Model Free Adaptive Control Theory And Its Applications

Model free adaptive control (MFAC) theory is an adaptive control method which does not need to model the industrial process. Compared with traditional control methods based on modeling, MFAC is an advanced control strategy which based on information of Input/Output Data. It has parameter adaptability and structure adaptability. Based on the background, the study of MFAC theory and its applications in industrial processes is proposed. The thesis is organized as follows:Firstly, lots of simulation experiments on the application of MFAC in typical industrial processes have been done. The simulation results show that MFAC controller has excellent robustness, anti-jamming capability, parameter and structure adaptability.Secondly, a new MFAC control algorithm is deduced. Namely, weighted the slope of the system's output is introduced in the criterion function. In order to solve the control of a class integral system, a MFAC solution based on the new control algorithm with two feedback loops is proposed. Simulations show that it has good control quality. Its track performance and robustness is better than regular MFAC.Thirdly, combined with Human Simulated Intelligent Control (HSIC) strategy, an optimized MFAC is proposed. The HSIC is upon the MFAC, where MFAC is used to control directly. The information of dynamic process is used to heuristic and instinct inference in HSIC. The HSIC chooses the appropriate controller from improved MFAC and regular one continuously. The simulation results show that HIS-MFAC has good dynamic performance and excellent robustness. It can restrain overshoot and accelerate system response simultaneously. This lays a foundation for industrial control on field.