Intelligent Scheme for Partial Discharge Signals Denoising in Gas-Insulated Equipment via Improved Zebra Optimization

Accurately detecting partial discharge (PD) signals released from gas-insulated equipment is crucial to identifying potential insulation faults. However, current PD denoising methods that lack the capability to adaptively adjust model parameters based on signal characteristics, thus, limiting their generalization ability and accuracy in complex test environments. To address this issue, this article proposes a parameter-adaptive PD denoising strategy based on an intelligent optimization algorithm framework. The strategy involves constructing a PD denoising program that combines signal decomposition, entropy criteria, and filtering analysis. To enhance the applicability and ability to handle signals with diverse properties, an adaptive strategy based on an improved zebra optimization algorithm (IZOA) framework is proposed. In this regard, the simulation and experimental results demonstrate the effectiveness of the method in denoising composite signals with complex components, by which to promote the accurate detection of PD.