K. Mathioudakis

Adaptive Simulation of Gas Turbine Performance

Greater use is being made of dynamic simulation of energy systems as a design tool for selecting control strategies and establishing operating procedures. This paper discusses the dynamic modeling of a gas-fired combined-cycle power plant with a gas turbine, a steam turbine, and an alternator-all rotating on a common shaft. A waste-heat boiler produces steam at two pressures using heat from the gas turbine flus gas. The transient behavior of the system predicted by the model for various upset situations appears physically reasonable and satisfactory for the operating constraints

Bayesian Network Approach for Gas Path Fault Diagnosis

A method for solving the gas path analysis problem of jet engine diagnostics based on a probabilistic approach is presented. The method is materialized through the use of a Bayesian Belief Network (BBN). Building a BBN for gas turbine performance fault diagnosis requires information of a stochastic nature expressing the probability of whether a series of events occurred or not. This information can be extracted by a deterministic model and does not depend on hard to find flight data of different faulty operations of the engine. The diagnostic problem and the overall diagnostic procedure are first described. A detailed description of the way the diagnostic procedure is set-up, with focus on building the BBN from an engine performance model, follows. The case of a turbofan engine is used to evaluate the effectiveness of the method. Several simulated and benchmark fault case scenarios have been considered for this reason. The examined cases demonstrate that the proposed BBN-based diagnostic method composes a powerful tool. This work also shows that building a diagnostic tool, based on information provided by an engine performance model, is feasible and can be efficient as well.

Setting Up of a Probabilistic Neural Network for Sensor Fault Detection Including Operation With Component Faults

The diagnostic ability of probabilistic neural networks (PNN) for detecting sensor faults on gas turbines is examined. The structure and the features of a PNN, for sensor fault detection, are presented. It is shown that with the proposed formulation, a powerful tool for sensor fault identification is produced. A particular feature of the PNN produced is the ability to detect sensor faults even in the presence of engine component malfunction, as well as on deteriorated engines. In such situations, the size of bias that can be identified increases. The way to establish the limits of sensor bias that can be detected is presented along with results from application to test cases with realistic noise magnitudes. The diagnostic procedure proposed here is also supported by an engine performance model. The data used for setting up and testing the PNN are generated by such a model.

Turbofan Performance Deterioration Tracking Using Nonlinear Models and Optimization Techniques

A method of identifying the gradual deterioration in the components of jet engines is presented. It is based on the use of an engine model which has the capability to adapt component condition parameters, so that measured quantities are matched. The main feature of the method is that it gives the possibility to identify performance deviations in a number of parameters larger than the number of measured quantities. This is achieved by optimizing a cost function which incorporates not only measurement matching terms, but also terms expressing various constraints resulting from the physical knowledge of the deterioration process. Time series of data representing deterioration scenarios are used to demonstrate the methods capabilities. The test case considered is a twin spool partially mixed turbofan, representative of present-day large civil aeroengines. Implementation aspects, related to both the measurement set and the identification algorithms are discussed. An interpretation of the output of the method in function of different parameters entering the diagnostic problem is presented.

Performance analysis of industrial gas turbines for engine condition monitoring

Abstract This paper presents methods of analysing aerothermodynamic performance measurement data for the purpose of assessing the condition of the components of a gas turbine. Features of the methods are analysed in function of the available measurements and ways of extracting as much information as possible from a given measurement set are discussed. The principles discussed are highlighted by presenting results from application to data from operating industrial gas turbines. Particular applications discussed are the identification of deposits on the blades of a gas turbine used for power generation and the monitoring of compressor fouling on another industrial gas turbine.

Optimal Measurement and Health Index Selection for Gas Turbine Performance Status and Fault Diagnosis

In this paper, the authors present a method for defining the health estimation parameters and the measurements that must be used when a monitoring system for an engine is being set up. The particular engine layout, the available measuring instruments, and the accuracy by which data can be collected are the factors taken into account. The particular health condition estimation factors that have to be used are defined as a function of this information and the desired depth of fault identification. A fast selection procedure based on the method of singular value decomposition is presented. The uncertainty in the estimations is also derived, thus giving an additional element of information useful for decision making. The proposed method, together with adaptive performance modeling, provides a self-sufficient tool, which can be applied for setting up and subsequent exploitation of a health monitoring expert system. The advantage of the procedure is that it provides a frame of application, allowing quick implementation in a new engine of interest, other than the ones previously considered.

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.

Incorporating Neural Networks Into Gas Turbine Performance Diagnostics

Possibilities of incorporating neural nets in different tasks of a gas turbine performance diagnostic procedure are investigated. The purpose is to examine how neural nets can be implemented and what advantages they may offer. First, the possibility to constitute a performance model by using neural nets is considered. Different modes of operation are examined and the neural net architectures for achieving better accuracy are discussed. Subsequently, different problems of fault detection and identification are considered. Classification of faults is performed on the basis of diagnostic parameters produced by adaptive modelling. Both sensor faults and actual engine component faults are examined. A decision logic based on several neural nets is proposed. At a first level it is decided whether a fault exists, and if yes, checks are performed in order to identify the fault in as much detail as possible. Summarizing, the paper discusses different aspects of neural net implementation, in an effort to provide guidelines for application of this type of technique in the field of gas turbine diagnostics.Copyright

Combining Advanced Data Analysis Methods for the Constitution of an Integrated Gas Turbine Condition Monitoring and Diagnostic System

This paper presents principles for the constitution of gas turbine monitoring and diagnostic systems which:a. are integrated, namely manage all the tasks essential for achieving a diagnosis (measurement, analysis, interpretation, historical data management etc.)b. employ different kind of processing methods in order to cover an extensive range of engine conditions, including direct data evaluation and data consistency checks, thermodynamic analysis, vibration analysis.The requirements to be fulfilled by an industrial gas turbine monitoring system are briefly reviewed and ways to achieve them are discussed, indicating how they can be materialized by implementation of specific techniques. Techniques previously derived by the group of the authors are implemented, and the merits they offer when used in combination are discussed. Features of a system, materialized according to the principles discussed, into an operating industrial gas turbine is presented. On-line application of advanced analysis techniques, such as adaptive performance modeling is discussed, on the basis of observations of the collected data.Data collected from an engine operating in the field are presented to substantiate the matters discussed, and cases of successful fault identification are shown.Copyright

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.