Font Size: a A A

Research And Design Of Automatic Warning System For Electric Vehicle Battery Compartment Fire Based On Multi-sensor Fusion

Posted on:2021-12-12Degree:MasterType:Thesis
Country:ChinaCandidate:F Y ZhangFull Text:PDF
GTID:2492306470983239Subject:Vehicle Engineering
Abstract/Summary:PDF Full Text Request
With the development of new energy vehicles,electric vehicles play an increasingly important role in people’s travel,and the safety issues that they bring with it have also received more and more attention.Among them,the battery safety problems of electric vehicles are particularly significant.The existing battery management system(BMS)on electric vehicles can provide effective protection in the normal working state of the battery,but it is still easy to cause dangerous accidents such as fire in the car under some extreme conditions.In view of the above problems,this thesis carries out the research of the automatic warning system for the fire of the battery compartment of the electric vehicle,aiming to realize the identification and prediction of the fire in the battery compartment by detecting the environmental information in the battery compartment,and realize the early warning function.The main contents are as follows:(1)Analysis of lithium battery temperature characteristics and battery compartment ignition characteristics.Analyze the different performance characteristics of power batteries in high temperature and low temperature environments.Through experiments,the battery temperature changes caused by battery charging and discharging at different environmental temperatures are studied,and the cause of thermal runaway of the battery is analyzed.Through the lithium battery ignition experiment and the fire simulation software Pyrosim is used to simulate the ignition characteristics of the battery compartment,the environmental variable parameters used for system monitoring are analyzed.(2)Design of fire detection algorithm.A fire detection model based on multi-sensor fusion is proposed,which preprocesses information in the data layer,introduces BP neural network in the feature layer for training and learning and predicts fire,and introduces DS evidence theory in the decision layer to get the final result.(3)Design of automatic warning system for battery compartment fire,including system hardware design and system software design.The heat dissipation structure of the battery pack in the battery compartment is designed.Based on the real battery compartment design physical model,the semi-physical experiment verification is carried out.The results show that the battery compartment fire early warning system designed in this paper has the advantages of low false alarm rate and more sensitive detection of fire.
Keywords/Search Tags:multi-sensor fusion, automatic fire warning, fire detection model, battery compartment, BP neural network, DS evidence theory
PDF Full Text Request
Related items