【摘要】 将高频率分辨力谱估计技术与优化算法相结合而提出一种新的异步电动机转子故障检测方法。针对两种典型的高频率分辨力谱估计技术——多重信号分类(multiple signalclassification,MUSIC)与旋转不变信号参数估计技术(estimation of signal parameters via rotational invariancetechnique,ESPRIT),应用模拟转子故障的定子电流信号测试其频率分辨力、精度等性能,结果表明:即使对于短时信号,二者仍具高频率分辨力,可以准确地分辨定子电流信号中转子故障特征分量、主频分量之频率;但对其幅值、初相角,仅能提供"粗糙"估计。为此,尝试以优化算法——模拟退火算法(simulated annealing algorithm,SAA)与模式搜索算法(pattern search algorithm,PSA)确定各分量的幅值与初相角。同时,分别对MUSIC与ESPRIT、SAA与PSA做了性能对比,遴选优者并应用于转子故障检测。最后,针对转子断条故障进行实验,结果表明:基于高频率分辨力谱估计技术与优化算法的异步电动机转子故障检测方法有效、可行,即使在负载波动、噪声等干扰严重情况下仍然适用。更多还原

【Abstract】 This paper presented a detection method for rotor fault in induction motors,which was based on high frequency resolution spectrum estimation technique and optimization algorithm.Multiple Signal Classification(MUSIC) and Estimation of Signal Parameters via Rotational Invariance Technique(ESPRIT),which were the typical spectrum estimation techniques with high frequency resolution,had been tested with simulated stator current signal of an induction motor with rotor fault.It shows that even with a short-time measurement data,MUSIC and ESPRIT are both capable of correctly identifying the frequencies of the rotor fault characteristic components but with a low accuracy on the amplitudes and initial phases of those components.Typical optimization algorithms,Simulated Annealing Algorithm(SAA) and Pattern Search Algorithm(PSA) are then used to determine their amplitudes and initial phases and shows satisfactory results.In addition,the overall performance comparison between MUSIC and ESPRIT,along with that between SAA and PSA,had been made to pick out the superior,ESPRIT and PSA,for detecting rotor fault in induction motors.Finally,experiments on an induction motor were conducted to demonstrate the effectiveness of the presented method in detecting rotor fault with short-time measurement data.It proves that the proposed method is a promising choice for rotor fault detection in induction motors operating with small slip and fluctuant load.更多还原