CN112836967B - New energy automobile battery safety risk assessment system - Google Patents
New energy automobile battery safety risk assessment system Download PDFInfo
- Publication number
- CN112836967B CN112836967B CN202110146869.7A CN202110146869A CN112836967B CN 112836967 B CN112836967 B CN 112836967B CN 202110146869 A CN202110146869 A CN 202110146869A CN 112836967 B CN112836967 B CN 112836967B
- Authority
- CN
- China
- Prior art keywords
- new energy
- energy automobile
- data
- safety risk
- alarm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000012502 risk assessment Methods 0.000 title claims abstract description 31
- 238000011156 evaluation Methods 0.000 claims abstract description 26
- 238000004140 cleaning Methods 0.000 claims abstract description 19
- 238000007477 logistic regression Methods 0.000 claims abstract description 18
- 238000000034 method Methods 0.000 claims abstract description 16
- 230000004927 fusion Effects 0.000 claims abstract description 10
- 238000000605 extraction Methods 0.000 claims abstract description 6
- 238000012795 verification Methods 0.000 claims description 16
- 238000001207 Hosmer–Lemeshow test Methods 0.000 claims description 4
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 238000000546 chi-square test Methods 0.000 claims description 4
- 238000013450 outlier detection Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 2
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 150000001875 compounds Chemical class 0.000 description 2
- 239000000178 monomer Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000011002 quantification Methods 0.000 description 2
- 238000000611 regression analysis Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000008358 core component Substances 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
Images
Classifications
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- 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
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- 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
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Operations Research (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Geometry (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Secondary Cells (AREA)
Abstract
The invention discloses a new energy automobile battery safety risk assessment system, which comprises the following modules: the vehicle driving information acquisition module is used for acquiring driving data of the new energy vehicle and cleaning the data; the index extraction module is used for carrying out binary Logistic regression analysis on the driving data and the three dynamic indexes after data cleaning, and determining characteristic indexes related to new energy automobile alarming and influence weights of the characteristic indexes on corresponding alarming information; and the battery safety risk evaluation module is used for building a Bayesian network model according to the determined relationship between the characteristic indexes and the corresponding alarm information, verifying the information fusion degree of the characteristic indexes identified by binary Logistic regression and the alarm information, verifying the safety risk evaluation result of each type of alarm condition, and finally outputting the quantitative result of the safety evaluation of the new energy automobile battery. The method can quantitatively evaluate the battery safety risk state of the new energy automobile and accurately identify the safety state of the battery.
Description
Technical Field
The invention relates to the field of new energy automobile driving safety, in particular to a new energy automobile battery safety risk assessment system.
Background
After decades of rapid development, the automobile industry has occupied a prominent position in national economy in China, and has become one of the pillar industries in China, and particularly in recent years, the arrangement of automobile production and sales in China has become the first world. The development of the traditional automobile industry also brings various problems that the dependence on external petroleum is high and the environmental pollution is continuously intensified. The world energy safety and environmental protection increase the demands on the technical research and development and rapid development of new energy automobiles, and promote various countries to develop new energy automobiles from prospective perspectives and strategic sights. For China, the new energy automobile industry, one of strategic emerging industries, has become an important breakthrough for energy conservation and emission reduction, exciting economy and changing the industrial structure in China.
In the face of the arrival of the electric era of automobiles, how to ensure the safety is a main concern, and the power battery is used as a core component of the new energy automobile, the requirements of accurate quantitative estimation of the safety state and protection of a safety structure are increasingly highlighted, so that the accurate quantitative estimation of the safety state of the power battery and the improvement of the durability, the safety and the reliability of the new energy automobile become very important.
At present, the evaluation method for the battery safety of the new energy automobile at home and abroad mainly comprises the steps of grading by experts, and carrying out qualitative analysis on risks through personal experience and subjective judgment, so that the quantification is less; in the national standard GB/T32960.3-2016, due to the fact that the fault level of alarming is defined by the manufacturer, relevant scholars find out through previous collected driving data research that the battery safety state of the new energy automobile cannot be accurately identified under the condition that the fault occurs and the alarming is carried out, and therefore accurate evaluation on the battery safety of the new energy automobile cannot be carried out.
Disclosure of Invention
The invention aims to provide a new energy automobile battery safety risk assessment system to solve the problem that the new energy automobile battery safety cannot be accurately assessed.
In order to solve the technical problem, the invention provides a technical scheme that: this new energy automobile battery safety risk assessment system includes:
the vehicle driving information acquisition module is used for acquiring driving data of the new energy vehicle and cleaning the data;
the index extraction module is used for carrying out binary Logistic regression analysis on driving data subjected to data cleaning and three dynamic indexes, namely Euro volt, single battery voltage difference and single battery temperature difference, aiming at the two classification problems of whether to alarm or not, and determining characteristic indexes related to new energy automobile alarm and influence weights of the characteristic indexes on corresponding alarm information;
and the battery safety risk evaluation module is used for respectively taking the characteristic indexes as a data layer, taking the alarm information as an alarm layer and taking the safety evaluation result as an index layer according to the relation between the determined characteristic indexes and the corresponding alarm information, building a Bayesian network model, verifying the information fusion degree of the characteristic indexes identified by binary Logistic regression and the alarm information, verifying the safety risk evaluation result of each type of alarm information, and finally outputting the quantitative result of the safety evaluation of the new energy automobile battery.
According to the scheme, the vehicle driving information acquisition module acquires driving data based on national standards through a T-BOX sensor installed in the new energy automobile.
According to the scheme, the data cleaning specifically adopts a piecewise linearity difference method to complete missing values.
According to the scheme, the vehicle driving information acquisition module performs outlier detection on the driving data subjected to data cleaning by adopting an MAD algorithm, calculates a judgment coefficient D, and eliminates abnormal data with the judgment coefficient larger than D.
According to the scheme, the vehicle driving information acquisition module adopts an undersampling method based on data density to the driving data after data cleaning, and the proportion of each type of alarm information to the normal driving data is kept at 1: 4.
According to the scheme, the specific operation of the binary Logistic regression analysis is as follows:
in the formula，PjOf alarm type, beta0Is a constant, p is the number of characteristic indexes, xqAs influencing factor, betaqA weight coefficient being a characteristic index;
the binary Logistic regression analysis includes the chi-square test and the Hosmer-Lemeshow test.
According to the scheme, the battery safety risk assessment module respectively defines the non-overlapping part of the alarm information and the normal data in the data layer as a fault state and a Good state, and defines the overlapping part of the alarm information and the normal data as a Fair; the alarm layer is divided into two states according to whether an alarm is given or not; the index layer is divided into four states of no risk, acceptable risk, tolerable risk and unacceptable risk according to the definition of the state standard on the fault level.
According to the scheme, the Bayesian network model is built through a GeNIe platform, a condition probability table of an index layer is determined through the Bayesian network model, the safety state of the new energy automobile is quantitatively evaluated according to the condition probability table, and information fusion degree verification and safety risk evaluation result verification are performed by using a verification module of a GenNIe platform.
According to the scheme, the information fusion degree verification is to verify the cause and effect of the characteristic indexes identified by the binary Logistic regression and the alarm information, and specifically to test whether the ROC curve area output by the GeNIe platform is larger than 0.5.
According to the scheme, the safety risk assessment result verification is to verify the safety risk assessment result for each type of alarm information, specifically, when certain alarm is determined to occur, whether the output safety risk assessment result is consistent with the risk state interval value divided according to the fault grade is verified.
The invention has the beneficial effects that: the method comprises the steps that driving data of a new energy automobile are obtained through a vehicle driving information acquisition module and are subjected to data cleaning; determining characteristic indexes related to new energy automobile alarm and influence weights of all indexes on the alarm through an index extraction module; finally outputting a quantitative result of the new energy automobile battery safety evaluation through a battery safety risk evaluation module; through the steps, the safety risk of the new energy automobile battery can be accurately and quantitatively evaluated.
Furthermore, the missing values are supplemented through a piecewise linearity difference method, and outlier detection is carried out on the supplemented data through an MAD algorithm, so that efficient and rapid screening processing on the driving data is realized.
Furthermore, the ratio of each type of alarm information to the normal driving data is set to be 1:4, so that the finally output new energy automobile battery safety evaluation quantification result is more stable and accurate.
Drawings
FIG. 1 is a system framework diagram of one embodiment of the present invention;
FIG. 2 is a flowchart of a vehicle driving information collection module according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating the operation of the index extraction module according to an embodiment of the present invention;
fig. 4 is a flowchart illustrating the operation of the battery safety risk assessment module according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present disclosure more apparent, the technical solutions of the embodiments of the present disclosure will be described clearly and completely with reference to the drawings of the embodiments of the present disclosure. It is to be understood that the described embodiments are only a few embodiments of the present disclosure, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the described embodiments of the disclosure without any inventive step, are within the scope of protection of the disclosure.
Referring to fig. 1, the new energy automobile safety risk assessment system includes:
the vehicle driving information acquisition module is used for acquiring driving data of the new energy vehicle and cleaning the data;
the index extraction module is used for carrying out binary Logistic regression analysis on driving data subjected to data cleaning and three dynamic indexes, namely Euro volt, single battery voltage difference and single battery temperature difference, aiming at the two classification problems of whether to alarm or not, and determining characteristic indexes related to new energy automobile alarm and influence weights of the characteristic indexes on corresponding alarm information;
and the battery safety risk evaluation module is used for respectively taking the characteristic indexes as a data layer, taking the alarm information as an alarm layer and taking the safety evaluation result as an index layer according to the relation between the determined characteristic indexes and the corresponding alarm information, building a Bayesian network model, verifying the information fusion degree of the characteristic indexes identified by binary Logistic regression and the alarm information, verifying the safety risk evaluation result of each type of alarm information, and finally outputting the quantitative result of the safety evaluation of the new energy automobile battery.
Further, referring to fig. 2, the vehicle driving information collecting module collects driving data based on national standards through a T-BOX sensor installed in the new energy vehicle.
Further, the data cleaning specifically adopts a piecewise linearity difference method to complete missing values:
Xk＝{x1k,x2k,···,xnk}
in the formula, XkIs a set of data metrics; k is the number of data indexes; n is the number of elements of the characteristic index;
xik＝Ni-1kxi-1k+Ni+1kxi+1k,i＝(0,1,2,…,n)
in the formula (I), the compound is shown in the specification,
xikis a missing value.
Further, the vehicle driving information acquisition module performs outlier detection on the driving data subjected to data cleaning by adopting an MAD algorithm, and calculates the median of the data:
in the formula (I), the compound is shown in the specification,
is a characteristic indexXkA median of (d); if n is an odd number, the median is an observed value sorted to the middle; when n is an even number, the median is ordered as
And
average of the observed values of (a);
calculating the MAD:
in the formula, MAD is a set X of characteristic indexeskSubtracting the characteristic index X from each element in the listkB is a constant, typically b is 1.4826;
in order to detect abnormal values in the observed data, each observed value x needs to be calculatednkThe determination coefficient of (1):
identifying the observed value x when the decision coefficient D exceeds a given threshold value DnkAccording to a large number of scientific experiments and engineering practices, abnormal data are removed by taking d to be 2.5.
Furthermore, the vehicle driving information acquisition module adopts an undersampling method based on data density to the driving data after data cleaning, and the proportion of each type of alarm information to the normal driving data is kept at 1:4, so that the result obtained by subsequent analysis is more stable and accurate.
Further, the driving data processed by the undersampling method is subjected to standardization processing to eliminate dimensional difference, the driving data is changed into Gaussian distribution with a mean value of 0 and a variance of 1 to obtain a standard matrix, and the minimum density of the data is set as follows:
where m represents the number of samples contained around a data point in the circular region (excluding the data point), r is the radius of the circular region, and m is 6 and r is 0.005 in this embodiment; the method comprises the following steps that T is used as a critical value for dividing data density, an area formed by sample points with the data density larger than or equal to T is defined as a high-density data cluster, low-density data clusters smaller than T are eliminated, and all the high-density data clusters are obtained through calculation by means of a DBSCAN clustering algorithm in the embodiment;
and then calculating the average value corresponding to each high-density data cluster, finding the sample closest to the value in each high-density data cluster, taking the sample as the center of the data cluster, reserving some sample data close to the pseudo center according to the requirement, deleting the sample far away from the pseudo center, and finally obtaining a new majority sample set.
Further, referring to fig. 3, a binary Logistic regression analysis method is adopted to perform regression analysis on the cleaned driving data and three dynamic indexes, namely, ohm per volt, the voltage difference of the battery monomer and the temperature difference of the battery monomer, so as to determine the characteristic indexes related to the new energy vehicle alarm and the influence weight of each characteristic index on the corresponding alarm information; the operation of performing binary Logistic regression analysis on each type of alarm information specifically comprises the following steps:
in the formula, PjOf alarm type, beta0Is a constant, p is the number of characteristic indexes, xqAs a factor, betaqQ is a constant, q is 1,2, 3, …, p;
in the process of regression analysis, chi-square test and Hosmer-Lemeshow test are required, namely the significance (sig) of the chi-square test is less than 0.05, and the significance (sig) of the Hosmer-Lemeshow test is more than 0.05;
further, the battery safety risk assessment module takes the characteristic indexes as data layers, the alarm information as an alarm layer and the safety evaluation result as an index layer according to the determined characteristic indexes and the relation of the characteristic indexes and the corresponding alarm information, so that a Bayesian network model is built; because the bayesian network inputs not a specific numerical value but a state data, the present embodiment divides the actual data of each layer into a state, and for the data layer, according to the non-overlapping part of the alarm and the normal data, the data layer is respectively defined as: both fault and Good states; for data overlap, the present embodiment defines a new state Fair; aiming at an alarm layer, two states are divided according to whether an alarm is given or not; for the index layer, the definition of the fault level based on the national standard is divided into four states of no risk, acceptable risk, tolerable risk and unacceptable risk.
Further, based on the division of different states, calculating prior probability according to the frequency of each characteristic index occurring in different states, and weighting the data of the data layer and the alarm layer according to the weight of the characteristic index and the frequency of occurring alarm in the binary Logistic regression equation, wherein a membership function is adopted to calculate the conditional probability of the alarm layer and the index layer, the specific steps are as follows:
s1, sequentially assigning 3 state levels Good, Fair and fault of the data layer in the Bayesian network to be 1-3, and recording as
(c is 1,2, 3 represents 3 different state grades, q is 1,2, …, p represents different characteristic indexes), and the state calculation formulas of an alarm layer and an index layer can be obtained in the same way;
s2, assigning a fixed weight beta to each characteristic indexq；
S3, calculating the state score of the alarm information directly connected with each characteristic index:
in the formula, sigmaqβ＝1；
S4, converting the state score into the conditional probability of different states of the alarm layer and the index layer by using a score-probability conversion formula:
in the formula, muiMembership functions, f, representing different state classes1(x) To f6(x) Determined by the formula (1), wherein the formula (2) is a score-probability conversion formula for calculating the conditional probability of the alarm layer, the formula (3) is a score-probability conversion formula for calculating the conditional probability of the alarm layer, and s is respectively taken to ensure that the calculated data are within a 95% confidence interval of a function1To s6The values of (A) are: 1.05, 1.95, 2.05, 2.95, 3.05 and 3.95.
Further, referring to fig. 4, in the embodiment, a bayesian network model is built on a gemie platform, a condition probability table of an index layer is finally determined, the safety state of the new energy vehicle is quantitatively evaluated according to the condition probability table, and a verification module of the gemie platform is utilized to perform information fusion degree verification and safety risk evaluation result verification;
the method comprises the steps of verifying feature indexes identified by binary Logistic regression and cause and effect of alarm, specifically judging state division of a data layer in a built Bayesian network, and judging whether alarm can be accurately generated or not, wherein the verification of information fusion degree is to verify whether the area of an ROC curve output by a GeNIe platform is larger than 0.5 or not;
and verifying the safety risk assessment result, namely verifying the safety risk assessment result aiming at each type of alarm condition, specifically when certain alarm is determined to occur, verifying whether the output safety risk assessment result is consistent with the risk state interval value divided according to the fault grade, and finally outputting the safety risk assessment result of the new energy automobile battery after the two kinds of verification are passed.
In summary, the invention provides a system for performing safety risk assessment on a new energy automobile battery, which is used for quantitatively assessing the safety risk of the new energy automobile battery by acquiring driving data of the new energy automobile, cleaning the data, extracting characteristic indexes associated with alarm by using a correlation analysis method, building a safety risk assessment model and performing model verification.
The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. New energy automobile battery safety risk evaluation system, its characterized in that: the system comprises:
the vehicle driving information acquisition module is used for acquiring driving data of the new energy vehicle and cleaning the data;
the index extraction module is used for carrying out binary Logistic regression analysis on driving data subjected to data cleaning and three dynamic indexes, namely Euro volt, single battery voltage difference and single battery temperature difference, aiming at the two classification problems of whether to alarm or not, and determining characteristic indexes related to new energy automobile alarm and influence weights of the characteristic indexes on corresponding alarm information;
and the battery safety risk evaluation module is used for respectively taking the characteristic indexes as a data layer, taking the alarm information as an alarm layer and taking the safety evaluation result as an index layer according to the relation between the determined characteristic indexes and the corresponding alarm information, building a Bayesian network model, verifying the information fusion degree of the characteristic indexes identified by binary Logistic regression and the alarm information, verifying the safety risk evaluation result of each type of alarm information, and finally outputting the quantitative result of the safety evaluation of the new energy automobile battery.
2. The new energy automobile battery safety risk assessment system of claim 1, characterized in that: the vehicle driving information acquisition module acquires driving data based on national standards through a T-BOX sensor installed in the new energy automobile.
3. The new energy automobile battery safety risk assessment system of claim 1, characterized in that: the data cleaning specifically adopts a piecewise linearity difference method to complete missing values.
4. The new energy automobile battery safety risk assessment system of claim 1, characterized in that: the vehicle driving information acquisition module adopts an MAD algorithm to perform outlier detection on driving data subjected to data cleaning, calculates a judgment coefficient D, and eliminates abnormal data of which the judgment coefficient is larger than D.
5. The new energy automobile battery safety risk assessment system of claim 1, characterized in that: the vehicle driving information acquisition module adopts an undersampling method based on data density to the driving data after data cleaning, and the proportion of each type of alarm information to the normal driving data is kept at 1: 4.
6. The new energy automobile battery safety risk assessment system of claim 1, characterized in that: the specific operation of the binary Logistic regression analysis is as follows:
in the formula, PjOf alarm type, beta0Is a constant, p is the number of characteristic indexes, xqAs influencing factor, betaqA weight coefficient being a characteristic index;
the binary Logistic regression analysis includes the chi-square test and the Hosmer-Lemeshow test.
7. The new energy automobile battery safety risk assessment system of claim 1, characterized in that: the battery safety risk evaluation module respectively defines the non-overlapping part of the alarm information and the normal data in the data layer as a fault state and a Good state, and defines the overlapping part of the alarm information and the normal data as a Fair; the alarm layer is divided into two states according to whether an alarm is given or not; the index layer is divided into four states of no risk, acceptable risk, tolerable risk and unacceptable risk according to the definition of the state standard on the fault level.
8. The new energy automobile battery safety risk assessment system of claim 1, characterized in that: the Bayesian network model is built through a GeNIe platform, a condition probability table of an index layer is determined through the Bayesian network model, the safety state of the new energy automobile is quantitatively evaluated according to the condition probability table, and information fusion degree verification and safety risk evaluation result verification are performed by using a verification module of a GenNIe platform.
9. The new energy automobile battery safety risk assessment system of claim 1, characterized in that: the information fusion degree verification is to verify the cause and effect of the characteristic indexes identified by the binary Logistic regression and the alarm information, and specifically to test whether the ROC curve area output by the GeNIe platform is larger than 0.5.
10. The new energy automobile battery safety risk assessment system of claim 1, characterized in that: and the safety risk assessment result verification is to verify the safety risk assessment result for each type of alarm information, and specifically to verify whether the output safety risk assessment result is consistent with the risk state interval value divided according to the fault grade when certain alarm is determined to occur.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110146869.7A CN112836967B (en) | 2021-02-03 | 2021-02-03 | New energy automobile battery safety risk assessment system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110146869.7A CN112836967B (en) | 2021-02-03 | 2021-02-03 | New energy automobile battery safety risk assessment system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112836967A CN112836967A (en) | 2021-05-25 |
| CN112836967B true CN112836967B (en) | 2022-07-08 |
Family
ID=75932555
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110146869.7A Active CN112836967B (en) | 2021-02-03 | 2021-02-03 | New energy automobile battery safety risk assessment system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112836967B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115330177B (en) * | 2022-08-10 | 2024-02-13 | 北京市科学技术研究院 | Urban public facility security risk assessment method based on causal learning |
| CN116754967A (en) * | 2023-04-24 | 2023-09-15 | 中广核新能源安徽有限公司固镇分公司 | Method and system for online evaluation of electrochemical cells of an energy storage power station |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20140106142A (en) * | 2013-02-26 | 2014-09-03 | 한국항공대학교산학협력단 | System, apparatus and method for prognosticating failure of battery based on charge voltage characteristics |
| CN105404936A (en) * | 2015-11-24 | 2016-03-16 | 中国电力科学研究院 | Method for evaluating health index of distribution transformer |
| CN106651408A (en) * | 2015-10-28 | 2017-05-10 | 苏宁云商集团股份有限公司 | Data analysis method and apparatus |
| CN109270461A (en) * | 2018-10-15 | 2019-01-25 | 四川长虹电器股份有限公司 | Fault detection method based on Bayesian network |
| CN111967192A (en) * | 2020-08-24 | 2020-11-20 | 哈尔滨理工大学 | Naive Bayes-based battery safety degree estimation method |
-
2021
- 2021-02-03 CN CN202110146869.7A patent/CN112836967B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20140106142A (en) * | 2013-02-26 | 2014-09-03 | 한국항공대학교산학협력단 | System, apparatus and method for prognosticating failure of battery based on charge voltage characteristics |
| CN106651408A (en) * | 2015-10-28 | 2017-05-10 | 苏宁云商集团股份有限公司 | Data analysis method and apparatus |
| CN105404936A (en) * | 2015-11-24 | 2016-03-16 | 中国电力科学研究院 | Method for evaluating health index of distribution transformer |
| CN109270461A (en) * | 2018-10-15 | 2019-01-25 | 四川长虹电器股份有限公司 | Fault detection method based on Bayesian network |
| CN111967192A (en) * | 2020-08-24 | 2020-11-20 | 哈尔滨理工大学 | Naive Bayes-based battery safety degree estimation method |
Non-Patent Citations (2)
| Title |
|---|
| Kui Yang ; Xuesong Wang ; Rongjie Yu.A Bayesian dynamic updating approach for urban expressway real-time crash risk evaluation.《Transportation Research Part C: Emerging Technologies》.2018,P192-207. * |
| 基于贝叶斯修正Logistic回归模型的电梯风险评估研究;陆倩茵;《中国优秀博硕士学位论文全文数据库(硕士)经济与管理科学辑》;20190115;J145-134 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112836967A (en) | 2021-05-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113158947B (en) | Power battery health scoring method, system and storage medium | |
| CN112836967B (en) | New energy automobile battery safety risk assessment system | |
| CN115366683B (en) | Fault diagnosis strategy for multi-dimensional model fusion of power battery of new energy automobile | |
| CN113762329A (en) | Method and system for constructing state prediction model of large rolling mill | |
| CN116381514B (en) | Cell differential pressure early warning method, device, storage medium and equipment | |
| CN110633729A (en) | Driving risk hierarchical clustering method for intelligent networking vehicle group test | |
| CN114048468A (en) | Intrusion detection method, intrusion detection model training method, device and medium | |
| CN112685910A (en) | Complex equipment power pack fault prediction method based on hybrid prediction model | |
| Li et al. | A Novel Method for Lithium‐Ion Battery Fault Diagnosis of Electric Vehicle Based on Real‐Time Voltage | |
| CN115098962A (en) | Method for predicting residual life of mechanical equipment in degradation state based on hidden half Markov model | |
| Wang et al. | Prognostics and health management system for electric vehicles with a hierarchy fusion framework: concepts, architectures, and methods | |
| Liu et al. | Online diagnosis and prediction of power battery voltage comprehensive faults for electric vehicles based on multi-parameter characterization and improved K-means method | |
| CN111381165A (en) | Vehicle power battery monitoring method, device and platform | |
| CN110119891B (en) | Traffic safety influence factor identification method suitable for big data | |
| CN110750848A (en) | Method for estimating remaining life of software-hardware degradation system by considering software operation | |
| CN117033891A (en) | Traffic accident severity assessment method based on single vehicle traffic accident database | |
| CN116679221A (en) | Battery abnormality detection and battery maintenance management method and device | |
| CN116061690A (en) | Safety early warning method and device in electric automobile charging process | |
| CN106372450A (en) | High-speed train axle temperature prediction method based on stepwise regression analysis | |
| AU2021104319A4 (en) | A system for traction inverter fault detection and a method thereof | |
| Borucka | Method of testing the readiness of means of transport with the use of semi-Markov processes | |
| CN113705920A (en) | Generation method of water data sample set for thermal power plant and terminal equipment | |
| Tian et al. | STLRF‐Stack: A fault prediction model for pure electric vehicles based on a high dimensional imbalanced dataset | |
| CN113887569B (en) | Accident frequent region distinguishing method based on region scale | |
| CN116703431A (en) | Vehicle assessment method, device, equipment and storage medium |
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 |