智能科学与技术学报 ›› 2019, Vol. 1 ›› Issue (2): 163-170.doi: 10.11959/j.issn.2096-6652.201925

• 常规论文 • 上一篇    下一篇

基于人工智能技术的火电厂燃煤锅炉智能燃烧优化研究及应用

牛培峰1,马云鹏2(),张欣欣1,胡晓宾1   

  1. 1 燕山大学电气工程学院,河北 秦皇岛 066004
    2 天津商业大学信息工程学院,天津300134
  • 修回日期:2019-06-03 出版日期:2019-06-20 发布日期:2019-09-09
  • 作者简介:牛培峰(1958- ),男,吉林舒兰人,燕山大学教授,主要研究方向为人工智能技术、复杂系统过程控制。|马云鹏(1989- ),男,河北沧州人,博士,天津商业大学讲师,主要研究方向为机器学习、模式识别、复杂系统过程控制。|张欣欣(1991- ),女,河北邢台人,燕山大学博士生,主要研究方向为神经网络的稳定性分析、模式识别、机器学习等。|胡晓宾(1983- ),男,内蒙古乌兰察布人,燕山大学博士生,主要研究方向为人工智能、数据挖掘与建模。
  • 基金资助:
    国家自然科学基金资助项目(61573306)

Research and application on combustion optimization of coal-fired boiler in thermal power plant based on artificial intelligence technology

Peifeng NIU1,Yunpeng MA2(),Xinxin ZHANG1,Xiaobin HU1   

  1. 1 School of Electrical Engineering,Yanshan University,Qinhuangdao 066004,China
    2 School of Information Engineering,Tianjin University of Commerce,Tianjin 300134,China
  • Revised:2019-06-03 Online:2019-06-20 Published:2019-09-09
  • Supported by:
    The National Natural Science Foundation of China(61573306)

摘要:

为了降低火电厂燃煤锅炉的氮氧化合物排放浓度和锅炉煤耗,提出了样本增量量子神经网络和改进型量子蜂群算法。样本增量量子神经网络可以动态建立燃煤锅炉的氮氧化合物排放浓度和锅炉煤耗的综合模型,并且可实现模型滚动优化。基于建立的综合模型,通过应用改进型量子蜂群算法优化一二次风量、燃煤量和各二次风门开度来实现锅炉燃烧优化。基于上述两种方法,开发了一套燃煤锅炉智能燃烧优化软件,并应用于某热电厂330 MW锅炉上,测试结果表明,氮氧化合物排放浓度和锅炉煤耗均有不同程度的降低,说明建模方法和优化算法是有效的。

关键词: 氮氧化合物排放, 锅炉, 神经网络, 量子计算, 蜂群算法

Abstract:

In order to reduce the NOxemission concentration and coal consumption of coal-fired boilers in thermal power plants,the sample increment quantum neural network and an improved quantum bee colony algorithm were proposed.The quantum sample incremental feed-forward neural network can dynamically establish a comprehensive model of Nitrogen and Oxygen emission concentration and boiler coal consumption of coal-fired boiler,and can realize rolling optimization of the model.Based on the established comprehensive model,the optimization of the primary and secondary air volume and coal and the opening degree of each secondary air valve were realized by using the improved quantum bee colony algorithm.Based on the above two methods,a set of intelligent combustion optimization software for coal-fired boiler was developed and applied to the 330 MW boiler of a thermal power plant.The test results show that the Nitrogen and Oxygen emission concentration and the coal consumption of the boiler have been reduced in varying degrees.It is shown that the modeling method and the optimization algorithm are effective.

Key words: NOxemission, boiler, neural network, quantum computation, bee colony algorithm

中图分类号: 

No Suggested Reading articles found!