状态空间模型的双层结构预测控制算法

双层结构预测控制是指先进行设定值优化、再进行设定值跟踪的预测控制.在已有的双层结构动态矩阵控制的基础上,本文给出基于状态空间模型的双层结构预测控制算法.该算法基于干扰模型和新定义的开环预测值,给出了新的开环预测模块.该开环预测模块采用Kalman滤波方法得到操作变量、被控变量的开环动、稳态预测值.基于这些开环预测值,稳态目标计算模块的基本原理同双层结构动态矩阵控制,但是具体细节上遵循状态空间方法.动态控制模块基于稳态目标计算提供的操作变量、被控变量的稳态目标(设定值),采用二次规划算法计算控制作用.仿真算例证实了该算法的有效性.

Double-layered model predictive control of state-space model

The so-called double-layered model predictive control (MPC) performs firstly the setpoint optimization, then the setpoint tracking. Based-on the existing double-layered dynamic matrix control, this paper gives an algorithm for double-layered MPC based on the state-space model. Based on the disturbance model and the newly defined open-loop predictions, this algorithm proposes a new open-loop prediction module. This open-loop prediction module adopts the Kalman filter to obtain the open-loop dynamic/steady-state predictions of manipulated/controlled variables (MVs/CVs). Based on these open-loop predictions, the steady-state target calculation (SSTC) module is the same as in double-layered dynamic matrix control, but its details obey the state-space method. Based on the steady-state targets (setpoints) of MVs/CVs provided by SSTC, the dynamic control module computes the control moves by solving the quadratic programming. The numerical example verifies the effectiveness of the proposed algorithm.