针对工程实际中所存在的测试仪器本身的饱和限幅影响 ,提出所谓“类随机电报”问题 ,并对测试仪器不同饱和深度所造成的“类随机电报”过程的自相关函数进行了计算机仿真研究 ;本文以类随机电报为背景 ,成功地提取了弱周期性故障信号 ;文章最后给出结论 :当振动噪声较强而受测试仪器饱和特性一定程度的限幅时 ,仍然可以采用循环自相关分析方法提取淹没于其中的周期性微弱故障信号 ,从而实现往复运动机械装置的故障早期诊断 In view of the influence of saturation limit of test instrument itself in engineering practice, the so-called “quasi-random telegram” problem is proposed and the autocorrelation function of “quasi-random telegraph” caused by different saturation depths of the test instrument is simulated by computer In this paper, a class of stochastic telegraph is used as the background to extract the weak periodic fault signal successfully. Finally, the conclusion is drawn that the cyclic autocorrelation analysis can still be used when the vibration noise is strong and limited to a certain extent by the saturation characteristics of the test instrument Method to extract the periodic weak fault signal submerged therein, so as to realize the early fault diagnosis of reciprocating mechanical devices