Gas turbine diagnostics that relies on gas path measurements is a well-developed area with many algorithms developed. They follow two general approaches, data-driven, and physics-based. The first approach uses deviations of monitored variables from their baseline values. A diagnostic decision is traditionally made in the space of these deviations (diagnostic features) by pattern recognition techniques, for example, artificial neural networks. The necessary fault classes can be constructed from deviation vectors (patterns) using the displays of real faults, and the approach has a theoretical possibility to exclude a complex physics-based model and its inherent errors from a diagnostic process. For the second approach known as a gas path analysis, a nonlinear physics-based model (a.k.a. thermodynamic model) is an integral part of a diagnostic process. The thermodynamic model (or the corresponding linear model) relates monitored variables with operational conditions and model’s internal quantities called fault parameters. The identification of the thermodynamic model on the basis of known measurements of the monitored variables and operational conditions allows estimating unknown fault parameters. The knowledge of these parameters drastically simplifies a final diagnostic decision because great values of these parameters indicate damaged engine components and give us the measure of damage severity. As the diagnostic decision seems to be simple, the studies following this approach are usually completed by the analysis of fault parameter estimation accuracy, and complex pattern recognition techniques are not employed. Instead, simple tolerance-based fault detection and isolation is sometimes performed. It is not clear from known comparative studies which of the two approaches is more accurate, and the issue of seems to be challenging. This paper tries to solve this problem, being grounded on the following principles. We consider that a key difference of the second approach is a transformation from the diagnostic space of the deviations of monitored variables to the space of fault parameters. To evaluate the influence of this transformation on diagnostic accuracy, the other steps of the approaches should be equal. To this end, the pattern recognition technique employed in the data-driven approach is also included in the physics-based approach where it is applied to recognize fault parameter patterns instead of a tolerance-based rule. To realize and compare the data-driven and modified physics-based approaches, two corresponding diagnostic procedures differing only by the mentioned transformation have been developed. They use the same set of deviation vectors of healthy and faulty engines as input data and finally compute true classification rates that are employed to compare the procedures. The results obtained for different cases of the present comparative study show that the classification rates are practically the same for these procedures, and this is true for both fault detection and fault isolation. That is, correct classification does not depend on the mentioned transformation, and both approaches are equal from the standpoint of the classification accuracy of engine states.

In an effort to better compare particular gas turbine diagnostic solutions and recommend the best solution, the software tool called Propulsion Diagnostic Method Evaluation Strategy (ProDiMES) has been developed. This benchmarking platform includes a simulator of the aircraft engine fleet with healthy and faulty engines. The platform presents a public approach, at which different investigators can verify and compare their algorithms for the diagnostic stages of feature extraction, fault detection, and fault identification. Using ProDiMES, some different diagnostic solutions have been compared so far. This study presents a new attempt to enhance a gas turbine diagnostic process. A data-driven algorithm that embraces the mentioned three diagnostic stages is verified on the basis of ProDiMES. At the feature extraction stage, this algorithm uses a polynomial model of an engine baseline to compute deviations of actual gas path measurements from the corresponding values of a healthy engine. At the fault detection and fault identification stages, a common classification for fault detection and fault identification is firstly constructed using deviation vectors (patterns). One of the three chosen pattern recognition techniques then performs both fault detection and fault identification as a common process. Numerous numerical experiments have been conducted to select the best configurations of the baseline model, a pertinent structure of the fault classification, and the best recognition technique. The experiments were accompanied by a computational precision analysis for each component of the proposed algorithm. The comparison of the final diagnostic ProDiMES metrics obtained under the selected optimal conditions with the metrics of other diagnostic solutions shows that the proposed algorithm is a promising tool for gas turbine monitoring systems.

The paper is devoted to the use of mathematical simulation to investigate the possibilities of ensuring the admissible thermal mode of gas turbine packages equipped with aircraft and marine derivative gas turbine engines. The method proposed for complex heat transfer simulation in the gas turbine packages includes some models. A generalized mathematical model is formed to describe the thermophysical processes taking place in the gas turbine packages. A particular mathematical model of gas turbine engine casing heat transfer and a method to correct the boundary conditions are also developed. These models have been validated with the data collected from the heat transfer measurements in simple objects and from full-scale tests of turbo-compressor units. The proposed method of complex heat transfer simulation has been used to evaluate a temperature state of the gas turbine packages, in particular to ensure the effectiveness of covering the engine casing by thermal insulation.

The problem of pneumatic volumes constituting a gas turbine engine gas path and their effect on an engine transient belongs to the topical problems of modern turbomachinery engineering. Many researchers have been trying to find the most resource efficient method of volume modeling. The associated urgent problem is the integration of the volume models with an engine model. Following the world trends, the present paper deals with the pneumatic volume simulation that considers different dynamic factors described with the mass, energy, and momentum conservation laws. The developed dynamic volume models were examined and compared. The comparison cases included temperature and pressure change at the volume entrance and pressure change at the volume discharge. To reduce the computation time, the linearization of model equations was also proposed and proved. Based on the comparison results, the guidelines and recommendations on the volume effect mathematical modeling and the volume model integration with the engine model have been made.

One of the principle purposes of gas turbine diagnostics is the estimation and monitoring of important unmeasured quantities such as engine thrust, shaft power, and engine component efficiencies. There are simple methods that allow computing the unmeasured parameters using measured variables and gas turbine thermodynamics. However, these parameters are not good diagnostic indices because they strongly depend on engine operating conditions but in a less degree are influenced by engine degradation and faults. In the case of measured gas path variables, deviations between measurements and an engine steady state baseline were found to be good indicators of engine health. In this paper, the deviation computation and monitoring are extended to the unmeasured parameters. To verify this idea, the deviations of compressor and turbine efficiencies as well as a high pressure turbine inlet temperature are examined. Deviation computations were performed at steady states for both baseline and faulty engine conditions using a nonlinear thermodynamic model and real data. These computational experiments validate the utility of the deviations of unmeasured variables for gas turbine monitoring and diagnostics. The thermodynamic model is used in this paper only to generate data, and the proposed algorithm for computing the deviations of unmeasured parameter can be considered to be a data-driven technique. This is why the algorithm is not affected by inaccuracies of a physics-based model, is not exigent to computer resources, and can be used in on-line monitoring systems.

Algorithms for predicting the remaining lifetime of an engine play an important role in gas turbine monitoring systems. This paper addresses the improvement of models to determine the thermal boundary conditions that are necessary to calculate engine lifetime in critical hot components. Two methods for model development are compared. The first method uses physics-based models. The second method formulates the models based on a similarity concept. The object of analysis is a cooled blade of a high-pressure turbine. Two unmeasured thermal boundary conditions are considered: the heating temperature and the heat transfer coefficient. Instrumental and truncation errors are estimated for each model and 10 faulty conditions are considered to take into account the existing engine-to-engine differences and performance deterioration. The blade temperature and the thermal stress at the critical points are calculated using the results obtained by the developed models as boundary conditions. The results of the comparison show that the physics-based models are more robust to power plant faults. The best models for the heating temperature and the heat transfer coefficient were chosen. It is shown that the accuracy of the heating temperature model is more important for reliable lifetime prediction.

The steady-state and transient engine performances of gas turbine control system development are usually evaluated by applying full thermodynamic engine models. Most models only address the operating range between the idle and maximum power points, but more recently, they also address a sub-idle operating range. The lack of information about the component maps at the sub-idle modes creates major challenges for the starting system and control system designers. A common method to cope with the problem extrapolates the performances of the engine components to the sub-idle operation range. Precise extrapolation is a challenge to be studied by many scientists. As a rule, many scientists are only concerned about particular aspects of the problem such as the lighting combustion chamber or the turbine operation under the turned-off conditions of the combustion chamber. However, there are no known reports about a model that considers all of these mentioned aspects and simulates the engine starting. To synthesize a thermodynamic model of starting, most known methods require the performance of the components in the sub-idle range. The proposed paper addresses a new method that simulates the engine starting. The method substitutes the non-linear thermodynamic model with a linear dynamic model, which is supplemented with a simplified static model. The latter model is the set of direct relations between parameters that are used in the control algorithms instead of commonly used component performances. Specifically, the static model consists of simplified relations between the gas path parameters and the corrected rotational speed. The paper also describes an algorithm for model synthesis and its practical application to real data.

A modern gas turbine engine (GTE) is a complex non-linear dynamic system with the mutual effect of gas-dynamic and thermal processes in its components. The engine development requires the precise real-time simulation of all main operating modes. One of the most complex operating modes for modeling is “cold stabilization”, which is the rotors acceleration without completely heated up the turbine elements. The dynamic heating problem is a topical practical issue. Solving the problem requires coordinating a gas-path model with heat and stress models, which is also a significant scientific problem. The phenomenon of interest is the radial clearances change during engines operation and its influence on engines static and dynamic performances. To consider the clearance change, it is necessary to synthesize the quick proceeding stress-state models (QPSSM) of a rotor and a casing for the initial temperature and dynamic heating. The unique feature of the QPSSM of GTEs is separate equation sets, which allow the heat exchange between structure elements and the gas (air) and the displacements of the turbine rotor and the casing. This ability appears as a result of determining the effect of each factor on different structural elements of the engine. The presented method significantly simplifies the model identification, which can be performed based on a precise calculation of the unsteady temperature fields of the structural elements and the variation of the radial clearance. Thus, the present paper addresses a new method to model the engine dynamics considering its heating up. The method is based on the integration of three models: the gas-path dynamics model, the clearance dynamics model and the model of the clearance effect on the efficiency. The paper also comprises the program implementation of the models. The method was tested by applying to a particular turbofan engine.

Diagnostics is an important aspect of a condition based maintenance program. To develop an effective gas turbine monitoring system in short time, the recommendations on how to optimally design every system algorithm are required. This paper deals with choosing a proper fault classification technique for gas turbine monitoring systems. To classify gas path faults, different artificial neural networks are typically employed. Among them the Multilayer Perceptron (MLP) is the mostly used. Some comparative studies referred to in the introduction show that the MLP and some other techniques yield practically the same classification accuracy on average for all faults. That is why in addition to the average accuracy, more criteria to choose the best technique are required. Since techniques like Probabilistic Neural Network (PNN), Parzen Window (PW) and k -Nearest Neighbor (K-NN) provide a confidence probability for every diagnostic decision, the presence of this important property can be such a criterion. The confidence probability in these techniques is computed through estimating a probability density for patterns of each concerned fault class. The present study compares all mentioned techniques and their variations using as criteria both the average accuracy and availability of the confidence probability. To compute them for each technique, a special testing procedure simulates numerous diagnosis cycles corresponding to different fault classes and fault severities. In addition to the criteria themselves, criteria imprecision due to a finite number of the diagnosis cycles is computed and involved into selecting the best technique.

Efficiency of gas turbine monitoring systems primarily depends on the accuracy of employed algorithms, in particular, pattern recognition techniques to diagnose gas path faults. In investigations many techniques were applied to recognize gas path faults, but recommendations on selecting the best technique for real monitoring systems are still insufficient and often contradictory. In our previous works, three recognition techniques were compared under different conditions of gas turbine diagnosis. The comparative analysis has shown that all these techniques yield practically the same accuracy for each comparison case. The present contribution considers a new recognition technique, Probabilistic Neural Network (PNN), comparing it with the techniques previously examined. The results for all comparison cases show that the PNN is not practically inferior to the other techniques. With this inference, the recommendation is to choose the PNN for real monitoring systems because it has an important advantage of providing confidence estimation for every diagnostic decision made.

This paper addresses the problem of estimation of unmeasured gas turbine engine variables using statistical analysis of measured data. Possible changes of an engine health condition and lack of information about these changes caused by limited instrumentation are taken into account. Engine thrust is under consideration as one of the most important unmeasured parameters. Two common methods of aircraft gas turbine engine (GTE) thrust monitoring and their errors due to health condition changes are analyzed. Additionally, two mathematical techniques that allow reducing in-flight thrust estimation errors in the case of GTE deterioration are suggested and verified in the paper. They are a ridge trace and a principal component analysis. A turbofan engine has been chosen as a test case. The engine has five measured variables and 23 health parameters to describe its health condition. Measurement errors are simulated using a generator of random numbers with the normal distribution. The engine is presented in calculations by its nonlinear component level model (CLM). Results of the comparison of thrust estimates computed by the CLM and the proposed techniques confirm accuracy of the techniques. The regression model on principal components has demonstrated the highest accuracy.

Gas turbine diagnostic algorithms widely use fault simulation schemes, in which measurement errors are usually given by theoretical random number distributions, like the Gaussian probability density function. The scatter of simulated noise is determined on the basis of known information on maximum errors for every sensor type. Such simulation differs from real diagnosis because instead of measurements themselves the diagnostic algorithms work with their deviations from an engine baseline. In addition to simulated measurement inaccuracy, the deviations computed for real data have other error components. In this way, simulated and real deviation errors differ by amplitude and distribution. As a result, simulation-based investigations might result in too optimistic conclusions on gas turbine diagnosis reliability. To understand error features, deviations of real measurements are analyzed in the present paper. To make error presentation more realistic, it is proposed to extract an error component from real deviations and to integrate it in fault description. Finally, the effect of the new noise representation mode on diagnostic reliability is estimated. It is shown that the reliability change due to inexact error simulation can be significant.

Efficiency of gas turbine condition monitoring systems depends on quality of diagnostic analysis at all its stages such as feature extraction (from raw input data), fault detection, fault identification, and prognosis. Fault identification algorithms based on the gas path analysis may be considered as an important and sophisticated component of these systems. These algorithms widely use pattern recognition techniques, mostly different artificial neural networks. In order to choose the best technique, the present paper compares two network types: a multilayer perceptron and a radial basis network. The first network is being commonly applied to recognize gas turbine faults. However, some studies note high recognition capabilities of the second network. For the purpose of the comparison, both networks were included into a special testing procedure that computes for each network the true positive rate that is the probability of a correct diagnosis. Networks were first tuned and then compared using this criterion. Same procedure input data were fed to both networks during the comparison. However, to draw firm conclusions on the networks’ applicability, comparative calculations were repeated with different variations of these data. In particular, two engines that differ in an application and gas path structure were chosen as a test case. By way of summing up comparison results, the conclusion is that the radial basis network is a little more accurate than the perceptron, however the former needs much more available computer memory and computation time.

Life usage algorithms constitute one of the principal components of gas turbine engines monitoring systems. These algorithms aim to determine the remaining useful life of gas turbines based on temperature and stress estimation in critical hot part elements. Knowing temperatures around these elements is therefore very important. This paper deals with blades and disks of a high pressure turbine (HPT). In order to monitor their thermal state, it is necessary to set thermal boundary conditions. The main parameter to determine is the total gas temperature in relative motion at the inlet of HPT blades T w * . We propose to calculate this unmeasured temperature as a function of measured gas path variables using gas path thermodynamics. Five models with different thermodynamic relations to calculate the temperature T w * are proposed and compared. All temperature models include some unmeasured parameters that are presented as polynomial functions of a measured power setting variable. A nonlinear thermodynamic model is used to calculate the unknown coefficients included in the polynomials and to validate the models considering the influence of engine deterioration and operating conditions. In the validation stage, the polynomial’s inadequacy and the errors caused by the measurement inaccuracy are analyzed. Finally, the gas temperature models are compared using the criterion of total accuracy and the best model is selected.

Gas turbine health monitoring includes the common stages of problem detection, fault identification, and prognostics. To extract useful diagnostic information from raw recorded data, these stages require a preliminary operation of computing differences between measurements and an engine baseline, which is a function of engine operating conditions. These deviations of measured values from the baseline data can be good indicators of engine health. However, their quality and success of all diagnostic stages strongly depend on an adequacy of the baseline model employed and, in particular, on mathematical techniques applied to create it. To create the baseline model we have applied polynomials and the least square method for computing their coefficients over a long period of time. Some methods were proposed to enhance such a polynomial-based model. The resulting accuracy was sufficient for reliable monitoring gas turbine deterioration effects. The polynomials previously investigated enough are used in the present study as a standard for evaluating artificial neural networks, a very popular technique in gas turbine diagnostics. The focus of this comparative study is to verify whether the use of networks results in a better description of the engine baseline. Extensive field data of two different industrial gas turbines were used to compare these two techniques in various conditions. The deviations were computed for all available data and quality of the resulting deviations plots was compared visually. A mean error of the baseline model was an additional criterion for the comparing the techniques. To find the best network configurations many network variations were realized and compared with the polynomials. Although the neural networks were found to be close to the polynomials in accuracy, they could not exceed the polynomials in any variation. In this way, it seems that polynomials can be successfully used for engine monitoring, at least for the analyzed gas turbines.

In modern gas turbine health monitoring systems, the diagnostic algorithms based on gas path analysis may be considered as principal. They analyze gas path measured variables and are capable of identifying different faults and degradation mechanisms of gas turbine components (e.g. compressor, turbine, and combustor) as well as malfunctions of the measurement system itself. Gas path mathematical models are widely used in building fault classification required for diagnostics because faults rarely occur during field operation. In that case, model errors are transmitted to the model-based classification, which poses the problem of rendering the description of some classes more accurate using real data. This paper looks into the possibility of creating a mixed fault classification that incorporates both model-based and data-driven fault classes. Such a classification will combine a profound common diagnosis with a higher diagnostic accuracy for the data-driven classes. A gas turbine power plant for natural gas pumping has been chosen as a test case. Its real data with cycles of compressor fouling were used to form a data-driven class of the fouling. Preliminary qualitative analysis showed that these data allow creating a representative class of the fouling and that this class will be compatible with simulated fault classes. A diagnostic algorithm was created based on the proposed classification (real class of compressor fouling and simulated fault classes for other components) and artificial neural networks. The algorithm was subjected to statistical testing. As a result, probabilities of a correct diagnosis were determined. Different variations of the classification were considered and compared using these probabilities as criteria. The performed analysis has revealed no limitations for realizing a principle of the mixed classification in real monitoring systems.

Monitoring algorithms analyzing measured gas path variables provide invaluable insight into gas turbine operating health. Some useful information about a gas turbine and its measurement system can be obtained from a direct analysis of raw measurements. To draw more comprehensive diagnostic information, deviations are usually calculated as discrepancies between the measured and baseline values of monitored variables. The deviations can serve as good indicators of different engine degradation mechanisms. However, there are many negative factors that tend to mask degradation effects. For a long period of time we have analyzed quality of gas path measurement data and a deviation accuracy problem of a gas turbine power plant driving a natural gas pipeline compressor. Possible error sources were examined and some methods were proposed to improve the accuracy of deviation calculations. This paper looks at maintenance data of another object, the General Electric LM2500 gas turbine used as a drive of an electric generator. The data cover prolonged periods of axial compressor fouling with washings between them, and provide valuable information for a deviation examination. In order to reduce deviation errors, the paper considers different cases of the abnormal functioning of the sensors and baseline model inadequacy and proposes measures to avoid them. As a result of these and previous efforts, the deviations have become good fouling indicators. They are capable to quantify the increase of exhaust gas temperature (EGT) and, consequently, to improve planning axial compressor washings.

This paper presents an investigation of a conventional gas turbine diagnostic process and its generalization. A usual sequence of diagnostic actions consists of two stages: monitoring (fault detection) and subsequent proper diagnosis (fault identification). Such an approach neither implies fault identification nor uses the information about incipient faults unless the engine is recognized as faulty. In previous investigations for engine steady state operation conditions we addressed diagnostics problems without their relation with the monitoring process. Fault classes were given by samples of patterns generated by a static gas turbine performance model. This fault simulation took into account faults of varying severity including incipient ones. A diagnostic algorithm employed artificial neural networks to identify an actual fault. In the present paper we consider the monitoring and diagnosis as joint processes extending our previous approach over both of them. It is proposed to form two classes for the monitoring using the above-mentioned classes constructed for the diagnosis. A two-shaft industrial gas turbine has been chosen to test the proposed integrated approach to monitoring and diagnosis. A general recommendation following from the presented investigation is to identify faults simultaneously with fault detection. This permits accumulating preliminary diagnoses before the engine faulty condition is detected and a rapid final diagnosis after the fault detection.

Operating conditions (control variables and ambient conditions) of gas turbine plants and engines vary considerably. The fact that health monitoring has to be uninterrupted creates the need for a run time diagnostic system to operate under any conditions. The diagnostic technique described in this paper utilizes the thermodynamic models in order to simulate gaspath faults and uses neural networks for the faults localization. This technique is repeatedly executed and the diagnoses are registered. On the basis of these diagnoses and beforehand known faults, the correct diagnosis probabilities are then calculated. The present paper analyses the influence of the operating conditions on a diagnostic process. In the technique, different options are simulated of a diagnostic treatment of the measured values obtained under variable operating conditions. The mentioned above probabilities help to compare these options. The main focus of the paper is on the so called multipoint (multimode) diagnosis that groups the data from different operating points (modes) to set only a single diagnosis.