分层结构多模型预测控制在630MW火电机组协调系统中的应用  

Application of Hierarchically—organized Multi-model Predictive Control in 630MW Power Unit Coordination Control System

作　　者：陈钊 乔侨 Chen Zhao

机构地区：[1]国投钦州发电有限公司,广西钦州535000 [2]南京英纳维特自动化科技有限公司,江苏南京210000

出　　处：《工业控制计算机》2020年第10期4-6,10,共4页Industrial Control Computer

摘　　要：为提升630MW火电机组协调系统控制品质,提出了分层结构多模型预测控制。首先由现场阶跃响应试验数据,辨识建立被控对象在不同负荷点的传递函数模型集。为保证计算结果的实时性,同时实现多个子系统的平滑过渡,通过性能指标函数的加权来实现多模型的整合。最后考虑到子模型个数对控制性能的影响,根据机组运行负荷范围、控制偏差及偏差变化率的大小,组合不同层次、不同数量的子模型集,以适应机组运行时的不同控制目标。工程应用表明,分层结构多模型预测控制能有效提升协调控制系统的负荷调节品质,同时保证关键参数的稳定性。In order to improve control quality of 630 MW thermal power unit CCS,the hierarchically-organized multi-model predictive control is designed in this paper.Firstly,on-site step response test data is used to identify and establish the transfer function model set of the controlled object at different load points.To ensure the real-time performance of the calculation results and achieve smooth transition of multiple subsystems at the same time,the integration of multiple models is achieved by weighted sum of GPC performance index function.Finally,considering the influence of sub-model number,different levels with different sub-model sets are applied according to operating load range,control deviation and deviation change rate.

关 键 词：协调控制系统 预测控制 多模型 分层结构 

分 类 号：TM621[电气工程—电力系统及自动化]