MUMBAI, India, Jan. 9 -- Intellectual Property India has published a patent application (202531115778 A) filed by Mr. Tabish Shanu; and Dr. Ambarisha Mishra, Purnia, Bihar, on Nov. 23, 2025, for 'transformer fault classification system using wavelet scattering feature extraction and multiclass support vector machine (msvm) technique.'
Inventor(s) include Mr. Tabish Shanu; and Dr. Ambarisha Mishra.
The application for the patent was published on Jan. 9, under issue no. 02/2026.
According to the abstract released by the Intellectual Property India: "The present invention provides an intelligent and highly reliable transformer fault classification system that integrates a wavelet scattering feature extraction framework with a multiclass support vector machine (MSVM) classifier to achieve accurate and noise-resistant fault diagnosis. Electrical current or voltage waveforms obtained from transformer terminals are preprocessed and transformed through a multi-layer wavelet scattering network, which generates hierarchical, translation-invariant, and stable scattering coefficients that capture essential fault-related characteristics. These coefficients are then classified using an optimized MSVM algorithm capable of distinguishing multiple types of transformer faults, including turn-to-turn faults, magnetizing inrush currents, winding deformation, insulation failures, core saturation, and external short circuits. The system operates in real time, integrates seamlessly with existing SCADA and substation communication protocols, and supports predictive maintenance through long-term data storage. By combining advanced multi-resolution signal analysis with robust machine learning techniques, the invention significantly enhances transformer monitoring, improves diagnostic precision, and strengthens the reliability and resilience of modern power systems."
Disclaimer: Curated by HT Syndication.
