On-board Fault Diagnosis of a Laboratory Mini SR-30 Gas Turbine Engine

TL;DR

This paper develops a machine learning-based fault diagnosis system for a laboratory gas turbine engine, enabling real-time detection of fuel and sensor faults through comparative analysis of classification techniques.

Contribution

It introduces a data-driven, passive fault diagnosis approach using machine learning classifiers specifically tailored for gas turbine fault detection.

Findings

Support vector classifier outperforms other methods

The scheme effectively detects faults in real-time

Simulation results demonstrate high accuracy and reliability

Abstract

Inspired by recent progress in machine learning, a data-driven fault diagnosis and isolation (FDI) scheme is explicitly developed for failure in the fuel supply system and sensor measurements of the laboratory gas turbine system. A passive approach of fault diagnosis is implemented where a model is trained using machine learning classifiers to detect a given set of fault scenarios in real-time on which it is trained. Towards the end, a comparative study is presented for well-known classification techniques, namely Support vector classifier, linear discriminant analysis, K-neighbor, and decision trees. Several simulation studies were carried out to demonstrate and illustrate the proposed fault diagnosis scheme's advantages, capabilities, and performance.