| Through summaring and analysing the hazards of lifting appliances, the study establishes the safety evaluation model based on intrinsically safety theory and associated analytical methods, researches safety evaluation methods based on neural work using information entropy and fuzzy mathematics theory, builds a safety evaluation web-system based on B/S structure for lifting machinery.The main work has been done in this paper includes:(1) Through summing up the hazard factors of lifting appliances and analyzing the strengths and weaknesses of the common safety assessment and prediction methods, a completive and comprehensive safety evaluation index system is built based on the "people-machine-environment" thinking, which divides the hazard factors into four aspects:the equipment body, the organization, the intrinsic safety culture, and the emergency&troubleshooting.(2) Because of the fact that the traditional safety evaluation methods over-reliance on the expert experience and their intermediate parameters and the final result commonly have many subjective factors, this paper designed a novel evaluation method, using the neural network to realize simulation for crane safety evaluation based on comparing probabilistic neural network, RBF neural network and LM-BP neural network, to reduce subjective factors in evaluation and enhance the accuracy and credibility of the evaluation.(3) The paper established an online safety evaluation system for lifting appliances based on the evaluation index system and evaluation method that proposed previously, using the UML language to design system, taking Visual Studio2010as the development platform, C#as the development language and SQL Server2005as the database.(4) In this paper, the information entropy theory and fuzzy mathematics have been used to amend the uncertain data encountered in the study and to reduce the noise in order to enhance the reliability of the research work and the accuracy of the evaluation results. |
PDF Full Text Request