电站燃煤锅炉NO_x排放神经网络模型  

Artificial Neural Networks Model of NO_x Emissions in a Power Station Coal-fired Boiler

作　　者：许昌[1] 吕剑虹[1] 

机构地区：[1]东南大学动力系

出　　处：《锅炉技术》2005年第2期13-17,共5页Boiler Technology

摘　　要：人工神经网络具有联想、记忆、并行计算、自适应、自学习、适于处理非线性问题等优点。电站燃煤锅炉NOx 排放规律非常复杂,很难对其进行建模。提出电站燃煤锅炉NOx 排放量的神经网络模型,该神经网络模型具有可以预测各一次风粉单元NOx 生成量、锅炉NOx 排放量、网络隐节点数少、泛化能力强、鲁棒性好、学习速度快的优点。所提出的模型可以为大型电站锅炉通过燃烧系统自动调整或结构改造降低 NOx 排放提供依据。Artificial neural networks (ANN) have a lot of advantages, such as association, recollection, parallel calculation, self-adaption, self-learning, and fit to deal with non-linearity problems. NO x emissions law of coal-fired power station boiler is very complex. It is very difficult to establish the model. This paper presents an ANN model of NO x emissions in a coal-fired power station boiler. The model possess many advantages such as predicting NO x emissions of each first wind and coal powder unite, NO x emission of boiler, few hidden nodes, strong generalization, good robust and fast learning. The model presented can provide reference for huge power station boiler to reduce NO x emissions by auto-adjustment or structure reform.

关 键 词：燃煤电站 锅炉 氮氧化物 人工神经网络 

分 类 号：TK223[动力工程及工程热物理—动力机械及工程]