摘要
传统的金属冶炼产品产量预测方法主要通过对历史数据的统计处理,总结数据之间的规律实现产量预测。在预测时考虑的因素过少,导致预测结果误差较大、预测耗时长。针对上述问题,研究金属冶炼中产品产量预测的数学建模方法。使用聚类算法识别金属冶炼过程中的不同工况后,使用灰色预测模型根据不同工况下影响产品产量的指标进行预测。将预测值作为神经网络输入,利用数据样本训练神经网络后,完成产品产量预测模型构建,实现对产品产量的准确预测。对比实验结果表明,应用研究的预测方法的预测精度更高,且预测效率至少提升了65.2%,具有优越性。
The traditional method of output prediction of metal smelting products is mainly based on the statistical processing of historical data,summarizing the rules between the data to achieve output prediction.Too few factors are considered in the prediction,resulting in large prediction error and long prediction time.In view of the above problems,the mathematical modeling method of product output prediction in metal smelting is studied.After using clustering algorithm to identify different working conditions in metal smelting process,the grey prediction model is used to predict the indexes that affect the product output under different working conditions.The predicted value is taken as the input of neural network,and the neural network is trained by data samples to complete the construction of product output prediction model,so as to realize the accurate prediction of product output.The experimental results show that the prediction accuracy of the application research method is higher,and the prediction efficiency is improved by at least 65.2%.
作者
宋继凯
SONG Ji-kai(Tianshui Health School of Gansu Province,Tianshui 741000,China)
出处
《世界有色金属》
2021年第6期11-12,共2页
World Nonferrous Metals
关键词
金属冶炼
产品产量
预测
数学建模
神经网络
metal smelting
product output
prediction
mathematical modeling
neural network