基于初值优化的自适应最稀疏时频分析方法

发布时间：2018-04-19 00:08

本文选题：故障诊断 + 自适应最稀疏时频分析　；参考：《湖南大学学报(自然科学版)》2017年08期

【摘要】：自适应最稀疏时频分析(adaptive and sparsest time-frequency analysis,ASTFA)是一种新的时频分析方法,该方法需要事先确定较为准确的初始值,缺乏自适应性.针对ASTFA存在的问题,提出了基于初值优化的ASTFA方法.该方法使用残余量的能量作为优化目标函数,使用不同的初始值对信号进行分解,当残余量的能量最小时,则认为该初始值为最优初始值.因此,该方法能够自适应地寻找最优的初始值,增加了ASTFA方法的自适应性.采用仿真信号将该方法与原ASTFA方法进行对比,结果表明该方法能自适应地得到更准确的分解结果.对仿真信号和滚动轴承故障数据进行分析,结果表明ASTFA在抑制端点效应和模态混淆、抗噪声性能、提高分量的准确性等方面要优于经验模态分解(empirical mode decomposition,EMD),并能有效应用于滚动轴承故障诊断.
[Abstract]:Adaptive and sparsest time-frequency Analysis (ASTFAA) is a new time-frequency analysis method, which requires accurate initial values in advance and lacks self-adaptability.Aiming at the problem of ASTFA, a ASTFA method based on initial value optimization is proposed.In this method, the residual energy is used as the optimization objective function, and different initial values are used to decompose the signal. When the residual energy is minimum, the initial value is considered to be the optimal initial value.Therefore, the method can find the optimal initial value adaptively and increase the adaptability of the ASTFA method.The simulation signal is used to compare the method with the original ASTFA method. The results show that the method can get more accurate decomposition results adaptively.The simulation signal and rolling bearing fault data are analyzed. The results show that ASTFA can suppress the end-point effect and modal confusion, and can resist noise.It is better to improve the accuracy of components than empirical mode decomposition (EMD) and can be effectively applied to the fault diagnosis of rolling bearings.
【作者单位】：湖南大学汽车车身先进设计制造国家重点实验室;
【基金】：国家科技支撑计划资助项目(2015BAF32B03) 国家自然科学基金资助项目(51375152,51575168) 智能型新能源汽车国家2011协同创新中心、湖南省绿色汽车2011协同创新中心资助项目~~
【分类号】：TH17

【相似文献】