Research On Heat Treatment Furnace Modeling And Optimization Based On Dynamic Fuzzy Neural Network

Heat treatment furnace of silicon steel stators and rotors is an important equipment of continuous annealing and bluing process in the production process of motor. Reasonable settings of every district temperature and the protection gas concentration and other parameters of heat treatment furnace, can not only ensure production quality of stators and rotors, but also can reduce the energy consumption of heat treatment furnace. Therefore, research on heat treatment furnace modeling and optimization has an important application value.Taking the tunnel bluing furnace for heat treatment as the application background, the main work and achievements of this thesis are as follows:1) A complex system modeling method based on dynamic fuzzy neural network was proposed. Through the analysis of the object, we selected appropriate model and input variables, and realized modeling of nonlinear dynamic systems. Model prediction simulations verified the feasibility of the modeling method.2) Based on dynamic fuzzy neural network model, particle swarm algorithm was applied to optimize the operation of the production process. By adjusting the premise parameters, analyzing the input data, we can find out the parameters effect output obviously, then optimize the parameters by using particle swarm optimization algorithm.3) Based on operational data which were collected from the heat treatment furnace, we got on dynamic fuzzy neural network modeling and research of operations optimization with particle swarm. The results of the research on modeling and optimization showed that the output by optimizing protection gas concentration was better and the error had been reduced to an acceptable range. According to the optimized operation parameters we can determine whether silicon steel stators and rotors in heat treatment process is abnormal or not, so as to take effective control and treatment to ensure the quality of product. This study shows that the research of heat treatment furnace modeling and optimization based on dynamic fuzzy neural network has a certain value which can be used to guide the actual production operations.