Research on Data Fusion Model of Safety Monitoring for Oil Depot

Faced with the problem of safety monitoring of oil depot under complex environment, a multi-source data fusion model was constructed with combination of fuzzy neural network and particle swarm optimization. Pretreatment methods of monitoring data were given for eliminating interference. Multi-source data fusion algorithm based on fuzzy neural network was proposed, in which the fuzzy reasoning rules were embedded. Particle Swarm Optimization algorithm was used to train fuzzy neural network weights, prune redundancy link, and optimize data fusion fuzzy rule base. Simulation verification shows that this model could realize security status assessment of oil depot effectively, and reduce the false alarm rate and missing rate, proving to have strong practical value. http://www.sj-ce.org/paperInfo.aspx?ID=153