CN109685266A - A kind of lithium battery bin fire prediction method and system based on SVM - Google Patents
A kind of lithium battery bin fire prediction method and system based on SVM Download PDFInfo
- Publication number
- CN109685266A CN109685266A CN201811573513.6A CN201811573513A CN109685266A CN 109685266 A CN109685266 A CN 109685266A CN 201811573513 A CN201811573513 A CN 201811573513A CN 109685266 A CN109685266 A CN 109685266A
- Authority
- CN
- China
- Prior art keywords
- fire
- probability
- lithium battery
- battery bin
- svm
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
- G08B17/06—Electric actuation of the alarm, e.g. using a thermally-operated switch
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Emergency Management (AREA)
- Development Economics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Game Theory and Decision Science (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Alarm Systems (AREA)
Abstract
The invention discloses a kind of lithium battery bin fire prediction method and system based on SVM, fire prediction method is the following steps are included: using the ambient temperature information of lithium battery bin as original sample, according to the Characteristics of Temperature Field value in lithium battery bin as modeling sample；Svm classifier model is constructed based on modeling sample；Fire probability model is constructed, probabilistic model construction method is to be mapped to the value of the distance of test sample to the svm classifier model hyperplane [0,1] using Sigmoid, and the result of mapping is the probability that fire occurs；Present system includes: temperature information acquisition module, svm classifier model construction module, fire probability model construction module, display module, fire alarm module.The present invention applies to machine learning and embedded development in the prediction of fire, preferably incorporates historical data in prediction while guaranteeing real-time, so that prediction result is more accurate and reaction is rapider.
Description
Technical field
The present invention relates to fire prediction technique and systems in a kind of lithium battery bin, and in particular to one kind based on SVM (support to
Amount machine) lithium battery bin fire prediction method and system.
Background technique
Current driving force battery catches fire has become electric car and leading cause of fire occurs for hybrid vehicle
One of, once it is on fire, it will seriously endanger vehicle occupant, the security of the lives and property, generation while causing huge economic losses
Severe social influence, this will hinder the popularization and use process of national new-energy automobile.Therefore, lithium can be rapidly and accurately predicted
Battery compartment fire becomes a very important project of current electric car safety.Traditional fire detector relies primarily on biography
Sensor is monitored the combustion products such as temperature, smog, is judged as generation fire, traditional prediction side when exceeding certain threshold value
The method reaction time is too long, and lithium battery bin belongs to confined space and battery work will also generate certain heat, traditional prediction
Method rate of false alarm is also higher.
Summary of the invention
The purpose of the present invention is to provide a kind of lithium battery bin fire prediction method and system based on SVM, for current
The incomplete problem of deficiency and system function of fire prediction method, the present invention apply to machine learning and embedded development
In the prediction of fire, by establishing fire prediction model, historical data is preferably incorporated to prediction while guaranteeing real-time
In, so that prediction result is more accurate and reaction is rapider.
In order to achieve the above objectives, the present invention adopts the following technical scheme:
A kind of lithium battery bin fire prediction method based on SVM, comprising the following steps:
If S1, dry temperature sensor is arranged in lithium battery bin, with the environment of the collected lithium battery bin of temperature sensor
Temperature information is original sample；
S2, lithium battery bin Characteristics of Temperature Field value is extracted according to original sample；
S3, using Characteristics of Temperature Field value as modeling sample, construct svm classifier model；
S4, building fire probability model, using Sigmoid by test sample to svm classifier model hyperplane away from
From value be mapped to [0,1], the result of mapping be fire occur probability.
Further, the temperature sensor effective monitoring scope covers environment in entire lithium battery bin, and single temperature
Sensor sphere of action is not overlapped.
Further, using the mean value of original sample and variance as Characteristics of Temperature Field value in S2, specific formula is as follows:
The mean value of every group of temperature data in original sample are as follows:
The variance of every group of temperature data in original sample are as follows:
Wherein, TiRespectively indicate i-th of temperature sensor temperature value in collected coverage area, n indicate temperature
The number of sensor.
Further, MATLAB Data Analysis Platform is used using modeling sample as the input of model training sample in S3
And the relative program packet of LIBSVM constructs svm classifier model.
Further, the value of the distance of test sample to svm classifier model hyperplane is mapped using Sigmoid in S4
To [0,1], the result of mapping is the probability that fire occurs, specific formula is as follows:
Distance of the test sample to svm classifier model hyperplane are as follows: g (x)=ω x+b
Wherein, x is test sample Characteristics of Temperature Field value, and b is classification thresholds, and ω is weight vector；
Probability is converted by Sigmoid:
Wherein g (x) is when for timing, for characteristic value above hyperplane, the probability acquired is the probability that fire occurs；Work as g
(x) when being negative, for characteristic value below hyperplane, the probability acquired is the probability that fire does not occur；When g (x) is zero, characteristic value
On hyperplane, show fire occurs and the probability of fire does not occur respectively to account for 50%.
Further, after finding out the probability that fire occurs, different alarm signals is issued according to probability interval, specifically such as
Under:
When fire probability section (0,0.2] when, issue warning yellow signal；
When fire probability section (0.2,0.8] when, issue Amber Alert signal；
When fire probability section (0.8,1] when, issue warning red signal.
A kind of lithium battery bin fire prediction system based on SVM, including temperature information acquisition module, the SVM being sequentially connected
Disaggregated model constructs module and fire probability model construction module；
The temperature information acquisition module is used to acquire the ambient temperature information of lithium battery bin as original sample, and root
Lithium battery bin Characteristics of Temperature Field value is extracted according to original sample；
The svm classifier model construction module constructs svm classifier mould for the Characteristics of Temperature Field value based on original sample
Type；
The fire probability model construction module is used to utilize Sigmoid by test sample to svm classifier model
The value of the distance of hyperplane is mapped to [0,1], and the result of mapping is the probability that fire occurs.
Further, display module is connected in fire probability model construction module, the display module is used for
Display in real time the result of temperature information acquisition module and fire probability model construction module.
Further, fire alarm module, the fire alarm are connected in fire probability model construction module
Module is for issuing different alarm signals, specific rules according to the result of fire probability model construction module are as follows:
When fire probability section (0,0.2] when, issue warning yellow signal；
When fire probability section (0.2,0.8] when, issue Amber Alert signal；
When fire probability section (0.8,1] when, issue warning red signal.
Compared with prior art, the invention has the following beneficial technical effects:
Present invention combination machine learning algorithm and embedded development, and propose a kind of extracting method of Characteristics of Temperature Field, lead to
Excess temperature sensor acquires the environment temperature of lithium battery bin, and then extracts and obtain lithium battery bin temperature profile value, and utilizes SVM points
Class device classifies to historical sample Characteristics of Temperature Field value, and historical data is made full use of while guaranteeing real-time, improves and knows
The accuracy of other fire, in addition the present invention is by building fire probability model, using Sigmoid by test sample to SVM
The value of the distance of disaggregated model hyperplane is mapped to [0,1], and the result of mapping is the probability that fire occurs, so that predicting
Cheng Gengjia is rapid, the results show, and this method can improve the accuracy and speed of fire identification to a certain extent.
Further, temperature sensor effective monitoring scope of the present invention covers environment in entire lithium battery bin, to guarantee temperature
Each temperature sensor data does not interact when spending field feature extraction, and single temperature sensor sphere of action is not overlapped.
Further, the present invention can according to occur fire probability locating for section issue different alarm signals, with
Staff is reminded to take precautions against.
Present system acquires lithium battery bin temperature information and Extracting temperature field characteristic value using temperature information acquisition module,
Model is constructed using Characteristics of Temperature Field value by svm classifier model construction module and historical sample Characteristics of Temperature Field value is divided
Class makes full use of historical data while guaranteeing real-time, improves the accuracy of identification fire, passes through fire probability mould
Type construct module the value of the distance of test sample to the svm classifier model hyperplane is mapped to using Sigmoid [0,
1], the result of mapping is the probability that fire occurs, and it is convenient rapidly to predict.
Further, present system by be arranged display module can display in real time temperature information acquisition module and
Fire probability model construction module as a result, observing convenient for staff.
Further, present system, can be according to locating for the probability that fire occurs by setting fire alarm module
Section issues different alarm signals, to remind staff to take precautions against.
Detailed description of the invention
Fig. 1 is the flow chart of the lithium battery bin fire prediction method the present invention is based on SVM；
Fig. 2 is the structural schematic diagram of the lithium battery bin fire prediction system the present invention is based on SVM.
Wherein: 1, temperature information acquisition module；2, svm classifier model construction module；3, fire probability model construction
Module；4, display module；5, fire alarm module.
Specific embodiment
Present invention is further described in detail with reference to the accompanying drawing:
Referring to Fig. 1, a kind of lithium battery bin fire prediction method based on SVM, comprising the following steps: with the ring of lithium battery bin
Border temperature information is original sample, using the Characteristics of Temperature Field value in lithium battery bin as modeling sample；Based on the modeling sample
This building svm classifier model；Fire probability model is constructed, the probabilistic model construction method is will using Sigmoid
The value of distance of test sample to the svm classifier model hyperplane is mapped to [0,1], and the result of mapping is fire hair
Raw probability.
The temperature profile value, is laid out using multiple temperature sensors, and covers sensor effective monitoring scope
Environment in entire lithium battery bin is covered, each sensing data does not interact when to guarantee that Characteristics of Temperature Field is extracted, single to sense
Device sphere of action is not overlapped, and collected data are as original sample, and the mean value and variance of original sample are as Characteristics of Temperature Field
Value.
The svm classifier model uses MATLAB data point using temperature profile value as the input of model training sample
(program bag is that a set of of Taiwan Univ. professor Lin Zhiren exploitation is used for SVM mould to the relative program packet of analysis platform and LIBSVM
Program bag formula identification and returned, download address are http://www.csie.ntu.edu.tw/~cjlin/) SVM points of building
Class model, using default parameters.
The fire probability model construction module is used to utilize Sigmoid by test sample to svm classifier model
The value of the distance of hyperplane is mapped to [0,1], and the result of mapping is the probability that fire occurs.
Referring to fig. 2, a kind of lithium battery bin fire prediction system based on SVM includes temperature information acquisition module 1, SVM points
Class model constructs module 2, fire probability model construction module 3, display module 4, fire alarm module 5；
The temperature information acquisition module 1 is used to acquire the ambient temperature information of lithium battery bin as original sample；
The svm classifier model construction module 2 constructs svm classifier mould for the temperature profile value based on original sample
Type；
The fire probability model construction module 3 is used to divide test sample to the SVM using Sigmoid
The value of the distance of class model hyperplane is mapped to [0,1], and the result of mapping is the probability that fire occurs；
The display module 4 is for displaying in real time temperature information acquisition module 1 and fire probability model structure
Model the result of block 4；
The fire alarm module 5 is different for being issued according to the result of fire probability model construction module 4
Alarm signal, specific rules are as follows:
When fire probability section (0,0.2] when, issue warning yellow signal；
When fire probability section (0.2,0.8] when, issue Amber Alert signal；
When fire probability section (0.8,1] when, issue warning red signal；
Operation of the present invention process is described in detail below:
As shown in Figure 1, the present invention is based on the lithium battery bin fire prediction method of SVM the following steps are included:
Step S1, using the ambient temperature information of lithium battery bin as original sample.
Specifically, it is laid out using multiple temperature sensors, and sensor effective monitoring scope is made to cover entire lithium electricity
Environment in the storehouse of pond, each sensing data does not interact when to guarantee that Characteristics of Temperature Field is extracted, single sensor sphere of action
It is not overlapped, collected data are as original sample.
Step S2, lithium battery bin Characteristics of Temperature Field value are extracted, and obtained characteristic value is as svm classifier model modeling sample.
Specifically, for the mean value and variance of original sample as temperature profile value, formula is as follows:
The mean value of every group of temperature data in original sample are as follows:
The variance of every group of temperature data in original sample are as follows:
Wherein, TiRespectively indicate i-th of temperature sensor temperature value in collected coverage area, n indicate temperature
The number of sensor.
Step S3 constructs svm classifier model based on the modeling sample.
Specifically, using Characteristics of Temperature Field value as the input of model training sample, using MATLAB Data Analysis Platform with
And the program bag of LIBSVM constructs svm classifier model, using default parameters.
Step S4 constructs fire probability model, using Sigmoid by test sample to the svm classifier model
The value of the distance of hyperplane is mapped to [0,1], and the result of mapping is the probability that fire occurs, specific formula is as follows:
Distance of the test sample to svm classifier model hyperplane are as follows: g (x)=ω x+b
Wherein, x is test sample Characteristics of Temperature Field value, and b is classification thresholds, and ω is weight vector；
Probability is converted by Sigmoid:
Wherein g (x) is when for timing, for characteristic value above hyperplane, the probability acquired is the probability that fire occurs；Work as g
(x) when being negative, for characteristic value below hyperplane, the probability acquired is the probability that fire does not occur；When g (x) is zero, characteristic value
On hyperplane, show fire occurs and the probability of fire does not occur respectively to account for 50%.
As shown in Fig. 2, the lithium battery bin fire prediction system of the invention based on SVM include temperature information acquisition module 1,
Svm classifier model construction module 2, fire probability model construction module 3, display module 4, fire alarm module 5.
Temperature information acquisition module 1 is used to acquire the ambient temperature information of lithium battery bin as original sample.
Specifically, it is laid out using multiple temperature sensors, and sensor effective monitoring scope is made to cover entire lithium electricity
Environment in the storehouse of pond, each sensing data does not interact when to guarantee that Characteristics of Temperature Field is extracted, single sensor sphere of action
It is not overlapped, collected data are as original sample.
Therefore, temperature information acquisition module 1 executes following operation:
(1) acquisition lithium battery bin is occurring fire and the ambient temperature information of fire is not occurring as original sample respectively,
Using ambient temperature information when fire occurs as positive sample, the ambient temperature information of fire does not occur as negative sample.
(2) mean value and variance of original sample are as Characteristics of Temperature Field value.
Svm classifier model construction module 2 is connected with temperature information acquisition module 1, for the temperature field based on original sample
Characteristic value constructs svm classifier model.
Specifically, input of the temperature profile value as model training sample, using MATLAB Data Analysis Platform and
The relative program packet of LIBSVM constructs svm classifier model, using default parameters.
Fire probability model construction module 3 is connected with svm classifier model construction module 2, for utilizing Sigmoid will
The value of distance of test sample to the svm classifier model hyperplane is mapped to [0,1], and the result of mapping is fire hair
Raw probability.
Specifically, test sample is calculated first to the distance of svm classifier model hyperplane, recycles Sigmoid that will test
The value of distance of sample to the svm classifier model hyperplane is mapped to [0,1], and the result of mapping is what fire occurred
Probability.
Display module 4 is connected with fire probability model construction module 3, for displaying in real time the temperature information
The result of acquisition module and the fire probability model construction module.
Specifically, the data of each temperature sensor and fire in lithium battery bin is displayed in real time with liquid crystal display to occur generally
The probability that rate model construction module 3 exports.
Fire alarm module 5 is connected with fire probability model construction module 3, general for being occurred according to the fire
The result of rate model construction module issues different alarm signals.
Specifically, when fire probability section (0,0.2] when, issue warning yellow signal；
When fire probability section (0.2,0.8] when, issue Amber Alert signal；
When fire probability section (0.8,1] when, issue warning red signal.
Particular embodiments described above has carried out technical solution disclosed by the invention and beneficial achievement further detailed
Describe in detail it is bright, related researcher it should be understood that its according to can to foregoing embodiments describe technical solution modify, or
Its some technical characteristics is replaced on an equal basis, and these modifications or substitutions, it does not separate the essence of the corresponding technical solution originally
The scope of inventive technique scheme.
Claims (9)
1. a kind of lithium battery bin fire prediction method based on SVM, which comprises the following steps:
If S1, dry temperature sensor is arranged in lithium battery bin, with the environment temperature of the collected lithium battery bin of temperature sensor
Information is original sample；
S2, lithium battery bin Characteristics of Temperature Field value is extracted according to original sample；
S3, using Characteristics of Temperature Field value as modeling sample, construct svm classifier model；
S4, building fire probability model, using Sigmoid by the distance of test sample to svm classifier model hyperplane
Value is mapped to [0,1], and the result of mapping is the probability that fire occurs.
2. a kind of lithium battery bin fire prediction method based on SVM according to claim 1, which is characterized in that the temperature
It spends sensor effective monitoring scope and covers environment in entire lithium battery bin, and single temperature sensor sphere of action is not overlapped.
3. a kind of lithium battery bin fire prediction method based on SVM according to claim 1, which is characterized in that will in S2
The mean value and variance of original sample as Characteristics of Temperature Field value, specific formula is as follows:
The mean value of every group of temperature data in original sample are as follows:
The variance of every group of temperature data in original sample are as follows:
Wherein, TiRespectively indicate i-th of temperature sensor temperature value in collected coverage area, n indicate temperature sensor
Number.
4. a kind of lithium battery bin fire prediction method based on SVM according to claim 1, which is characterized in that in S3 with
Input of the modeling sample as model training sample uses MATLAB Data Analysis Platform and the relative program packet structure of LIBSVM
Build svm classifier model.
5. a kind of lithium battery bin fire prediction method based on SVM according to claim 1, which is characterized in that sharp in S4
The value of the distance of test sample to svm classifier model hyperplane is mapped to [0,1] with Sigmoid, the result of mapping is
The probability that fire occurs, specific formula is as follows:
Distance of the test sample to svm classifier model hyperplane are as follows: g (x)=ω x+b
Wherein, x is test sample Characteristics of Temperature Field value, and b is classification thresholds, and ω is weight vector；
Probability is converted by Sigmoid:
Wherein g (x) is when for timing, for characteristic value above hyperplane, the probability acquired is the probability that fire occurs；When g (x) is
When negative, for characteristic value below hyperplane, the probability acquired is the probability that fire does not occur；When g (x) is zero, characteristic value is super
In plane, show fire occurs and the probability of fire does not occur respectively to account for 50%.
6. a kind of lithium battery bin fire prediction method based on SVM according to claim 5, which is characterized in that when finding out
After the probability that fire occurs, different alarm signals is issued according to probability interval, specific as follows:
When fire probability section (0,0.2] when, issue warning yellow signal；
When fire probability section (0.2,0.8] when, issue Amber Alert signal；
When fire probability section (0.8,1] when, issue warning red signal.
7. a kind of lithium battery bin fire prediction system based on SVM, which is characterized in that including the temperature information acquisition being sequentially connected
Module (1), svm classifier model construction module (2) and fire probability model construction module (3)；
The temperature information acquisition module (1) is used to acquire the ambient temperature information of lithium battery bin as original sample, and root
Lithium battery bin Characteristics of Temperature Field value is extracted according to original sample；
The svm classifier model construction module (2) constructs svm classifier mould for the Characteristics of Temperature Field value based on original sample
Type；
The fire probability model construction module (3) is used to surpass test sample to svm classifier model using Sigmoid
The value of the distance of plane is mapped to [0,1], and the result of mapping is the probability that fire occurs.
8. a kind of lithium battery bin fire prediction system based on SVM according to claim 7, which is characterized in that fire hair
It is connected with display module (4) on raw probabilistic model building module (3), the display module (4) is for displaying in real time temperature
The result of information acquisition module (1) and fire probability model construction module (3).
9. a kind of lithium battery bin fire prediction system based on SVM according to claim 7, which is characterized in that fire hair
It is connected with fire alarm module (5) on raw probabilistic model building module (3), the fire alarm module (5) is used for according to fire
The result of calamity occurrence Probability Model building module (4) issues different alarm signals, specific rules are as follows:
When fire probability section (0,0.2] when, issue warning yellow signal；
When fire probability section (0.2,0.8] when, issue Amber Alert signal；
When fire probability section (0.8,1] when, issue warning red signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811573513.6A CN109685266A (en) | 2018-12-21 | 2018-12-21 | A kind of lithium battery bin fire prediction method and system based on SVM |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811573513.6A CN109685266A (en) | 2018-12-21 | 2018-12-21 | A kind of lithium battery bin fire prediction method and system based on SVM |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109685266A true CN109685266A (en) | 2019-04-26 |
Family
ID=66188849
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811573513.6A Pending CN109685266A (en) | 2018-12-21 | 2018-12-21 | A kind of lithium battery bin fire prediction method and system based on SVM |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109685266A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110988714A (en) * | 2019-11-11 | 2020-04-10 | 天津大学 | Lithium battery fire hazard comprehensive testing system and method |
| CN113085555A (en) * | 2021-03-24 | 2021-07-09 | 东风汽车集团股份有限公司 | Warning method for battery pack after thermal runaway |
| CN113391209A (en) * | 2021-05-26 | 2021-09-14 | 江苏小牛电动科技有限公司 | Method, device and system for predicting health state of battery and battery |
| CN118038623A (en) * | 2024-04-11 | 2024-05-14 | 山东引领电子科技有限公司 | Automobile fire alarm system and method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103886344A (en) * | 2014-04-14 | 2014-06-25 | 西安科技大学 | Image type fire flame identification method |
| CN106355812A (en) * | 2016-08-10 | 2017-01-25 | 安徽理工大学 | Fire hazard prediction method based on temperature fields |
| CN107085904A (en) * | 2017-03-31 | 2017-08-22 | 上海事凡物联网科技有限公司 | Forest fire danger class decision method and system based on single classification SVM |
| CN107634857A (en) * | 2017-09-22 | 2018-01-26 | 安徽师范大学 | Fault Model structure and appraisal procedure based on SVM |
-
2018
- 2018-12-21 CN CN201811573513.6A patent/CN109685266A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103886344A (en) * | 2014-04-14 | 2014-06-25 | 西安科技大学 | Image type fire flame identification method |
| CN106355812A (en) * | 2016-08-10 | 2017-01-25 | 安徽理工大学 | Fire hazard prediction method based on temperature fields |
| CN107085904A (en) * | 2017-03-31 | 2017-08-22 | 上海事凡物联网科技有限公司 | Forest fire danger class decision method and system based on single classification SVM |
| CN107634857A (en) * | 2017-09-22 | 2018-01-26 | 安徽师范大学 | Fault Model structure and appraisal procedure based on SVM |
Non-Patent Citations (3)
| Title |
|---|
| 基于支持向量机的图像型火灾探测算法: "杨娜娟，王慧琴，马宗方", 《计算机应用》 * |
| 张绪祥，高扬: "基于HMM的锂电池仓火灾预测算法研究", 《汽车实用技术》 * |
| 沈兰荪，蔡轶珩，张新峰: "《中医舌象的采集与分析》", 30 April 2007, 沈兰荪，蔡轶珩，张新峰 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110988714A (en) * | 2019-11-11 | 2020-04-10 | 天津大学 | Lithium battery fire hazard comprehensive testing system and method |
| CN113085555A (en) * | 2021-03-24 | 2021-07-09 | 东风汽车集团股份有限公司 | Warning method for battery pack after thermal runaway |
| CN113085555B (en) * | 2021-03-24 | 2022-06-03 | 东风汽车集团股份有限公司 | Warning method for battery pack after thermal runaway |
| CN113391209A (en) * | 2021-05-26 | 2021-09-14 | 江苏小牛电动科技有限公司 | Method, device and system for predicting health state of battery and battery |
| CN118038623A (en) * | 2024-04-11 | 2024-05-14 | 山东引领电子科技有限公司 | Automobile fire alarm system and method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109685266A (en) | A kind of lithium battery bin fire prediction method and system based on SVM | |
| CN103824420B (en) | Fatigue driving identification system based on heart rate variability non-contact measurement | |
| CN107320115B (en) | Self-adaptive mental fatigue assessment device and method | |
| CN104802737A (en) | Mobile phone based vehicle abnormality driving behavior detection method | |
| CN103559508B (en) | A kind of based on continuous Adaboost video vehicle detection method | |
| CN109034111A (en) | A kind of driver's hand based on deep learning is from steering wheel detection method and system | |
| CN108345666A (en) | A kind of vehicle abnormality track-detecting method based on time-space isolated point | |
| CN105615878A (en) | Fatigue driving electroencephalographic monitoring method | |
| CN109583508A (en) | A kind of vehicle abnormality acceleration and deceleration Activity recognition method based on deep learning | |
| CN109686050A (en) | Environment inside car monitoring and pre-alarming method based on cloud service and deep neural network | |
| CN102682601A (en) | Expressway traffic incident detection method based on optimized support vector machine (SVM) | |
| CN108682119A (en) | Method for detecting fatigue state of driver based on smart mobile phone and smartwatch | |
| CN106779215B (en) | Machine learning-based power grid global delay situation sensing method | |
| CN104978570A (en) | Incremental learning based method for detecting and identifying traffic sign in traveling video | |
| CN109598174A (en) | The detection method and its device and system of driver status | |
| CN102707713B (en) | Fault diagnosis system and method for automobile safety air bag assembly working procedure | |
| CN106792883A (en) | Sensor network abnormal deviation data examination method and system | |
| Li et al. | Detecting stealthy cyberattacks on automated vehicles via generative adversarial networks | |
| CN108710637A (en) | Taxi exception track real-time detection method based on time-space relationship | |
| CN114235050A (en) | Marine environment monitoring and early warning method, device and system | |
| CN110210382A (en) | A kind of face method for detecting fatigue driving and device based on space-time characteristic identification | |
| Wang et al. | Smartphone sensors-based abnormal driving behaviors detection: Serial-feature network | |
| CN114863170A (en) | Deep learning-based new energy vehicle battery spontaneous combustion early warning method and device | |
| CN104200599B (en) | Bus multisensor fire detection method | |
| CN202548932U (en) | Vehicle-mounted smart traveling data recorder |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190426 |