| The ball mills is one of the key equipments of grinding process in ore dressing plants and pulverizing system in power plants. Its running state effects main economic and technologic targets of the process systems directly, such as production ability and energy consumption. Ball mill is a complex system with nonlinear, time-variation, large delay and multivariable coupling feature, and its operating conditions often vary violently. It's difficult to build a precise mathematic model of the ball mill and to acquire the ideal control performance with classical control methods. The automation of ball mill has become the obstacle for its commercial application. From these problems, this thesis researches the predictive control strategy and its application on the ball mill. The main works of the thesis are as follows:(1) The principle, dynamic performance and industrial applications of the ball mill are briefly introduced. Some control strategies for ball mill are researched and summarized systemically.(2) An overview of Model predictive control (MPC),Generalized predictive control (GPC) and Predictive functional control (PFC) algorithms is given. PFC algorithm based on finite step response (FSR) model for single input and single output (SISO) system is introduced.(3) A PFC algorithm based on finite step response (FSR) model is extended to multivariable case for the complex characteristic of the ball mill. Simulation results on a pilot ball mill grinding circuit show that the PFC based on multivariable system has better control performance than the decoupled multi-loop PID control.(4) Based on analyzing the dynamic characteristic and modeling of ball coal mill, the generalized predictive controller is designed to improve the security and economics of the process. Simulation results show that the control strategy can achieve better control performance of setpoint tracking, disturbance rejection and robustness to model mismatch.The last part is summary and perspective. |
PDF Full Text Request