基于粒度计算与模糊规则的钢卷仓储吞吐量长期预测  

Long-term prediction of steel coil storage throughput based on particle size calculation and fuzzy rules

作　　者：姬克 黄细霞[1] 卢占标 鲍佳松 JI Ke;HUANG Xixia;LU Zhanbiao;BAO Jiasong(Key Laboratory of Marine Technology and Control Engineering,Shanghai Maritime University,Shanghai 201306,China)

机构地区：[1]上海海事大学航运技术与控制工程交通行业重点实验室,上海201306

出　　处：《传感器与微系统》2021年第5期124-127,共4页Transducer and Microsystem Technologies

基　　金：国家自然科学基金资助项目(61304186)。

摘　　要：为了实现对钢卷仓储吞吐量的长期预测,以便帮助钢铁物流园区提前规划库位分配和装备准备,降低物流成本和物流园区的空置率,提出基于粒度计算和模糊规则的钢卷仓储吞吐量长期预测模型。通过时间序列分解模型将原始时间序列分解,分别将分解后的数据划分为多个数据粒并对其进行聚类,根据信息粒的类别建立模糊逻辑关系。根据模糊规则实现对未来7天的预测并不断迭代实现对四周的吞吐量预测。选用某无水港2014年至2018年的吞吐量数据进行验证,实验结果表明:所提出的预测方法结果能够满足钢铁物流规划需求,长期预测精度高于ARIMA模型。In order to realize long-term prediction of steel coil storage throughput,so as to help steel logistics parks plan storage location allocation and equipment preparation in advance,reduce logistics costs and the vacancy rate of logistics parks,a method is presented to predict the throughput of steel coil storage based on particle size calculation and fuzzy rules.This method decomposes the original time series by time series decomposition model,divides the decomposed data into multiple information granules and clusters them,and establishes fuzzy logic relations according to the classification of information granules.According to the fuzzy rules,the throughput of the next 7 days is forecasted and the throughput of the four weeks is forecasted.The throughput data of a waterless port from 2014 to 2018 are selected for verification.The experimental results show that the results of the proposed prediction method can meet the requirements of steel logistics planning,and the long-term prediction accuracy is higher than that of ARIMA model.

关 键 词：粒度计算 模糊规则 钢卷吞吐量预测 时间序列分析 长期预测 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]