In this article, a hypersensitive multispectral partial discharge (PD) optical sensor array was developed, by which the optical pulses in seven independent bands can be acquired simultaneously. By using this sensor array, the multispectral pulses for three typical PDs in gas insulated system were obtained experimentally and analyzed with phase-based (phase-resolved) and nonphase-based (spectral-ratio-based) multispectral characteristics, respectively. It indicates that the multispectral characteristics produced by a specific discharge defect provide unique spectral signatures in discharge mode as well as stage evolution. Based on the intrinsic relationship between the discharge feature and optical emission spectrum, we adopted the classification algorithms and spectral-ratio-reserved multispectral characteristics to implement pattern recognition as well as assessment on the three typical PDs, which obtained the hit ratios exceeding 91%. In principle, such detection approach also supports the phase-independent PD diagnosis especially for dc power equipment.

A hydropower generator (HPG) is the key equipment for power grid peaking and frequency modulation, whose faults are usually in the form of vibration. Hence, it is of great significance to measure the vibration trend of an HPG which can contribute to achieving advanced management and predictive maintenance, thus improving the stability of the power system and enhancing the economic efficiency. For this purpose, a novel measuring model for the vibrational trend of an HPG based on optimal variational mode decomposition (OVMD) and a least squares support vector machine (LSSVM) improved with chaotic sine cosine algorithm optimization (CSCA) is proposed in this paper. To begin with, the mode number and Lagrange multiplier updating step of the variational mode decomposition (VMD) are determined using the center frequency observation method and the proposed least squares error index (LSEI), thus achieving the OVMD decomposition; after which the non-stationary vibration sequence is decomposed into a set of intrinsic mode functions (IMFs). Then, the inputs and outputs of the LSSVM model for the corresponding IMF are deduced by phase space reconstruction. Subsequently, the LSSVM predictor optimized by the improved sine cosine algorithm (SCA) with the combination of chaotic variables is employed to predict each IMF. Finally, the ultimate measuring results of the original trend are calculated by accumulating all the predicted IMFs. Furthermore, the validity of the proposed method is confirmed by an engineering application as well as comparative analyses.