A Frequency Mixing Processing Method for PD Signals Acquired by Multiple Different Type of Sensors

Multi-methods-combined detection of partial discharge (PD) can improve detection sensitivity and efficiency. This paper proposes a frequency mixing processing method for PD signals acquired by different type of sensors. In this method, the multi-band signal is tuned to the unified intermediate frequency (IF), so that the multi-source signal can be processed uniformly and with a low sample rate. This paper focuses on the optimal frequency band selection and detection sensitivity enhancement based on multi-source signals. The augmented simulation (ASim) results show that the selection of center frequency and bandwidth plays a decisive role in signal extraction. In practical detection, the center frequency with high SNR should be chosen first. Then the bandwidth is determined according to the characteristics of the detection system and the noise distribution. Based on the proposed framework, the detection sensitivity can be effectively improved by using multi-source signals.