摘要
针对旋转机械耦合故障的诊断问题,提出一种基于EMD(Empirical Mode Decomposition)和分形盒维数的诊断方法。该方法结合EMD对非线性信号处理的自适应性和分形盒维数能对非线性行为定量描述的特点,先对故障信号进行EMD处理,得到含有故障特征的本征模式函数(Intrinsic Mode Function,IMF),然后求出各IMF的盒维数,通过盒维数的比较分析进行故障诊断。构造了含有裂纹-碰摩-松动耦合故障的转子-轴承系统动力学模型,用龙格-库塔法求出故障模型振动信号。通过对耦合故障信号进行分析,得到耦合故障特征向量,并与传统的边界谱诊断方法比较,证明该方法对旋转机械耦合故障诊断的有效性和优越性。
Aiming at coupling faults diagnosis for rotary machinery, a diagnosis method based on EMD (empirical mode decomposition) and fractal box dimension was proposed. In the method, taking advantage of the adaptability of EMD in processing nonlinear signal and the characteristic of fractal box dimension that can quantitatively describe nonlinear behaviors, the fault signal was processed by EMD to obtain intrinsic mode functions (IMF) containing fault characteristics, then to solve box dimension of IMFs and to detect fault through comparing and analyzing the box dimension. A dynamic model of the rotor-bearing system was established in order to consider the coupling fault of crack, rub-impact and pedestal looseness. The simulatod vibration signal of the fault model was obtained by using Runge-Kutta method. Through analyzing the coupling fault signal, getting the coupling fault feature vectors and comparing with the results of traditional boundary spectrum diagnosis method, the validity of the method proposed and its benefit were proved.
出处
《振动与冲击》
EI
CSCD
北大核心
2013年第15期209-214,共6页
Journal of Vibration and Shock
基金
国家自然科学基金(51104129)
河北省自然科学基金资助项目(E2011203153)
高等学校博士学科点专项科研基金(20101333120004)
关键词
故障诊断
经验模式分解
分形盒维数
耦合故障
Dynamic random access storage
Failure analysis
Rotating machinery
Runge Kutta methods
Signal processing
