摘要
为了简化约束存在时多变量广义预测控制算法的设计与实现,依据对角CARIMA模型的结构特点,将多输入多输出对象的参数辨识和模型预报问题转化为一系列多输入单输出子对象的参数辨识和模型预报问题.推导了输入输出的约束形式及优化求解过程,简化了多变量对象的参数辨识、模型预报、目标函数和约束条件系数矩阵的计算.在由DCS控制的非线性液位装置上的对比实验结果表明了该方法的有效性.
A constrained mulitivariable generalized predictive control algorithm （CMGPC） for CARIMA model with diagonal matrics A and C is developed to simplify the design and implementation of CMGPC when considering of input and output constraints. The identification and model output prediction problems of the multivariable process with rn inputs and n outputs are transformed into that of a series of subprocesses with m inputs and single output by utilizing the structure characteristic of the diagonal CARIMA model. The input and output constraints and optimization problem are analyzed in detail. This method simplifies system identification, output prediction and the computation of the objective function coefficients and constraint matrics. Comparative experiment results obtained from a nonlinear liquid system controlled by DCS show the effectiveness of the method.
出处
《控制与决策》
EI
CSCD
北大核心
2009年第3期330-334,341,共6页
Control and Decision
基金
国家863计划项目(2007AA04Z162)
辽宁省科学技术计划项目(2006219013)
辽宁省高校创新团队计划项目(2006T089)
