Research On Safety Evaluation Of Vehicle Operating Status & Its Application

Safety evaluation of vehicle running status is always the hot spot in rail safety research. It has great effect on preventing traffic accidents of freight train and reducing serious accidents in train movement. It plays an important role in safety and steady operation and management in railway.Freight train becomes one of the main mode in railway transportation, which plays a pivotal position. In case of accidents, it would cause very serious loss in social consequence and property. Therefore, it received increasing attention.Based on non-integrality analysis of current research of vehicle running status and the theory of "Human-Equipment-Environment", assuming people be in a reliable state, will not lead to any risk, In-depth study of the environmental factors and the equipment factors of vehicle operation, this thesis put forward a comprehensive evaluation approach combined by natural environment and equipment operation. The approach creatively takes vehicle's hotbox into consideration, and uses it as an important influencing factor. And then this thesis establishes a safety evaluation framework of vehicle running status via the approach, which is a maiden attempt in this domain.Based on the analysis of the traditional BP neural network, this thesis promotes the it with the sigmoid activation function, and proposes a sectional fuzzy BP neural network which consists of three parts and has string-shaped structure. The improved algorithm is better to evaluate such complex system with huge volume of data.The result shows that the method can achieve an effective evaluation of the freight train operating status, and in practice to expand the applications.On the basis of the above-mentioned research work, this thesis develops a special safety evaluation system of vehicle running status, which could gradually accumulate experience database of safety evaluation, enhance the applicability of evaluation approach.