《计算机应用研究》|Application Research of Computers

基于灰色粗糙集与BP神经网络的设备故障预测

Equipment fault prediction based on grey rough set and BP neural network

免费全文下载 (已被下载 次)  
获取PDF全文
作者 郭宇,杨育
机构 重庆大学 机械传动国家重点实验室,重庆 400030
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2017)09-2642-04
DOI 10.3969/j.issn.1001-3695.2017.09.017
摘要 为更有效地预测设备故障,提出一种基于灰色粗糙集与BP神经网络的设备故障预测模型。用灰色关联分析和粗糙集理论分别对二维故障决策表进行横向和纵向两个维度的约简,将冗余的数据和属性去掉,并将约简后的数据输入到BP神经网络,预测设备故障。最后以地铁信号设备故障预测为例进行实例验证,结果表明该模型预测误差更小,预测准确率更高。
关键词 灰色关联分析;粗糙集;BP神经网络;约简;故障预测
基金项目 国家自然科学基金资助项目(71571023)
本文URL http://www.arocmag.com/article/01-2017-09-017.html
英文标题 Equipment fault prediction based on grey rough set and BP neural network
作者英文名 Guo Yu, Yang Yu
机构英文名 StateKeyLaboratoryofMechanicalTransmission,ChongqingUniversity,Chongqing400030,China
英文摘要 In order to predict equipment failure more effectively, this paper proposed a model of equipment fault prediction based on the grey rough set and BP neural network. By use of grey incidence analysis and rough set theory, it reduced a two-dimensional fault decision table from both horizontal and vertical dimensions, and removed the redundant data and attributes of the decision table, after the reduction, input the data to the BP neural network to predict the equipment failure. Finally, it carried out a case study on the fault prediction of subway signal equipment, and the results show that the model has smaller prediction error and higher accuracy.
英文关键词 grey incidence analysis; rough set; BP neural network; reduction; fault prediction
参考文献 查看稿件参考文献
  [1] 艾红, 周东华. 动态系统的故障预测方法[J] . 华中科技大学学报:自然科学版, 2009, 37(S1):222-225.
[2] 褚青青, 肖涵, 吕勇, 等. 基于多重分形理论与神经网络的齿轮故障诊断[J] . 振动与冲击, 2015, 34(21):15-18.
[3] Sun Wenjun, Shao Siyu, Zhao Rui, et al. A sparse auto-encoder-based deep neural network approach for induction motor faults classification[J] . Measurement, 2016, 89:171-178.
[4] 孟宗, 胡猛, 谷伟明, 等. 基于LMD多尺度熵和概率神经网络的滚动轴承故障诊断方法[J] . 中国机械工程, 2016, 27(4):433-437. [5] 赵劲松, 李元, 邱彤. 一种基于小波变换与神经网络的传感器故障诊断方法[J] . 清华大学学报:自然科学版, 2013, 53(2):205-209, 221.
[6] Jahromi A T, Er M J, Li Xiang, et al. Sequential fuzzy clustering based dynamic fuzzy neural network for fault diagnosis and prognosis[J] . Neurocomputing, 2016, 196:31-41.
[7] Rigatos G, Siano P. Power transformers’ condition monitoring using neural modeling and the local statistical approach to fault diagnosis[J] . International Journal of Electrical Power & Energy Systems, 2016, 80:150-159.
[8] Wang Yingmin, Zhang Fujun, Cui Tao, et al. Fault diagnosis for manifold absolute pressure sensor(MAP) of diesel engine based on Elman neural network observer[J] . Chinese Journal of Mechanical Engineering, 2016, 29(2):386-395.
[9] 徐贵斌, 周东华. 基于在线学习神经网络的状态依赖型故障预测[J] . 浙江大学学报:工学版, 2010, 44(7):1251-1254, 1320.
[10] Bilski P. Data set preprocessing methods for the artificial intelligence-based diagnostic module[J] . Measurement, 2014, 54:180-190.
[11] Li Xinli, Yao Wanye, Qingjie, et al. Fault diagnosis of wind turbine based on rough set and BP network[C] //Proc of the 3rd International Conference on Mechatronics, Robotics and Automation. Paris:Atlantis Press, 2015:877-883.
[12] Yang Qiujing. Study on computer network application layer fault diagnosis based on RSNN:advanced materials research[C] //Advanced Materials Research. Zurich:Trans Tech Publications Ltd, 2014:1423-1426.
[13] 刘思峰, 蔡华, 杨英杰. 灰色关联分析模型研究进展[J] . 系统工程理论与实践, 2013, 33(8):2041-2046.
[14] 余亮, 边馥苓. 粗糙神经网络在森林火灾预警中的应用[J] . 武汉大学学报:信息科学版, 2006, 31(8):720-723.
[15] 侯智, 余忠华. 采用主成分分析—信息熵法评价人机系统方案[J] . 机械设计与研究, 2009, 25(6):15-17, 21.
[16] 马宗杰, 刘华文. 基于奇异值分解—偏最小二乘回归的多标签分类算法[J] . 计算机应用, 2014, 34(7):2058-2060, 2089.
[17] 于洪, 王国胤, 姚一豫. 决策粗糙集理论研究现状与展望[J] . 计算机学报, 2015, 38(8):1628-1639.
[18] 徐章艳, 刘作鹏, 杨炳儒, 等. 一个复杂度为max(O(|C||U|), O(|C|2|U/C|))的快速属性约简算法[J] . 计算机学报, 2006, 29(3):391-399.
[19] 苏航. 基于BP神经网络的地铁信号设备故障预测[D] . 广州:华南理工大学, 2013.
收稿日期 2016/6/21
修回日期 2016/8/5
页码 2642-2645
中图分类号 TP206.3
文献标志码 A