基于BP-神经网络的发动机装配车间物料需求预测
首发时间:2019-04-29
摘要:汽车发动机装配物料种类繁多、数量大,需求具有非线性、非平稳的特征,依靠传统经验法很难做出合理有效的预测来应对生产,给企业带来不可估量的风险。本文首先从物料信息来源追溯,再结合车间物联网技术对物料数据进行采集上传,生成物料需求任务,同时采用BP神经网络对历史月份数据训练学习,达到要求的误差极限,输入样本预测下月物料需求量和安全库存。最后结合实际问题验证其合理性。
For information in English, please click here
Material Demand Prediction in Engine Assembly Workshop Based on BP-Neural Network
Abstract:Automobile engine assembly materials have many kinds and large quantities, and the demand is non-linear and non-stationary. It is difficult to make reasonable and effective prediction based on traditional empirical method to deal with production, which brings inestimable risks to enterprises. Firstly, this paper traces the material information sources, and then collects and uploads the material data with the Internet of Things technology in the workshop to generate the material demand task. At the same time, BP neural network is used to train and learn the historical monthly data, which achieves the required error limit, and the input sample is used to predict the material demand and safe inventory in the next month. Finally, the rationality of the method is verified by practical problems.
Keywords: Material Requirement Internet of Things BP Neural Network Data Acquisition
基金:
引用
No.****
动态公开评议
共计0人参与
勘误表
基于BP-神经网络的发动机装配车间物料需求预测
评论
全部评论0/1000