Intelligent Data-Driven Diagnosis of Incipient Interturn Short Circuit Fault in Field Winding of Salient Pole Synchronous Generators

Abstract

This paper examines if machine learning (ML) and signal processing can be used for on-line condition monitoring to reveal inter-turn short circuit fault (ITSC) in the field winding of salient pole synchronous generators (SPSG). This was done by creating several ML classifiers to detect ITSC faults. A data set for ML was built using power spectral density of the air gap magnetic field extracted by fast Fourier transform (FFT), discrete wavelet transform energies, and time series feature extraction based on scalable hypothesis tests (TSFRESH) to extract features from measurements of SPSG operated under several different severities of ITSC fault. Using this data set, a wide range of classifiers were trained to detect the presence of ITSC faults. The classifiers evaluated were logistic regression, K-nearest neighbours, radial basis function support vector machine (SVM), linear SVM, XGBoost decision tree forest, multi-layer perceptron (MLP), and a stacking ensemble classifier including all of the aforementioned. The classifiers were optimised using hyperparameter grid searches. In addition, some feature selection and reduction algorithms were assessed such as random forest feature selection, TSFRESH feature selection, and principal component analysis. This resulted in a classifier capable of detecting 84.5% of samples containing ITSC fault, with a 92.7% chance that fault detections are correct.

Opinions on power grid infrastructure planning with investment risks

“…Infrastructure project schedules can be characterized by investment and construction completion rate curves. Hence, Tsfresh and feature filtering methods are used to identify vital sequential features affecting power grid infrastructure planning from investment and construction completion rate curves (Ehya et al, 2022). Based on hypothesis testing, the feature filtering method is applied to evaluate the correlation between the project duration T and the sequential feature x extracted using the Tsfresh method.…”

Non-Contact Supervision of COVID-19 Breathing Behaviour With FMCW Radar and Stacked Ensemble Learning Model in Real-Time

“…Furthermore, the research of [90] using MFCC FE demonstrated that an MLP as a meta-classifier improved the overall performance of the models. In addition, [91] also used MLP as one of their meta-classifier. Thus, we employed an MLP as our metaclassifier.…”

A Feature-Extraction-Based Lightweight Convolutional and Recurrent Neural Networks Adaptive Computing Model for Container Terminal Liner Handling Volume Forecasting

“…For one thing, the feature extraction of CTO-LGC running log is implemented and performed by a Python package that is time series feature extraction on the basis of scalable hypothesis tests named as tsfresh [65], and the extracted features with tsfresh has been used for multiple types of tasks, for instance, classification, compression, forecasting, detection, recognition, and diagnosis [66][67][68][69][70]. For PSAC, the tsfresh package plays an important role in the deep learning data preparation process.…”