Iterative Learning Control Of Batch Processes

How to improve the control of batch processes is a challenging problem for control community because of nonlinearity, time variation, time delay and modeling errors. The control problem on this system has great value in the research of theory and practice. This dissertation mainly studies affine nonlinear batch process with time delay and disturbance based on the open-loop PD-type iterative learning control algorithm with forgetting factor. The following contents are included:(1) This dissertation uses the open-loop PD-type learning algorithm with forgetting factor and gives the convergence condition of spectral radius form for a class of affine nonlinear systems. Then the convergence of the learning algorithm is given based on operator theory.(2) This dissertation discusses the convergence in three cases according to time delay: the open-loop PD-type learning algorithm with forgetting factor is given for a class of affine nonlinear systems with state delay, the open-loop PD-type learning algorithm of advanced type with forgetting factor is given for a class of affine nonlinear systems with state delay and control delay. The convergence condition of spectral radius form is given respectively. Then the convergence of the learning algorithm is given based on operator theory.(3) The effectivity of the learning algorithm and the convergence condition is proved through the simulation. The result of the simulation between PID and ILC proves the superiority of the method in trajectory tracking and interference suppression.(4) ILC algorithm is applied in Continuous Stirred Tank Reactor (CSTR for short) which is the key control object in the production process of Unsaturated Polyester Resin--a kind of typical chemical processes. The result of the simulation between PID and ILC proves that ILC algorithm can obtain a better control effect.