发电技术 ›› 2023, Vol. 44 ›› Issue (6): 809-816.DOI: 10.12096/j.2096-4528.pgt.22108

• 发电及环境保护 • 上一篇    下一篇

基于最小二乘支持向量机的电站锅炉高效率低NO x 的多目标优化研究

梁中荣1, 蓝茂蔚2,3, 郑国1, 何荣强1, 屈可扬3, 甘云华3   

  1. 1.湛江电力有限公司,广东省 湛江市 524099
    2.中国能源建设集团广东省电力设计研究院有限公司, 广东省 广州市 510663
    3.华南理工大学电力学院,广东省 广州市 510640
  • 收稿日期:2023-02-23 出版日期:2023-12-31 发布日期:2023-12-28
  • 通讯作者: 甘云华
  • 作者简介:梁中荣(1974),男,高级工程师,研究方向为火电机组智能控制优化技术研究与应用,13560538632@126.com
    甘云华(1979),男,博士,教授,研究方向为新型燃烧、高效传热、能源系统智能优化等,本文通信作者,ganyh@scut.edu.cn
  • 基金资助:
    国家自然科学基金项目(52376108);广东省基础与应用基础研究基金项目(2020B1515020040)

Study on Multi-Objective Optimization of High-Efficiency and Low-NO x Emissions of Power Station Boilers Based on Least Squares Support Vector Machines

Zhongrong LIANG1, Maowei LAN2,3, Guo ZHENG1, Rongqiang HE1, Keyang QU3, Yunhua GAN3   

  1. 1.Zhanjiang Electric Power Co. , Ltd. , Zhanjiang 524099, Guangdong Province, China
    2.China Energy Engineering Group Guangdong Electric Power Design Institute Co. , Ltd. , Guangzhou 510663, Guangdong Province, China
    3.School of Electric Power Engineering, South China University of Technology, Guangzhou 510640, Guangdong Province, China
  • Received:2023-02-23 Published:2023-12-31 Online:2023-12-28
  • Contact: Yunhua GAN
  • Supported by:
    National Natural Science Foundation of China(52376108);Basic and Applied Basic Research Foundation of Guangdong Province(2020B1515020040)

摘要:

针对锅炉燃烧系统的多目标优化,在所建立的锅炉燃烧系统预测模型的基础上,分别采用加权-粒子群算法和多目标粒子群优化(multi-objective particle swarm optimization,MOPSO)算法优化锅炉系统的可调整运行参数,以实现锅炉高效率低NO x 排放。分析表明,2种优化算法所得的运行参数相近,趋势与燃烧特性分析和燃烧调整试验结果相符合,说明智能算法优化电站锅炉燃烧系统有效可行。但是加权-粒子群优化算法主观依赖性严重,难以选取合适的权值,优化时间长且结果少;而MOPSO算法优化时间远远小于加权-粒子群算法优化时间,并且优化结果更多,优化效率更高,更有利于指导锅炉的实际运行。

关键词: 电站锅炉, 多目标优化, 加权-粒子群算法, 多目标粒子群优化 (MOPSO)

Abstract:

Aiming at the multi-objective optimization of boiler combustion system, on the basis of the established prediction model of boiler combustion system, the weighted-particle swarm algorithm and the multi-objective particle swarm optimization (MOPSO) algorithm were used to optimize the adjustable operating parameters of the boiler, which can realize the operating state of the boiler with high efficiency and low NOx emission. The analysis shows that the operating parameters obtained by the two optimization algorithms are similar, and the trend is consistent with the combustion characteristics analysis and combustion adjustment test results. It indicates that the intelligent algorithm is effective and feasible to optimize the combustion system of the power plant boiler. However, the weighted-particle swarm optimization algorithm has serious subjective dependence. It is difficult to select appropriate weights, and the optimization time is long and the results are few. However, the optimization time of the MOPSO algorithm is far less than the optimization time of the weighted-particle swarm optimization algorithm, the optimization results are more, and the optimization efficiency is higher. Therefore, the MOPSO algorithm is more beneficial to guide the actual operation of the boiler.

Key words: power station boiler, multi-objective optimization, weighted-particle swarm optimization, multi-objective particle swarm optimization (MOPSO)

中图分类号: