Modeling And Control Simulation Of Mixture Making Ingredient System With Long Time Delay

How to control mixture making ingredient system is researched in this paper. This system is characteristic of long time delay from the angle of process flow, which makes it difficult to control this system. At present, manpower controls mixture making ingredient in the field, which leads to lower the level of automation, increase the labor intensity of operators and reduce the productivity. An optimum control algorithm must be found for this system to improve the situation.First, the domestic and overseas research actuality of long time delay control is presented basically and the source of the project and the main research content in the paper is set forth. The process flow of mixture making ingredient system is described in detail. Design scheme is brought forward in view of control objective. Because of the character of long time delay and actual situation, the forecast control is adopted, namely process is identified not including the time-lag part and the output of the gained model is controlled. Modeling is very important. Several methods of parameter estimation and the initialization method of RLS are introduced. RLS is used to model by analyzing the identification methods. After obtained the model of this system, control algorithms are presented. The type, comparison, analysis, formula and the controller design of self-tuning control are introduced with the origin, actuality, development direction, merits and the design of fuzzy control. The principle and implementation of identification-free algorithm for adaptive control are also presented. Finally, which algorithm is the optimum for the system is made certain by analyzing the simulation outcome of the algorithms referred in this paper. Self-tuning control algorithm can meet control objective not only the process is constant parameter but also with disturb from the angle of the response curves, so it is adopted in the paper.After self-tuning control algorithm is simulated so many times, it is testified that the output can track quickly the set value and the overshoot is small. The effect is in good working order.