华南理工大学学报(自然科学版) ›› 2006, Vol. 34 ›› Issue (4): 96-100.

• 动力与电气工程 • 上一篇    下一篇

基于HGA-ANN配风方式的锅炉飞灰含碳量优化

刘定平 陈敏生 马晓茜   

  1. 华南理工大学 电力学院,广东 广州 510640
  • 收稿日期:2005-05-11 出版日期:2006-04-25 发布日期:2006-04-25
  • 通信作者: 刘定平(1965-),男,副教授,主要从事燃烧优化与控制研究 E-mail:liudingping@126.com
  • 作者简介:刘定平(1965-),男,副教授,主要从事燃烧优化与控制研究

HGA-ANN-Based Optimiztion of Unburned Carbon in Fly Ash from Utility Boiler Influenced by Air Distribution M ode

Liu Ding-ping  Chen Min-sheng  Ma Xiao-qian   

  1. College of Electric Power,South China Univ.of Tech.,Guangzhou 510640,Guangdong,China
  • Received:2005-05-11 Online:2006-04-25 Published:2006-04-25
  • Contact: 刘定平(1965-),男,副教授,主要从事燃烧优化与控制研究 E-mail:liudingping@126.com
  • About author:刘定平(1965-),男,副教授,主要从事燃烧优化与控制研究

摘要: 利用人工神经网络(ANN)进行锅炉飞灰含碳量建模,并分析二次风配风方式对飞灰含碳量的敏感性影响,同时采用混合遗传算法HGA与复合形法对运行工况寻优,获得各种工况下二次风开度的优化调整方式.应用某台300MW 机纽的现场试验数据进行仿真计算,结果表明该方法可以指导运行人员进行二次风开度的优化调整,降低飞灰含碳量,同时也解决了锅炉变工况下运行参数基准值的确定问题.

关键词: 飞灰含碳量, 二次风配风, 神经网络, 混合遗传算法, 敏感性分析

Abstract:

The content of the unburned carbon in fly ash from utility boiler is modeled by ANN (Artificial Neural Network),and a sensitivity analysis of the effect of secondary air distribution mode on the unburned carbon is car.fled out.Meanwhile,the hybrid genetic algorithm (HGA)and the compound form method are employed to search for the optimum solution to the neural network model.thus obtaining optimized distribution modes of secondary air adapting to various operation conditions.By the simulation of the in.site experimental results of a 300 MW unit.it is finally concluded that the proposed distribution mode of secondary air is feasible and helps to reduce the content of unburned carbon in fly ash.Moreover,the operation standard values of boiler parameters in different oDeration conditions can be determ ined by the proposed method.

Key words: unburned carbon content, secondary air distribution, neural network, hybrid genetic algorithm, sensi-tivity analysis