Abstract:At present, aero engine fault diagnosis is mainly based on the steady-state condition at the cruise phase, and the gas path parameters in the entire flight process are not effectively used. At the same time, high quality steady-state monitoring measurements are not always available and as a result the accuracy of diagnosis might be affected. There is a recognized need for real-time performance diagnosis of aero engines operating under transient conditions, which can improve their condition-based maintenance. Recent studies have demonstrated the capability of the sequential model-based diagnostic method to predict accurately and efficiently the degradation of industrial gas turbines under steady-state conditions. Nevertheless, incorporating real-time data for fault detection of aero engines that operate in dynamic conditions is a more challenging task. The primary objective of this study is to investigate the performance of the sequential diagnostic method when it is applied to aero engines that operate under transient conditions while there is a variation in the bypass ratio and the heat soakage effects are taken into consideration. This study provides a novel approach for quantifying component degradation, such as fouling and erosion, by using an adapted version of the sequential diagnostic method. The research presented here confirms that the proposed method could be applied to aero engine fault diagnosis under both steady-state and dynamic conditions in real-time. In addition, the economic impact of engine degradation on fuel cost and payload revenue is evaluated when the engine under investigation is using hydrogen. The proposed method demonstrated promising diagnostic results where the maximum prediction errors for steady state and transient conditions are less than 0.006% and 0.016%, respectively. The comparison of the proposed method to a benchmark diagnostic method revealed a 15% improvement in accuracy which can have great benefit when considering that the cost attributed to degradation can reach up to $702,585 for 6000 flight cycles of a hydrogen powered aircraft fleet. This study provides an opportunity to improve our understanding of aero engine fault diagnosis in order to improve engine reliability, availability, and efficiency by online health monitoring.

Aero engine health monitoring, diagnostics and prognostics for condition-based maintenance: an overview

An Integrated Monitoring, Diagnostics, and Prognostics System for Aero-Engines under Long-Term Performance Deterioration

Performance-based health monitoring, diagnostics and prognostics for condition-based maintenance of gas turbines: A review

A Data-Driven Aero-Engine Degradation Prognostic Strategy

Aircraft Gas Turbine Engine Health Monitoring System by Real Flight Data

Gas Turbine Engine Health Management: Past, Present, and Future Trends

Aircraft on-condition reliability assessment based on data-intensive analytics

Research on Architecture Design of the Aeroengine Health Management System

A data-driven degradation prognostic strategy for aero-engine under various operational conditions

A data-driven approach for health status assessment and remaining useful life prediction of aero-engine

A Data-Driven Approach for Assessing Aero-Engine Health Status

Performance-Based Gas Turbine Health Monitoring, Diagnostics, and Prognostics: A Survey

A Comprehensive Review of Key Technologies for Enhancing the Reliability of Aero-Engines

Aircraft Engine Prognostics Based on Informative Sensor Selection and Adaptive Degradation Modeling with Functional Principal Component Analysis

A dynamic performance diagnostic method applied to hydrogen powered aero engines operating under transient conditions

Explainable Data-Driven Method Combined with Bayesian Filtering for Remaining Useful Lifetime Prediction of Aircraft Engines Using NASA CMAPSS Datasets

Aero Engine Remaining Useful Life Prediction: A Survey

Dynamic Health Monitoring of Aero-Engine Gas-Path System Based on SFA-GMM-BID

Recent trends and challenges in predictive maintenance of aircraft’s engine and hydraulic system

Intelligent Combined Neural Network and Kernel Principal Component Analysis Tool for Engine Health Monitoring Purposes

Neural network approach for a combined performance and mechanical health monitoring of a gas turbine engine