连续加热炉运行状态神经网络融合分析

Neural network fusion analysis of the running states for continuous heating furnace

连续加热炉是一个具有大惯性、纯滞后和分布参数的非线性系统,其温度场分布由温度燃烧控制系统决定,而炉内温度场直接影响被加热钢坯质量、烧损及能耗大小。本文利用加热炉测控系统实时测量数据和炉内钢坯温度预报模型,通过神经网络信息融合方法,对连续加热炉的运行工况进行综合分析,其分析结果对在线优化加热炉运行参数具有重要参考作用。其分析方法对类似的设备和生产线有较普遍的参考价值。

The continuous heating furnace is a n onlinear system with large inertia,net lag and distributed parame-ters,whose temperature field distribution is determined by combustion syste m,but its temperature field is related to the quality,combustion loss of the h eated steel slab and energy consume .Makes a full analysis of the heating furnace running state with application of neural network data fusion method and real time database of the distributed measurement and control system and t he temperature prediction model of h eated slab ,whose analysis results present an important reference to op timize the running parameters of the continuous heating furnace .the method is adapted to the homologous device and production line.