模型预测控制在线优化算法评估与选择  

Assessment and Selection of Online Optimization Algorithms for Model Predictive Control

作　　者：夏浩[1] 杨月彩 XIA Hao;YANG Yue-cai(College of Control Science and Engineering,Dalian University of Technology,Dalian 116024,China)

机构地区：[1]大连理工大学控制科学与工程学院

出　　处：《控制工程》2018年第8期1505-1510,共6页Control Engineering of China

基　　金：国家自然科学基金面上项目(61273098)

摘　　要：模型预测控制的应用越来越广泛,其在线优化问题的求解算法也越来越多样,不同的算法具有不同的性能,因此评估算法性能并对给定的模型预测控制问题选择合适的在线优化算法极其重要。针对种类繁多的模型预测控制优化算法,归纳出适当的评估标准,实现对各种算法多方面的性能评价;给出一个基本的算法选择方案,阐明如何依据实际问题选择合适的在线优化算法,并通过多个实例进行具体说明;最后对多种实例问题进行仿真计算,并通过数据实现比较。With the wide applications of model predictive control, a variety of algorithms for solving online quadratic optimization problems have been proposed. As different algorithms have different characteristics, it is important to assess the performance of these algorithms and to select the appropriate online optimization algorithm for a given model predictive control problem. This paper outlines the appropriate assessment criteria so that the performance of various algorithms can be fairly assessed. Then a guideline to select the suitable algorithms for actual model predictive control problems is provided and described in detail. Furthermore, the calculation results from a number of typical model predictive control problems with different optimization algorithms are listed and compared.

关 键 词：模型预测 优化算法 评估 选择 

分 类 号：TP173[自动化与计算机技术—控制理论与控制工程]