文献类型：期刊文献

中文题名：旋转机械振动故障相似性系数的优化挖掘方法

英文题名：Optimization of Rotating Machinery Vibration Fault Similarity Coefficient Mining Method

作者：陈晖

机构：[1]贵州理工学院信息工程学院

第一机构：贵州理工学院电气与信息工程学院

年份：2016

卷号：32

期号：4

起止页码：126-129

中文期刊名：科技通报

外文期刊名：Bulletin of Science and Technology

收录：CSTPCD;;北大核心:【北大核心2014】；

语种：中文

中文关键词：故障数据挖掘;相似性系数;粒子群算法

外文关键词：fault data mining;;similarity coefficient;;particle swarm algorithm

摘要：针对旋转机械振动故障数据分类挖掘的效率极低、误差率大的问题。为此,提出基于相似性系数与粒子群算法融合的旋转机械振动故障数据优化挖掘方法。以故障数据之间的差异性为依据,对故障数据进行中心化、无量化及标准化处理,以此保证故障数据变量的统一性,为故障数据挖掘提供便利;依据相似系数理论,构建异常旋转机械振动故障数据库挖掘的数学模型,并采用粒子群算法对该模型进行求解,计算旋转机械振动故障数据库挖掘模型的最优解,实现并行数据库故障数据精确挖掘。实验结果表明,采用改进算法进行旋转机械振动故障数据优化挖掘,能够提高挖掘的速度与精度,提高算法鲁棒性,满足了机械振动故障数据库实际的应用需求。
For rotating machinery vibration fault data classification mining efficiency is extremely low, the error rate of big problems. To this end, and particle swarm optimization (pso) algorithm is proposed based on similarity coefficient of rotating machine vibration fault data fusion optimization of mining method. Based on differences between failure data, the fault data centralized, no quantitative and standardized treatment, assure the unity of the failure data variables, provide convenience for the fault data mining;Abnormal based on the theory of similarity coefficient, build the mathematical model of rotating machinery vibration fault database mining, and USES the particle swarm algorithm to solve the model, calculation of rotating machinery vibration fault database mining model's optimal solution, realize the parallel database data mining. The experimental results show that the improved algorithm for rotating machinery vibration fault data optimization of mining, can improve the speed and precision of the mining, improve the robustness of the algorithm, satisfying the actual application requirements of mechanical vibration fault database.