Predictive Diagnosis of High-Power Transformer Faults by Networking Vibration Measuring Nodes With Integrated Signal Processing

Abstract

This paper addresses the problem of predictive diagnostic in high-power transformers. Particularly, this paper is focused on their use in uninterruptible power supply systems for safety critical applications, such as railway interlocking signaling installations. With respect to the state of the art, where typically only thermal and electrical faults are monitored, this work proposes a distributed network of measuring nodes where also vibration-based mechanical stress diagnosis is implemented. Mechanical degradation is tracked through vibration measures, using multichannel accelerometers, with sensitivity down to 0.5 mg and local signal processing in the transformed frequency domain, up to 1 kHz. A compact hardware-software implementation of the nodes is also presented. The performances of the diagnostic system are assessed through experimental measurements on real three-phase high-power transformers used in railway applications.