论文部分内容阅读
采用机械故障诊断学中的时域识别方法，通过对往复式压缩机填料密封气腔内气体压力波动的分析，对其建立自回归模型，根据所建模型的ＡＩＣ指标，从数值上识别填料密封的正常与异常状态；结合对压力波形的频谱分析，判断故障的类型，利用各组填料密封腔中压力的分配情况，判断出故障产生的位置．经实验证明，采用填料密封腔中气体压力信号诊断技术及３种手段（时域识别、频谱分析、压力分配识别）相结合的诊断方法，可以准确判断填料工作的机械故障
Using the method of time domain identification in mechanical fault diagnosis, the autoregressive model was established by analyzing the gas pressure fluctuation in the gas seal of packing compressor seal. According to the AIC index of the model, the packing seal was numerically identified Of the normal and abnormal state; combined with the pressure waveform of the spectrum analysis to determine the type of fault, the use of packing packing chamber pressure distribution, to determine the location of the fault. The experiment proves that the diagnosis method of gas pressure signal in packing seal cavity and the combination of three methods (time domain identification, spectrum analysis and pressure distribution identification) can accurately determine the mechanical failure of packing work