scholarly journals Assessment of Dynamic Bayesian Models for Gas Turbine Diagnostics, Part 2: Discrimination of Gradual Degradation and Rapid Faults

Machines ◽  
2021 ◽  
Vol 9 (12) ◽  
pp. 308
Author(s):  
Valentina Zaccaria ◽  
Amare Desalegn Fentaye ◽  
Konstantinos Kyprianidis
Keyword(s):  
Gas Turbine ◽  
Gas Turbines ◽  
Dynamic Bayesian Networks ◽  
Performance Model ◽  
Diagnostic Methods ◽  
Sensor Faults ◽  
Layer System ◽  
Isolation And Identification ◽  
Multiple Faults ◽  
Abrupt Faults

There are many challenges that an effective diagnostic system must overcome for successful fault diagnosis in gas turbines. Among others, it has to be robust to engine-to-engine variations in the fleet, it has to discriminate between gradual deterioration and abrupt faults, and it has to identify sensor faults correctly and be robust in case of such faults. To combine their benefits and overcome their limitations, two diagnostic methods were integrated in this work to form a multi-layer system. An adaptive performance model was used to track gradual deterioration and detect rapid or abrupt anomalies, while a series of static and dynamic Bayesian networks were integrated to identify component degradation, component abrupt faults, and sensor faults. The proposed approach was tested on synthetic data and field data from a single-shaft gas turbine of 50 MW class. The results showed that the approach could give acceptable accuracy in the isolation and identification of multiple faults, with 99% detection and isolation accuracy and 1% maximum error in the identified fault magnitude. The approach was also proven robust to sensor faults, by replacing the faulty signal with an estimated value that had only 3% error compared to the real measurement.

Hybrid Modeling of Heavy Duty Gas Turbines for On-Line Performance Monitoring

2014 ◽  
Author(s):  
Thomas Palmé ◽  
Francois Liard ◽  
Dan Cameron
Keyword(s):  
Gas Turbine ◽  
Surrogate Model ◽  
Gas Turbines ◽  
Performance Monitoring ◽  
Performance Model ◽  
Heavy Duty ◽  
Operational Parameters ◽  
Actual Performance ◽  
Real Gas ◽  
The Real

Due to their complex physics, accurate modeling of modern heavy duty gas turbines can be both challenging and time consuming. For online performance monitoring, the purpose of modeling is to predict operational parameters to assess the current performance and identify any possible deviation between the model’s expected performance parameters and the actual performance. In this paper, a method is presented to tune a physical model to a specific gas turbine by applying a data-driven approach to correct for the differences between the real gas turbine operation and the performance model prediction of the same. The first step in this process is to generate a surrogate model of the 1st principle performance model through the use of a neural network. A second “correction model” is then developed from selected operational data to correct the differences between the surrogate model and the real gas turbine. This corrects for the inaccuracies between the performance model and the real operation. The methodology is described and the results from its application to a heavy duty gas turbine are presented in this paper.

Validation of an Advanced Diagnostic Methodology for the Identification and Classification of Gas Turbine Sensor Faults by Means of Field Data

2019 ◽  
Author(s):  
Lucrezia Manservigi ◽  
Mauro Venturini ◽  
Giuseppe Fabio Ceschini ◽  
Giovanni Bechini ◽  
Enzo Losi
Keyword(s):  
Gas Turbine ◽  
Gas Turbines ◽  
Data Cleaning ◽  
Sensitivity Analyses ◽  
Sensor Faults ◽  
Sensor Fault ◽  
Sensor Fault Detection ◽  
Optimal Setting ◽  
Integrated Diagnostics

Abstract Sensor fault detection is a crucial aspect for raw data cleaning in gas turbine industry. To this purpose, a comprehensive approach for Improved Detection, Classification and Integrated Diagnostics of Gas Turbine Sensors (named I-DCIDS) was developed by the authors to detect and classify several classes of fault. For single-sensors or redundant/correlated sensors, the I-DCIDS methodology can identify seven classes of fault, i.e. Out of Range, Stuck Signal, Dithering, Standard Deviation, Trend Coherence, Spike and Bias. Since the considered faults are detected by means of a methodology which relies on basic mathematical laws and user-defined parameters, sensitivity analyses are carried out in this paper on I-DCIDS parameters to derive some rules of thumbs about their optimal setting. The sensitivity analyses are carried out on four heterogeneous and challenging datasets with redundant sensors installed on Siemens gas turbines.

Increasing Diagnostic Effectiveness by Inclusion of Fuel Composition and Water Injection Effects

2002 ◽  
Author(s):  
K. Mathioudakis ◽  
N. Aretakis ◽  
A. Tsalavoutas
Keyword(s):  
Gas Turbine ◽  
Gas Turbines ◽  
Water Injection ◽  
Diagnostic Technique ◽  
Diagnostic Methods ◽  
Working Fluid ◽  
Fluid Properties ◽  
Industrial Gas Turbines ◽  
The Impact ◽  
Engine Condition

The paper presents an analysis of the effect of changing the fuel on the performance of industrial gas turbines and examines the impact of such a change on methods used for engine condition assessment and fault diagnostics. A similar analysis is presented for the effects of water injection in the combustion chamber (which is usually done for reducing NOx emissions). First, the way of incorporating the effect of fuel changes and water injection into a computer model of gas turbine performance is described. The approach employed is based on the change of (a) working fluid properties, (b) turbomachinery components performance. The model is then used to derive parameters indicative of the “health” of a gas turbine and thus diagnose the presence of deterioration or faults. The impact of ignoring the presence of an altered fuel or injected water is shown to be of a magnitude that would render a diagnostic technique that does not incorporate these effects ineffective. On the other hand, employing the appropriate physical modeling makes the diagnostic methods robust and insensitive to such changes, being thus able to provide useful diagnostic information continuously during the use of a gas turbine.

Multiple Faults Detection of Gas Turbine by MLP Neural Network

2009 ◽  
Author(s):  
Gustavo R. Matuck ◽  
Joa˜o Roberto Barbosa ◽  
Cleverson Bringhenti ◽  
Isaias Lima
Keyword(s):  
Neural Network ◽  
Gas Turbine ◽  
Gas Turbines ◽  
Back Propagation ◽  
Optimization Techniques ◽  
Multiple Faults ◽  
Error Back Propagation ◽  
Artificial Neural ◽  
Hidden Layer ◽  
Simulator Program

This paper describes a procedure to measure the performance of detection and isolation of multiple faults in gas turbines using artificial neural network and optimization techniques. It is on a particular form of artificial neural networks, the traditional multi-layer perceptron (MLP). Error back-propagation and different activation functions are used. The main goal is to recognize single, double and triple faults in a turboshaft engine, whose performance data were output from a gas turbine simulator program, tuned to represent the engine running at an existing power station. MLP network is a nonlinear interpolation function usually made of input layer, hidden-layer and output-layer, with different neuronal units, but in this work, only one hidden-layer was used. Weights were altered by error back-propagation from the initial values established from a seed fixed between 0 and 1. The activation function in the MLP algorithm is the sigmoid function. The best moment to stop the training process and avoid the over fitting problem was chosen by cross-validation. Optimization of convergence error was achieved using the momentum criteria and reducing the oscillation problem in all nets trained. Several configurations of the neural network have been compared and evaluated, using several noise graduations incorporated to the data, aiming at finding the network most suitable to detect and isolate multiple faults in gas turbines. Based on the results obtained it is inferred that the procedure reported herein may be applied to actual systems in order to assist in maintenance programs, at least.

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

2003 ◽  
Vol 125 (3) ◽  
pp. 634-641 ◽  
Author(s):  
C. Romesis ◽  
K. Mathioudakis
Keyword(s):  
Fault Detection ◽  
Gas Turbines ◽  
Probabilistic Neural Network ◽  
Engine Performance ◽  
Performance Model ◽  
Test Cases ◽  
Sensor Faults ◽  
Sensor Fault ◽  
Sensor Fault Detection ◽  
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.

Part-Load Performance of Gas Turbines: Part II — Multi-Point Adaptation With Compressor Map Generation and GA Optimization

2012 ◽  
Author(s):  
E. Tsoutsanis ◽  
Y. G. Li ◽  
P. Pilidis ◽  
M. Newby
Keyword(s):  
Gas Turbine ◽  
Gas Turbines ◽  
Combined Cycle ◽  
Performance Model ◽  
Part Load ◽  
Design Performance ◽  
Turbine Performance ◽  
Map Generation ◽  
Compressor Map ◽  
Performance Adaptation

Accurate gas turbine performance simulation is a vital aid to the operational and maintenance strategy of thermal plants having gas turbines as their prime mover. Prediction of the part load performance of a gas turbine depends on the quality of the engine’s component maps. Taking into consideration that compressor maps are proprietary information of the manufacturers, several methods have been developed to encounter the above limitation by scaling and adapting component maps. This part of the paper presents a new off-design performance adaptation approach with the use of a novel compressor map generation method and Genetic Algorithms (GA) optimization. A set of coefficients controlling a generic compressor performance map analytically is used in the optimization process for the adaptation of the gas turbine performance model to match available engine test data. The developed method has been tested with off-design performance simulations and applied to a GE LM2500+ aeroderivative gas turbine operating in Manx Electricity Authority’s combined cycle power plant in the Isle of Man. It has been also compared with an earlier off-design performance adaptation approach, and shown some advantages in the performance adaptation.

scholarly journals On-Board Condition Management for Aircraft Gas Turbines

1992 ◽  
Author(s):  
G. W. Gallops ◽  
F. D. Gass ◽  
M. H. Kennedy
Keyword(s):  
Control System ◽  
Gas Turbine ◽  
Condition Monitoring ◽  
System Architecture ◽  
Gas Turbines ◽  
Performance Model ◽  
Performance Models ◽  
Turbine Performance ◽  
Secondary Air ◽  
System Operating

A revolutionary approach to gas turbine condition monitoring is made possible by the recent development of accurate real-time gas turbine performance models. This paper describes an approach for an integrated condition management system operating concurrently with the gas turbine control system for improved availability, safety and economy. This paper considers the system subject to the requirements and constraints of aircraft gas turbines. A system architecture is described based on a primary, gas path performance model with supplementary models representing the secondary air, fuel and lubrication systems and the rotor system dynamics. Measurement and processing requirements for the system are defined. Preflight, in-flight and postflight application and analysis by the gas turbine operator are discussed.

Performance Deterioration Modeling in Aircraft Gas Turbine Engines

1998 ◽  
Vol 120 (2) ◽  
pp. 344-349 ◽  
Author(s):  
A. V. Zaita ◽  
G. Buley ◽  
G. Karlsons
Keyword(s):  
Gas Turbine ◽  
Gas Turbines ◽  
Operating Time ◽  
Performance Model ◽  
Turbine Engines ◽  
Gas Turbine Engines ◽  
Engine Simulation ◽  
Performance Deterioration ◽  
Data Correlations ◽  
Design Characteristics

Steady-state performance models can be used to evaluate a new engine’s baseline performance. As a gas turbine accumulates operating time in the field, its performance deteriorates due to fouling, erosion, and wear. This paper presents the development of a model for predicting the performance deterioration of aircraft gas turbines. The model accounts for rotating component deterioration based on the aircraft mission profiles and environmental conditions and the engine’s physical and design characteristics. The methodology uses data correlations combined with a stage stacking technique for the compressor and a tip rub model, along with data correlations for the turbine to determine the amount of performance deterioration. The performance deterioration model interfaces with the manufacturer’s baseline engine simulation model in order to create a deteriorated performance model for that engine.

Computer Models for Education on Performance Monitoring and Diagnostics of Gas Turbines

2002 ◽  
Vol 30 (3) ◽  
pp. 204-218 ◽  
Author(s):  
K. Mathioudakis ◽  
A. Stamatis ◽  
A. Tsalavoutas ◽  
N. Aretakis
Keyword(s):  
Gas Turbine ◽  
Gas Turbines ◽  
Performance Monitoring ◽  
Measurement Data ◽  
Performance Model ◽  
Operating Conditions ◽  
Engine Operation ◽  
Monitoring And Diagnostics ◽  
Engine Components ◽  
The Way

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

Modeling and Simulation of Gas Turbine System on a Virtual Test Bed (VTB)

2012 ◽  
Author(s):  
Eshwarprasad Thirunavukarasu ◽  
Ruixian Fang ◽  
Jamil A. Khan ◽  
Roger Dougal
Keyword(s):  
Gas Turbine ◽  
Dynamic Modeling ◽  
Gas Turbines ◽  
Simulation Software ◽  
Performance Model ◽  
Simulation Environment ◽  
Test Bed ◽  
Turbine Engine ◽  
Virtual Test ◽  
Virtual Test Bed

Gas Turbine is a complex system and highly non linear in its overall performance. In order to study its impact on electric power quality under various load conditions, it is essential to create a high quality performance model of gas turbine to simulate its behavior in real time efficiently. This paper focuses on dynamic modeling of generic gas turbine model using alternate simulation environment for better feasibility. The model is developed on a virtual test bed which is an advanced dynamic simulation environment and can run a dynamic co-simulation effectively. The approach is by developing mathematical models of individual components of gas turbines and utilizing component performance map matching method to run the simulation. The paper discusses briefly about the VTB simulation environment and its use for dynamic modeling of gas turbine. The simulation studies carried out include design condition, off design condition and transient conditions. Working model of twin shaft turbine engine using compressor and turbine maps are validated with established gas turbine simulation software and results are shown.

Sign in / Sign up

Export Citation Format

Share Document