Advanced Spectral Estimators and Neural Network Systems for Engine Health Monitoring

Period of Performance: 04/14/1998 - 10/14/1998


Phase 1 SBIR

Recipient Firm

Orincon Corp.
4770 Eastgate Mall
San Diego, CA 92121
Principal Investigator


The failure of machine components is expected but difficult to predict. Worst-case turbine engine inspection and maintenance procedures are very expensive, and all parts of a given type must be replaced after a fixed number of hours, regardless of the fatigue and wear that they have experienced. A less restrictive high-fatigue schedule can reduce costs, but may result in a larger number of catastrophic failures. By analyzing the vibrational modes and performance parameters of an operating turbine engine, one can determine the extent of degredation due to fatigue without the necessity of dismantling the engine and performing a detailed inspection of its components. This "condition-based" maintenance will result in large cost savings by extending the useful life of mechanical components, extending normal maintenance cycles, and safeguarding against premature component failure. ORINCON proposes to develop a diagnostic system for automated detection and localization of mechanical faults in an operating turbine engine. Our engine health diagnostic (EHD) system uses advanced spectral estimation techniques, model-based condition algorithms, and neural network processing will be integrated to detect mechanical flaws and predict failures. The proposed system will combine operational engine parameters and vibration data from engine accelerometers using ORINCON's advanced spectral estimation techniques, neural net processing, and multisensor fusion. The result of this effort, an automated diagnostic system to detect and localize small flaws in rotating machinery, has broad commercial application. Prevention of catastrophic failure and the significant savings to be realized from on-condition maintenance are benefits that have high value across the spectrum of military and civilian operations involving gas turbine engines and other rotating machinery. There is immediate interest in this technology from GE, a major producer of turbine engines.