基于改进型BP网络的染纱能耗软测量

Soft-sensing for Dyeing Machine Energy Consumption Based on Improved BP Neural-network

染纱生产过程是一个典型的间歇生产过程,具有非线性、时变性、输入和输出量都较多等特点,难以建立准确的数学模型。同时,在线测量仪表价格昂贵。将改进型BP神经网络算法应用于车间能耗的测量,以车间蒸汽能耗为例建立车间能耗BP神经网络软测量模型。仿真结果表明,建立的软测量模型能实现蒸汽能耗的实时测量和估计,并为后续进行车间能源管理提供了良好的基础。

Dyeing process is a typical batch production process, which has the characteristics of nonlinearity and time-varying, It is difficult to build accurate models by traditional methods, and on-line measurement instrument is very expensive. The improved BP neural-network algorithm was applied in the. measurement for energy consumption of workshop, a relative BP neural-network soft sensing model of steam consumption was established. The simulation results show that the soft sensing model can achieve good realtime measurement and estimation, which provides a good foundation for follow-up workshop energy management.