应用神经网络方法预测生物质的热值  

Forecasting heating value of biomass with method of artificial neural network

作　　者：矫常命[1] 蔡均猛[2] 何芳[1] 

机构地区：[1]山东理工大学轻工与农业工程学院,山东淄博255049 [2]上海理工大学,上海200093

出　　处：《可再生能源》2006年第2期25-27,共3页Renewable Energy Resources

基　　金：国家"863"项目(2001AA514030;2003AA514030);山东省自然科学基金项目(2004ZX21)

摘　　要：应用人工神经网络方法对生物质的热值进行了预测,网络的训练数据集来自美国Biomass Feedstock Composition and Property Database of U.S.D epartment of Energy。神经网络以生物质的工业分析结果作为输入数据,采用56组数据对网络进行训练,以7组数据对网络进行验证,对网络输出值与实际值进行比较,相对误差在0.08%以内。人工神经网络成功地预测各种生物质的热值,说明人工神经网络能够处理生物质的热值与工业分析各组分间的非线性关系。To forecast the heat value of biomass with method of Artificial Neural Network, the data for training network was got from ＂Biomass Feedstock Composition and Property Database of U.S. Department of Energy＂. The data from the approximate analysis of biomass was regarded as the input of the artificial neural network. The network was trained with 56 group of data and proved with 7 groups. Then the output of the network was compared with the real value, and the relative error was less than 0.08%. The heat value was successful to be forecasted by the artificial neural network, and it shows that artificial neural network can deal with non-linear relation between the heat value of biomass and the composition of its approximate analysis.

关 键 词：生物质 热值 神经网络 

分 类 号：S216.2[农业科学—农业机械化工程] G303[农业科学—农业工程]