改进的广义预测控制及其在温度系统中的应用

Improved generalized predictive control and its application in temperature system

温度控制在很多生产过程中与生产效率、产品质量和能耗等重大技术经济指标相关联;针对目前温度控制系统中存在大滞后、大惯性以及时变等难题,提出一种新型的广义预测控制算法(AM-GPC),该算法采用基于辅助模型的递推最小二乘法对带积分的受控自回归滑动平均模型(CARIMA)进行高精度参数估计,并将其应用于温度控制系统中。仿真结果表明提出的AM-GPC算法较传统的广义预测控制算法具有参数估计收敛快、自适应能力强以及跟踪能力强等特点。

In many production processes,temperature control is associated with production efficiency,product quality,energy consumption and other major technical and economic indexes directly.Aiming at the problems of large delay,great inertia and time-varying in temperature control system,this paper presents a new auxiliary model recursive least squares generalized predictive control （AM-GPC） algorithm.The algorithm uses the auxiliary model based recursive least Squares method to conduct high precision parameter estimation for the controlled autoregressive integrated moving average （CARIMA） model; and the algorithm was applied to a temperature control system.Simulation results show that the AM-GPC algorithm proposed in this paper is better than traditional GPC algorithm and has the advantages of fast convergence of parameter estimation,strong adaptive ability,strong tracking capability and etc.