基于平移窗信号自相关分析的振动稳定性研究

Vibrational Stability Based on Autocorrelation Analysis of Translating Window Signals

刘广孚 周凯迪 朱赫 冯连雷 姚爱军

LIUGuangfu ZHOUKaidi ZHUHe FENGLianlei YAOAijun

中国石油大学(华东)信息与控制工程学院, 山东 青岛 266580 贵州航天电器股份有限公司, 贵州 贵阳 550009 中国石化胜利油田东辛采油厂, 山东 东营 257000

College of Information and Control Engineering, China University of Petroleum(East China), Qingdao, Shandong 266580, China Guizhou Aerospace Electronics Co. Ltd., Guiyang, Guizhou 550009, China Dongxin Oil Production Plant, Shengli Oilfiled, SINOPEC, Dongying, Shandong 257000, China

针对目前潜油电泵的振动信号稳定性分析方法存在的不足，提出了基于平移窗信号自相关分析方法，并定义了稳定指数用于定量评价振动信号的稳定性。首先利用滑动叠加平均法获得一个能够最大程度地代表振动信号周期性特征的窗信号；然后将该窗信号在振动信号上平移并进行自相关分析，从而获得自相关系数序列；最后对自相关系数序列进行处理得到稳定指数。仿真及实验结果表明，本方法可有效地评判信号是否存在振动幅度不稳定或振动周期不稳定的现象，并成功地检验出具有潜在故障的电泵，验证了该方法在评价振动信号稳定性方面的可行性和有效性。

Targeting the limitations of the current analysis methods for vibration signals of electric submersible pumps, an autocorrelation analysis method based on translating window signals is proposed, and a stability index is defined to evaluate the stability of the vibration signals quantitatively. First, a window signal that can best represent the periodic characteristics of the vibration signals is obtained using the moving average method. Subsequently, the window signal is translated on the vibration signals while autocorrelation analysis is performed to obtain the autocorrelation coefficient series. Finally, the autocorrelation coefficient series is processed to obtain the stability index. Simulation and experiment results show that this method can effectively judge whether the signal has an unstable vibration amplitude or period to determine the electric pump with potential fault successfully, which verifies the feasibility and effectiveness of the method in evaluating the stability of vibration signals.

10.11885/j.issn.1674-5086.2018.03.26.02