By Prof. Ir. Dr. Hazlee Azil Illias (Universiti Malaya)
KUALA LUMPUR, 10 June 2026 - The Universiti Malaya High Voltage Research Group (UMHVRG) congratulates Ciptian Weried Priananda for successfully completing his PhD studies in Universiti Malaya with Distinction. His thesis entitled “Detection Transformer with Customized Convolutional Neural Network Backbone and Flattening Augmentation For Multi-Source Partial Discharge Recognition In Rotating Machines” proposes an innovative approach that utilizes Deep Neural Networks (DNNs) to diagnose single-source and multi-source PD faults in rotating machines. The system integrates advanced image processing techniques to extract distinctive phase-resolved partial discharge (PRPD) patterns and employs sophisticated DNN architectures, particularly the Detection Transformer (DETR), to perform accurate detection, localization, and classification of PD sources. PRPD data are obtained from a lab-scale replica of a rotating machine’s insulation system under various controlled defect conditions. These data are then processed into annotated PRPD images for training, with Density-Based Spatial Clustering of Applications with Noise (DBSCAN) clustering used to filter noise and improve data quality. Evaluation results demonstrate that the proposed system delivers high detection accuracy and consistent classification performance, even in challenging scenarios involving multiple overlapping PD sources.
Apart from writing a thesis report, Ciptian also published 3 journal papers in IEEE Transactions on Dielectrics and Electrical Insulation, presented 3 papers in conferences and filed a patent.
Once again, congratulations Ciptian! UMHVRG members wish you all the best in your future endeavors…
Board Of Examiners Meeting for Ciptian Weried Priananda
