一种基于MPC结构评价基准的预测控制器  

Predictive Controller Using MPC Structure Evaluation Benchmark

作　　者：周会[1] 苏成利[1] 李平[1] 孙小平[1] 

机构地区：[1]辽宁石油化工大学信息与控制工程学院,辽宁抚顺113001

出　　处：《控制工程》2015年第5期927-932,共6页Control Engineering of China

基　　金：国家自然科学基金(61203021);辽宁省科技攻关项目(2011216011)

摘　　要：针对线性高斯二次型(LQG)基准在经济性能评估算法中不可达问题,提出基于模型预测控制(MPC)结构评价基准的经济性能评估算法。该算法利用扰动简化的过程模型设计控制器,真实的过程模型用来计算带有输入输出方差约束的MPC性能曲线,并以此建立经济性能评估优化命题,在对优化结果分析讨论的基础上得到经济性能评价结果,有效的避免了因模型差异带来的计算误差,仿真结果表明了该算法更具可达性。As linear quadratic Gaussian （LQG）-based benchmark used in the economic performance evaluation algorithm is inaccessible, a new method based on model prediction control （MPC） structure evaluation benchmark is proposed for economic performance assessment. The algorithm uses the process model with simplified disturbance to design controller while the real process model is used to calculate the MPC performance curve with input and output variance constraints, based on which the economic performance evaluation optimization problem is built, and economic performance evaluation results are gotten based on the analysis of the optimization results. The proposed algorithm effectively avoids the calculation error caused by model, the simulation results show that the algorithm is more accessible.

关 键 词：经济性能评估 MPC 结构评价基准 模型预测控制 LQG 基准 

分 类 号：TP29[自动化与计算机技术—检测技术与自动化装置]