Fuzzy Logic Combined Logistic Regression Methodology for Gas Turbine First-Stage Nozzle Life Prediction

Ji Hong Yan; Nobuhiro Isobe; Jay Lee

doi:10.4028/www.scientific.net/AMM.10-12.583

Paper Titles

The Design of Software Platform of Data Transformation and Integration Based on J2EE Technology
p.563

A Study of Cutting State Monitoring System Based on Virtual Instrument
p.568

A Hybrid Feature Selection Based on Ant Colony Optimization and Probabilistic Neural Networks for Bearing Fault Diagnostics
p.573

Depth Control Method of AFM-Based Nanomachining with Diamond Tip in Deflection Mode
p.578

Fuzzy Logic Combined Logistic Regression Methodology for Gas Turbine First-Stage Nozzle Life Prediction
p.583

A New Method with Supersonic Focalize Decomposition and Chemical Flocculate Floating-Air to Process the Dyeing and Printing Wastewater
p.588

Research and Realization of Auto-Calibration Technique Based on Structured Grating 3D Measuring System
p.594

Study on Substrate Temperature Field during its Swing Movement in Miniature HFCVD System
p.599

Research on Wear Behavior of Tools while Infeed Cutting Austempered Ductile Iron (ADI) with Ceramic Tools
p.605

HomeApplied Mechanics and MaterialsApplied Mechanics and Materials Vols. 10-12Fuzzy Logic Combined Logistic Regression...

Fuzzy Logic Combined Logistic Regression Methodology for Gas Turbine First-Stage Nozzle Life Prediction

Abstract:

Significant aspects of intelligent maintenance include the abilities to diagnose impending failures, prognose the remaining useful lifetime of the process and schedule maintenance operations so that uptime is maximized. Prognosis is probably the most difficult of the three issues leading to total intelligent maintenance. This paper describes a fuzzy logic combined logistic regression method of fatigue severity assessment and remaining useful life prediction of gas turbine hot components. Logistic regression method is proposed to derive fuzzy logic rule base using historical maintenance running records and engineers’ experience. Implementation of the prognostic methodology presents a great opportunity to significantly enhance current engine health monitoring capabilities and risk management practices.