Journal of Hebei University of Water Resources and Electric Engineering

Journal of Hebei University of Water Resources and Electric Engineering ›› 2021, Vol. 31 ›› Issue (4): 60-65.DOI: 10.16046/j.cnki.issn2096-5680.2021.04.010

• Technology Theory and Application • Previous Articles     Next Articles

Research on Demand Forecasting of Rarely—used Spare Parts Based on Deep Artificial Neural Network

ZHOU Zi-qiang1, JIANG Jiu-chao2   

  1. 1.I&C Department, Liaoning Hongyanhe Nuclear Power Co., Ltd., 127 Nanshan Road, Zhongshan District, 116000, Dalian, Liaoning,  China;
    2.Deparentment of Electrical Automation, Hebei University of Water Resources and Electric Engineering, 061001, Cangzhou, Hebei, China
  • Received:2021-07-23 Revised:2021-08-19 Online:2021-12-31 Published:2022-01-06

基于深度神经网络的不常用备件需求预测研究

周自强1, 姜久超2   

  1. 1.辽宁红沿河核电有限公司仪控处,辽宁省大连市中山区南山路127号 116000;
    2.河北水利电力学院  电气自动化系,河北省沧州市黄河西路49号 061001
  • 作者简介:周自强(1987- )，男，河北藁城人，工程师，现主要从事仪表维修及维修策略优化研究。Email:770993079@qq.com
  • 基金资助:
    河北省高校水利自动化与信息化应用技术研发中心科研项目

PDF

Abstract: In order to realize the demand forecast of rarely-used spare parts and solve the problem of resource waste caused by the lack of scientific basis when making purchasing plan of rarely-used spare parts, LSTM and Bi-LSTM neural networks are used to study the demand forecast of rarely-used spare parts. This paper establishes the deep neural networks for forecasting the demand of rarely-used spare parts, including LSTM and Bi-LSTM neural networks. Through the case analysis and the comparison of prediction results, it is proved that the two kinds of deep neural networks can effectively forecast the demand of rarely-used spare parts.

Key words: rarely-used spare parts, deep neural networks, demand forecast, LSTM, Bi-LSTM

摘要：

为了解决采购计划制定时无科学依据而造成资源浪费的问题，采用LSTM和Bi-LSTM神经网络实现对不常用备件需求预测进行研究，建立了不常用备件需求预测的深度神经网络，通过实例分析和对预测结果的对比，证明两种神经网络能够实现对不常用备件需求的有效预测。


关键词: 不常用备件, 深度神经网络, 需求预测, LSTM, Bi-LSTM

CLC Number: