铸造技术2018，Vol.39Issue(5)：1028-1032,5.DOI:10.16410/j.issn1000-8365.2018.05.021
基于深度置信网络的高炉炉况故障分类方法的研究
Research on Fault Classification of Blast Furnace Condition Based on Deep Belief Network
摘要
Abstract
With the continuous development of intelligent manufacturing and computer technology, the modern industrial systems had tended to be complicated and intelligent, which caused the original industrial system fault diagnosis methods to meet the bottleneck. In recent years, deep learning method shown the distinctive advantages and potential in terms of feature extraction and pattern recognition. According to the complex characteristics of the blast furnace smelting system, and combining with the advantages of deep learning in dealing with complex distribution data and extracting features, a kind of blast furnace condition classification method were proposed based on the deep belief network model, the actual sample data were analyzed. The results show that this method is suitable for the classification of blast furnace condition, and it has strong feature extraction ability and fault tolerance characteristic, and it has better performance than BP neural network and support vector machine method.关键词
深度学习/高炉炉况/深度置信网络/特征提取/分类Key words
deep learning/blast furnace condition/DBN/feature extraction/classification分类
信息技术与安全科学引用本文复制引用
赵辉,赵德涛,岳有军,王红君..基于深度置信网络的高炉炉况故障分类方法的研究[J].铸造技术,2018,39(5):1028-1032,5.基金项目
天津市科技支撑计划重点项目: 基于全流程优化控制与系统节能思想的钢铁企业先进能源管理系统 (2013ZCZDGX03800) （2013ZCZDGX03800）
