旋转机械振动故障相似性系数的优化挖掘方法

发布时间：2018-01-29 17:25

本文关键词： 故障数据挖掘相似性系数粒子群算法　出处：《科技通报》2016年04期 　论文类型：期刊论文

【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。
[Abstract]:Aiming at the problem of low efficiency and large error rate of classification mining for vibration fault data of rotating machinery. Based on similarity coefficient and particle swarm optimization algorithm, an optimal mining method for vibration fault data of rotating machinery is proposed. Based on the difference of fault data, fault data is centralized. No quantization and standardized processing, so as to ensure the unity of fault data variables, and provide convenience for fault data mining; According to the similarity coefficient theory, the mathematical model of abnormal rotating machinery vibration fault database mining is constructed, and the particle swarm optimization algorithm is used to solve the model. The optimal solution of rotating machinery vibration fault database mining model is calculated, and the accurate mining of parallel database fault data is realized. The experimental results show that the improved algorithm is used to optimize the rotating machinery vibration fault data mining. It can improve the speed and precision of mining, improve the robustness of the algorithm, and meet the practical application needs of mechanical vibration fault database.
【作者单位】：贵州理工学院信息工程学院;
【分类号】：TP311.13;TH17
【正文快照】： 0引言随着数据库技术的发展与普及,数据库的规模不断扩大,对大型旋转机械数据库中的故障数据进行挖掘,可以保证大型旋转机械设备振动监测与故障诊断的准确性。但对于故障数据的分类挖掘实现过程复杂,直接影响了故障数据挖掘效果与正常的应用需求。如何有效的进行故障数据优化

【相似文献】