N. Aretakis

Turbocharger Unstable Operation Diagnosis Using Vibroacoustic Measurements

The possibility to detect unstable operating condition (stall or surge) of an automotive turbocharger using vibration or acoustic measurements is studied. An experimental study is performed, in order to acquire and analyze test data, to find out whether vibration or acoustic measurements can be correlated to aerothermodynamic operating condition. An instrumentation set allowing the definition of the operating point on the map of the compressor of the turbocharger is used. Hot wires at the compressor inlet serve as flow condition indicators and provide a clear indicator of the presence or not of instabilities, such as rotating stall or surge. Accelerometers are mounted on the casing and microphones are placed in the vicinity of the compressor casing, to measure vibration and sound emission. Data covering an extensive range of the compressor performance map have been collected and analyzed. Signal features from the different measuring instruments are discussed. Using such features, a bi-parametric criterion is established for determination of whether the compressor operates in the stable part of its performance characteristic or in the presence of unstable operation phenomena (rotating stall, surge). The possibility of generalizing the validity of observations is supported, by presenting results from testing a second turbocharger, which is shown to exhibit similar behavior.

Classification of radial compressor faults using pattern-recognition techniques

Abstract An application of pattern-recognition techniques for the classification of faults in a radial compressor is presented. A number of mechanical alterations, simulating faults, are introduced in a test compressor. They include the insertion of an inlet obstruction, an obstruction in a diffuser passage, variation of impeller tip clearance and impeller fouling. Two kinds of measurements, namely sound emission and casing vibration, are examined. Three kinds of pattern-recognition techniques with increasing complexity are used in order to classify the examined faults correctly according to engine condition. The possibility of using each one of these techniques for diagnosing faults in a radial compressor is also examined. It is demonstrated that minor faults, which do not affect performance, can be identified using the proposed techniques.

Wavelet Analysis for Gas Turbine Fault Diagnostics

The application of wavelet analysis to diagnosing faults in gas turbines is examined in the present paper. Applying the wavelet transform to time signals obtained from sensors placed on an engine gives information in correspondence to their Fourier transform. Diagnostic techniques based on Fourier analysis of signals can therefore be transposed to the wavelet analysis. In the paper the basic properties of wavelets, in relation to the nature of turbomachinery signals, are discussed. The possibilities for extracting diagnostic information by means of wavelets are examined, by studying the applicability to existing data from vibration, unsteady pressure, and acoustic measurements. Advantages offered, with respect to existing methods based on harmonic analysis, are discussed as well as particular requirements related to practical application.

GAS TURBINE FAULT DIAGNOSIS FROM FAST RESPONSE DATA USING PROBABILISTIC METHODS AND INFORMATION FUSION

The paper covers firstly the use of probabilistic neural networks for the classification of spectral fault signatures obtained from fast response data (sound, vibration, unsteady pressure). The method is compared to other alternatives, such as geometrical and statistical pattern recognition. The effectiveness of the method is demonstrated by presenting the results from application to data from a radial compressor and an industrial gas turbine. Further, probabilistic methods are used to perform information fusion. The outcomes of different diagnostic methods are used as a first level of diagnostic inference, and are fed to two different fusion processes which are based on i) Probabilistic Neural Networks and ii) Bayesian Belief Networks. It is demonstrated that these fusion processes provide powerful tools for effective fault classification.Copyright

Radial Compressor Fault Identification Using Dynamic Measurement Data

The exploitation of different unsteady quantity measurements for identifying mechanical alterations on a radial compressor with a vaned diffuser is examined in this paper. Measurements of sound emission, casing vibration and unsteady inner wall pressure are performed. The mechanical alterations considered have been chosen in order to reproduce or simulate faults in the compressor. They include the insertion of an inlet obstruction, an obstruction in a diffuser passage, variation of impeller tip clearance, and impeller fouling. Processing these measurement data leads to the derivation of fault signatures which can be utilized for identifying them. The suitability of measuring each of the above physical quantities is discussed with respect to their sensitivity to particular faults. The dependence of the fault signatures on operating point is also examined. It’s demonstrated that minor faults which do not affect compressor operation and are not detectable by performance monitoring, can possibly be detected by the proposed methodology.Copyright

Derivation of signatures for faults in gas turbine compressor blading

Abstract A procedure for the derivation of signatures for faults in the blades of a gas turbine is presented here. A variety of blade faults, corresponding to changes in the angle or spacing of one or more blades, are examined. A fluid dynamic simulation model is used to derive the unsteady pressure signals sensed by a stationary transducer for the cases of faulty blades. The blade fault signatures are derived by processing these signals using Fourier techniques. The features of blade fault signatures are studied. Finally, the possibilities they offer for the discrimination and identification of different possible blade faults are also examined.

Blade Fault Recognition Based on Signal Processing and Adaptive Fluid Dynamic Modelling

An approach for identification of faults in blades of a gas turbine, based on physical modelling is presented. A measured quantity is used as an input and the deformed blading configuration is produced as an output. This is achieved without using any kind of “signature”, as is customary in diagnostic procedures for this kind of faults. A fluid dynamic model is used in a manner similar to what is known as “inverse design methods”: the solid boundaries which produce a certain flow field are calculated by prescribing this flow field. In the present case a signal, corresponding to the pressure variation on the blade-to-blade plane, is measured. The blade cascade geometry that has produced this signal is then produced by the method. In the paper the method is described and applications to test cases are presented. The test cases include theoretically produced faults as well as experimental cases, where actual measurement data are shown to produce the geometrical deformations which existed in the test engine.© 1997 ASME

Computer models for education on performance monitoring and diagnostics of gas turbines

The paper discusses how performance models can be incorporated in education on the subject of gas turbine performance monitoring and diagnostics. A particular performance model, built for educational purposes, is employed to demonstrate the different aspects of this process. The way of building a model is discussed, in order to ensure the connection between the physical principles used for diagnostics and the structure of the software. The first aspect discussed is model usage for understanding gas turbine behaviour under different operating conditions. Understanding this behaviour is essential, in order to have the possibility to distinguish between operation in ‘healthy’ and ‘faulty’ engine condition. A graphics interface is used to present information in different ways such as operating line, operating points on component maps, interrelation between performance variables and parameters. The way of studying faulty engine operation is then presented, featuring a novel comparison to existing simulation programs. Faults can be implanted into different engine components and their impact on engine performance studied. The notion of fault signatures on measured quantities is explained. The model has also a diagnostic capability, allowing the introduction of measurement data from faulty engines and providing a diagnosis, namely a picture of how the performance of engine components has deviated from a ‘healthy’ condition

Classification of Radial Compressor Faults Using Pattern Recognition Techniques

Abstract An application of pattern-recognition techniques for the classification of faults in a radial compressor is presented. A number of mechanical alterations, simulating faults, are introduced in a test compressor. They include the insertion of an inlet obstruction, an obstruction in a diffuser passage, variation of impeller tip clearance and impeller fouling. Two kinds of measurements, namely sound emission and casing vibration, are examined. Three kinds of pattern-recognition techniques with increasing complexity are used in order to classify the examined faults correctly according to engine condition. The possibility of using each one of these techniques for diagnosing faults in a radial compressor is also examined. It is demonstrated that minor faults, which do not affect performance, can be identified using the proposed techniques.

Derivation of Signatures for Faults in Gas Turbine Compressor Blading

Abstract A procedure for derivation of signatures for faults in blades of a gas turbine is presented. A variety of blade faults, corresponding to changes in angle or spacing of one or more blades are examined. A fluid dynamic simulation model is used to derive the unsteady pressure signals sensed by a stationary transducer for the cases of faulty blades. The blade fault signatures are derived by processing these signals using Fourier techniques. The features of blade fault signatures are studied. Finally, the possibilities they offer for discrimination and identification of different possible blade faults is also examined.