摘要
为了对能源需求进行更精确的预测,基于分数阶灰色模型和反向传播神经网络,建立基于误差补偿的分数阶灰色模型。随后利用河北省历年电力消费数据检验模型的精度,并将其与传统分数阶灰色模型作对比。最后对四川省煤炭类能源需求量进行了预测分析。结果表明,基于误差补偿的分数阶灰色模型具有更高的精确度与有效性,反向传播神经网络对分数阶灰色模型的误差补偿作用明显,该模型有效地模拟了原始数据的变化趋势,并预测出煤炭类能源的消费量将呈下降后趋于平稳的趋势。
In order to predict energy demand more accurately, a fractional-order grey model based on error compensation is established based on fractional-order grey model and back-propagation neural network. The accuracy of the model is tested with the power consumption data of Hebei province, and compared with the traditional fractional grey model. The coal energy demand in Sichuan province is forecasted and analyzed. Results show that the fractional order grey model based on the error compensation has higher accuracy and effectiveness, back propagation neural network has an obvious effect on error compensation of the fractional order grey model.The model effectively simulates the change trend of raw data,predicts that the coal type energy consumption will be leveled off after a downward trend.
作者
胡宇
陈兴志
黄子萌
苏铃麟
唐国鑫
郑克龙
Hu Yu;Chen Xingzhi;Huang Zimeng;Su Linglin;Tang Guoxin;Zheng Kelong(College of Science, Southwest University of Science and Technology, Sichuan Mianyang, 621010, China)
出处
《机械设计与制造工程》
2019年第5期88-92,共5页
Machine Design and Manufacturing Engineering
基金
西南科技大学大学生创新基金项目精准资助专项(jz18-048)
四川省科技厅重点研发项目(2017GZ0316)
关键词
反向传播神经网络
误差补偿
分数阶灰色模型
四川省煤炭类能源
平均相对误差绝对值
back propagation neural network
error compensation
fractional order grey model
Sichuan province coal category energy
mean absolute percentage error