摘要
为实现钢铁企业冷轧煤气消耗量的精准预测,提出了一种基于小波阈值和BP神经网络(BPNN)的预测模型。利用小波阈值对煤气消耗数据进行去噪预处理,筛选出合理数据,采用BP神经网络预测煤气消耗量。实验结果表明,与其他方法相比,小波阈值和BPNN模型的预测精度更高,为钢铁企业煤气合理调度、减少排放及提高能源利用率提供有力支撑。
A prediction model based on wavelet threshold and BP neural network(BPNN)is proposed for the accurate predic⁃tion of cold rolling gas consumption in iron and steel enterprises.The wavelet threshold is used to denoise the gas consumption data,and the reasonable data are selected.The BP neural network is used to predict the gas consumption.The experimental results show that the wavelet threshold and BPNN model has higher prediction accuracy,compared with other methods,which provides strong support for the rational scheduling of gas,the reduction of emission and the improvement of energy efficiency in iron and steel enter⁃prises.
作者
熊伟
张建喜
张守印
张宏健
Xiong Wei;Zhang Jianxi;Zhang Shouyin;Zhang Hongjian(College of Electrical Engineering,North China University of Science and Technology,Tangshan 063210;College of Mining Engineering,North China University of Science and Technology,Tangshan 063210)
出处
《现代计算机》
2023年第3期108-110,共3页
Modern Computer
关键词
冷轧煤气消耗
小波阈值
BP神经网络
预测模型
cold rolling gas consumption
wavelet threshold
BP neural network
prediction model