The Key Technologies Research Of Remote Monitoring And Fault Diagnosis In The Transport Ship For Hazardous Chemicals

With the development of both ship transporting and dangerous chemical industry in recent years, it is important to monitor the hazardous chemical’s leakage status and diagnosis faults during the whole process of transportation. Complex equipment of ships and numerous terminal data nodes raise new challenges for data processing and trouble diagnosing in the field of real-time, security and accuracy. In this paper, we focus on the data fusion technology based on routing protocols and key technologies such as dynamic fault set based on case-based reasoning, which satisfies the accuracy and security. The main content is as follows:(1) We analyze the demand of ship-management system, design the framework as well as hardware, software structures, and describe the key working process in detail.(2) For ship-management system problems in the process of data collection and transmission of remote monitoring, data fusion technology is studied and the DFACT algorithm is proposed. The proposed algorithm adopts the structure within the cluster data fusion tree and the method based on the combination of mobile agent model between clusters. Cluster selection was optimized by using particle swarm optimization algorithm in clustering way, which can reduce the gap within the cluster, optimize the network energy consumption, comprehensively consider the effect of the nodes energy and the surrounding nodes to select cluster head, cluster by constructing the data fusion tree in data transmission, as well as the transmission of data encryption, data fusion between cluster based on mobile agent under the model of the optimal path choice. Experiments show the proposed algorithm can carry on the fast data fusion to the node, reduce the energy consumption of nodes, prolong the network life cycle, and protect the security of data.(3) According to the dependencies of fault and symptoms, we put forward a case-based algorithm to diagnosis the fault dynamically. For dynamic fault set, the algorithm on the basis of time interval of fault continues to repair the prior probability of failure in the current time window, using the dependency model build by the Bayesian network. The most likely fault is obtained by the basic diagnostic hypothesis, on the basis of fault assumption; it improved the traditional fault case retrieval strategy of the ship, combined index technology, the similarity theory, narrowed scope, to quickly retrieve a want to speed the highest historical fault cases. Experiments show that the algorithm can meet the accuracy requirements of this system in fault diagnosis.(4) We have implemented the ship-management system, both hardware and software of monitoring and diagnosing. At the same time, the running instance of the system is introduced, which verifies that our research result can be applied to real life and work properly.