of the present invention;

[0077] Figure 7 is a flowchart of a method for determining a mapping relationship provided by an embodiment of the present invention;

[0078] Figure 8 is a flowchart of a method for determining a mapping relationship provided by an embodiment of the present invention;

[0079] Figure 9 1 is a schematic structural diagram of a lithium battery safety assessment device provided by an embodiment of the present invention;

[0080] Figure 10 is a structural diagram of a model building device provided by an embodiment of the present invention;

[0081] Figure 11 This is a structural diagram of a device for determining a fault mapping relationship provided by an embodiment of the present invention;

[0082] Figure 12An exemplary structural diagram of the electronic device is disclosed. DETAILED DESCRIPTION

[0083] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0084] Figure 1 This is a flow chart of a lithium battery safety assessment method provided by an embodiment of the present invention. This method is applied to scenarios where lithium batteries are safety assessed. Since lithium batteries are widely used, this method has many application scenarios. It can typically be used in the field of energy storage power stations. This method can be implemented by a lithium battery safety assessment device. Depending on its function and detection requirements, the device can be distributed in different locations. Common locations include the inside of a battery cell, in a battery management system (BMS), in a battery module or battery pack casing, and in connection parts. See Figure 1 , a lithium battery safety assessment method provided by an embodiment of the present invention includes:

[0085] S110: Obtain parameter information of the lithium battery to be evaluated within a period to be evaluated.

[0086] The evaluation period represents a specific time range for obtaining the evaluation parameters of the lithium battery. This time range can be fixed or adjusted according to different situations and conditions.

[0087] The parameter information to be evaluated refers to the parameter information used to perform safety evaluation on lithium batteries. The specific parameter information to be evaluated can be set according to the needs of the actual application scenario.

[0088] The parameter information to be evaluated can be a single data. To improve the evaluation accuracy, the parameter information to be evaluated can be time series information.

[0089] Taking the application scenario of an energy storage power station as an example, battery status information includes the temperature, voltage, charge status, discharge status, internal resistance, number of cycles, ambient humidity, etc. of the lithium battery, and the gas information in the battery compartment includes the concentration information of components such as oxygen, carbon dioxide, carbon monoxide, nitrogen oxides and volatile organic compounds.

[0090] S120: Input the parameter information to be evaluated into a safety assessment network model to obtain a warning fault type of the lithium battery.

[0091] The safety assessment network model is a pre-trained network model used for fault prediction. The early warning fault type refers to the predicted type of lithium battery fault.

[0092] S130. Match the warning fault type with the target confidence interval in a mapping relationship library, wherein the mapping relationship library stores known fault types and confidence intervals corresponding to the known fault types, and the confidence interval includes a parameter value interval of each evaluation parameter during abnormal operation.

[0093] The target confidence interval is the confidence interval that matches the warning fault type. A warning fault type may involve one or more parameters to be evaluated, with each evaluation parameter corresponding to a confidence interval. When a warning fault type involves multiple parameters to be evaluated, multiple target confidence intervals are associated with the warning fault type.

[0094] Known fault types are all known fault types or common fault types that have occurred in lithium batteries. Known fault types include warning fault types.

[0095] S140: Determine whether all the values of the parameters to be evaluated in the parameter information to be evaluated fall within the target confidence interval corresponding to the warning fault type.

[0096] S150: If it is determined that the values of the parameters to be evaluated all fall within their respective target confidence intervals, it is determined that a fault of the warning fault type has occurred.

[0097] The technical solution of the embodiment of the present invention matches and verifies the confidence interval evaluation information with the output results of the machine learning model. The two complement each other, improve the accuracy of the early warning, and avoid the limitations of a single evaluation method.

[0098] Figure 2 FIG. 1 is a flow chart of a lithium battery safety assessment method provided by another embodiment of the present invention. Figure 1 The present application is extended based on the embodiment shown Figure 2 The embodiment shown is described below in detail. Figure 2 The embodiment shown is Figure 1 The differences and similarities between the illustrated embodiments are not described in detail.

[0099] like Figure 2 As shown, in an embodiment of the present invention, after determining whether the values of the parameters to be evaluated in the parameter information to be evaluated all fall within their respective target confidence intervals, the lithium battery safety assessment method further includes:

[0100] S210: If it is determined that the values of the parameter to be evaluated do not all fall within the target confidence interval corresponding to the warning fault type, then the actual change trend of the parameter to be evaluated is compared with the standard change trend of the parameter to be evaluated under the warning fault type.

[0101] Among them, the actual change trend reflects the change trend of the parameter to be evaluated, and the standard change trend reflects the change trend of the evaluation parameter corresponding to the warning fault type.

[0102] Continuing with the application scenario of an energy storage power station as an example, when the warning fault type is a battery overheating fault type, when matching the set of voltage confidence intervals and carbon monoxide confidence intervals, the actual voltage change process collected this time is compared with the standard voltage change process corresponding to the overheating fault type, and the actual carbon monoxide change process collected this time is compared with the standard carbon monoxide change process corresponding to the overheating fault type.

[0103] S220: When the actual change trend of the parameter to be evaluated is consistent with the standard change trend, it is determined that a fault of the warning fault type has occurred.

[0104] It should be noted that when the actual change trend of the parameter to be evaluated does not fall within the standard change trend, it means that the warning result of the evaluation from the perspective of the safety assessment network model is the above-mentioned warning fault type. However, the evaluation from the perspective of the confidence interval cannot assess the above-mentioned warning fault type. The reasons for this situation may be as follows:

[0105] 1) The model parameters of the safety assessment network model are not suitable for the current operating conditions of the battery. For example, if the operating conditions change but the safety assessment network model is not updated, the accuracy of the safety assessment network model will decrease.

[0106] 2) Since the evaluation parameter information is time series information, the safety assessment network model may obtain the potential or early warning fault type through the parameter change trend, while the confidence interval does not predict the result in advance.

[0107] Based on this, when there is a conflict in the evaluation information between the two, verification is carried out through trend comparison.

[0108] Based on the above technical features, the technical solution of the embodiment of the present invention further compares and analyzes the changing trends of the evaluation parameters, and screens out situations that actually belong to the warning fault type from situations where all situations do not fall into their respective target confidence intervals, rather than directly determining situations where all situations do not fall into their respective target confidence intervals as situations where the warning fault type cannot be evaluated, thereby further improving the accuracy of the evaluation.

[0109] Figure 3 FIG. 1 is a flow chart of a lithium battery safety assessment method provided by another embodiment of the present invention. Figure 2 The present application is extended based on the embodiment shown Figure 3 The embodiment shown is described below in detail. Figure 3 The embodiment shown is Figure 2 The differences and similarities between the illustrated embodiments are not described in detail.

[0110] like Figure 3As shown, in an embodiment of the present invention, if it is determined that the values of the parameter to be evaluated do not all fall within the target confidence interval corresponding to the warning fault type, then after comparing the actual change trend of the parameter to be evaluated with the standard change trend of the parameter to be evaluated under the warning fault type, the lithium battery safety assessment method further includes:

[0111] S310: When the actual change trend of the parameter to be evaluated is inconsistent with the standard change trend, it is determined that a suspected fault has occurred, and the target confidence interval, the value of the parameter to be evaluated, the change trend comparison information of the parameter to be evaluated, and the warning fault type are sent to the client for fault confirmation.

[0112] If the actual change trend of the parameter to be evaluated is inconsistent with the standard change trend, it means that the actual change trend of the evaluation parameter value is different from the standard change trend. At this time, since the target confidence interval cannot evaluate the warning fault type, the trend comparison cannot evaluate the warning fault type either. It can be highly concluded that the lithium battery does not have this warning fault type. The reason is most likely that the safety assessment network model is used to assess the current operating status of the lithium battery and has insufficient accuracy. Therefore, the relevant information is fed back to the client, and the user actually evaluates whether the lithium battery has the warning fault type based on the feedback information. When it is confirmed that the safety assessment network model needs to be updated, the safety assessment network model is updated in a timely manner.

[0113] Based on this technical feature, the technical solution provided by the embodiment of the present invention enables timely manual intervention in situations where the type of warning fault cannot be assessed, thereby reducing safety risks.

[0114] Figure 4 FIG. 1 is a flow chart of a lithium battery safety assessment method provided by another embodiment of the present invention. Figure 1 The present application is extended based on the embodiment shown Figure 4 The embodiment shown is described below in detail. Figure 4 The embodiment shown is Figure 1 The differences and similarities between the illustrated embodiments are not described in detail.

[0115] like Figure 4 As shown, in an embodiment of the present invention, in order to improve the accuracy of fault prediction, before inputting the parameter information to be evaluated into the safety assessment network model, the lithium battery safety assessment method includes:

[0116] S410: Preprocess the parameter information to be evaluated, where the preprocessing includes at least one of data deduplication, missing value filling, and outlier detection.

[0117] Specifically, the preprocessing of the parameter information to be evaluated includes:

[0118] Removing duplicate information from the parameter information to be evaluated based on a hash algorithm or a timestamp method;

[0119] and / or, using multiple interpolation methods in combination with time series analysis techniques to infer missing values based on known data in the parameter information to be evaluated, and filling missing values in the parameter information to be evaluated, wherein the multiple interpolation methods include linear interpolation, spline interpolation, and weighted average based on adjacent data;

[0120] And / or, based on a set threshold, performing outlier detection on the parameter information to be evaluated, and removing or replacing the outlier.

[0121] Optionally, to improve the quality of the parameter information to be evaluated, the embodiments of the present invention may also include other preprocessing methods that can be conceived by those skilled in the art based on existing technologies. The above-mentioned algorithms are also only exemplary descriptions, and specific optimizations can be made based on the application scenario or technological development. The embodiments of the present invention are not limited to this.

[0122] Figure 5 This is a flow chart of a model building method provided by an embodiment of the present invention. Figure 1 The present application is extended based on the embodiment shown Figure 5 The embodiment shown is described below in detail. Figure 5 The embodiment shown is Figure 1 The differences and similarities between the illustrated embodiments are not described in detail.

[0123] like Figure 5 As shown, a model building method provided by an embodiment of the present invention includes:

[0124] S510: Obtain sample evaluation parameter information of a lithium battery during abnormal operation and a sample fault type of the sample evaluation parameter information.

[0125] S520: Using the sample evaluation parameter information as sample input and the sample fault type as a sample label, the initial neural network is trained to obtain a safety evaluation network model. Based on the model, any lithium battery safety evaluation method described in the embodiments of the present invention can be implemented.

[0126] The initial neural network can be any neural network, but since the evaluation parameter information contains time series data, such as changes in battery temperature and voltage over a period of time, RNNs and their variants are excellent choices. For example, the Long Short-Term Memory (LSTM) network is capable of effectively processing long-term dependencies in long-sequence data. Therefore, the initial neural network uses a network with LSTM capabilities to process long-term dependencies in the time series data of the sample evaluation parameter information.

[0127] The initial neural network can be configured based on actual needs. Optionally, the number of input layer neurons can be determined based on the number of evaluation parameters. The number of units in the LSTM hidden layer can be appropriately configured based on the complexity of the data and the length of the sequence. It should be noted that longer sequences require more units. The number of neurons in the output layer depends on the number of predefined fault types. The output of each neuron represents the probability of the corresponding fault type, and the sum of the probabilities of all output neurons is 1.

[0128] The number of fault types refers to the number of all possible fault types that may occur in lithium batteries under operating conditions, or the number of common fault types, which can be set according to actual needs.

[0129] Figure 6 This is a flow chart of a method for determining a mapping relationship provided by an embodiment of the present invention. Figure 1 The present application is extended based on the embodiment shown Figure 6 The embodiment shown is described below in detail. Figure 6 The embodiment shown is Figure 1 The differences and similarities between the illustrated embodiments are not described in detail.

[0130] like Figure 6 As shown, an embodiment of the present invention provides a method for determining a mapping relationship, including:

[0131] S610. Based on the known evaluation parameter information and known fault types in the fault records, statistics are collected on the fault records within the first preset period, a mapping relationship between the known fault types and the confidence interval set, and a mapping relationship between the known fault types and the single confidence interval is established, and the mapping relationship is stored in a mapping relationship library. Based on the mapping relationship library, any lithium battery safety assessment method described in the embodiments of the present invention can be implemented, wherein the confidence interval includes the parameter value range of each evaluation parameter during abnormal operation.

[0132] The known evaluation parameter information is parameter information included in the fault record and used to determine the fault type described in the fault record. The known fault type is the type of fault that occurred in the fault record. Both the known evaluation parameter information and the known fault type are known and stored in the fault record.

[0133] The first preset period may be the same as or different from the above-mentioned period to be evaluated.

[0134] Figure 7 This is a flow chart of a method for determining a mapping relationship provided by an embodiment of the present invention. Figure 6 The present application is extended based on the embodiment shown Figure 7 The embodiment shown is described below in detail. Figure 7 The embodiment shown is Figure 6 The differences and similarities between the illustrated embodiments are not described in detail.

[0135] like Figure 7 As shown, in an embodiment of the present invention, based on the known evaluation parameter information and known fault types in the fault records, statistics are collected on the fault records within a first preset period, and a mapping relationship between the known fault types and a set of confidence intervals, as well as a mapping relationship between the known fault types and a single confidence interval, is established, including:

[0136] S710 . Calculate, based on each fault record within a first preset period, the probability of occurrence of different fault types within confidence intervals corresponding to different evaluation parameters.

[0137] Continuing with the example of an energy storage power station, fault types can include overcharging, overdischarging, internal and external short circuits, etc. Evaluation parameters include battery status parameters and gas parameters. Battery status parameters include the lithium battery's temperature, voltage, charge state, discharge state, internal resistance, cycle count, and ambient humidity. Battery compartment gas parameters include the concentrations of oxygen, carbon dioxide, carbon monoxide, nitrogen oxides, and volatile organic compounds.

[0138] The confidence interval can be a parameter value interval of each evaluation parameter during abnormal operation, obtained by a technician in this field by performing an operation analysis on the battery operating conditions of the energy storage power station using simulation software of the energy storage power station.

[0139] In an energy storage power station, each battery with different operating conditions requires separate operational analysis to determine the parameter value ranges for each evaluation parameter under abnormal operation for batteries under the same operating conditions. In other words, batteries with different operating conditions have different parameter value ranges for each evaluation parameter under abnormal operation.

[0140] In addition, the confidence interval of each evaluation parameter of the same lithium battery will also change dynamically with the battery operating conditions.

[0141] S720 , sort the fault types in each confidence interval according to the probability values.

[0142] Within the confidence interval, the higher the probability of the fault type, the more closely the evaluation parameters of the confidence interval are related to the fault type, and the more closely the confidence interval is related to the fault type.

[0143] S730: Extract a preset number of known fault types that rank top in each confidence interval.

[0144] S740: Determine a set of confidence intervals having the same known fault type, and establish a mapping relationship between the known fault type and the confidence interval set, as well as a mapping relationship between the known fault type and a single confidence interval.

[0145] Among them, when there are confidence intervals with the same known fault type, it is necessary to establish a mapping relationship between the known fault type and the confidence interval set; when there are confidence intervals with different known fault types, it is necessary to establish a mapping relationship between the known fault type and a single confidence interval.

[0146] When there are multiple confidence intervals for a known fault type, it indicates that the multi-factor combination between the evaluation parameters corresponding to the multiple confidence intervals can more accurately locate the fault type.

[0147] For example, a single temperature parameter might only roughly determine whether the battery is overheating, but it's difficult to determine the extent of the overheating or the stage of the fault. For example, a temperature increase could be a temporary rise due to normal workload fluctuations, or it could be an early sign of a fault.

[0148] However, when combined with the CO concentration, if the temperature is just beginning to rise and the CO concentration also rises slightly, this may indicate that the fault is in the early stages, with mild signs of thermal runaway beginning to appear within the battery. As the temperature and CO concentration continue to rise, it can be inferred that the fault is gradually developing. When both are high, it is likely that the battery is already in the stage of severe overheating failure, and more serious damage or safety accidents are imminent. This comprehensive assessment of multiple factors can more accurately pinpoint the stage and extent of the fault.

[0149] From a fault warning perspective, relying solely on a single factor can lead to false alarms or missed warnings. For example, relying solely on temperature warnings can cause the temperature to temporarily exceed the normal range under certain operating conditions, such as high ambient temperatures or after a high-power battery discharge, even though the battery is not actually faulty. Similarly, carbon monoxide concentrations can fluctuate due to external factors, such as a brief ventilation system failure.

[0150] Therefore, when considering the correlation between the two and their confidence intervals, the probability of false alarms can be greatly reduced. For example, when outputting an overheating fault warning through the safety network assessment model, it can be verified that both the battery temperature and the carbon monoxide concentration are within the confidence interval, which can greatly reduce the probability of false alarms.

[0151] This application integrates confidence intervals with the output of machine learning models to form a feedback mechanism. This allows for dynamic assessment of battery safety status and timely updates of model parameters based on this information. However, confidence intervals are discrete features and cannot effectively analyze time series changes. Therefore, confidence intervals are not suitable for use solely to assess battery warning fault types.

[0152] Figure 8 This is a flow chart of a method for determining a mapping relationship provided by an embodiment of the present invention. Figure 7 The present application is extended based on the embodiment shown Figure 8 The embodiment shown is described below in detail. Figure 8 The embodiment shown is Figure 7 The differences and similarities between the illustrated embodiments are not described in detail.

[0153] like Figure 8 As shown, in an embodiment of the present invention, based on each fault record within the first preset period, the probability of occurrence of different fault types under the confidence intervals corresponding to different evaluation parameters is statistically analyzed, including:

[0154] S810: Extract each fault record within a first preset period from a fault feature library, where each fault record includes evaluation parameter information and a fault type corresponding to the fault.

[0155] Each fault record includes the evaluation parameter information corresponding to the fault and the corresponding fault type. For example, a fault record shows: battery voltage is 3.3V, battery internal resistance is 0.2Ω, carbon dioxide gas concentration is 15ppm, and the fault type is battery overheating.

[0156] S820: Count the fault types for the confidence interval of each evaluation parameter according to each fault record.

[0157] The counting process is as follows:

[0158] Determine whether each evaluation parameter value falls within its respective confidence interval in a single fault record.

[0159] When it falls into the confidence interval, the number of occurrences of the fault type in a single fault record in the confidence space is increased by 1.

[0160] For example, if the fault type of a single fault record is fault type B, and the battery voltage of 3.3V falls within the voltage confidence interval, the count for fault type B in the voltage confidence interval is incremented by 1. Assuming the count is 0, the count for fault type B in the voltage confidence interval becomes 1. Similarly, if the carbon dioxide concentration falls within the carbon dioxide confidence interval, the count for fault type B in the carbon dioxide confidence interval is incremented by 1. If the battery internal resistance does not fall within the battery internal resistance confidence interval, there is no need to increment the count for fault type B by 1.

[0161] S830. Count the total number of occurrences of all fault types within each confidence interval.

[0162] S840. Calculate the ratio between the number of occurrences of a single fault type in a single confidence interval and the total number of occurrences of all fault types in the single confidence interval to obtain the probability of occurrence of the single fault type in the single confidence interval.

[0163] Figure 9 This is a schematic diagram of the structure of a lithium battery safety assessment device provided by an embodiment of the present invention. Figure 9 The lithium battery safety assessment device provided by the embodiment of the present invention includes: an information acquisition module 10 , a fault prediction module 20 , an interval determination module 30 , an interval matching module 40 and a fault determination module 50 .

[0164] The information acquisition module 10 is used to obtain the parameter information of the lithium battery to be evaluated during the evaluation period;

[0165] A fault prediction module 20 is used to input the parameter information to be evaluated into a safety assessment network model to obtain a warning fault type of the lithium battery;

[0166] An interval determination module 30 is configured to match the warning fault type with a target confidence interval in a mapping relationship library; wherein the mapping relationship library stores known fault types and confidence intervals corresponding to the known fault types, and the confidence intervals include parameter value intervals for each evaluation parameter during abnormal operation;

[0167] An interval matching module 40 is used to determine whether all the values of the parameters to be evaluated in the parameter information to be evaluated fall within the target confidence interval corresponding to the warning fault type;

[0168] The fault determination module 50 is configured to determine that a fault of the warning fault type has occurred if it is determined that the values of the parameters to be evaluated all fall within their respective target confidence intervals.

[0169] Furthermore, the lithium battery safety assessment device further includes:

[0170] A trend comparison module is configured to, after determining whether all the values of the parameters to be evaluated in the parameter information to be evaluated fall within their respective target confidence intervals, compare the actual change trend of the parameters to be evaluated with the standard change trend of the parameters to be evaluated under the warning fault type if it is determined that all the values of the parameters to be evaluated do not fall within the target confidence interval corresponding to the warning fault type;

[0171] When the actual change trend of the parameter to be evaluated is consistent with the standard change trend, it is determined that a fault of the warning fault type has occurred.

[0172] Furthermore, the lithium battery safety assessment device further includes:

[0173] The fault reporting module is used to compare the actual change trend of the parameter to be evaluated with the standard change trend of the parameter to be evaluated under the warning fault type when not all of them fall into the category. When the actual change trend of the parameter to be evaluated and the standard change trend are inconsistent, it is determined that a suspected fault has occurred, and the target confidence interval, the value of the parameter to be evaluated, the change trend comparison information of the parameter to be evaluated and the warning fault type are sent to the client for fault confirmation.

[0174] Furthermore, the lithium battery safety assessment device further includes:

[0175] A preprocessing module, used in the method according to claim 1, characterized in that before the parameter information to be evaluated is input into the security assessment network model, the parameter information to be evaluated is preprocessed, and the preprocessing includes at least one of data deduplication, missing value filling, and outlier detection.

[0176] Specifically, the preprocessing of the parameter information to be evaluated includes:

[0177] Removing duplicate information from the parameter information to be evaluated based on a hash algorithm or a timestamp method;

[0178] and / or, using multiple interpolation methods in combination with time series analysis techniques to infer missing values based on known data in the parameter information to be evaluated, and filling missing values in the parameter information to be evaluated, wherein the multiple interpolation methods include linear interpolation, spline interpolation, and weighted average based on adjacent data;

[0179] And / or, based on a set threshold, performing outlier detection on the parameter information to be evaluated, and removing or replacing the outlier.

[0180] Figure 10 This is a schematic diagram of the structure of a model building device provided by an embodiment of the present invention. Figure 10 , a model building device provided by an embodiment of the present invention includes: a sample acquisition module 60 and a model training module 70.

[0181] The sample acquisition module 60 is used to obtain sample evaluation parameter information and sample fault types of the sample evaluation parameter information when the lithium battery is operating abnormally;

[0182] The model training module 70 is used to use the sample evaluation parameter information as sample input and the sample fault type as sample label to train the initial neural network to obtain a safety evaluation network model for implementing any lithium battery safety evaluation method described in any embodiment of the present invention.

[0183] Furthermore, the initial neural network adopts a network with long-short-term memory function to process the long-term dependency of the time series data in the sample evaluation parameter information.

[0184] Furthermore, the model building device further includes:

[0185] An input layer determination module is configured to use the sample evaluation parameter information as sample input and the sample fault type as a sample label to train the initial neural network, and determine the number of input layer neurons based on the number of evaluation parameter types in the sample evaluation parameter information before obtaining a safety assessment network model;

[0186] A hidden layer determination module, configured to determine the number of units in the hidden layer based on the data complexity and time series length of the sample evaluation parameter information;

[0187] A neuron determination module, configured to determine the number of neurons in the output layer according to the number of fault types in the sample fault types;

[0188] The neural network determination module is used to determine the initial neural network according to the number of neurons in the input layer, the number of units in the hidden layer, and the number of neurons in the output layer.

[0189] Figure 11 This is a structural diagram of a device for determining a fault mapping relationship provided by an embodiment of the present invention. Figure 11 , an embodiment of the present invention provides a device for determining a fault mapping relationship, the device including: a fault mapping module 80.

[0190] Among them, the fault mapping module 80 is used to perform statistics on the fault records within a first preset period based on the known evaluation parameter information and known fault types in the fault records, establish a mapping relationship between the known fault types and the confidence interval set, and a mapping relationship between the known fault type and a single confidence interval, store the mapping relationship in a mapping relationship library, and obtain a mapping relationship library for implementing any of the lithium battery safety assessment methods described in the embodiments of the present invention, wherein the confidence interval includes the parameter value interval of each evaluation parameter during abnormal operation.

[0191] Furthermore, the fault mapping module includes:

[0192] A probability statistics unit, configured to calculate, based on each fault record within a first preset period, the probability of occurrence of different fault types within confidence intervals corresponding to different evaluation parameters;

[0193] A sorting unit is used to sort the fault types in each confidence interval according to the probability value;

[0194] A type extraction unit, configured to extract a preset number of known fault types ranked top in each confidence interval;

[0195] The relationship establishing unit is used to determine a set of confidence intervals with the same known fault type, establish a mapping relationship between the known fault type and the confidence interval set, and a mapping relationship between the known fault type and a single confidence interval.

[0196] Furthermore, it is characterized in that the probability statistics unit is used to:

[0197] Extracting each fault record within a first preset period from the fault feature library, each fault record including evaluation parameter information and fault type corresponding to the fault;

[0198] According to each fault record, the fault type count is performed for the confidence interval of each evaluation parameter;

[0199] Count the total number of occurrences of all fault types within each confidence interval;

[0200] The probability of a single fault type occurring in a single confidence interval is obtained by calculating the ratio between the number of occurrences of a single fault type in a single confidence interval and the total number of occurrences of all fault types in a single confidence interval.

[0201] Furthermore, counting the fault types for the confidence interval of each evaluation parameter according to each fault record includes:

[0202] Determine whether each evaluation parameter value falls within its respective confidence interval in a single fault record;

[0203] When falling into the confidence interval, the number of occurrences of the fault type in a single fault record in the confidence space is accumulated.

[0204] In addition, some embodiments of the present application further provide an electronic device. The electronic device may be various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, etc. The electronic device may also be various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices.

[0205] The electronic device includes: one or more processors; and a memory storing computer program instructions, wherein the computer program instructions, when executed, enable the processor to perform the steps of the method provided in any one or more of the above embodiments. Figure 12 An exemplary structural diagram of the electronic device is disclosed. Figure 12 As shown, the electronic device includes: one or more processors 1101, a memory 1102, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. The various components are connected to each other using different buses and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed in the electronic device, including instructions stored in or on the memory to display graphical information of a GUI on an external input / output device (such as a display device coupled to the interface). In some other embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple electronic devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Among them, the components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described and / or required herein.

[0206] The electronic device may further include: an input device 1103 and an output device 1104. The processor 1101, the memory 1102, the input device 1103 and the output device 1104 may be connected via a bus or other means. Figure 12 The bus connection is taken as an example.

[0207] The input device 1103 can receive input digital or character information and generate key signal input related to user settings and function control of the electronic device, such as input devices such as a touch screen, a keypad, a mouse, a trackpad, a touch pad, an indicator stick, one or more mouse buttons, a trackball, and a joystick. The output device 1104 can include a display device, an auxiliary lighting device (e.g., an LED), and a tactile feedback device (e.g., a vibration motor). The display device can include, but is not limited to, a liquid crystal display (LCD), a light emitting diode (LED) display, and a plasma display. In some embodiments, the display device can be a touch screen.

[0208] To provide interaction with a user, the electronic device may be a computer. The computer may include: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball), through which the user can provide input to the computer. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0209] In the embodiments of the present application, a computer program / instruction is stored on a computer-readable medium. When executed by a processor, the computer program / instruction implements the steps of the method provided in any one or more of the above embodiments. The computer-readable medium may be included in the electronic device described in the above embodiments, or it may exist independently and not be incorporated into the device. The computer-readable medium carries one or more computer-readable instructions.

[0210] The memory 1102 can be used as a non-transitory computer-readable storage medium to store non-transitory software programs, non-transitory computer executable programs, and modules. The processor 1101 executes the non-transitory software programs, instructions, and modules stored in the memory 1102 to execute various functional applications and data processing of the server, thereby implementing the program instructions / modules corresponding to the method provided in any one or more of the above embodiments of the present application.

[0211] The memory 1102 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and applications required for at least one function; the data storage area may store data created based on the use of the electronic device, etc. In addition, the memory 1102 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 1102 may optionally include a memory remotely located relative to the processor 1101, and these remote memories may be connected to the electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0212] It should be noted that the computer-readable medium described in this application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared or semiconductor system, device or component, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this application, a computer-readable medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device.

[0213] Computer-readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology for information storage. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc-read only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

[0214] Computer program code for performing the operations of the present application can be written in one or more programming languages, or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on the remote computer or server. In cases involving a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0215] In the above-described embodiment, can realize wholly or in part by software, hardware, firmware or its arbitrary combination.For example, can adopt application-specific integrated circuit (ASIC), general computer or any other similar hardware device to realize.In certain embodiments, the software program of the present application can be carried out to realize above steps or function by processor.Similarly, the software program of the present application (comprising relevant data structure) can be stored in computer-readable recording medium, for example, RAM memory, magnetic or optical drive or floppy disk and similar device.In addition, some steps or functions of the present application can adopt hardware to realize, for example, as the circuit that cooperates with processor to perform each step or function.

[0216] The computer program product provided by the embodiment of the present application includes one or more computer programs / instructions, and when the computer program / instructions are executed by the processor, all or part of the process or function described in the embodiment of the present application is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instruction can be stored in a computer-readable storage medium, or transmitted from a computer-readable storage medium to another computer-readable storage medium. For example, the computer instruction can be transmitted from a website site, a computer, a server or a data center by wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or a data center that includes one or more available media integrations. The available medium can be a magnetic medium, (such as a floppy disk, a hard disk, a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid-state hard disk (SSD)).

[0217] The flowcharts or block diagrams in the accompanying drawings illustrate the possible architectures, functions and operations of the devices, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, program segment or part of code, and the module, program segment or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, as well as the combination of boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-specific system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0218] The scope of this application is defined by the appended claims rather than the foregoing description and is therefore intended to encompass within this application all changes that come within the meaning and range of equivalents of the claims. Any reference signs in the claims should not be construed as limiting the claims to which they relate. In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices stated in a device claim may also be implemented by one unit or device through software or hardware. Words such as "first" and "second" are only used to distinguish the description and do not indicate any particular order, nor should they be understood as indicating or implying relative importance.

[0219] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art may easily propose variations or substitutions within the technical scope disclosed in the present application, and such variations or substitutions shall be encompassed within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be subject to the scope of protection of the claims, and the above embodiments shall be regarded as exemplary and non-limiting.

Claims

1. A lithium battery safety assessment method, characterized in that: The method comprises: Obtaining parameter information of the lithium battery to be evaluated during the evaluation period; Inputting the parameter information to be evaluated into a safety assessment network model to obtain a warning fault type of the lithium battery; In a mapping relationship library, the warning fault type is matched with a target confidence interval; wherein the mapping relationship library stores known fault types and confidence intervals corresponding to the known fault types, and the confidence intervals include parameter value intervals of each evaluation parameter during abnormal operation; Determining whether all the values of the parameters to be evaluated in the parameter information to be evaluated fall within the target confidence interval corresponding to the warning fault type; If it is determined that the values of the parameters to be evaluated all fall within their respective target confidence intervals, it is determined that a fault of the warning fault type has occurred.

2. The method according to claim 1, characterized in that After determining whether all the values of the parameters to be evaluated in the parameter information to be evaluated fall within their respective target confidence intervals, the method further includes: If it is determined that the values of the parameter to be evaluated do not all fall within the target confidence interval corresponding to the warning fault type, then the actual change trend of the parameter to be evaluated is compared with the standard change trend of the parameter to be evaluated under the warning fault type; When the actual change trend of the parameter to be evaluated is consistent with the standard change trend, it is determined that a fault of the warning fault type has occurred.

3. The method according to claim 2, characterized in that When both parameters do not fall into the category, after comparing the actual change trend of the parameter to be evaluated with the standard change trend of the parameter to be evaluated under the early warning fault type, the method further includes: When the actual change trend of the parameter to be evaluated is inconsistent with the standard change trend, a suspected fault is determined to have occurred, and the target confidence interval, the value of the parameter to be evaluated, the change trend comparison information of the parameter to be evaluated, and the warning fault type are sent to the client for fault confirmation.

4. The method according to claim 1, wherein Before inputting the parameter information to be evaluated into the security assessment network model, the method further includes: The parameter information to be evaluated is preprocessed, and the preprocessing includes at least one of data deduplication, missing value filling, and outlier detection.

5. The method according to claim 4, characterized in that The preprocessing of the parameter information to be evaluated includes: Removing duplicate information from the parameter information to be evaluated based on a hash algorithm or a timestamp method; and / or, using multiple interpolation methods in combination with time series analysis techniques to infer missing values based on known data in the parameter information to be evaluated, and filling missing values in the parameter information to be evaluated, wherein the multiple interpolation methods include linear interpolation, spline interpolation, and weighted average based on adjacent data; And / or, based on a set threshold, performing outlier detection on the parameter information to be evaluated, and removing or replacing the outlier.

6. A model building method, characterized in that: The method comprises: Obtaining sample evaluation parameter information of a lithium battery during abnormal operation and a sample fault type corresponding to the sample evaluation parameter information; The sample evaluation parameter information is used as sample input, and the sample fault type is used as a sample label to train the initial neural network to obtain a safety evaluation network model for implementing the lithium battery safety evaluation method described in any one of claims 1 to 5.

7. The method according to claim 6, characterized in that The initial neural network uses a network with long-term and short-term memory functions to process the long-term dependencies of the time series data in the sample evaluation parameter information.

8. The method according to claim 6, characterized in that Before the sample evaluation parameter information is used as sample input and the sample fault type is used as a sample label to train the initial neural network to obtain the safety evaluation network model, the method further includes: Determining the number of neurons in the input layer according to the number of evaluation parameter types in the sample evaluation parameter information; Determining the number of units in the hidden layer based on the data complexity and time series length of the sample evaluation parameter information; Determining the number of neurons in the output layer according to the number of fault types in the sample fault types; The initial neural network is determined according to the number of neurons in the input layer, the number of units in the hidden layer, and the number of neurons in the output layer.

9. A method for determining a mapping relationship, characterized in that: The method comprises: According to the known evaluation parameter information and known fault types in the fault records, the fault records within the first preset period are statistically analyzed, and a mapping relationship between the known fault types and the confidence interval set, as well as a mapping relationship between the known fault type and the single confidence interval is established. The mapping relationship is stored in a mapping relationship library for implementing the lithium battery safety assessment method described in any one of claims 1 to 5, wherein the confidence interval includes the parameter value range of each evaluation parameter during abnormal operation.

10. The method according to claim 9, characterized in that The method of performing statistics on the fault records within the first preset period based on the known evaluation parameter information and the known fault types in the fault records, and establishing a mapping relationship between the known fault types and a set of confidence intervals, and a mapping relationship between the known fault types and a single confidence interval, includes: According to each fault record within the first preset period, the probability of occurrence of different fault types under the confidence intervals corresponding to different evaluation parameters is calculated; Sort the fault types in each confidence interval according to the probability value; Extracting the top preset number of known fault types for each confidence interval; A set of confidence intervals with the same known fault type is determined, and a mapping relationship between the known fault type and the set of confidence intervals, as well as a mapping relationship between the known fault type and a single confidence interval, is established.

11. A lithium battery safety assessment device, characterized in that: The device comprises: An information acquisition module is used to obtain parameter information of the lithium battery to be evaluated during the evaluation period; A fault prediction module is used to input the parameter information to be evaluated into a safety assessment network model to obtain a warning fault type of the lithium battery; An interval determination module is configured to match the warning fault type with a target confidence interval in a mapping relationship library; wherein the mapping relationship library stores known fault types and confidence intervals corresponding to the known fault types, and the confidence intervals include parameter value intervals for each evaluation parameter during abnormal operation; An interval matching module is used to determine whether the values of the parameters to be evaluated in the parameter information to be evaluated all fall within the target confidence interval corresponding to the warning fault type; The fault determination module is used to determine that a fault of the warning fault type has occurred if it is determined that the values of the parameters to be evaluated all fall within their respective target confidence intervals.

12. A model building device, characterized in that: The device comprises: A sample acquisition module, used to acquire sample evaluation parameter information and sample fault types of the sample evaluation parameter information when the lithium battery is operating abnormally; A model training module is used to take the sample evaluation parameter information as sample input and the sample fault type as sample label to train the initial neural network to obtain a safety evaluation network model for implementing the lithium battery safety evaluation method described in any one of claims 1-5.

13. A device for determining a fault mapping relationship, characterized in that: The device comprises: A fault mapping module is used to perform statistics on fault records within a first preset period based on known evaluation parameter information and known fault types in the fault records, establish a mapping relationship between known fault types and a set of confidence intervals, and a mapping relationship between known fault types and a single confidence interval, and store the mapping relationship in a mapping relationship library to implement the lithium battery safety assessment method described in any one of claims 1 to 5, wherein the confidence interval includes a parameter value range of each evaluation parameter during abnormal operation.

14. An electronic device, characterized in that: The electronic device comprises: one or more processors; and A memory storing computer program instructions, which, when executed, cause the processor to perform the steps of the method according to any one of claims 1 to 10.

15. A computer-readable storage medium, characterized in that The storage medium stores a computer program, and the computer program is used to execute the steps of the method according to any one of claims 1 to 10.

16. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

Images