A Real-World Application of Fuzzy Logic and Influence Coefficients for Gas Turbine Performance Diagnostics

Abstract

This paper presents an example of the use of fuzzy logic combined with influence coefficients applied to engine test-cell data to diagnose gas-path related performance faults. The approach utilizes influence coefficients, which describe the changes in measurable parameters due to changes in component condition such as compressor efficiency. Such approaches usually have the disadvantages of attributing measurement noise or sensor errors to changes in engine condition and do not have the ability to diagnose more faults than the number of measurement parameters that exist. These disadvantages usually make such methods impractical for anything but simulated data without measurement noise or errors. However, in this example, the influence coefficients are used in an iterative approach, in combination with fuzzy logic, to overcome these obstacles. The method is demonstrated using eight examples from real-world test-cell data.