Design And Research Of Monitoring And Inspection System For Intelligence Warehousing

With the acceleration of China's industrialization,dangerous chemicals logistics capacity increase sharply in recent years,hazardous goods logistics demand is rapid growth?According to the report of the 2010 China Warehousing Association's storage branch of dangerous goods,the output of China's dangerous chemicals has reached 14 million tons and ranks the top in the world.With the development of industrialization,the production and transportation of dangerous chemicals are increasing year by year.The logistics accidents of dangerous chemicals often occur in factories.After the serious fire and explosion accident of Tianjin port 8.12,all over the country strengthened the inspection of dangerous chemicals warehouse,and took effective measures to control such accidents.The risk factors affecting the storage safety of hazardous chemicals are the most essential factors that lead to accidents.Therefore,taking the rare-earth metal warehouse in hazardous chemicals warehouse as the research object,the mobile greedy algorithm of rare-earth metal warehouse is built by using the greedy algorithm.Aiming at the problems of spontaneous combustion and explosion accidents in rare-earth metal warehouse,a dynamic prediction method of the warehouse temperature based on improved extreme learning algorithm is proposed.The contents of the specific research work are as follows:First of all,the greedy genetic hybrid path optimization algorithm is proposed to optimize the route optimization problem of the rare earth metal warehouse inspection.The algorithm adds the greedy strategy to the operation of the genetic algorithm,which is used as a guide for searching the genetic selection operation.The mathematical model is set up whitch the warehouse mobile inspection,and a three dimensional mobile inspection and monitoring system based on the greedy genetic hybrid algorithm is designed.Secondly,in order to accurately predict the real time temperature of the rare earth metal warehouse,a rare-earth metal warehouse temperature prediction method based on improved limit learning machine algorithm is proposed.In this method,data mining technology is applied to denoise the samples of the temperature and time series in the rare-earth metal warehouse.The outlier data in the warehouse temperature samples are identified,and the mean data are processed by averaging the data.The missing data are processed by three spline difference?Thirdly,on the basis of data processing,a dynamic prediction model of rare-earth metal is established,warehouse temperature based on improved limit learning machine algorithm.The theory of structural risk minimization is introduced into the limit learning machine to overcome the over fitting problem in the prediction of the common limit learning machine algorithm.Finally,the simulation results show that the genetic greedy path optimization method avoids the premature convergence of the genetic algorithm and improves the convergence speed.Compared with BPNN,ELM and other prediction methods,the dynamic prediction method of rare-earth metal warehouse based on the improved extreme learning machine has higher prediction accuracy and shorter prediction time.It can provide a useful reference for the safe storage of rare-earth metal warehouse.