基于CARMA模型多变量系统的j步输出预估  

CARMA-model-based j-step-ahead Prediction for MIMO Systems

作　　者：程轶平[1] 

机构地区：[1]北京交通大学电子信息工程学院,北京100044

出　　处：《控制与决策》2006年第9期1050-1053,1058,共5页Control and Decision

基　　金：国家自然科学基金项目(60404010)

摘　　要：用于广义预测控制(GPC)的单变量j步预估器传统上是在多项式域采用丢番图方程导出的,针对与其等价的状态空间形式预估器,对该预估器的多变量扩展进行了z域分析,得到一个多项式域的基于CARM A模型的多变量j步预估器.该预估器可以简化多变量系统的GPC算法设计,在单变量情形,该预估器即为传统的GPC预估器,最后提供了传统GPC预估器与状态空间形式预估器等价性的严格证明.The SISO j-step-ahead predictor for generalized predictive control （GPC） is traditionally derived using the polynomial approach through Diophantine equations. An equivalent version of the predictor in state-space form is available in the literature. A z-domain analysis of the MIMO extension of the state-space predictor is carried out, then an MIMO j-step-ahead predictor in polynomial form based on the CARMA model is derived. The predictor can simplify GPC algorithm design for multivariable systems. In the SISO case the predictor is just the traditional GPC predictor. A rigorous proof of the equivalence between the traditional GPC predictor and the state-space predictor is proposed.

关 键 词：广义预测控制 多变量控制 CARMA模型 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]