论文部分内容阅读
针对输入执行机构故障及输出测量装置故障往往导致MPC(model predictive control)控制器无法实现控制目标的问题,通过对输入稳态与输出稳态关系的分析,提出将存在故障的输入或者输出从控制器的操作变量和被控输出中去除、改变控制器结构的变结构预测控制方法.由于输入故障变结构控制减少了控制器操作变量自由度导致输出稳态误差很大,故根据输出变量优先级重新计算输出设定点以保障重要输出优先满足控制要求.输出故障变结构控制采用结合输入变量稳态值目标跟踪的DMC(dynamic matrix control)算法,避免了输出传感器故障对系统的影响并且保障了被控输出的控制目标可达.利用Shell benchmark重油分馏塔模型仿真验证了本方法的有效性.
Aiming at the problems that the input actuator failure and the output measuring device fault often lead to the MPC controller unable to achieve the control target, the paper analyzes the relationship between the input steady state and the output steady state, and proposes to put the faulty input or output from the control The manipulator variable and the controlled output to remove and change the structure of the variable structure predictive control method.As the variable structure of the input fault control to reduce the degree of freedom of the controller operating variables lead to a large output steady-state error, it is based on the output variable priority Recalculate the output set point to ensure that the important output will be given priority to meet the control requirements Output Fault Variable Structure Control (DMC) algorithm based on steady-state target tracking with input variables avoids the influence of output sensor fault on the system and guarantees The control target of controlled output is reachable.The effectiveness of this method is verified by the simulation of the Shell benchmark heavy oil fractionation tower model.