Research On Model And Simulation For Shipborne Dangerous Goods Risk Early Warning Based On Wireless Sensor Network

Sea shipping is one of the most leading transportation ways for world trade, and more than half ship cargos belong to dangerous goods. As is well known, the security control and management to the transportation for dangerous cargos is very difficult to execute extremely well, which leads to the accidents happen frequently. Once an accident happens, it will cause serious threat to human life, natural environment and economy. As a result, the safety of water transportation for dangerous cargos has drawn common people’s attention. For present, a lot of safety management work has been done to dangerous goods transportation, but it does not go far enough to ensure transportation safety. So, it is an urgent issue to monitor the safety state of dangerous shipping cargo when they are in transit, which can realize the risk early warning and ensure the safety of dangerous cargo transportation. Aimed at this problem and focused on the package dangerous cargos and their state monitoring and risk early warning, this article will take researches form the following aspects:(1) In order to know well the security status of dangerous goods on board, wireless sensor network is used to monitor the environment information which indicates the safety of dangerous goods, and the wireless sensor network topology control method fit to ship is designed and tested by this paper.(2) Time series analysis is leaded into early warning for the safety of dangerous goods. In the analysis and early warning of dangerous goods security status information from single sensor, traditional difference method would lose the valuable information and influence the precision of early warning in the process of flatting time series. According to this situation, the exponential smoothing and autoregressive moving average are proposed. In the multi-sensor information fusion research, the paper proposes an algorithm for ship-borne dangerous goods in-transit security status information fusion based on co-integration theory, the algorithm plays the advantages of multi-sensor, can predict the dangerous goods status information accurately, and simultaneously, and diagnose whether the status is normal or not. (3) To master the evolution of the risk situation of the ship dangerous goods in abnormal state, Proposes that lead CFD numerical simulation method into the researches of ship packaged dangerous goods transport risk monitoring and early warning, in the case of hazardous gases in cargo hold, leakage and diffusion of hazardous gases are calculated and dynamic simulation by using this method. Meanwhile, combines the corresponding physical experiment to test the feasibility of the method, experimental results show that CFD numerical simulation method used in risk early warning of ship packaged dangerous goods transport is feasible and effective.(4) In order to provide guidance for ship’s officer to emergency disposal of shipping cargos’abnormal condition, a new way to judge the abnormal condition of shipping dangerous package goods is realized on the basis of combining technologies of the WSN monitoring and CFD numerical modeling. In this way, the WSN is used as the tool for monitoring dangerous goods state, and the CFD is used to do numerical model for the typical abnormal state in transit, then these two technologies are combined by the use of Pattern Recognition to judge the state of the shipping dangerous cargos. In the research of similarity measuring method of Pattern Recognition, aimed at the high complexity and low computational efficiency of the existing Dynamic Time Warp (DTW) algorithm, a pattern distance sliding window algorithm based on DTW is presented in this article.