Research On Electrode Control System Of Electric Arc Furnace Based On RBF Neural Network Inverse Identification

This paper takes a steel-making electric arc furnace control system of Masteel as the research object,and expounds the research and development process of the electric arc furnace electrode control system.The electrode control system of electric arc furnace is a complicated system that is changeable,nonlinear,strongly coupled,time-varying and has many interference factors.it is difficult to establish a more accurate mathematical model with traditional control theory,let alone to solve the decoupling problem.it is difficult to meet the requirements of production for electrode control.Aiming at this production problem,in the research and development of the electrode control system of electric arc furnace,the method of neural network on-line inverse identification and inverse control,that is,the electrode system control of electric arc furnace based on RBF network inverse identification,has realized the real-time on-line decoupling and control of the three-phase coupling system.According to the idea of inverse dynamic control,this paper puts forward an online self-learning control scheme combining RBF neural network inverse control and PID control.Real-time data of electrode current in electric arc furnace are treated as input samples of identifier,and the nearest neighbor clustering learning algorithm is applied to realize dynamic identification of inverse model of electrode control system in electric arc furnace.the inverse identification model obtained by dynamic identification is taken as a controller model,which is connected in series with the controlled object to form a normalized system with linear transmission relationship and decoupling.therefore,the control problem of the controlled object is transformed from nonlinear control to linear control.In this paper,RBF neural network based on nearest neighbor clustering algorithm is used to inverse identify and control the electrode system of three-phase electric arc furnace,and real-time and online decoupling and control of the electrode system of three-phase electric arc furnace are realized.Matlab is used to simulate and analyze the control of arc furnace electrode system based on RBF network inverse identification.the simulation results show that this method can better decouple and control the system,and at the same time enhance the robustness and anti-interference of the system.it verifies the effectiveness of the control strategy of arc furnace electrode system based on RBFnetwork inverse identification developed in this paper,achieves the expected design goal and has a wide range of engineering application prospects.