An Adaptive Internal Model Control Algorithm Of Research

With the high-tech development and application of the control system performance and industrial process control have become increasingly demanding. Requires not only control accuracy, more emphasis on the control of the robust, real-time, fault tolerance and the control parameters of the adaptive and learning ability. Internal model control as a kind of unique control system structure, or based on process mathematical model controller design of new control strategy. Because of its good control performance, simple design and analysis in the superiority of the system, therefore internal model control is a practical advanced control algorithm, is to improve the level of conventional control system design powerful tool. The development of internal model control is in constantly improving. By the conventional internal model control method for internal model control design to the intelligent development of internal model control method, more typical of the conventional design methods, such as two-step method, zero-pole cancellation method, predictive control law. Many scholars thought the internal model control and adaptive control, fuzzy decision making, neural network combined in complex process control has certain advantages, has been widespread concern.In this thesis, Adaptive Internal Model Control (AIMC) Algorithm was proposed and the simulation results were given. It's showed that Adaptive Internal Model Control Algorithm has good performance at modern mismatching within some range. The application of IMC in multivariable system with multiple time delays was discussed and Multivariable Internal Model control Algorithm proposed by Garcia and Morari was improved. At last, comparison of Improved Multivariable Internal Model Control Algorithm with time-delay Compensator proposed by Ogunnaike was given. Its simulation results on distillation column with multiple time delays showed performance of Improved Multivariable Internal Model Control is better.