Research On The Application Of Multivariable Predictive Controller In Main Steam Temperature Control System In200MW Unit Of Power Plant

The main steam temperature controlled object in200MW unit of power plant is a multiple inputs and multiple outputs coupling system, which has characteristics of long time-delayed, time-varying and nonlinear in operating conditions. Thus it is difficult to apply the conventional cascade PID control methods to achieve the desired control effect. By analyzing the technology characteristics of the main-steam temperature controlled object which proposes multi-variable model predictive control replacing the conventional cascade PID control method. In order to guarantee the main-steam temperature control system is long-term automatically operated within the specified range. One of the main focuses is to find effective method to deal with the problem of the main-steam temperature large deviation, which is caused by load changes and other unknown factors disturbances. Ultimately, security and economic issues are resolvedThere is one chapter to describe how to design the multivariable model predictive controller in this paper. According to the characteristics of the main steam temperature controlled object in200MW unit of power plant, then a reasonable control scheme which is multivariable model predictive control is proposed. The second stage desuperheater inlet temperature and the main-steam temperature are took as the controlled variables and separately controlled in this paper. The two-stage predictive controllers are used to control the first desuperheater and the second desuperheater separately. Then the appropriate steam temperature settings and regulation parameters are selected to achieve optimal control performance.In order to reflect the advanced of the predictive control, conventional cascade PID control method and the predictive control method are took to do comparison simulation experiments in this paper. The two control schemes are used to control the same object which is the main steam temperature in200MW unit of power plant on the computer simulation platform. By the experimental results show that the maximum deviation and the response speed of the main-steam temperature with predictive control system are better than that with conventional cascade PID control system.