论文部分内容阅读
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100 495t,节能约11 670 481kg标煤。
Aiming at the problem that the frequent fluctuations of surplus coal gas in iron and steel enterprises have a serious impact on the energy consumption and gas balance of self-provided power plant and it is difficult to predict by establishing a mechanism model, the HP (2) -Elman prediction model is established based on the properties of HP filter and Elman neural network. According to the characteristics of self-provided power plant, this paper establishes a fitting model to solve the economic operation load of boiler, and on this basis, optimizes the surplus gas. The application of the model shows that the average relative error of the prediction model is less than 2.8% and the average relative error of the gas supply of 30,45 and 60 points is 1.7%, 1.6% and 1.6% respectively. The optimized operation based on the prediction results can provide operation basis for the adjustment of gas counter and boiler load distribution of self-provided power plant. According to 330 days a year, it can produce steam of about 100 495t and energy saving about 11 670 481kg of standard coal.