Research On Thickness Control Of Hot Rolling Based On Improved Chaos Particle Swarm Optimization

The quality level of strip steel depends largely on the hot strip mill thickness control system.With the domestic economic development,the steel market is in an unprecedented fierce competition.So more companies focus on improving the quality of products to meet customer needs.But in recent years,China’s strip steel production capacity surplus,coupled with the strip plate thickness of the product,is not up to standard.So the pursuit of higher accuracy of the thickness of the field has become the focus of attention.After years of development,the traditional control theory based on the production line has been difficult to meet with the development of science and technology and product quality requirements of higher market.Some difficulties can not be dealt with to meet the industrial demand for a large number of steel.In order to make up the lackness of traditional control methods to meet the needs of the market,new control theory and method come into being.Aiming at improving the shortcomings of standard particle swarm optimization(PSO)algorithm,which is slow in convergence and is easy to fall into local optimum in optimization process,the chaotic particle swarm optimization algorithm is proposed to adjust the inertia weight according to the degree of premature convergence and individual fitness.It makes full use of chaos optimization and particle swarm optimization search characteristics to train the weight threshold of the neural network,which solves the problem that the BP neural network is slow to converge and easily fall into the local extremum.The simulation results showed that the algorithm not only improves the error precision,but also speeds up the training convergence.Using the improved chaotic particle swarm neural network to adjust the PID parameters for the thickness control system,it realizes the essential combination of the neural network and PID control law,thus an intelligent PID controller is constructed.Compared with the conventional PID controller,it is faster and more stable.This method is stable and feasible,and significantly improved the traditional PID controller.