In this paper a new method of pattern recognition based on approximate entropy and square demodulation analysis is presented.The core of this new method is to demodulate the vibration signal by square demodulation analysis,then the approximate entropy of the square demodulation signals are computed and compared.The study shows this new method could discriminate the normal and three fault signals distinctly.%提出了一种基于近似熵和平方解调分析的故障特征提取新方法.这种方法的核心是首先使用平方解调分析把调制的振动信号进行解调,然后计算并比较这些经平方解调后的信号的近似熵,从而实现故障的特征提取.研究表明,用这种方法提取信号特征,可以容易地将正常滚动轴承、内圈故障、外圈故障和滚子故障的信号区分.
展开▼