CN112092675A - Battery thermal runaway early warning method, system and server - Google Patents

Battery thermal runaway early warning method, system and server Download PDF

Info

Publication number
CN112092675A
CN112092675A CN202010899563.4A CN202010899563A CN112092675A CN 112092675 A CN112092675 A CN 112092675A CN 202010899563 A CN202010899563 A CN 202010899563A CN 112092675 A CN112092675 A CN 112092675A
Authority
CN
China
Prior art keywords
thermal runaway
battery
training
variables
sample data
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.)
Granted
Application number
CN202010899563.4A
Other languages
Chinese (zh)
Other versions
CN112092675B (en
Inventor
程硕
卢佳林
刘成龙
安胜伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Great Wall Motor Co Ltd
Original Assignee
Great Wall Motor Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Great Wall Motor Co Ltd filed Critical Great Wall Motor Co Ltd
Priority to CN202010899563.4A priority Critical patent/CN112092675B/en
Publication of CN112092675A publication Critical patent/CN112092675A/en
Application granted granted Critical
Publication of CN112092675B publication Critical patent/CN112092675B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • 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
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0635—Risk analysis of enterprise or organisation activities
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
    • H01M10/00—Secondary cells; Manufacture thereof
    • H01M10/42—Methods or arrangements for servicing or maintenance of secondary cells or secondary half-cells
    • H—ELECTRICITY
    • H01—ELECTRIC ELEMENTS
    • H01M—PROCESSES OR MEANS, e.g. BATTERIES, FOR THE DIRECT CONVERSION OF CHEMICAL ENERGY INTO ELECTRICAL ENERGY
    • H01M10/00—Secondary cells; Manufacture thereof
    • H01M10/42—Methods or arrangements for servicing or maintenance of secondary cells or secondary half-cells
    • H01M10/4285—Testing apparatus
    • 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
    • B60L2240/00—Control parameters of input or output; Target parameters
    • B60L2240/40—Drive Train control parameters
    • B60L2240/54—Drive Train control parameters related to batteries
    • B60L2240/545—Temperature
    • 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
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
    • Y02E60/10—Energy storage using batteries
    • 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)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • Theoretical Computer Science (AREA)
  • Educational Administration (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • General Business, Economics & Management (AREA)
  • Manufacturing & Machinery (AREA)
  • Chemical & Material Sciences (AREA)
  • Chemical Kinetics & Catalysis (AREA)
  • Electrochemistry (AREA)
  • General Chemical & Material Sciences (AREA)
  • Tourism & Hospitality (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Data Mining & Analysis (AREA)
  • Game Theory and Decision Science (AREA)
  • Evolutionary Biology (AREA)
  • Sustainable Development (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Sustainable Energy (AREA)
  • Power Engineering (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Secondary Cells (AREA)
  • Charge And Discharge Circuits For Batteries Or The Like (AREA)

Abstract

The invention provides a battery thermal runaway early warning method, a battery thermal runaway early warning system and a server. The method provided by the invention comprehensively considers the characteristic variables of the target battery, can be closer to the actual condition of the target battery, further enables the obtained thermal runaway risk score and the thermal runaway risk grade to be more objective, can accurately execute the corresponding early warning measure, and solves the problem that the battery thermal runaway risk of the electric automobile cannot be effectively monitored in the prior art.

Description

Battery thermal runaway early warning method, system and server
Technical Field
The invention relates to the technical field of new energy automobiles, in particular to a battery thermal runaway early warning method, a battery thermal runaway early warning system and a server.
Background
Currently, with the increasing global environmental protection problem, lithium ion batteries have been widely used in new energy fields such as electric vehicles and electrochemical energy storage.
However, due to high energy density and specific electrochemical characteristics, the lithium ion battery has hidden troubles in safety and stability, and thermal runaway of the battery is a main cause of ignition of the power battery. Once a lithium battery enters a thermal runaway state due to overcharge, overdischarge, internal short circuit and the like, a large amount of released heat can cause the decomposition of a Solid Electrolyte Interface (SEI) film of a negative electrode of the battery, the decomposition of an active material of a positive electrode and the oxidative decomposition of electrolyte, so that a large amount of gas is generated, and the explosion of the battery is caused, thereby endangering the personal safety and the property safety.
Disclosure of Invention
In view of this, the present invention provides a method, a system and a vehicle for early warning of thermal runaway of a battery, so as to solve the problem that the prior art cannot effectively monitor the risk of thermal runaway of the battery of an electric vehicle.
In order to achieve the purpose, the technical scheme of the invention is realized as follows:
a battery thermal runaway early warning method comprises the following steps:
acquiring a characteristic variable of a target battery;
inputting the characteristic variables of the target battery into a pre-trained thermal runaway classification model, and acquiring the probability, which is output by the thermal runaway classification model and used for representing that the target battery belongs to a thermal runaway category; the thermal runaway classification model is obtained by pre-training based on training sample data of sample batteries, and the probability that each target battery belongs to the thermal runaway category can be calculated based on different characteristic variables of different target batteries;
determining a thermal runaway risk score of the target battery according to the probability;
determining a thermal runaway risk grade corresponding to the thermal runaway risk score according to the thermal runaway risk score;
and executing early warning measures corresponding to the thermal runaway risk grade according to the thermal runaway risk grade.
Further, in the method, the thermal runaway classification model is obtained by training with the following method:
acquiring training sample data of a sample battery, wherein the training sample data comprises a characteristic variable and a label variable, and the label variable is a battery thermal runaway category;
inputting the characteristic variables and the label variables of the training sample data into a preset logistic regression model, and training the logistic regression model by adopting a preset model training method to obtain the thermal runaway classification model.
Further, the method further includes, after the inputting the characteristic variables and the tag variables of the training sample data into a preset logistic regression model and training the logistic regression model by using a preset model training method to obtain the thermal runaway classification model and before acquiring the characteristic variables of the target battery:
obtaining test sample data of a sample battery;
testing the accuracy of the thermal runaway classification model according to the test sample data;
and under the condition that the accuracy is lower than a first preset threshold, retraining the logistic regression model to obtain a thermal runaway classification model with the accuracy reaching the first preset threshold.
Further, in the method, before the feature variables and the tag variables of the training sample data are input to a preset logistic regression model and the logistic regression model is trained by using a preset model training method to obtain the thermal runaway classification model, the method further includes:
and cleaning the training sample data to remove abnormal data.
Further, in the method, the number of the feature variables of the training sample data is multiple, and the tag variable is determined by the multiple feature variables, which specifically includes:
calculating each characteristic variable of the training sample data according to a preset rule to obtain a single score;
summing the single scores according to corresponding preset weights, and determining the thermal runaway score of the target battery;
sequencing all sample batteries in the training sample data from small to large according to the thermal runaway scores;
and determining the sample batteries with preset percentage at the tail of the sequence as thermal runaway labels, and determining the rest sample batteries as non-thermal runaway labels.
Further, in the method, the feature variables of the training sample data include: at least one of the times of battery overcharge events, the times of battery overdischarge events, the times of battery high-temperature early warning events, the times of battery low-temperature early warning events, the times of cell temperature consistency abnormality, the times of cell voltage consistency abnormality and the times of state of charge value abnormality; the tag variable is of the non-thermal runaway type or the thermal runaway type.
Further, in the method, the inputting the feature variables and the tag variables of the training sample data into a preset logistic regression model, and training the logistic regression model by using a preset model training method to obtain the thermal runaway classification model includes:
coding and screening the characteristic variables, determining classification key variables, and determining classification key variables;
and inputting the classification key variable and the label variable into a preset logistic regression model, and training the logistic regression model by adopting a preset model training method to obtain the thermal runaway classification model.
Another objective of an embodiment of the present invention is to provide a battery thermal runaway early warning system, where the system includes:
the first acquisition module is used for acquiring characteristic variables of the target battery, wherein the characteristic variables are related variables of battery thermal runaway;
the second obtaining module is used for inputting the characteristic variables of the target battery into a pre-trained thermal runaway classification model and obtaining the probability, which is output by the thermal runaway classification model and used for representing that the target battery belongs to a thermal runaway category, of the target battery; the thermal runaway classification model is obtained by pre-training based on training sample data of sample batteries, and the probability that each target battery belongs to the thermal runaway category can be calculated based on different characteristic variables of different target batteries;
the first determination module is used for determining a thermal runaway risk score of the target battery according to the probability;
the second determination module is used for determining a thermal runaway risk grade corresponding to the thermal runaway risk score according to the thermal runaway risk score;
and the execution module is used for executing the early warning measure corresponding to the thermal runaway risk grade according to the thermal runaway risk grade.
Further, the system further comprises:
a third obtaining module, configured to obtain training sample data of a sample battery before obtaining a characteristic variable of a target battery, where the training sample data includes the characteristic variable and a tag variable, and the tag variable is a battery thermal runaway category;
and the training module is used for inputting the characteristic variables and the label variables of the training sample data into a preset logistic regression model, and training the logistic regression model by adopting a preset model training method to obtain the thermal runaway classification model.
Further, the system further comprises:
a third obtaining module, configured to obtain test sample data of the sample battery after the characteristic variable and the tag variable of the training sample data are input to a preset logistic regression model and the logistic regression model is trained by using a preset model training method to obtain the thermal runaway classification model and before the characteristic variable of the target battery is obtained;
the test module is used for testing the accuracy of the thermal runaway classification model according to the test sample data;
and the retraining module is used for retraining the logistic regression model under the condition that the accuracy is lower than a first preset threshold value so as to obtain a thermal runaway classification model with the accuracy reaching the first preset threshold value.
Further, the system further comprises:
and the data cleaning module is used for cleaning the training sample data to remove abnormal data before inputting the characteristic variables and the label variables of the training sample data into a preset logistic regression model and training the logistic regression model by adopting a preset model training method to obtain the thermal runaway classification model.
Further, in the system, the number of the feature variables of the training sample data is multiple, the system further includes a tag determination module, configured to determine the tag variable from the multiple feature variables, where the tag determination module specifically includes:
the calculating unit is used for calculating each characteristic variable of the training sample data according to a preset rule to obtain a corresponding single score;
the first determining unit is used for summing the single scores according to corresponding preset weights and determining the thermal runaway scores of the batteries;
the sorting unit is used for sorting all sample batteries in the training sample data from small to large according to the thermal runaway scores;
and the second determining unit is used for determining the sample batteries with the preset percentage at the tail of the sequence as the thermal runaway label and determining the rest sample batteries as the non-thermal runaway label.
Further, in the system, the training module includes:
a third determining unit, configured to perform coding processing and screening processing on the feature variables of the training sample data, and determine a classification key variable;
and the training unit is used for inputting the classification key variables and the label variables into a preset logistic regression model, and training the logistic regression model by adopting a preset model training method to obtain the thermal runaway classification model.
It is still another object of the present invention to provide a server, wherein the server is communicatively connected to a vehicle, the vehicle includes a battery, and the server includes the battery thermal runaway pre-warning system as described above.
Compared with the prior art, the battery thermal runaway early warning method, the battery thermal runaway early warning system and the battery thermal runaway early warning server have the following advantages:
the method comprises the steps of firstly obtaining characteristic variables of a target battery, then inputting the characteristic variables into a thermal runaway classification model, calculating the probability that the target battery belongs to a thermal runaway class, determining a thermal runaway risk score of the target battery according to the probability, determining a thermal runaway risk grade corresponding to the thermal runaway risk score according to the thermal runaway risk score, and then executing early warning measures corresponding to the thermal runaway risk grade according to the thermal runaway risk grade. Because the thermal runaway classification model can calculate the probability that each target battery belongs to the thermal runaway category based on different characteristic variables of different target batteries, the probability that the thermal runaway risk of the target battery occurs can be determined, the probability comprehensively considers the characteristic variables of the target battery and can be closer to the actual condition of the target battery, the obtained thermal runaway risk score and the thermal runaway risk grade are more objective, the corresponding early warning measure can be accurately executed, and the problem that the thermal runaway risk of the battery of the electric automobile cannot be effectively monitored in the prior art is solved.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate an embodiment of the invention and, together with the description, serve to explain the invention and not to limit the invention. In the drawings:
fig. 1 is a schematic flow chart of a battery thermal runaway early warning method according to an embodiment of the present invention;
fig. 2 is a schematic flow chart of a battery thermal runaway warning method according to a preferred embodiment of the invention;
fig. 3 is a schematic structural diagram of a battery thermal runaway early warning system according to an embodiment of the invention.
Detailed Description
Embodiments of the present application will be described in more detail below with reference to the accompanying drawings. While embodiments of the present application are illustrated in the accompanying drawings, it should be understood that the present application may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
The present invention will be described in detail below with reference to the embodiments with reference to the attached drawings.
Referring to fig. 1, a schematic flow chart of a battery thermal runaway early warning method provided by an embodiment of the present invention is shown, where the battery thermal runaway early warning method provided by the embodiment of the present invention includes steps S100 to S500.
In the embodiment of the invention, the method is applied to the server, the server is communicated with the vehicle-mounted communication terminal of the new energy vehicle, and the vehicle-mounted communication terminal can acquire data such as the running state, the battery working state and the driver operation of the vehicle in real time and upload the data to the server so that the server can analyze and monitor the vehicle condition and analyze related faults through cloud computing and big data technology.
And S100, acquiring characteristic variables of the target battery, wherein the characteristic variables are related variables of battery thermal runaway.
In the step S100, the target battery refers to a vehicle battery to be detected with a thermal runaway risk, and the characteristic variable may be abnormal early warning event data that may affect a thermal runaway state of the target battery, and for example, the characteristic variable may include: at least one of the times of battery overcharge events, the times of battery overdischarge events, the times of battery high-temperature early warning events, the times of battery low-temperature early warning events, the times of cell temperature consistency abnormality, the times of cell voltage consistency abnormality and the times of state of charge value abnormality.
The logic for calculating the number of the battery overcharge events is as follows:
in a charging process, recording an overcharge flag bit for a battery cell with the highest voltage value larger than a first voltage threshold; meanwhile, carrying out aggregation processing on the overcharged flag bits, and recording the overcharged behavior as 1 time when the aggregation number of the overcharged flag bits in one charging stroke reaches the threshold value of the overcharged flag bits; and accumulating the overcharge behavior times in all the charging strokes to obtain the overcharge event times of the battery. Wherein the first voltage threshold may be 4.173V.
The logic for calculating the number of the over-discharge events of the battery is as follows:
in the discharging process, under the condition that the state of charge value is smaller than the charge threshold value, recording an over-discharging zone bit by the battery cell of which the lowest voltage value is smaller than the second voltage threshold value; meanwhile, carrying out aggregation processing on the overdischarge flag bits, and recording the overcharging behavior for 1 time when the aggregation number of the overdischarge flag bits in one discharge stroke reaches the threshold value of the overdischarge flag bits; and accumulating the overdischarge behavior times in all the discharge strokes to obtain the times of the battery overdischarge events. The charge threshold may be 5%, and the first voltage threshold may be 3.350V.
The calculation logic of the number of the battery high-temperature early warning events is as follows:
in the discharging process, if the highest temperature of any one battery cell is greater than a first temperature threshold value, recording a high-temperature zone bit; simultaneously carrying out polymerization treatment on the high-temperature zone bits, and recording the high-temperature zone bits as 1-time high-temperature driving behavior when the polymerization quantity of the high-temperature zone bits in one discharge stroke reaches a high-temperature zone bit threshold value; and accumulating the times of high-temperature driving behaviors in all discharging strokes to obtain the times of the high-temperature early warning events of the battery. Wherein the first temperature threshold may be 55 ℃.
The calculation logic of the number of the low-temperature early warning events of the battery is as follows:
in the discharging process, if the lowest temperature of any single battery cell is smaller than a second temperature threshold value, recording a low-temperature zone bit; simultaneously carrying out polymerization treatment on the low-temperature zone bits, and recording the low-temperature driving behavior of 1 time when the polymerization quantity of the low-temperature zone bits in one discharge stroke reaches a low-temperature zone bit threshold value; and accumulating the low-temperature driving behavior times in all the discharging strokes to obtain the low-temperature early warning event times of the battery. Wherein the first temperature threshold may be-10 ℃.
The calculation logic of the abnormal times of the consistency of the cell temperature is as follows:
in the discharging process, if the temperature range of each battery cell is greater than a third temperature threshold value at any time, recording a temperature abnormal zone bit for 1 time; simultaneously carrying out polymerization treatment on the temperature abnormal zone bits, and recording as 1-time temperature consistency abnormality when the polymerization number of the temperature abnormal zone bits in one discharge stroke reaches the threshold value of the temperature abnormal zone bits; and accumulating the abnormal times of the temperature consistency in all the discharging strokes to obtain the abnormal times of the temperature consistency of the battery core. Wherein the third temperature threshold may be 4 ℃.
The calculation logic of the abnormal times of the cell voltage consistency is as follows:
in the discharging process, if the voltage range of each battery cell is larger than a third voltage threshold value at any time, recording a voltage abnormal zone bit for 1 time; simultaneously carrying out polymerization treatment on the voltage abnormal zone bits, and recording the voltage consistency abnormality for 1 time when the polymerization number of the voltage abnormal zone bits in one discharge stroke reaches the threshold value of the voltage abnormal zone bits; and accumulating the abnormal times of the voltage consistency in all the discharging strokes to obtain the abnormal times of the cell voltage consistency. Wherein the third voltage threshold may be 200 mV.
The logic for calculating the abnormal times of the state of charge value is as follows:
the state of charge value abnormality may include a state of charge value self-deviation abnormality, a state of charge value change abnormality, and a state of charge value drop abnormality.
The self-bias difference of the state of charge values usually means that the difference between a monitored state of charge value and an actual state of charge value reaches a first charge change threshold value in a charging process, the monitored state of charge value refers to the state of charge value obtained by directly detecting and calculating a battery, and the actual state of charge value refers to the state of charge value obtained by calculating charging and discharging data of an electricity utilization cabinet; and recording the self-deviation abnormality of the charge state value for 1 time when the difference value between the monitored charge state value and the actual charge state value for 1 time is detected to reach the first charge change threshold value. Illustratively, the first charge change threshold is 4%.
The abnormal change of the state of charge value refers to the fact that the difference value of the internal state of charge values of the batteries of the two adjacent vehicle condition data packets reaches a second state of charge change threshold value; and accumulating the times that the difference value of the internal charge state values of the batteries of the two adjacent vehicle condition data packets reaches a second charge change threshold value to obtain the self-bias difference frequent times of the charge state values. Illustratively, the second charge change threshold is 4%.
The abnormal decrease of the SOC value refers to a situation that the SOC value of the battery decreases too fast, and may be specifically divided into three situations, i.e., a low risk of Δ SOC, a medium risk of Δ SOC, and a high risk of Δ SOC, according to the decrease range from small to large. If the SOC value of the battery is connected for four times according to a preset time interval and the reduction amplitude is less than 0.5%, recording 1 time of low-risk SOC value reduction abnormal event; if the SOC value of the battery is connected for four times according to the preset time interval and the reduction range is 0.5% -1%, recording 1 time of abnormal event of reduction of the dangerous SOC value; and if the SOC value of the battery is connected for four times according to the preset time interval and the reduction amplitude is larger than 1%, recording 1 high-risk SOC value reduction abnormal event.
And accumulating the self-bias difference constant times of the state of charge values, the change abnormal times of the state of charge values and the reduction abnormal times of the state of charge values to obtain the abnormal times of the state of charge values.
Specifically, when the characteristic variable of the target battery is obtained, the relevant data of the target battery uploaded by the vehicle-mounted communication terminal can be obtained through the server firstly and serve as the battery data, and data which can affect the thermal runaway state of the battery generally has multiple types, so that the characteristic variable under the characteristic dimension can be extracted from the battery data based on multiple different characteristic dimensions, and therefore, the characteristic variables under the multiple characteristic dimensions are extracted, so that when the thermal runaway category is determined based on the characteristic variable in subsequent supplement, the target battery can be comprehensively analyzed from the perspective of different characteristic dimensions, and the final analysis result is more objective and reasonable.
S200, inputting the characteristic variables of the target battery into a pre-trained thermal runaway classification model, and acquiring the probability, which is output by the thermal runaway classification model and used for representing that the target battery belongs to a thermal runaway category, of the target battery; the thermal runaway classification model is obtained by pre-training based on training sample data of sample batteries, and the probability that each target battery belongs to the thermal runaway category can be calculated based on different characteristic variables of different target batteries.
In the step S200, the characteristic variables of the target battery may be input into a thermal runaway classification model trained in advance, and the thermal runaway classification model may calculate the probability that the target battery belongs to the thermal runaway class based on the characteristic variables. The thermal runaway classification model can be obtained by pre-training based on training sample data, and the probability that each target battery belongs to the thermal runaway category can be calculated based on different characteristic variables of different target batteries, so that after the probability that each target battery belongs to the thermal runaway category is calculated, whether the target battery specifically belongs to the thermal runaway category or the non-thermal runaway category in the thermal runaway category can be determined according to the probability.
And S300, determining the thermal runaway risk score of the target battery according to the probability.
In the step S300, the thermal runaway risk score is a probability high-low score of the thermal runaway risk of the battery, and reflects the risk degree of the thermal runaway of the battery. Determining a thermal runaway risk score according to the probability determined in the step S200, and also determining whether the target battery specifically belongs to a thermal runaway class or a non-thermal runaway class in the thermal runaway class according to the probability. In practical application, the probability that the target battery belongs to the thermal runaway class calculated by the thermal runaway classification model is multiplied by 100, and the thermal runaway risk score of the target battery is determined.
And S300, determining a thermal runaway risk grade corresponding to the thermal runaway risk score according to the thermal runaway risk score.
In the step S300, the thermal runaway risk levels set according to different thermal runaway risk scores are preset, that is, different thermal runaway risk scores all have corresponding thermal runaway risk levels. Specifically, the thermal runaway risk score may be divided into a plurality of intervals, each interval corresponds to one thermal runaway risk level, and the higher the thermal runaway risk score is, the higher the thermal runaway risk level corresponding to the interval is.
And S400, executing early warning measures corresponding to the thermal runaway risk grade according to the thermal runaway risk grade.
In the step S400, corresponding early warning measures need to be set in advance for different thermal runaway risk levels, that is, the thermal runaway risk levels and the early warning measures meet a preset corresponding relationship, where the early warning measures are measures for risk prevention, reminding and coping with the thermal runaway of the battery. After the thermal runaway risk level is determined according to the step S300, the corresponding pre-warning measure is determined according to the preset corresponding relationship, and the corresponding pre-warning measure is executed, so that the monitoring and pre-warning of the thermal runaway of the battery can be realized.
Compared with the prior art, the battery thermal runaway early warning method has the following advantages:
the method comprises the steps of firstly obtaining characteristic variables of a target battery, then inputting the characteristic variables into a thermal runaway classification model, calculating the probability that the target battery belongs to a thermal runaway class, determining a thermal runaway risk score of the target battery according to the probability, determining a thermal runaway risk grade corresponding to the thermal runaway risk score according to the thermal runaway risk score, and then executing early warning measures corresponding to the thermal runaway risk grade according to the thermal runaway risk grade. Because the thermal runaway classification model can calculate the probability that each target battery belongs to the thermal runaway category based on different characteristic variables of different target batteries, the probability that the thermal runaway risk of the target battery occurs can be determined, the probability comprehensively considers the characteristic variables of the target battery and can be closer to the actual condition of the target battery, the obtained thermal runaway risk score and the thermal runaway risk grade are more objective, the corresponding early warning measure can be accurately executed, and the problem that the thermal runaway risk of the battery of the electric automobile cannot be effectively monitored in the prior art is solved.
In practical applications, the characteristic variables need to be acquired by the server from the charging data and the discharging data of the target battery uploaded by the vehicle-mounted communication terminal.
In addition, because invalid or abnormal data with a certain probability exists in the data, the obtained data needs to be cleaned, and abnormal data is filtered.
Specifically, the data cleansing process includes step S1001: deleting data that the highest voltage of the battery monomer exceeds the voltage threshold range, deleting data that the lowest voltage of the battery monomer exceeds the voltage threshold range, deleting data that the highest temperature of the battery monomer exceeds the vehicle model calibration temperature threshold range, deleting data that the lowest temperature of the battery monomer exceeds the vehicle model calibration temperature threshold range, deleting data that the charging state is outside the signal list definition range, deleting data that the vehicle speed exceeds the vehicle speed threshold range, and deleting data that the state of charge value is greater than 100% and less than 0. The various thresholds can be defined according to actual conditions of different vehicle types.
Optionally, before the step S100, a thermal runaway classification model may be established in advance. Specifically, the process of establishing the thermal runaway classification model may include steps S101 to S102:
s101, obtaining training sample data of a sample battery, wherein the training sample data comprises characteristic variables and label variables, and the label variables are battery thermal runaway categories.
In the above step S101, the training sample data may include a feature variable of the sample battery and a label variable, where the feature variable may be a feature variable of a plurality of different feature dimensions of the sample battery. The characteristic variable may be data that may affect the target battery thermal runaway state, and may include, for example: at least one of the times of battery overcharge events, the times of battery overdischarge events, the times of battery high-temperature early warning events, the times of battery low-temperature early warning events, the times of cell temperature consistency abnormality, the times of cell voltage consistency abnormality and the times of state of charge value abnormality. The label variable represents the thermal runaway category of the sample battery, and can be a thermal runaway label or a non-thermal runaway label, wherein the thermal runaway label represents that the thermal runaway risk exists, and the non-thermal runaway label represents that the thermal runaway risk does not exist; wherein, the label of non-thermal runaway class can be represented by 0, and is used for 1 label of thermal runaway class, so that the label is digitalized, and the computer can conveniently identify and calculate.
In practical application, the batteries are batteries of the same vehicle type.
Optionally, in a specific embodiment, the feature variables of the training sample data include a plurality of feature variables, and the tag variables in the training sample are determined by the plurality of feature variables, which specifically includes steps S1011 to S1014.
Step S1011, calculating each characteristic variable of the training sample data according to a preset rule to obtain a corresponding single score.
In the step S1011, the preset rule is a preset calculation rule for calculating scores corresponding to the characteristic variables, and the step first calculates corresponding scores for different characteristic variables according to the preset rule, so as to calculate the thermal runaway score of the corresponding battery.
And step S1012, summing the single scores according to corresponding preset weights, and determining the thermal runaway scores of the batteries.
In the step S1012, it is necessary to set a corresponding weight for each characteristic variable of the battery in advance, that is, to set a corresponding scoring weight for different abnormal early warning events of the battery, that is, the preset weight, where the preset weight is specifically set according to actual conditions of the battery and the vehicle.
In step S1012, each characteristic variable of the battery is multiplied by a preset weight corresponding to the characteristic variable, and then added to obtain the thermal runaway score of the battery.
Specifically, the preset weighting coefficients of different abnormal early warning events are shown in table 1:
TABLE 1
Anomaly early warning event Weight coefficient
Overcharge 0.2
Over-discharge 0.1
Battery high temperature warning 0.1
Battery low temperature warning 0.1
Abnormal temperature consistency of vehicle battery core 0.3
Abnormal cell voltage uniformity 0.1
Delta SOC anomaly 0.1
In practical application, the setting of each weight coefficient needs to refer to the specific actual conditions of different vehicle types, and the weight coefficient of the early warning event with high discrimination is set to be relatively large.
And S1013, sorting all sample batteries in the training sample data from small to large according to the thermal runaway scores.
Because the training sample data includes data of a plurality of sample cells, the thermal runaway score of each sample cell in the training sample data can be calculated according to the step S1012, and in the step S1013, each sample cell in the training sample data is sorted from small to large according to the thermal runaway score to obtain a sample cell sequence.
Step 1014, determining the sample battery with the preset percentage at the tail of the sequence as a thermal runaway label, and determining the rest sample batteries as non-thermal runaway labels.
In the above step S1014, the sample battery located at the end of the battery sequence composed of the batteries in the training sample data according to the thermal runaway score and occupying a preset percentage of the total number of sample batteries in the training sample data is defined as the thermal runaway battery, that is, the label of the sample battery is set as the thermal runaway label, while the other sample batteries are defined as the non-thermal runaway batteries, that is, the normal batteries, and the corresponding labels are the non-thermal runaway labels.
Through the above specific implementation, the training sample data is divided into two types, one type represents a normal battery, that is, a vehicle corresponding to the normal battery can be represented by 0; another class, which represents a battery having a risk of thermal runaway, i.e. a vehicle corresponding to a battery comprising such a battery having a risk of thermal runaway, can be denoted by 1.
And S102, inputting the characteristic variables and the label variables of the training sample data into a preset logistic regression model, and training the logistic regression model by adopting a preset model training method to obtain the thermal runaway classification model.
In the step S102, the logistic regression model is trained by using the characteristic variables of the battery in the training sample data and the corresponding label variables, so as to obtain the thermal runaway classification model.
Specifically, for convenience of computer identification and calculation, feature vectors of corresponding dimensions may be constructed according to the feature variables obtained in step S102, and then the feature vectors and the label vectors corresponding to the training samples are input into a logistic regression model for training to obtain a thermal runaway classification model.
When constructing the feature vector according to the acquired feature variables, the feature variables also need to be cleaned first, and specifically, the cleaning can be realized by performing error correction processing on feature dimensions in which abnormal values appear in the feature variables.
Specifically, the data cleansing process includes: deleting data that the highest voltage of the battery monomer exceeds the voltage threshold range, deleting data that the lowest voltage of the battery monomer exceeds the voltage threshold range, deleting data that the highest temperature of the battery monomer exceeds the vehicle model calibration temperature threshold range, deleting data that the lowest temperature of the battery monomer exceeds the vehicle model calibration temperature threshold range, deleting data that the charging state is outside the signal list definition range, deleting data that the vehicle speed exceeds the vehicle speed threshold range, and deleting data that the state of charge value is greater than 100% and less than 0. The various thresholds can be defined according to actual conditions of different vehicle types.
In the present embodiment, the logistic regression model is trained by training sample data, and a thermal runaway classification model for determining the thermal runaway category of the vehicle battery is obtained.
Optionally, in an embodiment, the step S102 includes steps S1021 to S1022:
and S1021, coding and screening the characteristic variables of the training sample data, and determining classification key variables.
In the step S1021, the classification key variable is a feature variable that has a high degree of contribution to determining the battery thermal runaway category, and for example, the classification key variable is a feature variable that has a degree of contribution to determining the battery thermal runaway category that reaches a second preset threshold.
Because the value of the feature data under the discrete type feature dimension is often a discrete value, and the difficulty of extracting numerical features from the discrete value is often high for the model, in the step, the value in the discrete type feature data can be encoded to be vectorized, so that the model can conveniently extract the features in the value, and the processing effect is further improved
In the step S1021, the characteristic variables may be analyzed by using the Evidence Weight (WOE), the IV (Information Value), the multiple collinearity, and the like, to newly construct the characteristics or to find out the classification key variables affecting the thermal runaway, which is used for the subsequent training of the model.
Specifically, the IV value may be used to screen the classification-type features in the feature variables, and the classification-type features are converted into numerical-type features by encoding based on the WOE, after the WOE is encoded, the feature variables have standardized properties, and various values inside the feature variables can be directly compared with each other, and various values between different feature variables can also be directly compared with each other by the WOE, so that the variation or fluctuation condition of the WOE value inside the feature variables can be studied, and the contribution rate and relative importance of each feature variable are constructed by combining the coefficients fitted by the model, and further, the classification key variables can be determined.
Step S1022, inputting the classification key variables and the label variables into a preset logistic regression model, and training the logistic regression model by adopting a preset model training method to obtain the thermal runaway classification model.
In step S1022, the classified key variables and the corresponding label variables belonging to the same battery are input into the logistic regression model to train the logistic regression model, so as to obtain the thermal runaway classification model.
In the embodiment, in the training of the thermal runaway classification model by using the training sample data, the preset mathematical model is used for analyzing and obtaining the classification key variables, so that the logistic regression model can be trained more pertinently, and the reliability of the model training is improved.
Optionally, in an embodiment, after the step S102 and before the step S100, the method for warning a thermal runaway of a battery further includes steps S103 to S105.
And S103, obtaining test sample data of the sample battery.
The test sample data also includes a feature vector and a label vector of the sample battery, and the label vector may also be calculated according to the feature vector of the sample battery, and the specific manner may refer to the above steps S1011 to S1014, which is not described herein again.
And S104, testing the accuracy of the thermal runaway classification model according to the test sample data.
In the step S104, that is, the feature vector in each test sample data is input into the thermal runaway classification model, the test thermal runaway category of each test sample data is obtained, the number of correct test thermal runaway categories is determined according to the test thermal runaway category and the label vector, and finally, the ratio of the number of correct test thermal runaway categories to the total number of test training sample data can be determined as the accuracy of the thermal runaway classification model.
In practical application, before the test sample data is used to test the accuracy of the thermal runaway classification model, the test sample data may also be cleaned first to remove abnormal data, and the specific data cleaning manner may be the parameter of step S1001, which is not described herein again.
And S105, retraining the logistic regression model under the condition that the accuracy is lower than a first preset threshold value, so as to obtain a thermal runaway classification model with the accuracy reaching the first preset threshold value.
If the accuracy calculated in step S104 does not reach the first preset threshold, it indicates that the difference between the calculation result and the true value represented by the label vector is relatively large, i.e., it indicates that the current parameters in the thermal runaway classification model are not reasonable, and the current logical classification model cannot accurately determine the thermal runaway class, so that the logistic regression model needs to be retrained by using new training sample data to further adjust the parameters thereof, so as to obtain the thermal runaway classification model with the accuracy reaching the first preset threshold.
In the embodiment, the thermal runaway classification model is tested by using the test sample data, so that the thermal runaway classification model obtained by training is ensured, and the thermal runaway category can be more accurately determined.
In practical application, the training sample data and the test sample data are proportionally divided into a training set and a test set by the same batch of sample battery data, the training set corresponds to the training sample data, and the test set corresponds to the test sample data.
A flow diagram of a battery thermal runaway early warning method provided in a preferred embodiment of the present invention is shown in fig. 2, where the method includes steps S201 to S215.
Step S201, obtaining training sample data of a sample battery, wherein the training sample data comprises characteristic variables.
The above step S201 can refer to the description of step S101, and is not described herein again.
Step S202, cleaning the training sample data to remove abnormal data.
The above step S202 can refer to the description of step S1001, and is not described herein again.
Step S203, calculating each characteristic variable of the training sample data according to a preset rule to obtain a corresponding single score.
The above step S203 can refer to the description of step S1011, and will not be described herein.
And S204, summing the single scores according to corresponding preset weights, and determining the thermal runaway scores of the sample batteries.
Step S204 may refer to the description of step S1012, which is not described herein again.
And S205, sorting all sample batteries in the training sample data from small to large according to the thermal runaway scores.
The above step S205 can refer to the description of step S1013, and will not be described herein again.
And S206, determining the sample batteries with the preset percentage at the tail of the sequence as thermal runaway labels, and determining the rest sample batteries as non-thermal runaway labels.
The above step S206 can refer to the description of step S1014, which is not described herein again.
And S207, coding and screening the characteristic variables of the training sample data, and determining classification key variables.
Step S207 can refer to the description of step S1021, and is not described herein again.
And S208, inputting the classification key variables and the label variables into a preset logistic regression model, and training the logistic regression model by adopting a preset model training method to obtain the thermal runaway classification model.
The above step S208 can refer to the description of step S1022, and is not described herein again.
And S209, obtaining test sample data of the sample battery.
The above step S209 can refer to the description of step S103, and is not described herein again.
And S210, testing the accuracy of the thermal runaway classification model according to the test sample data.
The above step S210 can refer to the description of step S104, and is not described herein again.
And S211, retraining the logistic regression model under the condition that the accuracy is lower than a first preset threshold value, so as to obtain a thermal runaway classification model with the accuracy reaching the first preset threshold value.
The above step S211 can refer to the description of step S105, and is not described herein again.
And step S212, acquiring characteristic variables of the target battery.
The above step S212 can refer to the description of step S100, and is not described herein again.
Step S213, inputting the characteristic variables of the target battery into a thermal runaway classification model, calculating the probability that the target battery belongs to the thermal runaway class, and determining the thermal runaway risk score of the target battery according to the probability.
The above step S213 can refer to the description of step S200, and is not described herein again.
And S214, determining a thermal runaway risk grade corresponding to the thermal runaway risk score according to the thermal runaway risk score.
The above step S214 can refer to the description of step S300, and is not described herein again.
And S215, executing early warning measures corresponding to the thermal runaway risk level according to the thermal runaway risk level.
The above step S215 can refer to the description of step S400, and will not be described herein.
Compared with the prior art, the battery thermal runaway early warning method has the following advantages:
training a thermal runaway classification model by using training sample data, testing the accuracy of the thermal runaway classification model obtained by training by using test sample data, and finishing the training of the thermal runaway classification model when the accuracy of the thermal runaway classification model reaches a first preset threshold; and then acquiring a characteristic variable of the target battery, inputting the characteristic variable into a thermal runaway classification model, calculating the probability that the target battery belongs to a thermal runaway class, determining a thermal runaway risk score of the target battery according to the probability, determining a thermal runaway risk grade corresponding to the thermal runaway risk score according to the thermal runaway risk score, and executing an early warning measure corresponding to the thermal runaway risk grade according to the thermal runaway risk grade. Because the thermal runaway classification model can calculate the probability that each target battery belongs to the thermal runaway category based on different characteristic variables of different target batteries, the probability that the thermal runaway risk of the target battery occurs can be determined, the probability comprehensively considers the characteristic variables of the target battery and can be closer to the actual condition of the target battery, the obtained thermal runaway risk score and the thermal runaway risk grade are more objective, the corresponding early warning measure can be accurately executed, and the problem that the thermal runaway risk of the battery of the electric automobile cannot be effectively monitored in the prior art is solved.
Another objective of the present invention is to provide a battery thermal runaway early warning system, wherein, referring to fig. 3, fig. 3 shows a schematic structural diagram of a battery thermal runaway early warning system according to an embodiment of the present invention, the system includes:
the first obtaining module 31 is configured to obtain a characteristic variable of a target battery, where the characteristic variable is a variable related to battery thermal runaway;
a second obtaining module 32, configured to input the characteristic variable of the target battery to a pre-trained thermal runaway classification model, and obtain a probability that is output by the thermal runaway classification model and used for representing that the target battery belongs to a thermal runaway category; the thermal runaway classification model is obtained by pre-training based on training sample data of sample batteries, and the probability that each target battery belongs to the thermal runaway category can be calculated based on different characteristic variables of different target batteries;
a first determining module 33, configured to determine a thermal runaway risk score of the target battery according to the probability;
a second determining module 34, configured to determine, according to the thermal runaway risk score, a thermal runaway risk level corresponding to the thermal runaway risk score;
and the execution module 35 is configured to execute an early warning measure corresponding to the thermal runaway risk level according to the thermal runaway risk level. .
In the system according to the embodiment of the present invention, first, a first obtaining module 31 obtains a characteristic variable of a target battery, then, a second obtaining module 32 inputs the characteristic variable into a thermal runaway classification model, a probability that the target battery belongs to a thermal runaway class is calculated, then, a first determining module 33 determines a thermal runaway risk score of the target battery according to the probability, and then, a second determining module 34 determines a thermal runaway risk grade corresponding to the thermal runaway risk score according to the thermal runaway risk score, so that an executing module 35 executes an early warning measure corresponding to the thermal runaway risk grade according to the thermal runaway risk grade. Because the thermal runaway classification model can calculate the probability that each target battery belongs to the thermal runaway category based on different characteristic variables of different target batteries, the probability that the thermal runaway risk of the target battery occurs can be determined, the probability comprehensively considers the characteristic variables of the target battery and can be closer to the actual condition of the target battery, the obtained thermal runaway risk score and the thermal runaway risk grade are more objective, the corresponding early warning measure can be accurately executed, and the problem that the thermal runaway risk of the battery of the electric automobile cannot be effectively monitored in the prior art is solved.
Optionally, the system further comprises:
a third obtaining module, configured to obtain training sample data of a sample battery before obtaining a characteristic variable of a target battery, where the training sample data includes the characteristic variable and a tag variable, and the tag variable is a battery thermal runaway category;
and the training module is used for inputting the characteristic variables and the label variables of the training sample data into a preset logistic regression model, and training the logistic regression model by adopting a preset model training method to obtain the thermal runaway classification model.
Optionally, the system further comprises:
a third obtaining module, configured to obtain test sample data of the sample battery after the characteristic variable and the tag variable of the training sample data are input to a preset logistic regression model and the logistic regression model is trained by using a preset model training method to obtain the thermal runaway classification model and before the characteristic variable of the target battery is obtained;
the test module is used for testing the accuracy of the thermal runaway classification model according to the test sample data;
and the retraining module is used for retraining the logistic regression model under the condition that the accuracy is lower than a first preset threshold value so as to obtain a thermal runaway classification model with the accuracy reaching the first preset threshold value.
Optionally, the system further comprises:
and the data cleaning module is used for cleaning the training sample data to remove abnormal data before inputting the characteristic variables and the label variables of the training sample data into a preset logistic regression model and training the logistic regression model by adopting a preset model training method to obtain the thermal runaway classification model.
Further, in the system, the number of the feature variables is multiple, the system further includes a tag determination module configured to determine the tag variable from the multiple feature variables, and the tag determination module specifically includes:
the calculating unit is used for calculating each characteristic variable of the training sample data according to a preset rule to obtain a corresponding single score;
the first determining unit is used for summing the single scores according to corresponding preset weights and determining the thermal runaway scores of the batteries;
the sorting unit is used for sorting all sample batteries in the training sample data from small to large according to the thermal runaway scores;
and the second determining unit is used for determining the sample batteries with the preset percentage at the tail of the sequence as the thermal runaway label and determining the rest sample batteries as the non-thermal runaway label.
Further, in the system, the training module includes:
a third determining unit, configured to perform coding processing and screening processing on the feature variables of the training sample data, and determine a classification key variable;
and the training unit is used for inputting the classification key variables and the label variables into a preset logistic regression model, and training the logistic regression model by adopting a preset model training method to obtain the thermal runaway classification model.
It is still another object of the present invention to provide a server, wherein the server is communicatively connected to a vehicle, the vehicle includes a battery, and the server includes the battery thermal runaway pre-warning system as described above.
Compared with the prior art, the advantages of the battery thermal runaway early warning system, the server and the battery thermal runaway early warning method are the same, and are not repeated herein
In summary, according to the battery thermal runaway early warning method, the battery thermal runaway early warning system and the battery thermal runaway early warning server, the characteristic variable of the target battery is obtained firstly, then the characteristic variable is input into a thermal runaway classification model, the probability that the target battery belongs to the thermal runaway category is calculated, the thermal runaway risk score of the target battery is determined according to the probability, the thermal runaway risk grade corresponding to the thermal runaway risk score is determined according to the thermal runaway risk score, and then the early warning measure corresponding to the thermal runaway risk grade can be executed according to the thermal runaway risk grade. Because the thermal runaway classification model can calculate the probability that each target battery belongs to the thermal runaway category based on different characteristic variables of different target batteries, the probability that the thermal runaway risk of the target battery occurs can be determined, the probability comprehensively considers the characteristic variables of the target battery and can be closer to the actual condition of the target battery, the obtained thermal runaway risk score and the thermal runaway risk grade are more objective, the corresponding early warning measure can be accurately executed, and the problem that the thermal runaway risk of the battery of the electric automobile cannot be effectively monitored in the prior art is solved.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (10)

1. A battery thermal runaway early warning method is characterized by comprising the following steps:
acquiring a characteristic variable of a target battery, wherein the characteristic variable is a related variable of battery thermal runaway;
inputting the characteristic variables of the target battery into a pre-trained thermal runaway classification model, and acquiring the probability, which is output by the thermal runaway classification model and used for representing that the target battery belongs to a thermal runaway category; the thermal runaway classification model is obtained by pre-training based on training sample data of sample batteries, and the probability that each target battery belongs to the thermal runaway category can be calculated based on different characteristic variables of different target batteries;
determining a thermal runaway risk score of the target battery according to the probability;
determining a thermal runaway risk grade corresponding to the thermal runaway risk score according to the thermal runaway risk score;
and executing early warning measures corresponding to the thermal runaway risk grade according to the thermal runaway risk grade.
2. The method of claim 1, wherein the thermal runaway classification model is trained using:
acquiring training sample data of a sample battery, wherein the training sample data comprises a characteristic variable and a label variable, and the label variable is a battery thermal runaway category;
inputting the characteristic variables and the label variables of the training sample data into a preset logistic regression model, and training the logistic regression model by adopting a preset model training method to obtain the thermal runaway classification model.
3. The method according to claim 2, wherein the training sample data includes a plurality of feature variables, and the tag variable is determined from the plurality of feature variables, and specifically includes:
calculating each characteristic variable of the training sample data according to a preset rule to obtain a corresponding single score;
summing the single scores according to corresponding preset weights, and determining the thermal runaway scores of the batteries;
sequencing all sample batteries in the training sample data from small to large according to the thermal runaway scores;
and determining the sample batteries with preset percentage at the tail of the sequence as thermal runaway labels, and determining the rest sample batteries as non-thermal runaway labels.
4. The method of claim 2, wherein the feature variables of the training sample data comprise: at least one of the times of battery overcharge events, the times of battery overdischarge events, the times of battery high-temperature early warning events, the times of battery low-temperature early warning events, the times of cell temperature consistency abnormality, the times of cell voltage consistency abnormality and the times of state of charge value abnormality; the tag variable is of the non-thermal runaway type or the thermal runaway type.
5. The method according to claim 4, wherein the inputting the feature variables and the tag variables of the training sample data into a preset logistic regression model and training the logistic regression model by using a preset model training method to obtain the thermal runaway classification model comprises:
coding and screening the characteristic variables of the training sample data to determine classification key variables;
and inputting the classification key variable and the label variable into a preset logistic regression model, and training the logistic regression model by adopting a preset model training method to obtain the thermal runaway classification model.
6. A battery thermal runaway warning system, the system comprising:
the first acquisition module is used for acquiring characteristic variables of the target battery, wherein the characteristic variables are related variables of battery thermal runaway;
the second obtaining module is used for inputting the characteristic variables of the target battery into a pre-trained thermal runaway classification model and obtaining the probability, which is output by the thermal runaway classification model and used for representing that the target battery belongs to a thermal runaway category, of the target battery; the thermal runaway classification model is obtained by pre-training based on training sample data of sample batteries, and the probability that each target battery belongs to the thermal runaway category can be calculated based on different characteristic variables of different target batteries;
the first determination module is used for determining a thermal runaway risk score of the target battery according to the probability;
the second determination module is used for determining a thermal runaway risk grade corresponding to the thermal runaway risk score according to the thermal runaway risk score;
and the execution module is used for executing the early warning measure corresponding to the thermal runaway risk grade according to the thermal runaway risk grade.
7. The system of claim 6, further comprising:
a third obtaining module, configured to obtain training sample data of a sample battery before obtaining a characteristic variable of a target battery, where the training sample data includes the characteristic variable and a tag variable, and the tag variable is a battery thermal runaway category;
and the training module is used for inputting the characteristic variables and the label variables of the training sample data into a preset logistic regression model, and training the logistic regression model by adopting a preset model training method to obtain the thermal runaway classification model.
8. The system according to claim 7, wherein the training sample data includes a plurality of feature variables, the system further includes a label determination module configured to determine the label variable from the plurality of feature variables, and the label determination module specifically includes:
the calculating unit is used for calculating each characteristic variable of the training sample data according to a preset rule to obtain a corresponding single score;
the first determining unit is used for summing the single scores according to corresponding preset weights and determining the thermal runaway scores of the batteries;
the sorting unit is used for sorting all sample batteries in the training sample data from small to large according to the thermal runaway scores;
and the second determining unit is used for determining the sample batteries with the preset percentage at the tail of the sequence as the thermal runaway label and determining the rest sample batteries as the non-thermal runaway label.
9. The system of claim 8, wherein the training module comprises:
a third determining unit, configured to perform coding processing and screening processing on the feature variables of the training sample data, and determine a classification key variable;
and the training unit is used for inputting the classification key variables and the label variables into a preset logistic regression model, and training the logistic regression model by adopting a preset model training method to obtain the thermal runaway classification model.
10. A server, wherein the server is in communication connection with a vehicle, the vehicle comprises a battery, and the server comprises the battery thermal runaway pre-warning system according to any one of claims 6 to 9.
CN202010899563.4A 2020-08-31 2020-08-31 Battery thermal runaway early warning method, system and server Active CN112092675B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010899563.4A CN112092675B (en) 2020-08-31 2020-08-31 Battery thermal runaway early warning method, system and server

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010899563.4A CN112092675B (en) 2020-08-31 2020-08-31 Battery thermal runaway early warning method, system and server

Publications (2)

Publication Number Publication Date
CN112092675A true CN112092675A (en) 2020-12-18
CN112092675B CN112092675B (en) 2022-03-25

Family

ID=73757052

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010899563.4A Active CN112092675B (en) 2020-08-31 2020-08-31 Battery thermal runaway early warning method, system and server

Country Status (1)

Country Link
CN (1) CN112092675B (en)

Cited By (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113040711A (en) * 2021-03-03 2021-06-29 吾征智能技术(北京)有限公司 Cerebral stroke attack risk prediction system, equipment and storage medium
CN113341332A (en) * 2021-05-25 2021-09-03 三一重机有限公司 Battery thermal runaway early warning method and device, operation machine and electronic equipment
CN113437371A (en) * 2021-05-19 2021-09-24 湖南大学 Early warning system and early warning method for thermal runaway of lithium ion battery of new energy automobile
CN113689649A (en) * 2021-07-31 2021-11-23 华能新能源股份有限公司 Fire early warning method, fire early warning system and computer equipment
WO2022033348A1 (en) * 2020-08-10 2022-02-17 中兴通讯股份有限公司 Vehicle thermal runaway processing method and apparatus, vehicle, and computer readable storage medium
CN114137421A (en) * 2021-11-30 2022-03-04 蜂巢能源(上海)有限公司 Battery abnormality detection method, apparatus, device and storage medium
CN114179675A (en) * 2022-02-16 2022-03-15 深圳康普盾科技股份有限公司 Battery replacement cabinet safety control method and system based on Internet of things and storage medium
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
CN114295983A (en) * 2021-12-31 2022-04-08 上海洛轲智能科技有限公司 Battery thermal runaway early warning method and device, vehicle, equipment and storage medium
CN114325406A (en) * 2021-12-30 2022-04-12 重庆长安新能源汽车科技有限公司 Method and system for predicting thermal runaway of battery based on machine learning thinking
CN114312322A (en) * 2021-12-31 2022-04-12 中国第一汽车股份有限公司 Vehicle detection method and device
WO2022135520A1 (en) * 2020-12-24 2022-06-30 广州橙行智动汽车科技有限公司 Battery cell capacity anomaly determination method and apparatus, and vehicle and storage medium
WO2022174601A1 (en) * 2021-02-22 2022-08-25 华为技术有限公司 Method for training battery state prediction model, and related apparatus
CN115548480A (en) * 2022-11-01 2022-12-30 湖南东舟能源有限公司 Battery system and control method of battery system
CN115575830A (en) * 2022-10-10 2023-01-06 中国标准化研究院 Method and system for identifying inducement of thermal runaway of lithium battery
WO2023011066A1 (en) * 2021-08-03 2023-02-09 长城汽车股份有限公司 Performance test method for battery thermal management system, and related device
WO2023097507A1 (en) * 2021-11-30 2023-06-08 华为技术有限公司 Battery safety early warning method and apparatus
WO2023115988A1 (en) * 2021-12-22 2023-06-29 北京国家新能源汽车技术创新中心有限公司 Method for detecting internal short-circuit of traction battery
WO2023142065A1 (en) * 2022-01-29 2023-08-03 宁德时代新能源科技股份有限公司 Battery state evaluation method and apparatus, electronic device, and storage medium
CN117388725A (en) * 2023-03-13 2024-01-12 中国石油大学(华东) Early abnormality early warning method based on expansion force of lithium ion battery
WO2024113182A1 (en) * 2022-11-29 2024-06-06 宁德时代新能源科技股份有限公司 Battery risk level determination method, apparatus, storage medium and battery management system

Citations (26)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003303344A (en) * 2002-04-10 2003-10-24 Nec Corp Method and device for extracting image region
US20030225601A1 (en) * 2002-05-31 2003-12-04 Boehringer Ingelheim Pharmaceuticals, Inc. Method of performing risk analysis
US20090062970A1 (en) * 2007-08-28 2009-03-05 America Connect, Inc. System and method for active power load management
US20150081237A1 (en) * 2013-09-19 2015-03-19 Seeo, Inc Data driven/physical hybrid model for soc determination in lithium batteries
US20160243957A1 (en) * 2013-10-08 2016-08-25 Robert Bosch Gmbh Method for minimizing cell aging of a battery and/or battery comprising an apparatus for minimizing cell aging of the battery
EP3145748A1 (en) * 2014-05-20 2017-03-29 Jaguar Land Rover Limited Cooling system for vehicle device
CN108038631A (en) * 2017-12-29 2018-05-15 中国科学技术大学 A kind of lithium ion battery risk assessment method
KR101867553B1 (en) * 2017-07-21 2018-06-14 노진석 Device and method for managing drones
CN108445039A (en) * 2018-02-07 2018-08-24 清华大学 Power battery thermal runaway security performance prediction technique, device and computer readable storage medium
CN108656972A (en) * 2017-03-28 2018-10-16 长城汽车股份有限公司 A kind of electric vehicle charging/discharging thereof and device
US20190109470A1 (en) * 2011-03-17 2019-04-11 EVchip Energy Ltd. Battery pack system
US20190205821A1 (en) * 2018-01-03 2019-07-04 International Business Machines Corporation Automated identification, status monitoring and notification of stored items
CN109978229A (en) * 2019-02-12 2019-07-05 常伟 The method that the full battery core multi-point temperature of a kind of pair of power battery pack and tie point temperature carry out thermal runaway prediction
CN110095722A (en) * 2019-04-02 2019-08-06 清华大学 Power battery thermal runaway safety integrated evaluating method and system
CN110134567A (en) * 2019-04-30 2019-08-16 西北工业大学 Microprocessor nonuniform sampling heat distribution reconstructing method based on convolutional neural networks
CN110161414A (en) * 2019-06-13 2019-08-23 北京理工大学 A kind of power battery thermal runaway on-line prediction method and system
CN110370984A (en) * 2019-06-26 2019-10-25 清华大学 Power battery thermal runaway method for early warning
CN110400231A (en) * 2019-06-06 2019-11-01 湖南大学 A kind of electric energy measuring equipment crash rate predictor method weighting Nonlinear Bayesian
CN110752410A (en) * 2019-10-30 2020-02-04 上海理工大学 Method for rapidly sorting and recombining retired lithium batteries
CN110794305A (en) * 2019-10-14 2020-02-14 北京理工大学 Power battery fault diagnosis method and system
CN110888058A (en) * 2019-12-02 2020-03-17 西安科技大学 Algorithm based on power battery SOC and SOH joint estimation
US20200091752A1 (en) * 2018-09-13 2020-03-19 Honda Motor Co., Ltd. Electric power demand/supply management device
CN111007401A (en) * 2019-12-16 2020-04-14 国网江苏省电力有限公司电力科学研究院 Electric vehicle battery fault diagnosis method and device based on artificial intelligence
CN111239630A (en) * 2020-03-09 2020-06-05 江苏中兴派能电池有限公司 Energy storage battery service life prediction method and management system
DE102018132850A1 (en) * 2018-12-19 2020-06-25 Volkswagen Aktiengesellschaft Method of monitoring a battery
CN111391668A (en) * 2020-03-31 2020-07-10 威睿电动汽车技术(宁波)有限公司 Battery thermal runaway early warning processing method, device, equipment and storage medium

Patent Citations (26)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2003303344A (en) * 2002-04-10 2003-10-24 Nec Corp Method and device for extracting image region
US20030225601A1 (en) * 2002-05-31 2003-12-04 Boehringer Ingelheim Pharmaceuticals, Inc. Method of performing risk analysis
US20090062970A1 (en) * 2007-08-28 2009-03-05 America Connect, Inc. System and method for active power load management
US20190109470A1 (en) * 2011-03-17 2019-04-11 EVchip Energy Ltd. Battery pack system
US20150081237A1 (en) * 2013-09-19 2015-03-19 Seeo, Inc Data driven/physical hybrid model for soc determination in lithium batteries
US20160243957A1 (en) * 2013-10-08 2016-08-25 Robert Bosch Gmbh Method for minimizing cell aging of a battery and/or battery comprising an apparatus for minimizing cell aging of the battery
EP3145748A1 (en) * 2014-05-20 2017-03-29 Jaguar Land Rover Limited Cooling system for vehicle device
CN108656972A (en) * 2017-03-28 2018-10-16 长城汽车股份有限公司 A kind of electric vehicle charging/discharging thereof and device
KR101867553B1 (en) * 2017-07-21 2018-06-14 노진석 Device and method for managing drones
CN108038631A (en) * 2017-12-29 2018-05-15 中国科学技术大学 A kind of lithium ion battery risk assessment method
US20190205821A1 (en) * 2018-01-03 2019-07-04 International Business Machines Corporation Automated identification, status monitoring and notification of stored items
CN108445039A (en) * 2018-02-07 2018-08-24 清华大学 Power battery thermal runaway security performance prediction technique, device and computer readable storage medium
US20200091752A1 (en) * 2018-09-13 2020-03-19 Honda Motor Co., Ltd. Electric power demand/supply management device
DE102018132850A1 (en) * 2018-12-19 2020-06-25 Volkswagen Aktiengesellschaft Method of monitoring a battery
CN109978229A (en) * 2019-02-12 2019-07-05 常伟 The method that the full battery core multi-point temperature of a kind of pair of power battery pack and tie point temperature carry out thermal runaway prediction
CN110095722A (en) * 2019-04-02 2019-08-06 清华大学 Power battery thermal runaway safety integrated evaluating method and system
CN110134567A (en) * 2019-04-30 2019-08-16 西北工业大学 Microprocessor nonuniform sampling heat distribution reconstructing method based on convolutional neural networks
CN110400231A (en) * 2019-06-06 2019-11-01 湖南大学 A kind of electric energy measuring equipment crash rate predictor method weighting Nonlinear Bayesian
CN110161414A (en) * 2019-06-13 2019-08-23 北京理工大学 A kind of power battery thermal runaway on-line prediction method and system
CN110370984A (en) * 2019-06-26 2019-10-25 清华大学 Power battery thermal runaway method for early warning
CN110794305A (en) * 2019-10-14 2020-02-14 北京理工大学 Power battery fault diagnosis method and system
CN110752410A (en) * 2019-10-30 2020-02-04 上海理工大学 Method for rapidly sorting and recombining retired lithium batteries
CN110888058A (en) * 2019-12-02 2020-03-17 西安科技大学 Algorithm based on power battery SOC and SOH joint estimation
CN111007401A (en) * 2019-12-16 2020-04-14 国网江苏省电力有限公司电力科学研究院 Electric vehicle battery fault diagnosis method and device based on artificial intelligence
CN111239630A (en) * 2020-03-09 2020-06-05 江苏中兴派能电池有限公司 Energy storage battery service life prediction method and management system
CN111391668A (en) * 2020-03-31 2020-07-10 威睿电动汽车技术(宁波)有限公司 Battery thermal runaway early warning processing method, device, equipment and storage medium

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
陈天雨: "基于模型的动力电池系统多尺度热安全设计", 《中国机械工程》 *

Cited By (28)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022033348A1 (en) * 2020-08-10 2022-02-17 中兴通讯股份有限公司 Vehicle thermal runaway processing method and apparatus, vehicle, and computer readable storage medium
WO2022135520A1 (en) * 2020-12-24 2022-06-30 广州橙行智动汽车科技有限公司 Battery cell capacity anomaly determination method and apparatus, and vehicle and storage medium
WO2022174601A1 (en) * 2021-02-22 2022-08-25 华为技术有限公司 Method for training battery state prediction model, and related apparatus
CN113040711B (en) * 2021-03-03 2023-08-18 吾征智能技术(北京)有限公司 Cerebral apoplexy incidence risk prediction system, equipment and storage medium
CN113040711A (en) * 2021-03-03 2021-06-29 吾征智能技术(北京)有限公司 Cerebral stroke attack risk prediction system, equipment and storage medium
CN113437371A (en) * 2021-05-19 2021-09-24 湖南大学 Early warning system and early warning method for thermal runaway of lithium ion battery of new energy automobile
CN113341332A (en) * 2021-05-25 2021-09-03 三一重机有限公司 Battery thermal runaway early warning method and device, operation machine and electronic equipment
CN113689649A (en) * 2021-07-31 2021-11-23 华能新能源股份有限公司 Fire early warning method, fire early warning system and computer equipment
CN113689649B (en) * 2021-07-31 2022-11-15 华能新能源股份有限公司 Fire early warning method, fire early warning system and computer equipment
WO2023011066A1 (en) * 2021-08-03 2023-02-09 长城汽车股份有限公司 Performance test method for battery thermal management system, and related device
CN114137421B (en) * 2021-11-30 2023-09-19 章鱼博士智能技术(上海)有限公司 Battery abnormality detection method, device, equipment and storage medium
WO2023097507A1 (en) * 2021-11-30 2023-06-08 华为技术有限公司 Battery safety early warning method and apparatus
CN114137421A (en) * 2021-11-30 2022-03-04 蜂巢能源(上海)有限公司 Battery abnormality detection method, apparatus, device and storage medium
WO2023115988A1 (en) * 2021-12-22 2023-06-29 北京国家新能源汽车技术创新中心有限公司 Method for detecting internal short-circuit of traction battery
CN114325406A (en) * 2021-12-30 2022-04-12 重庆长安新能源汽车科技有限公司 Method and system for predicting thermal runaway of battery based on machine learning thinking
WO2023124220A1 (en) * 2021-12-31 2023-07-06 中国第一汽车股份有限公司 Vehicle detection method and device
CN114312322A (en) * 2021-12-31 2022-04-12 中国第一汽车股份有限公司 Vehicle detection method and device
CN114295983A (en) * 2021-12-31 2022-04-08 上海洛轲智能科技有限公司 Battery thermal runaway early warning method and device, vehicle, equipment and storage medium
CN114312322B (en) * 2021-12-31 2024-04-12 中国第一汽车股份有限公司 Vehicle detection method and device
WO2023142065A1 (en) * 2022-01-29 2023-08-03 宁德时代新能源科技股份有限公司 Battery state evaluation method and apparatus, electronic device, and storage medium
CN114179675A (en) * 2022-02-16 2022-03-15 深圳康普盾科技股份有限公司 Battery replacement cabinet safety control method and system based on Internet of things and storage medium
CN114240260B (en) * 2022-02-17 2022-05-20 北京航空航天大学 New energy group vehicle thermal runaway risk assessment method based on digital twinning
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
CN115575830A (en) * 2022-10-10 2023-01-06 中国标准化研究院 Method and system for identifying inducement of thermal runaway of lithium battery
CN115548480A (en) * 2022-11-01 2022-12-30 湖南东舟能源有限公司 Battery system and control method of battery system
CN115548480B (en) * 2022-11-01 2023-04-18 湖南东舟能源有限公司 Battery system and control method of battery system
WO2024113182A1 (en) * 2022-11-29 2024-06-06 宁德时代新能源科技股份有限公司 Battery risk level determination method, apparatus, storage medium and battery management system
CN117388725A (en) * 2023-03-13 2024-01-12 中国石油大学(华东) Early abnormality early warning method based on expansion force of lithium ion battery

Also Published As

Publication number Publication date
CN112092675B (en) 2022-03-25

Similar Documents

Publication Publication Date Title
CN112092675B (en) Battery thermal runaway early warning method, system and server
Guo et al. Online estimation of SOH for lithium-ion battery based on SSA-Elman neural network
Li et al. An indirect RUL prognosis for lithium-ion battery under vibration stress using Elman neural network
CN110794305A (en) Power battery fault diagnosis method and system
CN111007401A (en) Electric vehicle battery fault diagnosis method and device based on artificial intelligence
CN109116242B (en) Data processing method and device for power battery
EP3721499A1 (en) Method and system for online assessing state of health of a battery
CN112098850B (en) Lithium ion battery voltage fault diagnosis method and system based on SDO algorithm
CN109604186A (en) Power battery performance flexibility assesses method for separating
CN114818831B (en) Bidirectional lithium ion battery fault detection method and system based on multi-source perception
CN112924885B (en) Method for quantitatively diagnosing short circuit in battery based on peak height of incremental capacity curve
CN113933732A (en) New energy automobile power battery health state analysis method, system and storage medium
CN111257753B (en) Battery system fault diagnosis method
CN113281671A (en) Lithium ion battery remaining service life prediction method and system based on IGS-SVM
CN113696786A (en) Battery equalization method and system
CN115366683A (en) Fault diagnosis strategy for new energy automobile power battery multi-dimensional model fusion
CN116150572A (en) Automobile battery monomer consistency defect evaluation method based on cluster analysis
CN115061058A (en) Method and system for measuring and calculating gradient utilization residual life of retired power battery
Jeng et al. Evaluation of cell inconsistency in lithium-ion battery pack using the autoencoder network model
CN116819328A (en) Electric automobile power battery fault diagnosis method, system, equipment and medium
CN115166555A (en) Battery module internal resistance abnormity identification method and system for battery big data platform
CN114460481A (en) Energy storage battery thermal runaway early warning method based on Bi-LSTM and attention mechanism
CN113391214A (en) Battery micro-fault diagnosis method based on battery charging voltage ranking change
CN114646888A (en) Assessment method and system for capacity attenuation of power battery
CN114036647A (en) Power battery safety risk assessment method based on real vehicle data

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