Research And Realization On Key Technologies For Safety Check Of Transportation Of Railway Inflammable Goods

As China’s modern industrialization process has been accelerated, demands for transporting dangerous goods as important chemical raw materials have been increasing year by year. Railway is the main channel for the transportation of dangerous goods in China. However, due to the long transport route, changing transportation environments, and the physical and chemical characteristics of dangerous goods, once accidents happen in the transportation by railway, causalities along the railway route, environmental pollution, economic losses and significant social impact would be caused. At present, accidents often occur in the process to transport dangerous goods by railway. There is an urgent need to establish an effective monitoring system for the transportation of dangerous goods, so as to protect the safety of railway transportation.Power supply is usually unavailable for trucks used in the railway transportation. How to ensure the accuracy and reliability of the monitoring system and extend the life cycle of the system with the limited electric quantityis the main content of the design and research in this paper.Cotton is an important cash crop in China. Every year, a large amount of cotton is delivered to places around China by railway. In the transport process, it is quite possible that after being heated, cotton suffers spontaneous combustion, thus causing accidents. In this paper, cotton is taken as the research object. A system that can make real-time monitoring of the security status of the train carriages is designed, so as to avoid the spontaneous combustion of cotton in the transport process. In this paper, a mathematical model of the cotton combustion in the train carriages is built to make simulation of the combustion states of cotton, thus analyzing changes laws of smoke, temperature and so on when cotton suffers the spontaneous combustion in train carriages, providing theoretical support for the type selection and design, node deployment and system testing strategies of the sensor and offering data to support the sensor nodes to use the neural networks for data fusion of multi-sensor features. Based on the numerical simulation results of the cotton combustion in train carriages, the hardware and software design of the network security monitoring system with a wireless sensor is provided to cotton transportation. Moreover, research and analyses are made of the sampling mechanism of the monitoring mechanism, energy-saving programs and sensors of the system and multi-sensor data fusion. This system can monitor a variety of environmental information within train carriages. Compared with systems which can just monitor a certain environmental feature, this system can gain more accurate results, has a lower false alarm rate and provide more comprehensive feedback to accidents. Neural network, through the learning and training of change rules of characteristic quantitiesof variousenvironmental features, accesses to the features of cotton about its combustion in different stages. Using the neural network for data fusion can not only improve the accuracy of the system and lower the false alarm rate and the missing report rate, but also make judgment for the current accidents in the outbreak of fire, so as to ensure the following fire fighting. In this paper, attention is paid on research on data fusion of various sensors. A neural network data fusion model applied into nodes is established to use excellent generalization and fault-tolerating features of the neural network to improve the abilities of nodes to make judgments and analyses of environments in train carriages, reduce the transmission of redundant data among different nodes and extending the life cycle of the monitoring system.