基于多模型集成的铁矿粉库存量预测方法

来源期刊:中南大学学报(自然科学版)2011年第11期

论文作者:蔡雁 吴敏 王绍丽 王春生

文章页码:3399 - 3407

关键词:库存量预测;GM(1, 1)模型;ARIMA模型;集成模型

Key words:inventories prediction; gray model GM(1, 1); auto regressive integrated moving average model (ARIMA(p, d, q)); integrated model

摘    要:针对铁矿粉库存量预测问题,结合灰色系统模型与时间序列模型的优点,提出一种基于多模型集成的库存量集成预测方法。根据库存量历史数据,分别建立基于残差修正的等维新息GM(1, 1)模型与自回归积分移动平均模型ARIMA(p, d, q);采用基于信息熵的方法对2种模型进行加权集成;分别采用单一模型与集成模型对铁矿粉库存量进行预测。仿真验证结果表明:集成预测模型实现库存量的准确预测,在3种模型中预测结果最好。

Abstract: To realize effective prediction for iron mine powder, a multi-model based integrated prediction strategy was presented, which makes use of gray system model and time series model. According to the historical records on iron mine powder, two independent prediction models, the residual based equal dimension new information GM(1, 1), and auto regressive integrated moving average model (ARIMA(p, d, q)), were constructed respectively. Applying information entropy method, both models were weighted integrated to realize multi-model based integrated prediction. Finally, the integrated prediction model and both single prediction models were tested to predict iron mine powder. The comparison results show that the integrated prediction model is of higher precision than the other two.

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号