Application Research Of Fuzzy Neural Networks In The Grinding Control System

The grinding classification production process is an important part of the beneficiation flowsheet. Ores are grinded and classified through this process,useful minerals and gangue minerals are monomer dissociated,it creates conditions for the subsequent sorting operations. This thesis choose the typical two-stage grinding and classification of production process as the research object, CFNN is used for the grinding controller design, and simulation experiment is carried out in optimizing virtual equipment of grinding. The main contents of this thesis are as follows:First of all,it summarizes the domestic and foreign development and application situation of the grinding control in recent years and analyzes the characteristics and difficulties for control of the grinding classification production process.Then, through the research on the parameters of grinding process,it summarizes the related control law and ensures the granularity as the final control target and takes ore quantity,the adding water of the first ball mill entrance,the adding water of the spiral classifier,the flow of cyclone as the control variables. Identification of grinding system with neural network to establish the mathematical model.Aiming at these problems of complicated mechanism and many influence factors in typical two-stage grinding system,CFNN is used to design the grinding controller. By utilizing the processing ability of fuzzy control for fuzzy information and the strong learning ability of neural network,nonlinear and strong coupling in grinding control are well solved. By introducing compensatory fuzzy neural cells,the network with correct or erroneous fuzzy rules defined initially can be trained to posses higher fault tolerance and stability. Through the MATLAB simulation software to verify it,the result of simulation shows that ore granularity was well controlled in an ideal scope. It proved the effectiveness and practicality of CFNN to the grinding control.Finally,the control algorithm of CFNN is loaded into the grinding optimize virtual devices for semi-physical simulation,it got satisfied trendency chart of granularity and achieved good control effect.