CN118033443A - Single battery fault detection method, single battery fault detection equipment, single battery fault detection medium and single battery fault detection product - Google Patents
Single battery fault detection method, single battery fault detection equipment, single battery fault detection medium and single battery fault detection product Download PDFInfo
- Publication number
- CN118033443A CN118033443A CN202410243285.5A CN202410243285A CN118033443A CN 118033443 A CN118033443 A CN 118033443A CN 202410243285 A CN202410243285 A CN 202410243285A CN 118033443 A CN118033443 A CN 118033443A
- Authority
- CN
- China
- Prior art keywords
- voltage
- prediction window
- time points
- single battery
- battery
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title abstract description 28
- 238000000034 method Methods 0.000 claims abstract description 33
- 238000012549 training Methods 0.000 claims abstract description 27
- 239000000523 sample Substances 0.000 claims description 25
- 238000004590 computer program Methods 0.000 claims description 21
- 239000000178 monomer Substances 0.000 claims description 15
- 238000012216 screening Methods 0.000 claims description 5
- 238000003491 array Methods 0.000 claims description 3
- 238000004140 cleaning Methods 0.000 claims description 3
- 238000009413 insulation Methods 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 abstract description 4
- 230000002829 reductive effect Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 5
- HBBGRARXTFLTSG-UHFFFAOYSA-N Lithium ion Chemical compound [Li+] HBBGRARXTFLTSG-UHFFFAOYSA-N 0.000 description 4
- 230000000875 corresponding effect Effects 0.000 description 4
- 229910001416 lithium ion Inorganic materials 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 238000004880 explosion Methods 0.000 description 3
- 230000036961 partial effect Effects 0.000 description 3
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 2
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 229910052744 lithium Inorganic materials 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000032683 aging Effects 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 229910021389 graphene Inorganic materials 0.000 description 1
- 229910002804 graphite Inorganic materials 0.000 description 1
- 239000010439 graphite Substances 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000013021 overheating Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Landscapes
- Secondary Cells (AREA)
Abstract
The invention discloses a single battery fault detection method, single battery fault detection equipment, single battery fault detection media and single battery fault detection products, and relates to the technical field of battery fault prediction. According to the method, the LSTM model is used as a prediction model, so that effective modeling analysis can be carried out on time sequence data, and hidden rules and hidden trends can be found; the trend of the voltage of each single battery in the battery pack is reflected by utilizing the single voltage median, the voltage of each single battery is not required to be predicted, and the training and predicting speed of the LSTM model is improved; the predicted value of the median of the single voltage output by the LSTM model is compared with the actual voltage of the single battery to obtain a trend error, the predicted value and the actual value of the median of the single voltage are utilized to adjust a fault threshold value, and fault detection is carried out by combining the trend error, so that the robustness of fault detection is improved, the risk of thermal runaway of the electric vehicle can be timely and accurately found, the probability of accidents caused by the thermal runaway of the electric vehicle is reduced, and the driving safety of the electric vehicle is improved.
Description
Technical Field
The invention relates to the technical field of battery fault prediction, in particular to a single battery fault detection method, single battery fault detection equipment, single battery fault detection media and single battery fault detection products.
Background
With the urgent demands of global energy resource situation and environmental protection, new energy automobiles are increasingly important.
It is becoming the first choice for future modes of transportation. As a main energy source, a power battery is one of core components of a new energy automobile, and currently, the most widely used is a lithium ion power battery, which has the advantages of high energy density, low discharge rate, long service life and the like, and the lithium ion battery uses graphite as a negative electrode to replace metal lithium, so that the safety of the lithium ion battery is greatly improved. However, in extreme cases such as mechanical impact, high temperature, overcharge, etc., lithium ion batteries are prone to thermal runaway and even to explosion and fire. Nowadays, new energy automobiles frequently suffer accidents, thermal runaway caused by power battery faults is a main reason, and overcharge behavior in the working process of the batteries is one of main causes for triggering the thermal runaway of the power batteries. In general, overcharge may result in the following consequences for an electric vehicle: the aging process of the battery is accelerated, the capacity is attenuated, and the inconsistency among the battery units is increased, so that the phenomena of thermal runaway, even fire explosion and the like are easily caused.
The fault detection technology of the lithium battery can discover the fault condition of the battery as early as possible and take corresponding measures to ensure the safety and reliability of the battery. Once the battery is found to have faults, the problems of overheating, short circuit, overcharging or overdischarging of the battery and the like can be avoided through timely treatment, and potential fire or explosion risks are reduced.
Disclosure of Invention
The invention aims to provide a single battery fault detection method, equipment, medium and product, which can realize single battery fault detection and have the characteristics of high detection speed and high robustness.
In order to achieve the above object, the present invention provides the following solutions:
In a first aspect, the present invention provides a method for detecting a failure of a single battery, the method comprising:
Based on historical state data in the sliding window, acquiring a multi-step prediction result by adopting a trained LSTM model; the historical state data in the sliding window comprise the actual values of the total voltage, the SOC value and the single voltage median of the battery pack at different time points in the sliding window, and the multi-step prediction result comprises the predicted values of the single voltage median at different time points in the prediction window, wherein the single voltage median is the median of the voltages of all single batteries in the battery pack;
Detecting and obtaining actual voltages of all single batteries at different time points in a prediction window;
calculating fault thresholds at different time points in the prediction window according to the predicted values of the cell voltage median at different time points in the prediction window and the actual voltages of all the cells;
and determining the fault type of each single battery at different time points in the prediction window according to the predicted value of the single voltage median at different time points in the prediction window, the fault threshold and the actual voltage of each single battery.
Optionally, calculating the fault threshold value at different time points in the prediction window according to the predicted value of the median of the cell voltages at different time points in the prediction window and the actual voltage of each cell, specifically includes:
Calculating actual values of the median of the monomer voltages at different time points in the prediction window according to the actual voltages of the monomer batteries at different time points in the prediction window;
Calculating variable thresholds at different time points in the prediction window by using the following formula according to actual values and predicted values of the median of the monomer voltages at different time points in the prediction window;
Wherein Delta sim is a variable threshold value array, delta sim,t+0 and Delta sim,t+T-1 are variable threshold values of time points t+0 and t+T-1 in a prediction window respectively, T is the starting time of the prediction window, T is the length of the prediction window, V sim,t+0 and V sim,t+T-1 are actual values of the monomer voltage median of time points t+0 and t+T-1 in the prediction window respectively, And/>Predicted values of the median of the monomer voltages at time points t+0 and t+T-1 in the prediction window;
Calculating fault thresholds at different time points in the prediction window according to the variable thresholds at different time points in the prediction window; the fault threshold includes:
Upper threshold for primary failure: 0.12+Δ sim;
lower threshold for primary failure: -0.12+Δ sim;
upper threshold for secondary failure: 0.24+Δ sim;
Lower threshold for secondary failure: -0.24+Δ sim;
upper threshold for three-level fault: 0.36+Δ sim;
Lower threshold of three-level fault: -0.36+Δ sim.
Optionally, determining the fault type of each single battery at different time points in the prediction window according to the predicted value of the single voltage median at different time points in the prediction window, the fault threshold and the actual voltage of each single battery specifically includes:
Calculating the difference value between the actual voltage of each single battery at different time points in the prediction window and the predicted value of the median of the single voltages by using the following formula to obtain the trend error of each single battery at different time points in the prediction window;
Wherein error sim is a trend error array, V 1,t+0 and V 1,t+T-1 are actual voltages of the 1 st single battery at time points t+0 and t+t-1 in the prediction window, V M,t+0 and V M,t+T-1 respectively represent actual voltages of the M-th single battery at time points t+0 and t+t-1 in the prediction window, and M represents the number of single batteries in the battery pack;
according to the trend errors of the single batteries at different time points in the prediction window and the fault threshold values at different time points in the prediction window, determining the fault types of the single batteries at different time points in the prediction window by using the following formula;
Wherein, P m,t+t' is the failure type of the mth single cell at time point t+t ' in the prediction window, P 1、P2 and P 3 are the primary failure, the secondary failure and the tertiary failure respectively, error sim,m,t+t' is the trend error of the mth single cell at time point t+t ' in the prediction window, m=1, 2, M, T ' =0, 1, and T-1.
Optionally, based on the historical state data in the sliding window, obtaining a multi-step prediction result by using the trained LSTM model, and further including:
Acquiring historical charge and discharge data of a battery pack; the historical charge and discharge data includes: total current, total voltage, state of charge, cell voltage, state of charge, mileage, insulation resistance, and probe temperature of the battery pack;
constructing a training sample based on the historical charge-discharge data;
and training the LSTM model by using the training sample to obtain a trained LSTM model.
Optionally, the constructing a training sample based on the historical charge and discharge data specifically includes:
cleaning the historical charge and discharge data to obtain cleaned historical charge and discharge data;
Screening the cleaned historical charge and discharge data by adopting a correlation coefficient method to obtain data with the correlation with the median of the single voltage being greater than a preset threshold value, wherein the data is used as training data, and the training data comprises the total voltage, the SOC value and the single voltage of the battery pack;
constructing an input data structure and an output data structure based on the training data;
intercepting the input data structure by utilizing a sliding window to obtain an input data sample;
intercepting the output data structure by using a prediction window to obtain the actual output corresponding to the input data sample;
and constructing a training sample by using the data input sample and the actual output corresponding to the data input sample.
Optionally, the input data structure is:
The output data structure is as follows:
Wherein X u is an input data structure, Y u is an output data structure, totalV 0、total V1 and total V N are the total voltages of the battery packs at time 0, time 1 and time N respectively, SOC 0、SOC1 and SOC N are the SOC values of the battery packs at time 0, time 1 and time N respectively, V 1,0、V1,1、V1,N is the voltage value of the 1 st single battery in the battery packs at time 0, time 1 and time N respectively, V 2,0、V2,1 and V 2,N are the voltage values of the 2 nd single battery in the battery packs at time 0, time 1 and time N respectively, V M,0、VM,1 and V M,N are the voltage values of the Mth single battery in the battery pack at the time 0, the time 1 and the time N respectively, V 1,L and V 1,N+S-1 are the voltage values of the 1 st single battery in the battery pack at the time L and the time N+S-1 respectively, V 2,L and V 2,N+S-1 are the voltage values of the 2 nd single battery in the battery pack at the time L and the time N+S-1 respectively, V M,L and V M,N+S-1 are the voltage values of the Mth single battery in the battery pack at the time L and the time N+S-1 respectively, M is the number of single batteries in the battery pack, N is the number of arrays in the input data structure, L is the length of a sliding window, and S is the length of a prediction window.
In a second aspect, the present invention provides a computer device comprising: a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor executing the computer program to perform the steps of the method described above.
In a third aspect, the present invention provides a computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the above method.
In a fourth aspect, the invention provides a computer program product comprising a computer program which, when executed by a processor, carries out the steps of the method described above.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
The embodiment of the invention provides a single battery fault detection method, equipment, medium and product, wherein the method utilizes an LSTM model as a prediction model, can effectively model and analyze time sequence data, and can find hidden rules and trends; the trend of the voltage of each single battery in the battery pack is reflected by utilizing the single voltage median, the voltage of each single battery is not required to be predicted, and the training and predicting speed of the LSTM model is improved; the predicted value of the median of the single voltage output by the LSTM model is compared with the actual voltage of the single battery to obtain a trend error, the predicted value and the actual value of the median of the single voltage are utilized to adjust a fault threshold value, and fault detection is carried out by combining the trend error, so that the robustness of fault detection is improved, the risk of thermal runaway of the electric vehicle can be timely and accurately found, the probability of accidents caused by the thermal runaway of the electric vehicle is reduced, and the driving safety of the electric vehicle is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a method for detecting a single battery fault according to an embodiment of the present invention;
Fig. 2 is a schematic diagram of a method for detecting faults of a single battery according to an embodiment of the present invention;
FIG. 3 is a correlation coefficient screening feature diagram provided in an embodiment of the present invention;
FIG. 4 is a schematic diagram of a normal vehicle operation provided in an embodiment of the present invention;
FIG. 5 is a schematic illustration of a thermal runaway vehicle operating condition provided by an embodiment of the present invention;
FIG. 6 is an enlarged partial view of a primary failure of a thermal runaway vehicle operating condition provided by an embodiment of the present invention;
FIG. 7 is an enlarged partial view of a three-level fault of a thermal runaway vehicle operating condition provided by an embodiment of the present invention;
Fig. 8 is an internal structure diagram of a computer device according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The invention aims to provide a single battery fault detection method, equipment, medium and product, which can realize single battery fault detection and have the characteristics of high detection speed and high robustness.
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description.
Example 1
As shown in fig. 1, the method for detecting a single battery fault in this embodiment includes:
Step 101, obtaining a multi-step prediction result by adopting a trained LSTM model based on historical state data in a sliding window; the historical state data in the sliding window comprises the actual values of the total voltage, the SOC value and the median of the single voltages of the battery packs at different time points in the sliding window, and the multi-step prediction result comprises the predicted values of the median of the single voltages at different time points in the prediction window, wherein the median of the single voltages is the median of the voltages of all single batteries in the battery packs.
The LSTM model in the embodiment of the present invention is obtained based on historical data training, as shown in fig. 2, and the training method includes the following steps:
(1) And downloading historical charge and discharge data from the new energy automobile big data platform. The embodiment of the invention downloads the data of a normal vehicle and the data of a thermal runaway vehicle.
(2) And (5) data cleaning. The data for the obvious anomalies are excluded using threshold rules.
(3) And screening input characteristic parameters by a correlation coefficient method. The characteristics of total current, total voltage, state of charge, single voltage, state of charge, driving distance, insulation resistance and probe temperature are selected, the correlation coefficients thereof are calculated, fig. 3 is a correlation coefficient characteristic screening diagram, the absolute value of the correlation coefficient is considered to have strong correlation at 0.8-1, and then the total voltage, SOC and single voltage which are strongly correlated with the voltage of the single battery are selected as input characteristics to predict the single voltage median under the time sequence.
(4) And constructing a data structure.
Wherein X u is an input data structure, Y u is an output data structure, totalV 0、total V1 and total V N are the total voltages of the battery packs at time 0, time 1 and time N respectively, SOC 0、SOC1 and SOC N are the SOC values of the battery packs at time 0, time 1 and time N respectively, V 1,0、V1,1、V1,N is the voltage value of the 1 st single battery in the battery packs at time 0, time 1 and time N respectively, V 2,0、V2,1 and V 2,N are the voltage values of the 2 nd single battery in the battery packs at time 0, time 1 and time N respectively, V M,0、VM,1 and V M,N are the voltage values of the Mth single battery in the battery pack at the time 0, the time 1 and the time N respectively, V 1,L and V 1,N+S-1 are the voltage values of the 1 st single battery in the battery pack at the time L and the time N+S-1 respectively, V 2,L and V 2,N+S-1 are the voltage values of the 2 nd single battery in the battery pack at the time L and the time N+S-1 respectively, V M,L and V M,N+S-1 are the voltage values of the Mth single battery in the battery pack at the time L and the time N+S-1 respectively, M is the number of single batteries in the battery pack, N is the number of arrays in the input data structure, L is the length of a sliding window, and S is the length of a prediction window.
Illustratively, the training window, the sliding window and the prediction window are respectively 360, 6 and 6, the data input sample and the actual output (i.e. label) corresponding to the data input sample are intercepted in the data structure, and the training sample is constructed.
The data input sample may be intercepted in a non-repeated manner, or may be intercepted in a partially repeated manner, and the non-repeated interception is described below as an example.
Adopting a repeated interception-free mode, and obtaining an ith training sample as follows: (X i,Yi).
And training the LSTM model by using the training sample. Parameters of the LSTM model are shown in table 1.
TABLE 1 LSTM-based monomer voltage median prediction model parameter setting table
Step 102, detecting and obtaining the actual voltages of the single batteries at different time points in the prediction window.
And step 103, calculating fault thresholds at different time points in the prediction window according to the predicted values of the cell voltage median at different time points in the prediction window and the actual voltages of the cells.
Wherein the failure threshold comprises two parts (fixed threshold + variable threshold):
the upper and lower threshold values of the first-level fault are: + -0.12 + Delta sim.
The upper and lower threshold values of the secondary faults are: + -0.24 + delta sim.
The upper and lower threshold values of the three-level fault are: + -0.36 + delta sim.
Where + -0.12, + -0.24, and + -0.36 are fixed thresholds and Delta sim is a variable threshold.
Wherein Delta sim is a variable threshold value array, delta sim,t+0 and Delta sim,t+T-1 are variable threshold values of time points t+0 and t+T-1 in a prediction window respectively, T is the starting time of the prediction window, T is the length of the prediction window, V sim,t+0 and V sim,t+T-1 are actual values of the monomer voltage median of time points t+0 and t+T-1 in the prediction window respectively,And/>Predicted values of the median of the cell voltages at time points t+0 and t+T-1 within the prediction window, respectively.
If it isThe predicted value of the median of the cell voltage at time t+0 is greater than the actual value, indicating that the predicted value is overall higher, and the upper limit of the failure threshold is lowered and the lower limit is lowered accordingly, and Δ sim,t+0 is negative.
If it is(The predicted value indicating the median of the medium cell voltage at time t+0 is smaller than the actual value, and indicates that the predicted value is lower as a whole, and at this time, the upper limit of the failure threshold is raised, the lower limit is raised accordingly, and Δ sim,t+0 is positive.
And so on, the upper and lower limits of the fault threshold value under each frame time under the whole window are obtained.
Step 104, determining fault types of each single battery at different time points in the prediction window according to the predicted value of the single voltage median at different time points in the prediction window, the fault threshold and the actual voltage of each single battery, wherein the method specifically comprises the following steps:
Calculating the difference value between the actual voltage of each single battery at different time points in the prediction window and the predicted value of the median of the single voltages by using the following formula to obtain the trend error of each single battery at different time points in the prediction window;
Wherein error sim is a trend error array, V 1,t+0 and V 1,t+T-1 are actual voltages of the 1 st single battery at time points t+0 and t+t-1 in the prediction window, V M,t+0 and V M,t+T-1 respectively represent actual voltages of the M-th single battery at time points t+0 and t+t-1 in the prediction window, and M represents the number of single batteries in the battery pack.
And determining the fault types of the single batteries at different time points in the prediction window by using the following formula according to the trend errors of the single batteries at different time points in the prediction window and the fault threshold values at different time points in the prediction window.
Wherein, P m,t+t' is the failure type of the mth single cell at time point t+t ' in the prediction window, P 1、P2 and P 3 are the primary failure, the secondary failure and the tertiary failure respectively, error sim,m,t+t' is the trend error of the mth single cell at time point t+t ' in the prediction window, m=1, 2, M, T ' =0, 1, and T-1.
As shown in fig. 4, the difference (trend error) between the predicted values of the median of the single voltages subtracted from the voltages of all the single batteries of the normal vehicle is within the upper and lower limit thresholds of the dynamic primary fault, the fault early warning is not triggered during the whole operation period, the calculated amount is small, all the single voltages are not predicted at one time, the median of the single voltages in the next prediction window is only predicted, the fault interval is set by using the variable threshold, the training data amount is far less than the predicted training data amount of each single battery voltage, the dynamic adjustment can be performed according to the prediction effect and the real voltage condition, and the interpretation is high.
As shown in fig. 5, the predicted value of the median of the monomer voltage of the thermal runaway vehicle substantially matches the actual value in the whole prediction interval, and as can be seen from fig. 5, the predicted value of the median of the monomer voltage is lower than the actual value at about 9000 th sampling point, which indicates that the actual voltage is mostly above the predicted value of the median of the monomer voltage, and the primary, secondary and tertiary thresholds are adjusted overall; as can be seen from the partial enlarged view of the running condition of the thermal runaway vehicle shown in fig. 6, all the single voltages generate abrupt changes at the 8600 th sampling point, after the data of the section are not transmitted back to the large data platform and the data abrupt changes are eliminated, the 74-number single battery generates 193 first-order faults at the 9265-9549 sampling points, the 74-number single battery generates 12 first-order faults at the 9462-9473 sampling points and generates thermal runaway after the last three-level fault triggering is consistent with the result of analysis of the actual fault analysis report.
Example 2
Embodiment 2 of the present invention provides a computer apparatus, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, the processor executing the computer program to perform the steps of the method provided in embodiment 1.
The computer device may be a database, the internal structure of which may be as shown in fig. 8. The computer device includes a processor, a memory, an Input/Output interface (I/O) and a communication interface. The processor, the memory and the input/output interface are connected through a system bus, and the communication interface is connected to the system bus through the input/output interface. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device is used to store the pending transactions. The input/output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement the method in embodiment 1.
Example 3
Inventive embodiment 3 provides a computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the steps of the method provided by embodiment 1.
Example 4
Inventive embodiment 4 provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of the method provided by embodiment 1.
The object information (including, but not limited to, object device information, object personal information, etc.) and the data (including, but not limited to, data for analysis, stored data, presented data, etc.) according to the present invention are information and data authorized by the object or sufficiently authorized by each party.
Those skilled in the art will appreciate that implementing all or part of the above described methods may be accomplished by way of a computer program stored on a non-transitory computer readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, database, or other medium used in embodiments provided herein may include at least one of non-volatile and volatile memory. The nonvolatile Memory may include Read-only Memory (ROM), magnetic tape, floppy disk, flash Memory, optical Memory, high density embedded nonvolatile Memory, resistive random access Memory (ReRAM), magneto-resistive random access Memory (Magnetoresistive RandomAccess Memory, MRAM), ferroelectric Memory (Ferroelectric RandomAccess Memory, FRAM), phase change Memory (PHASE CHANGE Memory, PCM), graphene Memory, and the like. Volatile memory can include random access memory (RandomAccess Memory, RAM) or external cache memory, and the like. By way of illustration, and not limitation, RAM can be in various forms such as Static RandomAccessMemory, SRAM or dynamic Dynamic RandomAccess Memory (DRAM) and the like. The databases referred to in the embodiments provided herein may include at least one of a relational database and a non-relational database. The non-relational database may include, but is not limited to, a blockchain-based distributed database, and the like. The processor referred to in the embodiments provided in the present invention may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, or the like, but is not limited thereto.
The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.
The principles and embodiments of the present invention have been described herein with reference to specific examples, the description of which is intended only to assist in understanding the methods of the present invention and the core ideas thereof; also, it is within the scope of the present invention to be modified by those of ordinary skill in the art in light of the present teachings. In view of the foregoing, this description should not be construed as limiting the invention.
Claims (9)
1.A method for detecting a single battery fault, the method comprising:
Based on historical state data in the sliding window, acquiring a multi-step prediction result by adopting a trained LSTM model; the historical state data in the sliding window comprise the actual values of the total voltage, the SOC value and the single voltage median of the battery pack at different time points in the sliding window, and the multi-step prediction result comprises the predicted values of the single voltage median at different time points in the prediction window, wherein the single voltage median is the median of the voltages of all single batteries in the battery pack;
Detecting and obtaining actual voltages of all single batteries at different time points in a prediction window;
calculating fault thresholds at different time points in the prediction window according to the predicted values of the cell voltage median at different time points in the prediction window and the actual voltages of all the cells;
and determining the fault type of each single battery at different time points in the prediction window according to the predicted value of the single voltage median at different time points in the prediction window, the fault threshold and the actual voltage of each single battery.
2. The method for detecting a failure of a single cell according to claim 1, wherein calculating the failure threshold value at different time points in the prediction window based on the predicted value of the number of cell voltage medians at different time points in the prediction window and the actual voltage of each single cell, specifically comprises:
Calculating actual values of the median of the monomer voltages at different time points in the prediction window according to the actual voltages of the monomer batteries at different time points in the prediction window;
Calculating variable thresholds at different time points in the prediction window by using the following formula according to actual values and predicted values of the median of the monomer voltages at different time points in the prediction window;
Wherein Delta sim is a variable threshold value array, delta sim,t+0 and Delta sim,t+T-1 are variable threshold values of time points t+0 and t+T-1 in a prediction window respectively, T is the starting time of the prediction window, T is the length of the prediction window, V sim,t+0 and V sim,t+T-1 are actual values of the monomer voltage median of time points t+0 and t+T-1 in the prediction window respectively, And/>Predicted values of the median of the monomer voltages at time points t+0 and t+T-1 in the prediction window;
Calculating fault thresholds at different time points in the prediction window according to the variable thresholds at different time points in the prediction window; the fault threshold includes:
Upper threshold for primary failure: 0.12+Δ sim;
lower threshold for primary failure: -0.12+Δ sim;
upper threshold for secondary failure: 0.24+Δ sim;
Lower threshold for secondary failure: -0.24+Δ sim;
upper threshold for three-level fault: 0.36+Δ sim;
Lower threshold of three-level fault: -0.36+Δ sim.
3. The method for detecting a single battery fault according to claim 2, wherein determining the fault type of each single battery at different time points in the prediction window according to the predicted value of the single voltage median at different time points in the prediction window, the fault threshold value and the actual voltage of each single battery, specifically comprises:
Calculating the difference value between the actual voltage of each single battery at different time points in the prediction window and the predicted value of the median of the single voltages by using the following formula to obtain the trend error of each single battery at different time points in the prediction window;
Wherein error sim is a trend error array, V 1,t+0 and V 1,t+T-1 are actual voltages of the 1 st single battery at time points t+0 and t+t-1 in the prediction window, V M,t+0 and V M,t+T-1 respectively represent actual voltages of the M-th single battery at time points t+0 and t+t-1 in the prediction window, and M represents the number of single batteries in the battery pack;
according to the trend errors of the single batteries at different time points in the prediction window and the fault threshold values at different time points in the prediction window, determining the fault types of the single batteries at different time points in the prediction window by using the following formula;
Wherein, P m,t+t′ is the failure type of the mth single cell at time point t+t ' in the prediction window, P 1、P2 and P 3 are the primary failure, the secondary failure and the tertiary failure respectively, error sim,m,t+t′ is the trend error of the mth single cell at time point t+t ' in the prediction window, m=1, 2, M, T ' =0, 1, and T-1.
4. The method for detecting a single battery fault according to claim 1, wherein the step of obtaining a multi-step prediction result using the trained LSTM model based on the historical state data in the sliding window, further comprises:
Acquiring historical charge and discharge data of a battery pack; the historical charge and discharge data includes: total current, total voltage, state of charge, cell voltage, state of charge, mileage, insulation resistance, and probe temperature of the battery pack;
constructing a training sample based on the historical charge-discharge data;
and training the LSTM model by using the training sample to obtain a trained LSTM model.
5. The method for detecting a single battery fault according to claim 4, wherein the constructing a training sample based on the historical charge and discharge data specifically comprises:
cleaning the historical charge and discharge data to obtain cleaned historical charge and discharge data;
Screening the cleaned historical charge and discharge data by adopting a correlation coefficient method to obtain data with the correlation with the median of the single voltage being greater than a preset threshold value, wherein the data is used as training data, and the training data comprises the total voltage, the SOC value and the single voltage of the battery pack;
constructing an input data structure and an output data structure based on the training data;
intercepting the input data structure by utilizing a sliding window to obtain an input data sample;
intercepting the output data structure by using a prediction window to obtain the actual output corresponding to the input data sample;
and constructing a training sample by using the data input sample and the actual output corresponding to the data input sample.
6. The method for detecting a failure of a battery cell according to claim 5, wherein the input data structure is:
The output data structure is as follows:
Wherein X u is an input data structure, Y u is an output data structure, total V 0、total V1 and total V N are the total voltages of the battery packs at time 0, time 1 and time N respectively, SOC 0、SOC1 and SOC N are the SOC values of the battery packs at time 0, time 1 and time N respectively, V 1,0、V1,1、V1,N is the voltage value of the 1 st single battery in the battery packs at time 0, time 1 and time N respectively, V 2,0、V2,1 and V 2,N are the voltage values of the 2 nd single battery in the battery packs at time 0, time 1 and time N respectively, V M,0、VM,1 and V M,N are the voltage values of the Mth single battery in the battery pack at the time 0, the time 1 and the time N respectively, V 1,L and V 1,N+s-1 are the voltage values of the 1 st single battery in the battery pack at the time L and the time N+S-1 respectively, V 2,L and V 2,N+S-1 are the voltage values of the 2 nd single battery in the battery pack at the time L and the time N+S-1 respectively, V M,L and V M,N+s-1 are the voltage values of the Mth single battery in the battery pack at the time L and the time N+S-1 respectively, M is the number of single batteries in the battery pack, N is the number of arrays in the input data structure, L is the length of a sliding window, and S is the length of a prediction window.
7. A computer device, comprising: memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the method according to any of claims 1-6.
8. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, carries out the steps of the method according to any one of claims 1-6.
9. A computer program product comprising a computer program, characterized in that the computer program, when executed by a processor, implements the steps of the method according to any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410243285.5A CN118033443A (en) | 2024-03-04 | 2024-03-04 | Single battery fault detection method, single battery fault detection equipment, single battery fault detection medium and single battery fault detection product |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410243285.5A CN118033443A (en) | 2024-03-04 | 2024-03-04 | Single battery fault detection method, single battery fault detection equipment, single battery fault detection medium and single battery fault detection product |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118033443A true CN118033443A (en) | 2024-05-14 |
Family
ID=90993107
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410243285.5A Pending CN118033443A (en) | 2024-03-04 | 2024-03-04 | Single battery fault detection method, single battery fault detection equipment, single battery fault detection medium and single battery fault detection product |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118033443A (en) |
-
2024
- 2024-03-04 CN CN202410243285.5A patent/CN118033443A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111146514B (en) | Lithium ion battery module operation safety evaluation prediction method and system and electronic equipment | |
| EP3605127B1 (en) | Device and method for estimating battery resistance | |
| Cheng et al. | State‐of‐charge estimation with aging effect and correction for lithium‐ion battery | |
| CN111505532A (en) | Online detection method for early internal short circuit of series lithium battery pack based on SOC correlation coefficient | |
| CN112098850B (en) | Lithium ion battery voltage fault diagnosis method and system based on SDO algorithm | |
| CN111257753B (en) | Battery system fault diagnosis method | |
| US20220373609A1 (en) | State Value for Rechargeable Batteries | |
| CN113255764A (en) | Method, system and device for detecting electrochemical energy storage system fault by using machine learning | |
| CN115494400B (en) | Lithium battery lithium separation state online monitoring method based on ensemble learning | |
| KR20210000207A (en) | Method of detecting internal short-circuit cell | |
| CN115980585A (en) | Battery fault detection method and device, computer equipment and storage medium | |
| Ding et al. | Diagnosis of connection fault for parallel-connected lithium-ion batteries based on long short-term memory networks | |
| CN116819328A (en) | Electric automobile power battery fault diagnosis method, system, equipment and medium | |
| CN114839537A (en) | Battery state of charge prediction method and device | |
| CN117067921B (en) | Fault detection method of electric automobile and electric automobile | |
| CN113935423A (en) | Power battery fault early warning method and system for coupling fuzzy control rule with Elman neural network | |
| KR20220060931A (en) | Apparatus and method for managing battery | |
| Tang et al. | An aging-and load-insensitive method for quantitatively detecting the battery internal-short-circuit resistance | |
| CN116679213A (en) | SOH estimation method for electric vehicle power battery based on integrated deep learning | |
| CN118033443A (en) | Single battery fault detection method, single battery fault detection equipment, single battery fault detection medium and single battery fault detection product | |
| CN116299006A (en) | Method, device, equipment and storage medium for predicting health condition of battery pack | |
| CN112363058B (en) | Lithium ion battery safety degree estimation method and device based on impedance spectrum and Markov characteristic | |
| CN115825790B (en) | Early warning method, device and system for battery insulation fault and computer equipment | |
| Mu et al. | Estimation of State of Charge of Lithium‐Ion Batteries Based on Wide and Deep Neural Network Model | |
| CN115825755B (en) | Method for evaluating consistency of voltages of battery cells of energy storage battery |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |