CN116901707A - Power battery pack fault early warning method, system and vehicle - Google Patents
Power battery pack fault early warning method, system and vehicle Download PDFInfo
- Publication number
- CN116901707A CN116901707A CN202310704038.6A CN202310704038A CN116901707A CN 116901707 A CN116901707 A CN 116901707A CN 202310704038 A CN202310704038 A CN 202310704038A CN 116901707 A CN116901707 A CN 116901707A
- Authority
- CN
- China
- Prior art keywords
- battery pack
- consistency
- data
- degradation
- power 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
- 238000000034 method Methods 0.000 title claims abstract description 33
- 230000015556 catabolic process Effects 0.000 claims abstract description 88
- 238000006731 degradation reaction Methods 0.000 claims abstract description 88
- 239000000178 monomer Substances 0.000 claims description 24
- 238000012549 training Methods 0.000 claims description 18
- 239000011159 matrix material Substances 0.000 claims description 14
- 238000004422 calculation algorithm Methods 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 5
- 230000005856 abnormality Effects 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 230000009469 supplementation Effects 0.000 claims description 3
- 238000012423 maintenance Methods 0.000 abstract description 3
- 230000036541 health Effects 0.000 abstract description 2
- 238000009826 distribution Methods 0.000 description 9
- 238000012544 monitoring process Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 5
- 238000005070 sampling Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60L—PROPULSION OF ELECTRICALLY-PROPELLED VEHICLES; SUPPLYING ELECTRIC POWER FOR AUXILIARY EQUIPMENT OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRODYNAMIC BRAKE SYSTEMS FOR VEHICLES IN GENERAL; MAGNETIC SUSPENSION OR LEVITATION FOR VEHICLES; MONITORING OPERATING VARIABLES OF ELECTRICALLY-PROPELLED VEHICLES; ELECTRIC SAFETY DEVICES FOR ELECTRICALLY-PROPELLED VEHICLES
- B60L3/00—Electric devices on electrically-propelled vehicles for safety purposes; Monitoring operating variables, e.g. speed, deceleration or energy consumption
- B60L3/0023—Detecting, eliminating, remedying or compensating for drive train abnormalities, e.g. failures within the drive train
- B60L3/0046—Detecting, eliminating, remedying or compensating for drive train abnormalities, e.g. failures within the drive train relating to electric energy storage systems, e.g. batteries or capacitors
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/005—Testing of electric installations on transport means
- G01R31/006—Testing of electric installations on transport means on road vehicles, e.g. automobiles or trucks
- G01R31/007—Testing of electric installations on transport means on road vehicles, e.g. automobiles or trucks using microprocessors or computers
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/367—Software therefor, e.g. for battery testing using modelling or look-up tables
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/36—Arrangements for testing, measuring or monitoring the electrical condition of accumulators or electric batteries, e.g. capacity or state of charge [SoC]
- G01R31/392—Determining battery ageing or deterioration, e.g. state of health
Landscapes
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Power Engineering (AREA)
- Sustainable Energy (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Sustainable Development (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Hardware Design (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Secondary Cells (AREA)
- Tests Of Electric Status Of Batteries (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
Abstract
The invention provides a power battery pack fault early warning method, a power battery pack fault early warning system and a vehicle, and relates to the technical field of vehicles. The invention relates to a power battery pack fault early warning method, which comprises the following steps: acquiring real-time operation data of a battery pack, and inputting the real-time operation data into a preset prediction model to obtain a predicted amount of consistency degradation characteristics; and carrying out fault early warning according to the consistency degradation characteristic prediction quantity. According to the invention, the real-time operation data of the battery pack is input into a preset prediction model, the predicted quantity of the consistency degradation characteristics can be obtained, the fault early warning is carried out through the predicted quantity of the consistency degradation characteristics, for example, the time when the fault with quantifiable confidence coefficient is about to occur is given, namely, the future time health state of the battery pack is predicted when the battery pack can still be normally used, and the time of the future fault can be predicted on the premise of definite confidence coefficient, so that effective guiding suggestions can be given to maintenance planning.
Description
Technical Field
The invention relates to the technical field of vehicles, in particular to a power battery pack fault early warning method and system and a vehicle.
Background
Under the pressure of energy and environmental protection, the new energy automobile becomes the development direction of the future automobile, however, the battery fault is always a great hidden danger of the new energy automobile, and the development prospect of the new energy automobile is affected. The early fault diagnosis and early warning can prompt the vehicle owner to maintain the vehicle in time, so that the safety performance of the power battery is improved.
The prior art scheme only can qualitatively send out early warning information, so that accurate fault about-to-occur time cannot be given, and effective guiding advice cannot be given to maintenance planning.
Disclosure of Invention
The invention solves the problem of how to predict the time of future faults of the power battery pack on the premise of definite confidence.
In order to solve the problems, the invention provides a power battery pack fault early warning method, a power battery pack fault early warning system and a vehicle.
In a first aspect, the present invention provides a power battery pack fault early warning method, including:
acquiring real-time operation data of a battery pack, and inputting the real-time operation data into a preset prediction model to obtain a predicted amount of consistency degradation characteristics;
and carrying out fault early warning according to the consistency degradation characteristic prediction quantity.
Optionally, the training process of the prediction model includes:
acquiring historical operation data of the battery pack, and extracting consistency degradation characteristic data of the battery pack from the historical operation data;
and constructing a single consistency degradation data set of the battery pack according to the historical operation data and the consistency degradation characteristic data, and training the prediction model according to the single consistency degradation data set.
Optionally, the historical operating data includes cell temperatures, cell voltages, and cell states of charge of the battery pack, and the extracting the consistency degradation characteristic data of the battery pack from the historical operating data includes:
performing missing value supplementation and normalization processing on the historical operation data to form a data matrix containing the temperature of each monomer, the voltage of each monomer and the state of charge of each monomer;
and extracting the consistency degradation characteristic data from the data matrix.
Optionally, the extracting the consistency degradation characteristic data from the data matrix includes:
dividing a normal state cluster and a plurality of outlier state clusters by adopting a hierarchical clustering algorithm, wherein the outlier state clusters represent that the corresponding data matrix has abnormality;
and determining the consistency degradation characteristic data according to the center distances between the normal state clusters and the plurality of outlier state clusters.
Optionally, the historical operating data further includes a total current and a total voltage of the battery pack, and the training the predictive model from the cell consistency degradation dataset includes: and taking the total current, the total voltage and the consistency degradation characteristic at the current moment as inputs, taking the consistency degradation characteristic at a future moment as output, and training the prediction model.
Optionally, the performing fault pre-warning according to the predicted consistent degradation characteristic includes: and determining early warning fault occurrence time under preset confidence coefficient according to the consistency degradation characteristic pre-measurement and a preset upper control limit.
Optionally, the power battery pack fault early warning method further includes:
performing kernel density estimation, and taking a Gaussian radial basis function as a kernel function;
and adjusting the Gaussian radial basis function according to the consistency degradation characteristic data corresponding to the normal operation of the battery pack so as to determine the upper control limit.
Optionally, the power battery pack fault early warning method further includes: and constructing a confidence analysis model according to the consistency degradation characteristic data and the consistency degradation characteristic predicted quantity so as to determine the confidence.
In a second aspect, the present invention provides a power battery pack fault early warning system, including a computer readable storage medium storing a computer program and a processor, where the computer program is read and executed by the processor to implement the above power battery pack fault early warning method.
In a third aspect, the present invention provides a vehicle, including the above power battery pack failure warning system.
According to the invention, the real-time operation data of the battery pack is input into a preset prediction model, the predicted quantity of the consistency degradation characteristics can be obtained, the fault early warning is carried out through the predicted quantity of the consistency degradation characteristics, for example, the time when the fault with quantifiable confidence coefficient is about to occur is given, namely, the future time health state of the battery pack is predicted when the battery pack can still be normally used, and the time of the future fault can be predicted on the premise of definite confidence coefficient, so that effective guiding suggestions can be given to maintenance planning.
Drawings
Fig. 1 is a schematic flow chart of a power battery pack fault early warning method according to an embodiment of the invention;
fig. 2 is a schematic flow chart of a power battery pack fault early warning method according to an embodiment of the invention;
FIG. 3 is a schematic diagram of consistency degradation feature construction in accordance with an embodiment of the present invention;
FIG. 4 is a schematic diagram of predictive model training in accordance with an embodiment of the invention;
FIG. 5 is a diagram illustrating the determination of an upper control limit according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of an early warning according to an embodiment of the present invention;
fig. 7 is a schematic diagram of early warning time confidence analysis according to an embodiment of the present invention.
Detailed Description
In order that the above objects, features and advantages of the invention will be readily understood, a more particular description of the invention will be rendered by reference to specific embodiments thereof which are illustrated in the appended drawings.
As shown in fig. 1, an embodiment of the present invention provides a power battery fault early warning method, including:
and acquiring real-time operation data of the battery pack, and inputting the real-time operation data into a preset prediction model to obtain a predicted amount of the consistency degradation characteristic.
Specifically, as shown in fig. 2, after acquiring real-time operation data (i.e., actual operation data) of the battery pack, the real-time operation data is input into a preset prediction model to obtain a predicted amount of the consistency degradation characteristic.
And carrying out fault early warning according to the consistency degradation characteristic prediction quantity.
Specifically, referring to fig. 2, an upper control limit of the consistency degradation feature D with a given significance level α is calculated, and a predicted value and an upper control limit (Upper Control Limit, i.e., UCL) of the consistency degradation feature are determined, so as to send an early warning to a vehicle owner in advance of occurrence of a fault for a certain time, and a confidence level of the occurrence time of the early warning fault is defined, for example, the early warning informs that "the confidence level of occurrence of the fault after two hours is 90%, and the vehicle owner is requested to maintain as soon as possible".
Optionally, the training process of the prediction model includes:
and acquiring historical operation data of the battery pack, and extracting consistency degradation characteristic data of the battery pack from the historical operation data.
Specifically, as shown in fig. 2, monitoring data (i.e., historical life-cycle operation data) of a BMS (battery management system ) of a plurality of vehicles with power battery pack fault alarms during running are extracted, including a total current I, a total voltage V and a voltage V of each cell i Temperature T i State of charge SOC i And fault information, wherein i is a positive integer smaller than n, n is the total number of battery cells in each power battery pack, and the data corresponding to the battery packs of each automobile in the data are m sampling points in the time dimension.
Preprocessing the BMS monitoring data, and generating voltage V of each monomer i Temperature T i State of charge SOC i And extracting time sequence characteristics of consistency degradation of the battery pack, namely consistency degradation characteristic data, from the three data matrixes. Monitoring index for each monomer in battery packAnd (3) data mining is carried out, inconsistent information contained in the data is mined, and a consistency degradation model of the power battery pack, which considers temperature inconsistency, voltage inconsistency and charge state inconsistency, is established according to the inconsistent information.
And constructing a single consistency degradation data set of the battery pack according to the historical operation data and the consistency degradation characteristic data, and training the prediction model according to the single consistency degradation data set.
Specifically, with reference to fig. 2, a single consistency degradation data set of the battery pack is constructed by using historical operation data and consistency degradation characteristic data, and then a prediction model is built based on an LSTM network (a method such as a cyclic neural network, support vector regression, least squares regression and the like can also be adopted), so that the prediction of the consistency degradation characteristic D on the time sequence is realized.
Optionally, the historical operating data includes cell temperatures, cell voltages, and cell states of charge of the battery pack, and the extracting the consistency degradation characteristic data of the battery pack from the historical operating data includes:
and performing missing value supplementation and normalization processing on the historical operation data to form a data matrix containing the temperature of each monomer, the voltage of each monomer and the charge state of each monomer.
In the prior art, only battery state monitoring or fault diagnosis under a single characteristic is generally considered, and degradation of battery performance is often accompanied by electrical characteristics, thermodynamic characteristics and electrochemical characteristics, so that a single monitoring index is often not universal. Therefore, when the consistency degradation index is constructed, the embodiment performs feature level fusion on the voltage index, the temperature index and the charge state index, so that the constructed index has higher universality.
Specifically, a random forest algorithm can be adopted to supplement the missing value of the data of the battery management system, so as to obtain the original data of the preprocessed time-single serial number-monitoring index, namely, the voltage, temperature and state of charge time sequence data of each single unit of the power battery pack form a data matrix with the size of m x n x 3, and the data is normalized.
The hardware of the sensor and the communication equipment is limited in the links of data acquisition and transmission, so that the problems of frame loss, error and the like inevitably exist, and the problems of abnormal values, missing values and the like of a constructed data set are caused, so that the data are cleaned, the abnormal values are deleted, and the missing values are uniformly filled in order to improve the robustness of the constructed model.
And extracting the consistency degradation characteristic data from the data matrix.
Specifically, the above-mentioned obtained monitoring data of each single cell in the power battery pack includes three indexes of temperature, SOC and voltage, and the data form of each time sampling point is a single cell serial number-monitoring index, i.e. n sample points with three-dimensional characteristics are provided at each time sampling point, so as to form a data matrix X with size of n×3:
the multi-dimensional characteristic sample data X of the single batteries of each sampling point can be analyzed by adopting a clustering algorithm, the discrete degree of the sample points is quantized, and finally, the characteristic quantity D which represents the consistency degradation of each single battery of the battery pack along with time is constructed.
Optionally, the extracting the consistency degradation characteristic data from the data matrix includes:
and dividing a normal state cluster and a plurality of outlier state clusters by adopting a hierarchical clustering algorithm, wherein the outlier state clusters represent that the corresponding data matrix has an abnormality.
Specifically, as shown in fig. 3, when the hierarchical clustering algorithm is adopted, the distances between classes are measured by the average distance, and the sample x is calculated by the euclidean distance j And x k Distance d between jk ：
Since each monomer sample point at time i includes three features, an abnormal monomer may exhibit an outlier from one feature or multiple features, dividing hierarchical clustering result trees by 8 clusters includes 7 outlier state clusters and 1 normal state cluster (i.e., no index anomaly).
Wherein, in combination with the illustration of fig. 3, one index anomaly is any anomaly among the temperature of each monomer, the voltage of each monomer and the state of charge of each monomer, two index anomalies are two anomalies among the temperature of each monomer, the voltage of each monomer and the state of charge of each monomer, and three index anomalies are all anomalies among the temperature of each monomer, the voltage of each monomer and the state of charge of each monomer.
And determining the consistency degradation characteristic data according to the center distances between the normal state clusters and the plurality of outlier state clusters.
Specifically, as shown in fig. 3, the center distances L between the 7 outlier state clusters and the normal state clusters are weighted differently to measure the degree of dispersion D between the internal cells of the power battery pack i And D is taken as a characteristic quantity for describing consistency degradation of the single unit of the power battery pack:
D i ＝αL V,T,SOC +β∑ m,n∈C L m,n +γΣ m∈C L m ；
wherein, C= { V, T, SOC }, alpha, beta, gamma are weight coefficients, which are empirically given as 5,3,2, respectively.
From this, the degradation characteristic quantity D corresponding to the n monomers at the m sample time points in the time series data after the n monomers are clustered can be obtained.
Optionally, the historical operating data further includes a total current and a total voltage of the battery pack, and the training the predictive model from the cell consistency degradation dataset includes: and taking the total current, the total voltage and the consistency degradation characteristic at the current moment as inputs, taking the consistency degradation characteristic at a future moment as output, and training the prediction model.
Specifically, as shown in connection with FIG. 4, the current uniformity degradation characteristic D is represented by the total current I, the total voltage V i As input, degrade the feature D with consistency at some future time i+j Training the test model for output to realize the time sequence of the consistency degradation characteristic DThe above prediction, the training process includes: the data set is divided into 80% training set, 10% testing set and 10% verifying set, gradient descent algorithm is adopted, when the loss of the training set and the testing set is minimized, the super parameter training of the model is completed, and finally the verification is carried out on the rest data set.
Because the monitoring index of the BMS changes greatly along with the running working condition of the electric automobile in the running process of the automobile, a reliable early warning model is difficult to construct based on a simple model such as a statistical model or entropy analysis in the prior art, the current value of the total current and the consistency degradation index is taken as input and the future value of the consistency degradation index is taken as output when a data set is constructed, and the total current can sensitively reflect the driving behavior, so that a prediction model with strong robustness to the current change is constructed.
Optionally, the performing fault pre-warning according to the predicted consistent degradation characteristic includes: and determining early warning fault occurrence time under preset confidence coefficient according to the consistency degradation characteristic pre-measurement and a preset upper control limit.
Specifically, as shown in connection with fig. 6, performing fault pre-warning according to the predicted amount of consistency degradation characteristics includes: and determining early warning fault occurrence time under preset confidence according to the predicted quantity of the consistency degradation characteristic and a preset upper control limit, for example, when the predicted quantity of the consistency degradation characteristic exceeds the upper control limit, the early warning model gives out early warning of the battery pack fault to the vehicle owner and gives out the predicted fault occurrence time.
Optionally, the power battery pack fault early warning method further includes:
and (4) performing kernel density estimation, and taking the Gaussian radial basis function as a kernel function.
Specifically, in performing the kernel density estimation, a gaussian radial basis function is taken as the kernel function.
And adjusting the Gaussian radial basis function according to the consistency degradation characteristic data corresponding to the normal operation of the battery pack so as to determine the upper control limit.
Specifically, consistency degradation characteristic D data of the power battery pack in the normal running state is extracted from data of a plurality of new energy automobiles, and nuclear density estimation is carried out on probability density function parameters to obtain a probability density function p (x) of the consistency degradation characteristic D in the normal running state of the battery pack.
As shown in connection with fig. 5, given a significance level α of 0.01, the upper control limit D (α) for the consistency degradation feature D is calculated by:
P(D<D(α))＝99％；
the method comprises the steps of adopting a kernel density estimation method, taking a Gaussian radial basis function as a kernel function, and optimizing the Gaussian radial basis function through consistency degradation index data of a normal operation stage of a battery pack in actual operation to enable the Gaussian radial basis function to approach to actual probability distribution, wherein a significance level alpha is 0.01 to ensure that the control upper limit has 99% accuracy in the application of a model.
Optionally, the power battery pack fault early warning method further includes: and constructing a confidence analysis model according to the consistency degradation characteristic data and the consistency degradation characteristic predicted quantity so as to determine the confidence.
Specifically, the actual value of the consistency degradation characteristic is obtained in actual application, the distance between the actual value and the predicted value is regarded as a random variable, the proper probability distribution is obtained, a confidence analysis model of the predicted result is built, and the early warning fault occurrence time under the given confidence is obtained by combining the upper control limit.
When a confidence analysis model of a predicted result is constructed, a Weibull distribution (other distributions can be adopted) is selected to measure a probability density distribution function of a random variable between a predicted value and an actual value, and a Weibull distribution formula is as follows:
where x represents a random variable, m >0, m represents a shape parameter, η >0, η represents a scale parameter, γ >0, γ represents a position parameter, and the distribution is a two-parameter weibull distribution when γ=0.
The parameters of the weibull distribution are determined by the random variables of the distance between the predicted value and the actual value. Referring to fig. 7, the probability density distribution curve of the random variable moves with the consistency degradation index, and when the area exceeding the upper control limit is given by 10%, that is, the failure occurrence time prediction result has a confidence of 90%.
Another embodiment of the present invention provides a power battery pack fault early warning system, including a computer readable storage medium storing a computer program and a processor, where the computer program is read and executed by the processor to implement the above power battery pack fault early warning method.
Another embodiment of the present invention provides a vehicle, including the above power battery pack failure warning system.
Although the invention is disclosed above, the scope of the invention is not limited thereto. Various changes and modifications may be made by one skilled in the art without departing from the spirit and scope of the invention, and these changes and modifications will fall within the scope of the invention.
Claims (10)
1. The power battery pack fault early warning method is characterized by comprising the following steps of:
acquiring real-time operation data of a battery pack, and inputting the real-time operation data into a preset prediction model to obtain a predicted amount of consistency degradation characteristics;
and carrying out fault early warning according to the consistency degradation characteristic prediction quantity.
2. The power battery pack fault pre-warning method according to claim 1, wherein the training process of the prediction model comprises:
acquiring historical operation data of the battery pack, and extracting consistency degradation characteristic data of the battery pack from the historical operation data;
and constructing a single consistency degradation data set of the battery pack according to the historical operation data and the consistency degradation characteristic data, and training the prediction model according to the single consistency degradation data set.
3. The power battery pack failure warning method of claim 2, wherein the historical operating data includes cell temperatures, cell voltages, and cell states of charge of the battery pack, and wherein extracting consistent degradation characteristic data of the battery pack from the historical operating data includes:
performing missing value supplementation and normalization processing on the historical operation data to form a data matrix containing the temperature of each monomer, the voltage of each monomer and the state of charge of each monomer;
and extracting the consistency degradation characteristic data from the data matrix.
4. The power battery pack failure warning method of claim 3, wherein the extracting the consistency degradation characteristic data from the data matrix comprises:
dividing a normal state cluster and a plurality of outlier state clusters by adopting a hierarchical clustering algorithm, wherein the outlier state clusters represent that the corresponding data matrix has abnormality;
and determining the consistency degradation characteristic data according to the center distances between the normal state clusters and the plurality of outlier state clusters.
5. The power battery pack fault warning method of claim 3, wherein the historical operating data further comprises a total current and a total voltage of the battery pack, the training the predictive model from the cell consistency degradation dataset comprising: and taking the total current, the total voltage and the consistency degradation characteristic at the current moment as inputs, taking the consistency degradation characteristic at a future moment as output, and training the prediction model.
6. The power battery pack failure warning method according to claim 1, characterized in that the performing failure warning according to the consistency degradation characteristic prediction amount includes: and determining early warning fault occurrence time under preset confidence coefficient according to the consistency degradation characteristic pre-measurement and a preset upper control limit.
7. The power battery pack failure warning method according to claim 6, further comprising:
performing kernel density estimation, and taking a Gaussian radial basis function as a kernel function;
and adjusting the Gaussian radial basis function according to the consistency degradation characteristic data corresponding to the normal operation of the battery pack so as to determine the upper control limit.
8. The power battery pack failure warning method according to claim 2, further comprising: and constructing a confidence analysis model according to the consistency degradation characteristic data and the consistency degradation characteristic predicted quantity so as to determine the confidence.
9. A power battery pack fault warning system comprising a computer readable storage medium storing a computer program and a processor, the computer program when read and executed by the processor implementing the power battery pack fault warning method of any one of claims 1 to 8.
10. A vehicle comprising the power battery pack failure warning system of claim 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310704038.6A CN116901707A (en) | 2023-06-14 | 2023-06-14 | Power battery pack fault early warning method, system and vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310704038.6A CN116901707A (en) | 2023-06-14 | 2023-06-14 | Power battery pack fault early warning method, system and vehicle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116901707A true CN116901707A (en) | 2023-10-20 |
Family
ID=88363767
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310704038.6A Pending CN116901707A (en) | 2023-06-14 | 2023-06-14 | Power battery pack fault early warning method, system and vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116901707A (en) |
-
2023
- 2023-06-14 CN CN202310704038.6A patent/CN116901707A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | Research on electric vehicle charging safety warning model based on back propagation neural network optimized by improved gray wolf algorithm | |
| CN111584952B (en) | Method and system for online evaluation of electrochemical cells of energy storage power station | |
| CN112731159B (en) | Method for pre-judging and positioning battery faults of battery compartment of energy storage power station | |
| Guo et al. | Online estimation of SOH for lithium-ion battery based on SSA-Elman neural network | |
| Caliwag et al. | Hybrid VARMA and LSTM method for lithium-ion battery state-of-charge and output voltage forecasting in electric motorcycle applications | |
| CN113156917B (en) | Power grid equipment fault diagnosis method and system based on artificial intelligence | |
| CN108287317B (en) | Battery power prediction model generation method and system and power prediction method and system | |
| CN116154900B (en) | Active safety three-stage prevention and control system and method for battery energy storage power station | |
| CN109604186A (en) | Power battery performance flexibility assesses method for separating | |
| CN110146817A (en) | The diagnostic method of lithium battery failure | |
| CN113760670A (en) | Cable joint abnormity early warning method and device, electronic equipment and storage medium | |
| CN115356639A (en) | Intelligent health monitoring method and system for bidirectional lithium ion battery | |
| He et al. | Voltage abnormality-based fault diagnosis for batteries in electric buses with a self-adapting update model | |
| Yao et al. | Fault identification of lithium-ion battery pack for electric vehicle based on ga optimized ELM neural network | |
| CN116226469A (en) | Intelligent diagnosis method and system for energy storage equipment faults | |
| CN114460481A (en) | Energy storage battery thermal runaway early warning method based on Bi-LSTM and attention mechanism | |
| Kharlamova et al. | Evaluating Machine-Learning-Based Methods for Modeling a Digital Twin of Battery Systems Providing Frequency Regulation | |
| Chen et al. | Remaining useful life prediction of the lithium-ion battery based on CNN-LSTM fusion model and grey relational analysis | |
| CN114036647A (en) | Power battery safety risk assessment method based on real vehicle data | |
| CN117169761A (en) | Battery state evaluation method, apparatus, device, storage medium, and program product | |
| CN117630672A (en) | Method and device for predicting a device battery of a diagnostic device using a converter model | |
| Diao et al. | Research on Electric Vehicle Charging Safety Warning based on A-LSTM Algorithm | |
| CN116061690A (en) | Safety early warning method and device in electric automobile charging process | |
| CN116893366A (en) | Method and device for operating a system to detect anomalies in an electrical energy store of a device | |
| CN115856641A (en) | Method and device for predicting remaining charging time of battery and electronic equipment |
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 |