乘数型主蒸汽温度预测函数控制研究及应用

Research on and Application of Multiplier Predictive Function Control Over Main Steam Temperature

提出了一种乘数型预测函数控制方法,针对一阶加纯迟延对象,当采用阶跃函数时,可以得出预测函数最优控制律,该控制律中预测时域优化长度为指数系数。提出采用乘数系数的方法去计算最优控制律,将乘数系数称为预测调整系数,将修正后的被控对象输出简化为实际被控对象输出,获得乘数型预测函数控制最优控制律。对于主蒸汽温度控制系统来说,具有导前区和惰性区,常规的做法是进行串级控制,将主控制器改为乘数型预测函数控制器,同时提出一种减温水调节阀线性校正方法,工程实践证明文章所提出的方法简单有效,具有很强的推广应用价值。

This paper proposes a multiplier predictive function control method for the first order plus pure delay object. When the step function is adopted, the optimal control law is obtained. In this control law, the optimal time domain is exponential. Moreover, the paper also proposes multiplier coefficient and calls it predictive adjusting coefficient to simplify the corrected output of the controlled object as the output of the actual controlled object. Thus the optimal control law of multiplier predictive function control is obtained. Generally, the cascade control is adopted in the main steam temperature control system with leading and idle zones, and the main controller is replaced with the multiplier predictive function controller In this paper, a linear correction method for the desuperheater control valve is proposed. The practical applications show that the method presented in this paper is simple and effective, and valuable for popularization.