摘要
人工操作排除垃圾焚烧炉故障对操作员要求较高,且自动化程度低。应用BP神经网络方法,采用matlab软件编程建立垃圾焚烧炉过程控制模型,对垃圾焚烧炉两种典型故障的排除进行研究。在过程控制模型的建立过程中,采用神经网络集成,提高神经网络模型的泛化能力。最后以49组实际工况数据作为检验样本,检验误差率为7.612%和6.429%.检验结果表明神经网络集成可以提高模型的计算精度,该模型可以用于垃圾焚烧炉过程控制,提高设备的自动化程度。
Incineration of municipal solid waste (MSW) by manpower need high qualified workers. And the automation of this method is low. Matlab was used to establish a control of incinerating municipal solid waste model, which was based on a back propagation (BP) neural network. The model was about solving two classical faults of municipal solid waste incinerating. Neural network ensemble was used to improve generalized ability of the artificial neural network during the establishment of the model. At the end of the establishment of the model, 49 group real-time data were chosen for testing as test samples. The error rates of test result are 7. 612 % and 6.429 %. The test result shows neural network ensemble can improve the accuracy of the model. The model was useful in municipal solid waste incinerating process controlled and improved the automation of facility.
出处
《计算机与应用化学》
CAS
CSCD
北大核心
2008年第7期859-862,共4页
Computers and Applied Chemistry
基金
国家863项目子课题(2006AA04Z176).
关键词
BP神经网络
垃圾焚烧
过程控制
神经网络集成
BP neural network, MSW incineration, process control, neural network ensemble