摘要
利用BP神经网络模型来预测上海高校"十二五"期间的能源消耗水平。在这一模型中,将上海市GDP、城市居民可支配收入、高校学生人数、高校建筑面积、高校空调面积、高校科研经费,以及1999年-2011年上海市高校的总能耗作为输入层,输出层为2012年-2022年上海高校总能耗。首先采用1999年-2008年的样本数据对神经网络进行训练,接着采用2009年-2011年的样本数据对神经网络进行仿真,最后利用训练好的神经网络预测2012年-2022年上海市高校的总能耗。经过训练后的网络对于输入信号的仿真误差为0.007 774 7,表明BP神经网络模型可有效地预测上海高校的能耗水平。
This study is to predict energy needs and determine the future level of energy consumption of the universities at Shanghai by using BP artificial neural networks. In this model, Shanghai’s GDP, dis-posable income of urban residents, the number of college students, the floor area of college, the air condi-tioned area of college, scientific research funding, the energy consumption of Shanghai’s universities are used in the input layer of the network. Data from 1999 to 2008 are used for the training. Three years （2009, 2010 and 2011） are used only as test data to confirm this method. According to the results, the net energy consumption using the ANN technique has been predicted with acceptable accuracy. It is expected that BP artificial neural networks model is effective for predicting the energy consumption level of Shang-hai’s universities.
出处
《建筑节能》
CAS
2015年第1期92-97,共6页
BUILDING ENERGY EFFICIENCY
关键词
BP神经网络模型
上海高校
能耗预测
Back- propagation（BP） artificial neural networks
universities of Shanghai
predict of ener-gy consumption
