基于数据驱动与模式挖掘的循环流化床锅炉运行优化研究  

Operation optimization of CFB boiler based on data-driven and pattern mining

作　　者：贾继宁 吕勇哉[1] JIA Jining;LYU Yongzai(State Key Laboratory of Industrial Control Technology,Institute of Cyber-Systems and Control,Zhejiang University,Hangzhou 310027,China;College of Chemical Engineering,Zhejiang University of Technology,Hangzhou 310014,China)

机构地区：[1]浙江大学工业控制技术国家重点实验室智能系统与控制研究所,浙江杭州310027 [2]浙江工业大学化学工程学院,浙江杭州310014

出　　处：《热力发电》2020年第5期33-39,共7页Thermal Power Generation

基　　金：国家自然科学基金项目资助(51878614);浙江省教育厅科研项目资助(Y201326724)。

摘　　要：火电厂锅炉运行控制中存在可控运行参数多、机理建模困难、运行参数和设计值有偏离等问题,导致采用数据驱动的锅炉运行优化较为困难。为此,本文采用模式挖掘方法分析锅炉相关监控点分散控制系统(DCS)历史运行数据,构造合理的模糊划分矩阵,并进行矩阵运算,挖掘负荷变化下与锅炉高效运行匹配的模糊关联模式,从而指导锅炉运行优化。以东莞某热电厂130 t/h循环流化床(CFB)锅炉为研究对象,提取其DCS历史数据,采用本文方法对数据进行分析,挖掘出运行中隐含的控制优化策略。优化结果表明,该方法能够为锅炉经济运行提供快速合理的支持。In operation control of thermal power plant boilers,there are many problems,such as various controllable operation parameters,difficult mechanism modeling and deviation between operation parameters and design values,which make it difficult to optimize the operation of boilers driven by data.Thus,a pattern mining method is applied to analyze the historical operation data of the distributed control system(DCS)of the related boiler monitoring points.Reasonable fuzzy partition matrices are constructed and matrix operations are performed to mine the fuzzy association patterns that match the boiler's high-efficiency operation under load changes,thereby to guide the operation optimization of boiler.By taking the 130 t/h circulating fluidized bed boiler of a thermal power plant in Dongguan as the research object,the historical data of its DCS are extracted.The data are analyzed by the method of this paper and the control optimization strategy implied in operation is mined.The optimization results show that,this method can provide rapid and reasonable support for the economic operation of the boiler.

关 键 词：锅炉效率 CFB锅炉 控制优化 运行经济性 数据驱动 模式挖掘 

分 类 号：TK229.6[动力工程及工程热物理—动力机械及工程]