Study On Internal Model Control Strategy For Complicated System With Time Delay Process

In the chemical engineering, the complicated plant has the characteristics of long time delay, strong coupling and nonlinear. Thus, the complicated plant is difficult to realize the accuracy control. Consequently, the novel control theory is urgently developted to accommodate to the development requirement of the modern chemical engineering process. The paper focuses on internal model control theory from the aspects of the paremeter uncertainty and the structure uncertainty. And then, some solutions and improvements are proposed to deal with the problems of the robustness tuning strategy, the multivariable filter and the active rejection disturbance internal model control by LQG, FRD, SVD etc. The research contents are as follows:1. Design method of robust IMC-PID controller based on LQG tuning strategy is proposed to deal with the problem of small margin of dynamic performance for PID controller. Meanwhile, the proposed method can hold the problem that the dynamic performance of PID controller is largely affected by the gain variation. The core concept of the proposed control strategy is that the relationship between maximum sensitivity function Ms and IMC-PID controller and the relationship between maximum sensitivity function Ms and linear quadratic Gaussian are established, respectively. Simulation result shows the feasibility and availability of the the proposed control strategy, which fulfill the stable condition and the robustness condition for the control system.2. A novel multivariable IMC controller design method based on double-decomposition (FRD and SVD) is proposed to deal with the problem that the Pseudo-Moore matrix of the controlled plant is difficult to be calculated in the nonsquare system. Control strategy of multivariable IMC controller design method based on double-decomposition can availably overcome the model error of calculating the Pseudo-Moore matrix and improve the system robustness. In the design process, the reflexive generalised inverse matrix (RGIM) substitutes for the Pseudo-Moore matrix. And then, RGIM is used to design controller and reduce the algorithm complexity. The proposed method is successfully used into Woodberry model and Shell model, respectively. Simulation result shows the feasibility and availability of the the proposed control strategy.3. A novel two-degree of freedom control structure, named active disturbance rejection internal model control (ADRIMC), is proposed to fulfill that IMC control system has the ability of the active disturbance rejection. The method of robustness/performance tradeoff tuning is used to tune the controller parameters. And then, the stable condition and the robustness condition are analysed for ADRIMC control system. ADRIMC control strategy can weaken the dependence of the control system to the controlled plant, overcome the model error and improve the system robustness. However, it is known from the analyzing simulation result that the improvement of the anti-interference ability is at the cost of losing the partial dynamic performance in the ADRIMC control strategy. In other words, ADRIMC control strategy shows a tradeoff between the dynamic performance and the system robustness.4. In the multivariable control system, the decoupled model can not be directly used to design controller. Thus, the decoupled model needs to be reduced order. However, the reduce order error is generated in the reduced order process. The reduce order error can affect the system stability and weaken down the system robustness. Consequently, ADRIMC control strategy is brought in the design controller process of the multivariable system. The reduce order error is treeted as the inner disturbance by a two-degree of freedom structure to fulfill the active disturebance ability. The proposed control strategy is applied to the Alatiqi system and the OR system. Simulation result shows that the propsed method can availably deal with MIMO syste and possess a good disturbance rejection performance.5. A control system of complex chemical engineering process, named intelligent grey box-internal model control (IGB-IMC), is proposed to hold the tedious design process in the conventional control method. The purpose of the proposed control system can simplify the design process of the controller and satisfy the engineering requirement. IGB-IMC conrol system focuses on the concept of the internal model control strategy and combines the identification module with the control module. Thus, IGB-IMC conrol system consists of the identification module and the control module; and then, the control module consists of the model classifier module and the parameter tuning module. The advantages of the proposed control system have the simple structure and the easy and convenient operation. Simulation result on Jacketed stirred tank heate system (JSTH) shows that the the feasibility and availability of IGB-IMC conrol system.