非平稳工况下滚动轴承局部故障诊断方法研究
发布时间：2019-05-25 04:39
【摘要】：旋转机械广泛应用于工业生产中的重要设备,滚动轴承作为旋转机械的重要组成部件之一,其运转是否稳定与整个机械系统的安全运行有着密切的联系。然而,滚动轴承在其运行过程中往往会产生疲劳剥落等故障。故障早期信号比较微弱,通常淹没在背景噪声之中。同时,设备由于升降速等非平稳工况而引起的转速波动造成频率估计不准,这些问题严重制约着轴承故障的精准诊断。滚动轴承振源信号的分离方法和非平稳信号瞬时频率估计方法的研究,将有助于准确提取故障特征,进而实现对轴承故障准确的诊断,这对旋转机械保持良好的工作状态具有重要的意义。本文主要研究的内容有:(1)研究了滚动轴承故障信号特征的分离方法。通过EEMD削弱背景噪声;针对轴承组件传递路径对故障特征分离时的干扰现象,利用MED消除其不利影响,运用Teager-Kaiser能量算子解调分析方法从无关载波分量中提取出故障特征成分,根据信号频谱特征判断滚动轴承的故障类型。通过实验数据进行验证,结果表明该方法能有效地降低无关成分干扰并分离出故障特征。(2)对改进的NCT滚动轴承阶次跟踪方法进行了研究。分析了旋转机器非平稳信号的处理手段,阐明了将非平稳信号平稳化处理的关键和要点,通过NCT方法将非平稳信号刻画在时频面上,利用峰值搜索法估计出滚动轴承的瞬时频率,采用最小二乘拟合方法对该频率进行拟合,设定合适的阈值,通过循环估计得到恰当的瞬时频率;将得到的瞬时频率经过计算得到鉴相时标,以此时标为采样时刻,实现信号的重采样。通过仿真实验,验证了该方法无需转速计即可获得鉴相时标,其结果具有较高的精度。(3)研究了强噪声、变转速工况下滚动轴承故障诊断方法。提出了基于EEMD和改进型NCT强干扰下的非平稳信号处理技术。通过EEMD和MED联合降噪,实现对故障信号的有效分离;对分离后的信号进行基于改进型NCT的瞬时频率估计,获得鉴相时标,利用能量算子对信号进行解调分析;通过估计瞬时频率计算出相应鉴相时标,利用鉴相时标对解调信号进行阶比分析,根据幅频谱判断出故障类型。实验结果表明,在变转速工况下,该方法能够有效的进行滚动轴承故障诊断。
[Abstract]:Rotating machinery is widely used in important equipment in industrial production. As one of the important components of rotating machinery, the stability of rolling bearing is closely related to the safe operation of the whole mechanical system. However, fatigue spalling and other faults often occur in the running process of rolling bearings. The early signal of the fault is weak and usually submerged in the background noise. At the same time, the frequency estimation of the equipment caused by the fluctuation of rotating speed caused by non-stationary working conditions such as lifting speed is not accurate, which seriously restricts the accurate diagnosis of bearing faults. The research on the separation method of rolling bearing vibration source signal and the instantaneous frequency estimation method of non-stationary signal will be helpful to accurately extract the fault features and realize the accurate diagnosis of bearing fault. This is of great significance to keep the rotating machinery in good working condition. The main contents of this paper are as follows: (1) the separation method of fault signal characteristics of rolling bearing is studied. The background noise is weakened by EEMD. Aiming at the interference phenomenon of bearing assembly transmission path to fault feature separation, MED is used to eliminate its adverse effects, and Teager-Kaiser energy operator demodulation analysis method is used to extract fault feature components from independent carrier components. According to the signal spectrum characteristics, the fault type of rolling bearing is judged. The experimental results show that the method can effectively reduce the interference of independent components and separate the fault characteristics. (2) the improved NCT rolling bearing order tracking method is studied. This paper analyzes the processing method of non-stationary signal of rotating machine, expounds the key and key points of leveling processing of non-stationary signal, and describes the non-stationary signal on time-frequency surface by NCT method. The instantaneous frequency of rolling bearing is estimated by peak search method, the frequency is fitted by least square fitting method, the appropriate threshold is set, and the appropriate instantaneous frequency is obtained by cyclic estimation. The instantaneous frequency is calculated to obtain the phase discrimination time scale, which is used as the sampling time to realize the resampling of the signal. The simulation results show that the method can obtain phase discrimination time scale without rotating speed meter, and the results have high accuracy. (3) the fault diagnosis method of rolling bearing under strong noise and variable speed is studied. A non-stationary signal processing technology based on EEMD and improved NCT strong interference is proposed. Through the joint noise reduction of EEMD and MED, the fault signal is separated effectively, the instantaneous frequency of the separated signal is estimated based on the improved NCT, the phase discrimination time scale is obtained, and the signal is Demodulated and analyzed by energy operator. By estimating the instantaneous frequency, the corresponding phase discrimination time scale is calculated, and the order analysis of the demodulation signal is carried out by using the phase discrimination time scale, and the fault type is judged according to the amplitude and spectrum. The experimental results show that the method can effectively diagnose the fault of rolling bearings under the condition of variable speed.
【学位授予单位】：兰州理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TH133.33
[Abstract]:Rotating machinery is widely used in important equipment in industrial production. As one of the important components of rotating machinery, the stability of rolling bearing is closely related to the safe operation of the whole mechanical system. However, fatigue spalling and other faults often occur in the running process of rolling bearings. The early signal of the fault is weak and usually submerged in the background noise. At the same time, the frequency estimation of the equipment caused by the fluctuation of rotating speed caused by non-stationary working conditions such as lifting speed is not accurate, which seriously restricts the accurate diagnosis of bearing faults. The research on the separation method of rolling bearing vibration source signal and the instantaneous frequency estimation method of non-stationary signal will be helpful to accurately extract the fault features and realize the accurate diagnosis of bearing fault. This is of great significance to keep the rotating machinery in good working condition. The main contents of this paper are as follows: (1) the separation method of fault signal characteristics of rolling bearing is studied. The background noise is weakened by EEMD. Aiming at the interference phenomenon of bearing assembly transmission path to fault feature separation, MED is used to eliminate its adverse effects, and Teager-Kaiser energy operator demodulation analysis method is used to extract fault feature components from independent carrier components. According to the signal spectrum characteristics, the fault type of rolling bearing is judged. The experimental results show that the method can effectively reduce the interference of independent components and separate the fault characteristics. (2) the improved NCT rolling bearing order tracking method is studied. This paper analyzes the processing method of non-stationary signal of rotating machine, expounds the key and key points of leveling processing of non-stationary signal, and describes the non-stationary signal on time-frequency surface by NCT method. The instantaneous frequency of rolling bearing is estimated by peak search method, the frequency is fitted by least square fitting method, the appropriate threshold is set, and the appropriate instantaneous frequency is obtained by cyclic estimation. The instantaneous frequency is calculated to obtain the phase discrimination time scale, which is used as the sampling time to realize the resampling of the signal. The simulation results show that the method can obtain phase discrimination time scale without rotating speed meter, and the results have high accuracy. (3) the fault diagnosis method of rolling bearing under strong noise and variable speed is studied. A non-stationary signal processing technology based on EEMD and improved NCT strong interference is proposed. Through the joint noise reduction of EEMD and MED, the fault signal is separated effectively, the instantaneous frequency of the separated signal is estimated based on the improved NCT, the phase discrimination time scale is obtained, and the signal is Demodulated and analyzed by energy operator. By estimating the instantaneous frequency, the corresponding phase discrimination time scale is calculated, and the order analysis of the demodulation signal is carried out by using the phase discrimination time scale, and the fault type is judged according to the amplitude and spectrum. The experimental results show that the method can effectively diagnose the fault of rolling bearings under the condition of variable speed.
【学位授予单位】：兰州理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TH133.33
【相似文献】
相关博士学位论文 前4条
1 李永波;滚动轴承故障特征提取与早期诊断方法研究[D];哈尔滨工业大学;2017年
2 王晓龙;基于振动信号处理的滚动轴承故障诊断方法研究[D];华北电力大学(北京);2017年
3 刘尚坤;基于振动信号处理的旋转机械故障诊断方法研究[D];华北电力大学(北京);2017年
4 路振勇;航空发动机转子系统的动力学建模及非线性振动研究[D];哈尔滨工业大学;2017年
相关硕士学位论文 前10条
1 张学程;非平稳工况下滚动轴承局部故障诊断方法研究[D];兰州理工大学;2017年
2 杨治飞;滚动轴承的故障特征提取及剩余寿命预测研究[D];山东理工大学;2017年
3 李_┖，
本文编号：2485498
本文链接：https://www.wllwen.com/jixiegongchenglunwen/2485498.html
