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

Abstract

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 and more complicated and therefore demand more computational time although they are more flexible in applications, in particular for new gas turbine engines. 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 back propagation neural networks have been utilised 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 generalisation 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%. Overall, it can be concluded that the proposed neural network approach in creep life prediction is able to provide a good alternative to the more complicated model-based creep life prediction algorithms and can be applied to different types of gas turbine engines.

Gas Turbine Gas Path Fault Diagnosis based on Adaptive Nonlinear Steady-State Thermodynamic Model

“…From the linear GPA method developed by Urban in the late 1960's [13], a series of GPA methods were proposed, such as an adaptive nonlinear GPA method [7], artificial neural networks [3], rule-based expert system and rule-based fuzzy expert system [11], and genetic algorithm [4,12,17]. The merit of artificial intelligence methods, such as neural network, rough set [14], Bayesian network [9,10] and rule based expert system, is that they do not need a gas turbine performance model, as only the relation information between fault symptom and degradation is needed.…”

A methodology for energy savings verification in industry with application for a CHP (combined heat and power) plant

“…Gas turbines, with special focus on monitoring [38] and prediction [39] of performance, condition monitoring as an alternative to conventional scheduled maintenance [40], operation optimization of turbine compressor [41], health monitoring in military aircrafts engines [42] and turboshaft engines for helicopter propulsion [43], gas turbine diagnostics [44], fault isolation [45] with observation of model accuracy at partial load operation [46] and creep life prediction [47]. Boilers, with developed models aimed at predicting the performance of pulverized-coal boilers [48], predicting NO x emissions of CFB (circulating fluidised bed) boilers [49], estimating pollutant emissions in chain-grate stoker boilers [50], modelling NO x emissions in pulverized-coal boilers [51] with subsequent application of Genetic Algorithm for optimization purposes [52], predicting bottom ash depositing in pulverized-coal power plants [53] and reproducing the influence of fouling on the efficiency of biomass boilers [54].…”

Remaining Useful Life Prediction of Gas Turbine Engine using Autoregressive Model

“…These include support vector machines (SVM) [1,[7][8][9], Bayesian forecasting [10][11][12], Kalman Filters [10], state-space models [12], artificial neural networks [13][14][15][16], independent component analysis [17], regression techniques [18], Dempster-Shafer regression [1] and one parameter double exponential smoothing [10]. These techniques can be classified into two types: (i) physics based modelling approaches, and (ii) data driven models [5,[19][20][21].…”