Research on die fault signal of high speed screw cold heading machine based on tolerance matching in time domain and principal value analysis in frequency domain

Cold heading die is one of the key components of high speed screw cold heading machine.As the vulnerable parts, the die should be monitored and fault diagnosed. In this paper, the arithmetic average algorithm based on time domain is adopted to denoise the load signal of the mold, and then the load signal after denoising is processed by time-domain tolerance matching and frequance-domain principal value analysis, so as to identify faults. The simulation results show that the timedomain arithmetic average algorithm can effectively improve the signal-to-noise ratio (SNR), the obvious fault signals can be identified by time-domain tolerance matching, and the fault data close to the standard signal can be further judged by frequency domain principal value analysis. Tolerance matching in time domain and principal value analysis in frequency domain provide an effective means for diagnosing die faults of high speed screw cold heading machine.