High-Precision Partial Discharge Signal Reconstruction and Denoising Supported by Multi-in-One Sensors

This study addresses the limitations of traditional single-sensor systems in partial discharge (PD) signal acquisition and denoising, proposing a high-precision signal reconstruction and denoising method based on a multi-functional sensor. A multifunctional sensor system integrating current, voltage, temperature, and gas concentration is designed, combined with National Instruments PXIe-1073 acquisition cards and NI PXI-5122 synchronization modules to achieve sub-microsecond multichannel synchronized sampling. In terms of signal processing, a time-frequency analysis framework based on wavelet packet transform and adaptive reconstruction is proposed, alongside an EMD-VMD joint adaptive denoising algorithm, enhanced with principal component analysis for feature augmentation. Highfidelity signal restoration is achieved through interpolation enhancement, nonlinear least squares optimization, and multisource information fusion strategies. Experimental results based on field data from a 110 kV substation and laboratory simulations show that the system achieves a signal-to-noise ratio (SNR) of $31.7 \mathrm{~dB}, 74 \%$ higher than that of a single current sensor, with a pulse detection rate of $\mathbf{9 6 . 3 \%}$ and a false alarm rate reduced to $3.2 \%$. The proposed reconstruction method outperforms traditional methods in waveform fidelity and feature correlation, with a processing time reduction of about $30 \%$. The proposed approach significantly improves the synchronization and reconstruction accuracy of PD signals, with good real-time performance and robustness, providing effective technical support for online monitoring and intelligent diagnostics of power equipment insulation faults.