CN114734873B - Power battery monomer thermal runaway early warning method based on cloud online data - Google Patents
Power battery monomer thermal runaway early warning method based on cloud online data Download PDFInfo
- Publication number
- CN114734873B CN114734873B CN202210403841.1A CN202210403841A CN114734873B CN 114734873 B CN114734873 B CN 114734873B CN 202210403841 A CN202210403841 A CN 202210403841A CN 114734873 B CN114734873 B CN 114734873B
- Authority
- CN
- China
- Prior art keywords
- thermal runaway
- matrix
- failure probability
- battery
- battery cell
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 40
- 239000000178 monomer Substances 0.000 title claims abstract description 20
- 239000011159 matrix material Substances 0.000 claims abstract description 54
- 230000009467 reduction Effects 0.000 claims description 27
- 238000012549 training Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 7
- 230000008859 change Effects 0.000 claims description 7
- 230000006870 function Effects 0.000 claims description 7
- 238000010801 machine learning Methods 0.000 claims description 7
- 230000009466 transformation Effects 0.000 claims description 6
- 238000013135 deep learning Methods 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 5
- 230000006837 decompression Effects 0.000 claims description 3
- 238000005315 distribution function Methods 0.000 claims description 3
- 238000011946 reduction process Methods 0.000 claims description 3
- 239000013598 vector Substances 0.000 claims description 3
- 238000013528 artificial neural network Methods 0.000 description 3
- 125000004122 cyclic group Chemical group 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000002485 combustion reaction Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000002269 spontaneous effect Effects 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
- B60L58/00—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles
- B60L58/10—Methods or circuit arrangements for monitoring or controlling batteries or fuel cells, specially adapted for electric vehicles for monitoring or controlling batteries
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60Q—ARRANGEMENT OF SIGNALLING OR LIGHTING DEVICES, THE MOUNTING OR SUPPORTING THEREOF OR CIRCUITS THEREFOR, FOR VEHICLES IN GENERAL
- B60Q9/00—Arrangement or adaptation of signal devices not provided for in one of main groups B60Q1/00 - B60Q7/00, e.g. haptic signalling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/60—Other road transportation technologies with climate change mitigation effect
- Y02T10/70—Energy storage systems for electromobility, e.g. batteries
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Mechanical Engineering (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Sustainable Development (AREA)
- Transportation (AREA)
- Computing Systems (AREA)
- Power Engineering (AREA)
- Sustainable Energy (AREA)
- Mathematical Physics (AREA)
- Human Computer Interaction (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Secondary Cells (AREA)
- Tests Of Electric Status Of Batteries (AREA)
- Charge And Discharge Circuits For Batteries Or The Like (AREA)
Abstract
The invention relates to a power battery monomer thermal runaway early warning method based on cloud online data, which comprises the following steps: s1: extracting boundary characteristics of a battery module from the battery pack data collected by the cloud and forming a high-dimensional matrix; s2: extracting a low-dimensional feature matrix of the high-dimensional matrix, calculating the current failure probability based on the low-dimensional feature matrix, comparing the current failure probability with a preset failure probability threshold, and judging whether a thermal runaway risk exists currently; s3: when the thermal runaway risk is judged, calculating a contribution value of each dimension in the high-dimensional matrix to the failure probability, and determining a single battery corresponding to the boundary feature with the largest contribution of the failure probability as a high-risk single to be verified; s4: and analyzing the on-line voltage, temperature and SOC data of the high-risk monomer to be verified, and alarming according to the three-level different grades corresponding to the deviation degree value. According to the method, the thermal runaway of the power single battery is pre-warned, so that the stability and the safety of battery operation are improved.
Description
Technical Field
The invention belongs to the technical field of power batteries of new energy automobiles, and particularly relates to a power battery monomer thermal runaway early warning method based on cloud online data.
Background
In the actual running of the electric automobile, with the use of the electric automobile, the performance nonlinearity of the battery system is reduced due to the severe road surface condition, the environmental temperature and the dynamic change of the load, and further the problems of liquid leakage, insulation damage, internal micro short circuit and the like are unavoidable. The correlation between the current thermal runaway occurrence and the real-time working condition of the electric automobile is unknown, and the thermal runaway departure mechanism is not clear. Therefore, the method for monitoring the fault characteristics of the battery and evaluating the health state of the battery in time is a main means and an important method for preventing serious safety accidents such as spontaneous combustion, explosion and the like caused by further aging of the battery as far as possible. The method realizes the real-time state monitoring and accurate fault diagnosis of the battery management system, and further achieves the important practical significance of early warning of safety.
The current methods for realizing fault alarm of the battery management system mainly comprise three types: a judgment method based on a threshold value, a judgment method based on a physical model and a judgment method based on a data driving model. However, the verification and correction of the current thermal runaway early warning method are mostly completed under laboratory data, and further development is required to face the accuracy of the real random working condition. With the development of cloud technology, on-board BMS data can be transmitted to the cloud for storage and calculation in real time, on the basis of cloud online data, big data are mined by combining an intelligent algorithm, early warning of battery thermal runaway risks under real-time working conditions is achieved by combining a machine learning algorithm, and safe operation of an electric vehicle is guaranteed to be of great application value.
Disclosure of Invention
The technical problems to be solved are as follows:
aiming at the technical problems in the prior art, the invention provides a power battery monomer thermal runaway early warning method based on cloud online data, which is based on the cloud online data and various intelligent algorithms to realize the thermal runaway early warning positioning of the battery monomer in the real vehicle operation process, and the technical scheme of the invention is as follows:
a power battery monomer thermal runaway early warning method based on cloud online data comprises the following steps:
s1: extracting boundary characteristics of a battery module from battery pack data collected by a cloud end at a certain period of time, and forming a high-dimensional matrix from the boundary characteristics of the battery module;
S2: the dimension of the high-dimensional matrix is reduced through a machine learning algorithm, a low-dimensional feature matrix of the high-dimensional matrix is extracted, the current failure probability at each moment in the period is calculated based on the low-dimensional feature matrix, and the comparison is carried out according to a preset failure probability threshold and the current failure probability to judge whether the thermal runaway risk exists currently or not;
S3: when judging that the thermal runaway risk exists, decompressing the low-dimensional matrix in the step S2, calculating a contribution value of each dimension in the high-dimensional matrix in the step S1 to failure probability, and determining a single battery corresponding to the boundary feature with the largest contribution of the failure probability as a high-risk single to be checked;
S4: analyzing the on-line voltage, temperature and SOC data of the high-risk monomer to be verified, calculating the deviation degree of the on-line voltage, temperature and SOC data of the high-risk monomer to be verified in the current state and the predicted state by combining a pre-trained deep learning battery state prediction model, comparing the deviation degree value of the voltage, temperature and SOC data of the high-risk monomer to be verified with a preset three-level alarm threshold, and alarming according to the deviation degree value corresponding to three different levels.
Further, in step S1, the thermal runaway related parameters of the battery module include a voltage neel coefficient of the battery cell, a voltage change rate of the battery cell, a differential value of the voltage versus the temperature of the battery cell, a differential value of the voltage versus the state of charge of the battery cell, a temperature change rate of the battery cell, internal short circuit resistance of the battery cell, a voltage of the battery cell, a temperature of the battery cell, and a state of charge of the battery cell.
Further, in step S1, the boundary features extracted are the maximum, the next-maximum, the standard deviation, the extremely poor extracted for the thermal runaway related parameter
Further, in step S2, the machine learning algorithm includes a PCA algorithm with linear dimension reduction or a self-encoder algorithm with nonlinear dimension reduction.
Further, in step S2, the failure probability is calculated by calculating Hotelling T 2 statistic and SPE statistic of the low-dimensional matrix, thereby calculating a comprehensive indexAnd return to probability distribution/>Obtaining failure probability;
Further, in step S2, the failure probability is calculated, specifically, a matrix of a Hotelling T 2 statistic and an SPE statistic is calculated according to a normal battery training set dimension reduction process;
wherein, hotelling T 2 statistic is calculated according to the following formula:
Wherein X i is a boundary feature matrix of the i module at a certain moment, P k is a dimension-reduction transformation matrix, and is composed of principal component feature vectors, and S is a training sample set A diagonal matrix formed by principal component eigenvalues, k being the number of principal components;
SPE statistics are calculated as follows:
Wherein X i is a boundary feature matrix of the module group at a certain moment I, P k is a dimension reduction transformation matrix, I is a unit matrix, and k is the number of principal elements;
Solving based on Hotelling T 2 and SPE statistics according to comprehensive indexes The system failure rate function of the current moment of the sample can be obtained:
In the method, in the process of the invention, Delta 2 is the control limit value of Hotelling T 2 and SPE respectively,/>Is a symmetrical positive definite matrix, and the comprehensive index/>Coincidence probability distribution/>The degree of freedom is h, the coefficient is g, and the chi-square distribution is based on/>The probability distribution function can obtain the current failure probability function/>The failure probability h is calculated according to the following formula:
Further, in step S3, the calculation of the contribution value is different according to different dimension reduction methods, and for the linear dimension reduction method, the duty ratio of the module boundary feature to the low-dimension feature exceeding the control limit is calculated in an accumulated manner; and calculating the duty ratio of the low-dimensional fault dimension after decompression to different dimensions for the nonlinear dimension reduction method.
The invention has the beneficial effects that:
1. And (3) information fusion judgment: the early warning mechanism of the invention does not compare and judge the single variable, but judges after combining the high-dimensional characteristic data of the module to perform information fusion and dimension reduction, thereby avoiding the influence of the random error of the data of the single variable on the judgment and improving the robustness of the invention.
2. Multistage judgment, avoiding false alarm: in the invention, a plurality of layers of judgment logics are arranged, the boundary parameters of the battery module and the core state quantity of the single body are comprehensively evaluated, and the statistics judgment is carried out in a failure rate mode, so that false alarm is avoided as far as possible, and the practical value is improved.
3. High accuracy: according to the method, data dimension reduction and fitting are performed on mass driving data of the cloud platform, retraining of the model can be achieved based on cloud online data, and high accuracy is guaranteed by combining mass data sets. .
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 description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a workflow diagram of a power battery monomer thermal runaway early warning method based on cloud online data.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. 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 fall within the scope of the invention.
S1: and carrying out boundary feature extraction on the thermal runaway related parameter data of the battery module at a certain period of time collected by the cloud to form a data high-dimensional matrix. The thermal runaway related parameter may include a voltage nerl coefficient of the unit cell, a voltage change rate of the unit cell, a voltage-to-temperature differential value of the unit cell, a voltage-to-charge state differential value of the unit cell, a temperature change rate of the unit cell, internal short circuit resistance of the unit cell, a voltage of the unit cell, a temperature of the unit cell, a charge state of the unit cell, and the like, wherein the boundary feature may be a statistical parameter such as a maximum value, a suboptimal value, a standard deviation, a range, and the like of the thermal runaway related parameter.
S2: and reducing the dimension of the high-dimension matrix by a machine learning algorithm, and extracting the low-dimension characteristics of the high-dimension matrix. The current failure probability at each moment in the period is calculated based on the low-dimensional feature matrix. And judging whether the thermal runaway risk exists currently according to the given failure probability threshold value. The machine learning algorithm comprises a linear dimension reduction algorithm represented by PCA and a nonlinear deep learning dimension reduction algorithm represented by a self-encoder.
The low-dimensional features are set according to the required dimension reduction effect, and the lowest dimensional features are selected under the condition that the dimension reduction effect meets the actual requirement, for example, the low-dimensional features corresponding to 90% (determined according to the actual dimension reduction requirement) of the information are reserved in the ratio of covariance matrix eigenvalues in the PCA algorithm; the self-encoder algorithm ensures the reasonable corresponding low-dimensional characteristics of the loss function before and after the dimension reduction.
The failure probability calculation calculates the comprehensive index phi by calculating Hotelling T 2 statistic and SPE statistic of the low-dimensional matrix and returns to probability distributionSpecifically, the calculation method of Hotelling T 2 statistic and SPE statistic is calculated according to a normal battery training set dimension reduction process matrix, wherein Hotelling T 2 statistic is calculated according to the following formula:
Wherein X i is a boundary feature matrix of the i module at a certain moment, P k is a dimension-reduction transformation matrix, and is composed of principal component feature vectors, and S is a training sample set And a diagonal matrix formed by principal component eigenvalues, wherein k is the number of principal components.
The principal component feature values are terms known in the art, specifically, feature values of corresponding numbers of the matrix are selected according to dimension reduction, the feature values are sequentially selected from large to small, and the principal component is the feature number of dimension reduction.
SPE statistics are calculated as follows:
In the above formula, X i is a module boundary feature matrix at a certain moment I, P k is a dimension reduction transformation matrix, I is a unit matrix, k is the number of principal elements, and I refers to a certain moment.
Based on the solution of T 2 and SPE statistics, the system failure rate function of the current moment of the sample can be obtained according to the comprehensive index phi:
In the method, in the process of the invention, Delta 2 is the control limit value of Hotelling T 2 and SPE respectively,/>Is a symmetrical positive definite matrix, and the comprehensive index/>Coincidence probability distribution/>The degree of freedom is h, the coefficient is g, and the chi-square distribution is based on/>The probability distribution function can obtain the current failure probability function/>The failure probability h is calculated according to the following formula:
It should be noted that, the setting of the failure probability abnormality determination probability threshold is obtained by empirical determination, and can be determined by existing data, such as the ratio of the data set to be normally occupied.
S3: and when the thermal runaway risk is judged, calculating a contribution value of each dimension in the S1 high-dimensional matrix to the failure rate for the low-dimensional matrix provided by the S2, and determining the battery cell corresponding to the module boundary feature with the largest failure rate contribution. The monomer from which the dimension contributing the most is taken is marked as the high-risk monomer to be verified.
The calculation of the contribution value is different according to different dimension reduction algorithms, and the linear dimension reduction algorithm is used for carrying out accumulated calculation on the duty ratio of the module boundary characteristic to the low-dimension characteristic exceeding the control limit; for a nonlinear dimension reduction algorithm, the duty ratio of the low-dimension fault dimension after decompression to different dimensions can be calculated.
S4: and aiming at the battery cell positioned in the S3, analyzing the online voltage, the temperature and the SOC data of the battery cell, and calculating the deviation degree of the current state and the predicted state by combining a deep learning prediction model. And comparing the deviation degree of the voltage, the temperature and the SOC of the single body with a three-level alarm threshold value, and alarming according to the corresponding grade of the deviation degree.
The prediction model adopts a known prediction model, mainly comprises a cyclic neural network and control equation prediction, wherein the cyclic neural network is represented by LSTM and needs advanced training of a certain sample; the control equation prediction is performed on the basis of the parameter identification of a space state equation (represented by an equivalent circuit model); the two methods can also be fused to improve accuracy.
The three-level alarm threshold is preset according to different vehicle types and actual requirements.
The deep learning prediction model is used for predicting the state of a battery in the next second, the training process takes data of the type of the target vehicle before the time t as a model input through a data set of the type of the target vehicle, the data at the time t is taken as a model output, the sequence from 1 to the length of the data set is taken t, and the training set of the neural network is formed by utilizing the existing training method to complete training of the neural network model.
It is apparent that the above examples are given by way of illustration only and are not limiting of the embodiments. Other variations or modifications of the above teachings will be apparent to those of ordinary skill in the art. It is not necessary here nor is it exhaustive of all embodiments. While still being apparent from variations or modifications that may be made by those skilled in the art are within the scope of the invention.
Claims (7)
1. The utility model provides a power battery monomer thermal runaway early warning method based on high in the clouds online data which is characterized by comprising the following steps:
S1: extracting battery module boundary characteristics of battery module data collected by a cloud, and forming a high-dimensional matrix from the battery module boundary characteristics;
S2: performing dimension reduction on the high-dimensional matrix through a machine learning algorithm, extracting a low-dimensional feature matrix of the high-dimensional matrix, calculating the current failure probability at each moment in the period based on the low-dimensional feature matrix, comparing the current failure probability with a preset failure probability threshold value, and judging whether thermal runaway risk exists currently or not;
S3: when judging that the thermal runaway risk exists, decompressing the low-dimensional matrix in the step S2, calculating a contribution value of each dimension in the high-dimensional matrix in the step S1 to failure probability, and determining a single battery corresponding to the boundary feature with the largest contribution of the failure probability as a high-risk single to be checked;
S4: analyzing the on-line voltage, temperature and SOC data of the high-risk monomer to be verified, calculating the deviation degree of the on-line voltage, temperature and SOC data of the high-risk monomer to be verified in the current state and the predicted state by combining a pre-trained deep learning battery state prediction model, comparing the deviation degree value of the voltage, temperature and SOC data of the high-risk monomer to be verified with a preset three-level alarm threshold, and alarming according to the deviation degree value corresponding to different levels of the three levels.
2. The method of claim 1, wherein in step S1, the battery module data includes parameters related to thermal runaway, specifically including a voltage nerv coefficient of a battery cell, a voltage change rate of a battery cell, a differential value of a battery cell voltage versus temperature, a differential value of a battery cell voltage versus state of charge, a temperature change rate of a battery cell, internal short circuit internal resistance of a battery cell, a voltage of a battery cell, a temperature of a battery cell, and a state of charge of a battery cell.
3. The method for early warning of thermal runaway of a power battery cell based on cloud online data according to claim 1 or 2, wherein in step S1, the extracted boundary features are the maximum value, the next-to-maximum value, the standard deviation and the extreme difference extracted from the thermal runaway related parameters.
4. The method for early warning of thermal runaway of a power battery cell based on cloud online data according to claim 1, wherein in step S2, the machine learning algorithm includes a PCA algorithm for linear dimension reduction or a self-encoder algorithm for nonlinear dimension reduction.
5. The method for early warning of thermal runaway of a power battery cell based on cloud online data as claimed in claim 1, wherein in step S2, the failure probability is calculated by calculating a Hotelling T 2 statistic and an SPE statistic of the low-dimensional matrix, thereby calculating a comprehensive indexAnd return to probability distribution/>The failure probability is obtained.
6. The method for early warning of thermal runaway of a power battery monomer based on cloud online data according to claim 5, wherein in step S2, the failure probability is calculated, specifically, a Hotelling T 2 statistic and an SPE statistic are calculated according to a normal battery training set dimension reduction process matrix;
wherein, hotelling T 2 statistic is calculated according to the following formula:
Wherein X i is a boundary feature matrix of the i module at a certain moment, P k is a dimension-reduction transformation matrix, and is composed of principal component feature vectors, and S is a training sample set A diagonal matrix formed by principal component eigenvalues, k being the number of principal components;
SPE statistics are calculated as follows:
Wherein X i is a boundary feature matrix of the module group at a certain moment I, P k is a dimension reduction transformation matrix, I is a unit matrix, and k is the number of principal elements;
Solving based on Hotelling T 2 and SPE statistics according to comprehensive indexes The system failure rate function of the current moment of the sample can be obtained:
In the method, in the process of the invention, Delta 2 is the control limit value of Hotelling T 2 and SPE respectively,/>Is a symmetrical positive definite matrix, and the comprehensive index/>Coincidence probability distribution/>The degree of freedom is h, the coefficient is g, and the chi-square distribution is based on/>The probability distribution function can obtain the current failure probability function/>The failure probability h is calculated according to the following formula:
7. The method for early warning of thermal runaway of a power battery monomer based on cloud online data according to claim 1, wherein in step S3, the calculation of the contribution value is different according to different dimension reduction methods, and for a linear dimension reduction method, the duty ratio of the module boundary feature to the low-dimension feature exceeding the control limit is calculated in an accumulated manner; and calculating the duty ratio of the low-dimensional fault dimension after decompression to different dimensions for the nonlinear dimension reduction method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210403841.1A CN114734873B (en) | 2022-04-18 | 2022-04-18 | Power battery monomer thermal runaway early warning method based on cloud online data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210403841.1A CN114734873B (en) | 2022-04-18 | 2022-04-18 | Power battery monomer thermal runaway early warning method based on cloud online data |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114734873A CN114734873A (en) | 2022-07-12 |
CN114734873B true CN114734873B (en) | 2024-05-28 |
Family
ID=82280896
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210403841.1A Active CN114734873B (en) | 2022-04-18 | 2022-04-18 | Power battery monomer thermal runaway early warning method based on cloud online data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114734873B (en) |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115840157B (en) * | 2022-12-08 | 2023-08-22 | 斯润天朗(合肥)科技有限公司 | Lithium battery electrical performance index coordination analysis system based on EOF analysis |
CN116401585B (en) * | 2023-04-19 | 2023-11-10 | 江苏果下科技有限公司 | Energy storage battery failure risk assessment method based on big data |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106025405A (en) * | 2016-07-22 | 2016-10-12 | 北京航空航天大学 | Alarming device and method for quick monitoring of power battery failure |
DE102017107284A1 (en) * | 2017-04-05 | 2018-10-11 | Lisa Dräxlmaier GmbH | METHOD AND CONTROL DEVICE FOR MONITORING A PORTION NET OF A VEHICLE |
CN111624494A (en) * | 2020-04-20 | 2020-09-04 | 北京航空航天大学 | Battery analysis method and system based on electrochemical parameters |
CN112993426A (en) * | 2021-02-03 | 2021-06-18 | 武汉蔚能电池资产有限公司 | Power battery thermal runaway early warning system and method based on parking condition |
CN113752843A (en) * | 2021-11-05 | 2021-12-07 | 北京航空航天大学 | Power battery thermal runaway early warning device and method based on Saybolt physical system |
CN114240260A (en) * | 2022-02-17 | 2022-03-25 | 北京航空航天大学 | New energy group vehicle thermal runaway risk assessment method based on digital twinning |
CN114361617A (en) * | 2021-12-31 | 2022-04-15 | 重庆长安新能源汽车科技有限公司 | Power battery thermal runaway risk early warning method and early warning system |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP3657594A1 (en) * | 2018-11-22 | 2020-05-27 | Rolls-Royce Deutschland Ltd & Co KG | Method and device for detecting a thermal runaway in a battery module |
-
2022
- 2022-04-18 CN CN202210403841.1A patent/CN114734873B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106025405A (en) * | 2016-07-22 | 2016-10-12 | 北京航空航天大学 | Alarming device and method for quick monitoring of power battery failure |
DE102017107284A1 (en) * | 2017-04-05 | 2018-10-11 | Lisa Dräxlmaier GmbH | METHOD AND CONTROL DEVICE FOR MONITORING A PORTION NET OF A VEHICLE |
CN111624494A (en) * | 2020-04-20 | 2020-09-04 | 北京航空航天大学 | Battery analysis method and system based on electrochemical parameters |
CN112993426A (en) * | 2021-02-03 | 2021-06-18 | 武汉蔚能电池资产有限公司 | Power battery thermal runaway early warning system and method based on parking condition |
CN113752843A (en) * | 2021-11-05 | 2021-12-07 | 北京航空航天大学 | Power battery thermal runaway early warning device and method based on Saybolt physical system |
CN114361617A (en) * | 2021-12-31 | 2022-04-15 | 重庆长安新能源汽车科技有限公司 | Power battery thermal runaway risk early warning method and early warning system |
CN114240260A (en) * | 2022-02-17 | 2022-03-25 | 北京航空航天大学 | New energy group vehicle thermal runaway risk assessment method based on digital twinning |
Non-Patent Citations (1)
Title |
---|
锂离子电池热失控早期预警特征参数分析;李钊等;消防科学与技术;20200215(02);第8-11页 * |
Also Published As
Publication number | Publication date |
---|---|
CN114734873A (en) | 2022-07-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112092675B (en) | Battery thermal runaway early warning method, system and server | |
CN110161414B (en) | Power battery thermal runaway online prediction method and system | |
CN114734873B (en) | Power battery monomer thermal runaway early warning method based on cloud online data | |
CN104614630B (en) | A kind of recognition methods of battery micro-short circuit | |
CN113777515A (en) | Electric vehicle charging safety early warning method | |
CN113459894B (en) | Electric automobile battery safety early warning method and system | |
CN117341476B (en) | Battery differential pressure fault early warning method and system | |
CN115958957B (en) | Method and system for predicting charging overheat faults of power battery of electric automobile | |
CN104614632A (en) | Identification method for micro short circuit of battery | |
CN112363061A (en) | Thermal runaway risk assessment method based on big data | |
CN114580496A (en) | Self-detection method for single fault of lithium ion battery | |
CN114036647A (en) | Power battery safety risk assessment method based on real vehicle data | |
CN115494399A (en) | Intelligent identification method for individual difference risks of new energy automobile battery system | |
CN114460481B (en) | Bi-LSTM and attention mechanism-based energy storage battery thermal runaway early warning method | |
Xu et al. | Safety warning analysis for power battery packs in electric vehicles with running data | |
CN114545275A (en) | Indirect prediction method for remaining service life of lithium ion battery | |
Diao et al. | Research on Electric Vehicle Charging Safety Warning Based on A-LSTM Algorithm | |
CN117148200B (en) | PACK inconsistency monitoring method and device based on piecewise aggregation approximation and clustering | |
CN114509678A (en) | Power battery thermal runaway prediction method based on big data | |
CN113935423A (en) | Power battery fault early warning method and system for coupling fuzzy control rule with Elman neural network | |
CN116699407A (en) | Power battery safety risk early warning method based on safety entropy | |
CN116796285A (en) | Power battery thermal runaway advanced early warning method and system based on multi-source information fusion | |
CN116901707A (en) | Power battery pack fault early warning method, system and vehicle | |
CN115469226A (en) | Real-time safety early warning method for electric vehicle power battery based on operation big data | |
CN114839539A (en) | Lithium battery SOH estimation method based on multilevel sequence information self-adaptive fusion |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |