粒子群优化
方位(导航)
峰度
振动
主成分分析
小波
熵(时间箭头)
模式识别(心理学)
能量(信号处理)
最小二乘支持向量机
控制理论(社会学)
工程类
支持向量机
人工智能
计算机科学
数学
统计
算法
物理
量子力学
控制(管理)
作者
Chao Lu,Jie Chen,Rongjing Hong,Yang Feng,Yuanyuan Li
标识
DOI:10.1016/j.ymssp.2016.02.031
摘要
A novel prediction method is proposed based on least squares support vector machine (LSSVM) to estimate the slewing bearing׳s degradation trend with small sample data. This method chooses the vibration signal which contains rich state information as the object of the study. Principal component analysis (PCA) was applied to fuse multi-feature vectors which could reflect the health state of slewing bearing, such as root mean square, kurtosis, wavelet energy entropy, and intrinsic mode function (IMF) energy. The degradation indicator fused by PCA can reflect the degradation more comprehensively and effectively. Then the degradation trend of slewing bearing was predicted by using the LSSVM model optimized by particle swarm optimization (PSO). The proposed method was demonstrated to be more accurate and effective by the whole life experiment of slewing bearing. Therefore, it can be applied in engineering practice.
科研通智能强力驱动
Strongly Powered by AbleSci AI