太阳能热发电系统多模型加权预测控制研究  

Research on multi model weighted predictive control of solar thermal power generation system

作　　者：丁坤 孙亚璐 杨昌海 陈博洋 DING Kong;SUN Ya-lu;YANG Chang-hai;CHEN Bo-yang(Economic and Technological Research Institute,Gansu Electric Power Company,Lanzhou 730000,Gansu,China)

机构地区：[1]国网甘肃省电力公司经济技术研究院,甘肃兰州730000

出　　处：《西北师范大学学报（自然科学版）》2023年第6期43-49,共7页Journal of Northwest Normal University(Natural Science)

基　　金：国网甘肃省电力公司科技项目(522730230005)。

摘　　要：针对太阳能热发电过程中广泛存在的随机性和不确定性问题,文中采集实测的数据进行聚类后建立多模型,设计了多模型加权预测控制器.首先,采用模糊聚类算法对实测的太阳能热发电数据进行分类,再用递推最小二乘法建立系统的多模型作为预测模型;其次,取集热器入口温度、太阳辐照强度以及环境温度为扰动信号,以导热熔盐的流速为控制量控制集热器的出口温度,并针对不同的子模型设计预测控制器;第三,按照子控制器的加权策略得到最小的控制增量并对系统进行稳定性分析;最后,进行试验仿真分析,和目前多采用的单模型预测控制结果比较,加权多模型预测控制精度更高,滞后时间更短.In view of the randomness and uncertainty of the solar thermal power generation process,the measured actual data are collected to build a multi model after the clustering in this paper,and weighted multi model predictive controller is designed.Firstly,the fuzzy clustering algorithm is used to classify the measured data,and then the recursive least square method is used to establish the multi model of the system as the predictive model.Secondly,the measured collector input temperature,solar radiation and ambient temperature are taken as the disturbance signal,the flow of heat conductive and the output temperature are taken as the control variable and output,and predictive controllers are designed for different sub models.Thirdly,the optimal control increment is obtained according to the weighting strategy of the sub controller in order to obtain satisfactory results,the stability of the designed system is analyzed.Finally,simulation analysis is carried out,the results show that the weighted multi model predictive model(MMPC)has higher precision and shorter lag time than the single model predictive control(SMPC).

关 键 词：太阳能热发电 多模型预测控制 聚类多模型 线性菲涅尔 

分 类 号：TM727[电气工程—电力系统及自动化]