Bus Spontaneous-ignition Monitoring And Warning System Based On Data Fusion

There exist frequent bus spontaneous combustion fire accidents at present, which is due to the issues like the scarcity of the bus fire detection means, the complexity work environment of the vehicle itself and the difficulty of fire and rescue.This article focus on these three issues and conducted the following studies:Firstly, the bus spontaneous combustion mechanism was studied, and summarized that the bus spontaneous combustion fire accident mainly occurred in the engine compartment. The major causes are the short circuit of the electrical components in engine compartment and the high heat of mechanical friction. By the analysis of bus spontaneous combustion hazard sources, the article designs the bus spontaneous combustion monitoring and warning system, for the experimental data collection of engine compartment.Secondly, after the analysis of the combustion characteristics of the bus spontaneous combustion,. designed a bus engine compartment simulation platform for fire data collection. And through the colletion experiments of the spontaneous-ignition sensing parameters in engine compartments of three different fuels, the combustion characteristics of different materials are studied.Finally, the article makes research on the support vector machine model and data fusion methods, designs the data fusion method based on the support vector machine.and the performance of different material fire is tested by this method.. The experimental results show that by the data association, kernel function type selection and model parameter optimization, it can significantly improve the identification correct rate of the SVM model for the different material fire.This paper shows that in bus spontaneous combustion monitoring system, using support vector machine and the data fusion method of data association, can effectively identified the bus spontaneous combustion operating mode and the combustion sources, then to improve the bus safety.