基于HP(2)-Elman模型的钢铁企业富余煤气预测及优化调度

针对钢铁企业富余煤气的频繁波动对自备电厂能耗及煤气平衡影响严重,且难以通过建立机制模型进行预测的问题,依据HP滤波和Elman神经网络性质建立了HP(2)-Elman预测模型.并根据自备电厂能源利用的特点,建立拟合模型求解锅炉的经济运行负荷,在此基础上对富余煤气进行优化调度.模型应用表明:所建预测模型对煤气柜位预测平均相对误差小于2.8％,自备电厂煤气供入量30、45、60个点预测平均相对误差分别为1.7％、1.6％、1.6％.根据预测结果进行的优化调度可为煤气柜位调整及自备电厂锅炉负荷分配提供操作依据,一年按照330天计算,可多产蒸汽约100495t,节能约11670481kg标煤.

An HP(2)-Elman Model for Prediction and Scheduling on Affluent Gas in Steel Enterprises

Aiming at serious effect of the affluent gas fluctuate frequent on the power plant energy consumption and gas balance in iron and steel industry, and the difficult in prediction through using the mechanism modeling, an HP(2)-Elman model combined with the properties of HP filter, Elman neural network was established. The economic operation of the boiler load was calculated through the fitting model on basis of the characteristics of self-provided power plant energy utilization, and optimal scheduling was carried out. The simulation results show that the model of gas holder level forecast average relative error values are under 2. 8%, the supply of self-provided power plant 30, 45, 60 points gas forecast average relative error are 1. 7%, 1. 6%, 1. 6%. Then, a remarkable operating guidance can be provided for reasonable scheduling between gas holder level and the power plant boilers load distribution, and 11670481kg standard coal can be saved and more 100495t steam can be produced for 330 days per year.