Y. G. Li

Performance-analysis-based gas turbine diagnostics: A review

Abstract Gas turbine diagnostics has a history almost as long as gas turbine development itself. Early engine fault diagnosis was carried out based on manufacturer information supplied in a technical manual combined with maintenance experience. In the late 1960s, when L. A. Urban introduced gas path analysis, gas turbine diagnostics made a big breakthrough. Since then different methods have been developed and used in both aerospace and industrial applications. To date, a substantial number of papers have been published in this area. This paper intends to give a comprehensive review of performance-analysis-based methods available thus far for gas turbine fault diagnosis in the open literature.

An Adaptation Approach for Gas Turbine Design-Point Performance Simulation

Accurate simulation and understanding of gas turbine performance is very useful for gas turbine users. Such a simulation and performance analysis must start from a design point. When some of the engine component parameters for an existing engine are not available, they must be estimated in order that the performance analysis can be carried out. However, the initially simulated design-point performance of the engine using estimated engine component parameters may give a result that is different from the actual measured performance. This difference may be reduced with better estimation of these unknown component parameters. However, this can become a difficult task for performance engineers, let alone those without enough engine performance knowledge and experience, when the number of design-point component parameters and the number of measurable/target performance parameters become large. In this paper, a gas turbine design-point performance adaptation approach has been developed to best estimate the unknown design-point component parameters and match the available design-point engine measurable/target performance. In the approach, the initially unknown component parameters may be compressor pressure ratios and efficiencies, turbine entry temperature, turbine efficiencies, air mass flow rate, cooling flows, bypass ratio, etc. The engine target (measurable) performance parameters may be thrust and specific fuel consumption for aero engines, shaft power and thermal efficiency for industrial engines, gas path pressures and temperatures, etc. To select, initially, the design point component parameters, a bar chart has been used to analyze the sensitivity of the engine target performance parameters to the design-point component parameters. The developed adaptation approach has been applied to a design-point performance matching problem of an industrial gas turbine engine GE LM2500+ operating in Manx Electricity Authority (MEA), UK. The application shows that the adaptation approach is very effective and fast to produce a set of design-point component parameters of a model engine that matches the actual engine performance very well. Theoretically, the developed techniques can be applied to other gas turbine engines.

Gas Turbine Performance and Health Status Estimation Using Adaptive Gas Path Analysis

In gas turbine operations, engine performance and health status are very important information for engine operators. Such engine performance is normally represented by engine airflow rate, compressor pressure ratios, compressor isentropic efficiencies, turbine entry temperature, turbine isentropic efficiencies, etc., while the engine health status is represented by compressor and turbine efficiency indices and flow capacity indices. However, these crucial performance and health information cannot be directly measured and therefore are not easily available. In this research, a novel Adaptive Gas Path Analysis (Adaptive GPA) approach has been developed to estimate actual engine performance and gas path component health status by using gas path measurements, such as gas path pressures, temperatures, shaft rotational speeds, fuel flow rate, etc. Two steps are included in the Adaptive GPA approach, the first step is the estimation of degraded engine performance status by a novel application of a performance adaptation method, and the second step is the estimation of engine health status at component level by using a new diagnostic method introduced in this paper, based on the information obtained in the first step. The developed Adaptive GPA approach has been tested in four test cases where the performance and degradation of a model gas turbine engine similar to Rolls-Royce aero engine Avon-300 have been analyzed. The case studies have shown that the developed novel linear and nonlinear Adaptive GPA approaches can accurately and quickly estimate the degraded engine performance and predict the degradation of major engine gas path components with the existence of measurement noise. The test cases have also shown that the calculation time required by the approach is short enough for its potential online applications.

A gas turbine diagnostic approach with transient measurements

Abstract Most gas turbine performance analysis based diagnostic methods use the information from steady state measurements. Unfortunately, steady state measurement may not be obtained easily in some situations, and some types of gas turbine fault contribute little to performance deviation at steady state operating conditions but significantly during transient processes. Therefore, gas turbine diagnostics with transient measurement is superior to that with steady state measurement. In this paper, an accumulated deviation is defined for gas turbine performance parameters in order to measure the level of performance deviation during transient processes. The features of the accumulated deviation are analysed and compared with traditionally defined performance deviation at a steady state condition. A non-linear model based diagnostic method, combined with a genetic algorithm (GA), is developed and applied to a model gas turbine engine to diagnose engine faults by using the accumulated deviation obtained from transient measurement. Typical transient measurable parameters of gas turbine engines are used for fault diagnostics, and a typical slam acceleration process from idle to maximum power is chosen in the analysis. The developed diagnostic approach is applied to the model engine implanted with three typical single-component faults and is shown to be very successful.

Gas Path Fault Diagnosis of a Turbofan Engine From Transient Data Using Artificial Neural Networks

Transient and steady state data may contain the same essential fault information but some faults have been shown to be more easily detectable from transient data because the transient records provide significant diagnostic content especially as the fault effects are magnified under transient. Various traditional and conventional techniques such as fault trees, fault matrixes, gas path analysis and its variants have been applied to gas path fault diagnosis of gas turbines. Recently, artificial intelligence techniques such as artificial neural networks (ANN) as well as optimization techniques such as genetic algorithm (GA) are being explored for fault diagnosis activities. In this paper, a novel approach to gas path fault diagnosis is proposed. The method involves the use of ANN with engine transient data. A set of nested neural networks designed to estimate independent parameter (efficiencies and flow capacities) changes due to faults within single or multiple components of a turbofan engine are presented. The approach involves classification and approximation type networks. Measurements from the engine are first assessed by a trained network and if a fault is diagnosed, are then classified into two groups — those originating from sensor faults and those from component faults, by another trained network. Other trained networks continue the fault isolation process and finally the magnitude of the fault(s) is quantified. A computer simulation of the process shows that results from a batched process of these networks can be obtained in less than three seconds. Four of the gas path components — intermediate pressure compressor (IPC), high pressure compressor (HPC), high pressure turbine (HPT) and low pressure turbine (LPT) — and measurements from eight sensors are considered. Sensor noise and bias are also considered in this analysis. The comparison of fault signatures from a steady state and transient process show that diagnosis with transient data can improve the accuracy of gas turbine fault diagnosis.Copyright

Nonlinear Weighted-Least-Squares Estimation Approach for Gas-Turbine Diagnostic Applications

the summation of weighted square deviations between estimated and actual values of gas-turbine performance measurements. The measurement uncertainties associated with gas-path measurements are taken into account by using a weighting matrix. The concepts of fault cases and gas-path-analysis index are introduced. These provide a newdiagnosticapproach,enablingfaultisolationandenhancingtheconfidenceofdiagnosticresults.Thisdiagnostic approach allows the typically nonlinear gas-turbine thermodynamic-performance models to be directly used in condition monitoring while taking into account performance nonlinearities. An iterative calculation process is introducedtoobtainaconvergedestimationofenginedegradation.Thenonlinearweighted-least-squaresdiagnostic approachhasbeenappliedtoamodelindustrialgas-turbineenginetotestitseffectiveness.Thenumericaltestsofthe new diagnostic approach show that with appropriate selection of engine gas-path measurements, the method can be used effectively and successfully to predict gas-turbine performance degradation.

Fault Diagnosis of a Two Spool Turbo-Fan Engine Using Transient Data: A Genetic Algorithm Approach

Traditionally engine fault diagnosis has been performed at steady state conditions. There are several problems which can only be detected by transient data analysis like bearing fault, some control problems etc.. In addition, gas turbine performance deviation due to a component fault is more likely to be magnified during transients, when compared with the same parameter deviations at steady states. The specific approach used in this paper is to compare model-based information with measured data obtained from the engine during a slam acceleration. The measured transient data (from actual engine) is compared with a set of simulated data from the engine transient model, under similar operating conditions and known faults through a Cumulative Deviation. The Cumulative Deviations obtained from the comparisons are minimized for the best match using Genetic Algorithm. The Genetic Algorithm has been tailored to use real coding [1] method and to meet the requirements of the new procedure. The paper describes the application of the approach to a 2-spool turbofan engine and discusses the preliminary studies conducted.Copyright

Creep Life Prediction for Aero Gas Turbine Hot Section Component Using Artificial Neural Networks

Accurate and reliable component life prediction is crucial to ensure safety and economics of gas turbine operations. In pursuit of such improved accuracy and reliability, model-based creep life prediction methods have become more complicated and demand higher computational time. Therefore, there is a need to find an alternative approach that is able to provide a quick solution to creep life prediction for production engines while at the same time maintain the same accuracy and reliability as that of the model-based methods. In this paper, a novel creep life prediction approach using artificial neural networks is introduced as an alternative to the model-based creep life prediction approach to provide a quick and accurate estimation of gas turbine creep life. Multilayer feed forward backpropagation neural networks have been utilized to form three neural network–based creep life prediction architectures known as the range-based, functional-based, and sensor-based architectures. The new neural network creep life prediction approach has been tested with a model single-spool turboshaft gas turbine engine. The results show that good generalization can be achieved in all three neural network architectures. It was also found that the sensor-based architecture is better than the other two in terms of accuracy, with 98% of the post-test samples possessing prediction errors within ±0.4%.

Part-load performance of gas turbines - Part I: A novel compressor map generation approach suitable for adaptive simulation

Part-load performance prediction of gas turbines is strongly dependent on detailed understanding of engine component behavior and mainly that of compressors. The accuracy of gas turbine engine models relies on the compressor performance maps, which are obtained in costly rig tests and remain manufacturers proprietary information. The gas turbine research community has addressed this limitation by scaling default generic compressor maps in order to match the targeted off-design measurements. This approach is efficient in small range of operating conditions but becomes less accurate for wide range of operating conditions. In this part of the paper a novel method of compressor map generation which has a primary objective to improve the accuracy of engine models performance at part load conditions is presented. This is to generate a generic form of equations to represent the lines of constant speed and constant efficiency of the compressor map for a generic compressor. The parameters that control the shape of the compressor map have been expressed in their simplest form in order to aid the adaptation p rocess. The proposed compressor map generation method has the capacity to refine current gas turbine performance adaptation techniques, and it has been integrated into Cranfields PYTHIA gas turbine performance simulation and diagnostics software tool. Copyright © 2012 by ASME.

Gas Turbine Off-Design Performance Adaptation Using a Genetic Algorithm

Gas turbine gas path diagnostics is heavily dependent on performance simulation models accurate enough around a chosen diagnostic operating point, such as design operating point. With current technology, gas turbine engine performance can be predicted easily with thermodynamic models and computer codes together with basic engine design data and empirical component information. However the accuracy of the prediction is highly dependent on the quality of those engine design data and empirical component information such as component characteristic maps but such expensive information is normally exclusive property of engine manufacturers and only partially disclosed to engine users. Alternatively, estimated design data and assumed component information are used in the performance prediction. Yet, such assumed component information may not be the same as those of real engines and therefore poor off-design performance prediction may be produced. This paper presents an adaptive method to improve the accuracy of off-design performance prediction of engine models near engine design point or other points where detailed knowledge is available. A novel definition of off-design scaling factors for the modification of compressor maps is developed. A Genetic Algorithm is used to search the best set of scaling factors in order to adapt the predicted off-design engine performance to observed engine off-design performance. As the outcome of the procedure, new compressor maps are produced and more accurate prediction of off-design performance is provided. The proposed off-design performance adaptation procedure is applied to a model civil aero engine to test the effectiveness of the adaptive approach. The results show that the developed adaptive approach, if properly applied, has great potential to improve the accuracy of engine off-design performance prediction in the vicinity of engine design point although it does not guarantee the prediction accuracy in the whole range of off-design conditions. Therefore, such adaptive approach provides an alternative method in producing good engine performance models for gas turbine gas path diagnostic analysis.Copyright