摘要
球团回转窑煤粉供入量的预测对回转窑精准控制和建设智能工厂都至关重要。然而,煤粉供入量与众多因素有关,只依据煤粉的历史数据建立模型进行预测,效果难以满足实际生产需求。因此,建立了基于XGBoost的煤粉消耗量预测模型,对影响煤粉消耗量的众多因素进行筛选,确定影响煤粉消耗量的关键影响因素。以某球团厂为应用案例,开展了煤粉供入量预测研究。煤粉消耗量预测结果的平均相对误差为7.22%,生产状态平稳时的预测平均相对误差为6.1%,生产波动较大时的预测平均相对误差为8.9%,预测结果精度较高。
The prediction of pulverized coal supply in pelletizing rotary kilns is crucial for the precise control of rotary kilns and the construction of smart factories.However,it is difficult to achieve an accurate prediction of coal,which are influenced by various factors,just based on the historical data of coal consumption.Therefore,an XGBoost prediction model is established in this study.Many factors influencing the pulverized coal consumption are screened,and the key factors are determined.A pellet plant located in Liaoning Province,China,is used as a case study of the prediction of the supply of pulverized coal.The results show that the predicted values are in good agreement with the actual data.The mean absolute percentage error(MAPE)of the prediction of pulverized coal consumption is 7.22%for the whole investigated period,while the values of MAPE for smooth and unsmooth periods are 6.1%and 8.9%,respectively.
作者
刘新民
路明
刘旭
刘书含
孙文强
Liu Xinmin;Lu Ming;Liu Xu;Liu Shuhan;Sun Wenqiang(Northeastern University;Dagushan Pellet Plant,Ansteel Group Mining Co.,Ltd.;State Key Laboratory of Eco-Industry,Ministry of Ecology and Environment;Liaoning Engineering Research Center of Process Industry Energy Saving and Low-carbon Technologies)
出处
《冶金能源》
北大核心
2024年第2期30-35,共6页
Energy For Metallurgical Industry
基金
国家自然科学基金资助项目(52334008)。
