摘要
根据数据采用最小二乘法辨识得到制冷系统多变量传递函数模型,随后对制冷系统设计动态矩阵控制器,使系统蒸发温度和过热度在满足约束条件下达到控制要求;考虑到DMC控制器的控制时域、预测时域、控制量权重矩阵等参数直接影响系统的稳定性和输出跟踪效果,而一般试凑法选取参数不可避免地存在主观性和随机性,因此给出一种基于遗传算法的DMC参数寻优方法,并应用于制冷系统的控制问题中。最后的仿真试验表明了算法的有效性。
A multivariable transfer function model is obtained through least square method based on the experiment data, and then the DMC controller is established for the refrigeration system, which is used to track the superheat and evaporator temperature with reference without violate the constraints. Considering that the parameters, such as predictive horizon, control horizon and control variation weight matrix, they may affect the response of the system output, a genetic algorithm based parameters optimal tuning method is proposed in this paper. And the simulation result in the last part shows the efficiency of the algorithm.
出处
《信息技术》
2014年第8期18-21,共4页
Information Technology
基金
国家自然科学基金项目(61074016)
新世纪优秀人才支持计划(NCET-11-1051)
优秀青年基金(slg10008)
关键词
制冷系统
多变量
动态矩阵控制
遗传算法
refrigeration system
muhivariable
dynamic matrix control
genetic algorithm
