基于工况划分的火电机组运行多目标优化

详细信息    查看全文 | 推荐本文 |

英文篇名：Multi-Objective Optimization Based on Division of Working Conditions for Operation of Thermal Power Unit 作者：叶灵芝 ; 贾立 ; 宋鸣程 英文作者：YE Lingzhi;JIA Li;SONG Mingcheng;School of Mechanical and Electrical Engineering and Automation,Shanghai University; 关键词：电站运行优化 ; 工况划分 ; 数据挖掘 ; K-means算法 ; 运行参数 ; 多目标优化 ; 理想点法 ; NSGA-Ⅱ 英文关键词：Power station operation optimization;;Division of working conditions;;Data mining;;K-means algorithm;;Operating parameters;;Multi-objective optimization;;Ideal point method;;NSGA-II 中文刊名：ZDYB 英文刊名：Process Automation Instrumentation 机构：上海大学机电工程与自动化学院; 出版日期：2019-05-20 出版单位：自动化仪表 年：2019 期：v.40;No.453 基金：国家自然科学基金资助项目(61773251);; 上海市科委创新行动计划基金资助项目(16111106300、17511109400) 语种：中文; 页：ZDYB201905005 页数：6 CN：05 ISSN：31-1501/TH 分类号：29-34

摘要

针对火电机组运行优化过程中存在的工况变化大和目标参数最优值冲突的问题,提出了基于工况划分的火电机组运行多目标优化的方法。首先,根据机组的外部约束条件,提出了适用于工况划分的K-means改进算法,通过对初始聚类数K值和初始聚类中心的选择进行优化,得到快速、合理的机组负荷和煤质情况的工况划分结果。进而,采用多目标优化方法协调经济性指标和环保性指标的最优解,通过将NSGA-Ⅱ与理想点法结合,并使用新相似度度量方法,得到性能指标最优的运行参数。以某300 MW机组历史运行数据为试验对象,进行基于工况划分的火电机组运行多目标优化研究。仿真结果表明,所提出的K-means优化算法更快速、高效,得到的划分结果也更合理;划分后的每一工况的多目标优化结果也为实时操作提供了具体的优化运行指导。
        In optimization process of thermal power unit, there is problem that the working conditions change greatly and the target parameter optimal value conflicts. Thus, a multi-objective optimization based on division of working conditions for operation of thermal power plant is proposed. Firstly, according to the external constraints of the generator unit, an improved K-means algorithm suitable for division of working conditions is put forward..By optimizing the selection of initial clusters number K and initial clustering center, the fast and reasonable division result of the unit load and coal quality condition is obtained. Furthermore, the multi-objective optimization method is used to coordinate the optimal solutions between the economic indicators and the environmental protection indicators, to obtain the optimal solution. Through combining the NSGA-II and idea point method, and using new similarity metrics method, the operating parameters with optimal performance indicators are obtained.Taking the historical operation data of a 300 MW unit as the experimental object, the research on multi-objective optimization based on division of working conditions for thermal power unit operation is performed. Simulation results show that the improved K-means algorithm proposed is faster and more efficient; and the division results are more reasonable; the multi-objective optimal result of each working condition also provides specific optimization operation guide for real time operations.

引文

[1] 黄文成.基于数据挖掘技术的火电厂主运行参数优化研究[D].上海:上海大学,2017.
    [2] 孙群丽,谷俊杰,李丽.数据挖掘技术在锅炉中的运用[J].锅炉技术,2014,45(1):18- 21.
    [3] 李敬.火力发电厂实时数据采集与分析[D].南京:东南大学,2015.
    [4] RAO B S,VAISAKH K.Multi- objective adaptive clonal selection algorithm for solving environmental/economic dispatch and OPF problems with load uncertainty[J].International Journal of Electrical Power & Energy Systems,2013,53(1):390- 408.
    [5] 善林,李永森,胡笑旋,等.K- means 算法中的k 值优化问题研究[J].系统工程理论与实践,2006(2):97- 101.
    [6] 李芳.K- Means算法的k值自适应优化方法研究[D].合肥:安徽大学,2015.
    [7] 秦绪华,王秋平,陈志强.基于改进k- means算法的电站最有外部运行工况划分[J].热力发电,2017,46(6):28- 33.
    [8] 邓海,覃华,孙欣.一种优化初始中心的K- means聚类算法[J].计算机技术与发展,2013(11):42- 45.
    [9] 仝雪姣,孟凡荣,王志晓.对K- means初始聚类中心的优化[J].计算机工程与设计,2011,32(8):2721- 2788.
    [10]郑伟,刘达.基于全工况数据挖掘的多目标燃烧优化[J].热力发电,2017,46(6):69- 74.
    [11]陈碧云,韦杏秋,陈绍南,等.基于多种群遗传算法的电力系统多目标优化[J].电力系统及其自动化学报,2015,27(7):24- 29.
    [12]王大振,张丽香,苏银皎.改进理想点法及其限制火电厂污染物排放的应用[J].控制工程,2016,23(3):371- 377.
    [13]LI A D,HE Z,ZHANG Y.Bi- objective variable selection for key quality characteristics selection based on a modified NSGA- II and the ideal point method [J].Computers in Industry,2016.82:95- 103.
    [14]沈跃云,高小涛,章名耀.基于在线运行预测锅炉NOX排放浓度[J].中国电力,2011,44(12):64- 67.