IEEE Transactions on Power Delivery · 2001 · 188 citations · 8 references
ComputingComputer ScienceElectrical EngineeringArtificial Intelligence In EducationComputer VisionMachine LearningCondition MonitoringDiagnosisDiagnosticsElectrical System DiagnosticsFault DetectionNeural NetworksIncipient Fault DetectionMachine Learning For Fault DetectionNeural Network DiagnosticsNeural Network-based DiagnosisTransformer Oil Diagnostics
The study compares neural network efficiency for detecting early faults in power transformers. Neural networks were trained on five dissolved‑gas‑analysis criteria (Doernenburg, modified Rogers, Rogers, IEC, CSUS), then tested on new DGA data and compared with inspection results. Diagnosis success rates varied by criterion, ranging from 87 % to 100 %.
This paper presents a comparative study of neural network (NN) efficiency for the detection of incipient faults in power transformers. The NN was trained according to five diagnosis criteria commonly used for dissolved gas analysis (DGA) in transformer insulating oil. These criteria are Doernenburg, modified Rogers, Rogers, IEC and CSUS. Once trained, the neural network was tested by using a new set of DGA results. Finally, NN diagnosis results were compared with those obtained by inspection and an analysis. The study shows that NN rate of successful diagnosis is dependant on the criterion under consideration, with values in the range of 87-100%.
8
Neural networks: A comprehensive foundation
Ryszard Tadeusiewicz · Control Engineering Practice · 1995 · 7.3K citations
Kevin L. Moore · IEEE Potentials · 1992 · 319 citations