Accurate gas turbine performance models are crucial in many gas turbine performance analysis and gas path diagnostic applications. With current thermodynamic performance modeling techniques, the accuracy of gas turbine performance models at off-design conditions is determined by engine component characteristic maps obtained in rig tests and these maps may not be available to gas turbine users or may not be accurate for individual engines. In this paper, a nonlinear multiple point performance adaptation approach using a genetic algorithm is introduced with the aim to improve the performance prediction accuracy of gas turbine engines at different off-design conditions by calibrating the engine performance models against available test data. Such calibration is carried out with introduced nonlinear map scaling factor functions by “modifying” initially implemented component characteristic maps in the gas turbine thermodynamic performance models. A genetic algorithm is used to search for an optimal set of nonlinear scaling factor functions for the maps via an objective function that measures the difference between the simulated and actual gas path measurements. The developed off-design performance adaptation approach has been applied to a model single spool turbo-shaft aero gas turbine engine and has demonstrated a significant improvement in the performance model accuracy at off-design operating conditions.

At off-design conditions, engine performance model prediction accuracy depends largely on its component characteristic maps. With the absence of actual characteristic maps, performance adaptation needs to be done for good imitations of actual engine performance. A nonlinear multiple point genetic algorithm based performance adaptation developed earlier by the authors using a set of nonlinear scaling factor functions has been proven capable of making accurate performance predictions over a wide range of operating conditions. However, the success depends on searching the right range of scaling factor coefficients heuristically, in order to obtain the optimum scaling factor functions. Such search ranges may be difficult to obtain and in many off-design adaption cases, it may be very time consuming due to the nature of the trial and error process. In this paper, an improvement on the present adaptation method is presented using a least square method where the search range can be selected deterministically. In the new method, off-design adaptation is applied to individual off-design point first to obtain individual off-design point scaling factors. Then plots of the scaling factors against the off-design conditions are generated. Using the least square method, the relationship between each scaling factor and the off-design operating condition is generated. The regression coefficients are then used to determine the search range of the scaling factor coefficients before multiple off-design points performance adaptation is finally applied. The developed adaptation approach has been applied to a model single-spool turboshaft engine and demonstrated a simpler and faster way of obtaining the optimal scaling factor coefficients compared with the original off-design adaptation method.

Fault classification has become one of the main features in gas turbine health monitoring. Hence techniques such as gas path analysis, artificial neural networks, expert systems, fuzzy logic and many others have been developed for this purpose in the past. In this paper, an alternative rough set based diagnostic method using enhanced fault signatures combined with three fault classification frameworks for gas turbine fault classification have been introduced, i.e. Framework 1 with a single step to classify single and dual component faults, Framework 2 with the first step to identify weather it is a single or dual component faults and the second step to identify the individual faults, and Framework 3 with the first step to identify faults associated with component types and the second step to identify the individual faults. Such frameworks have been applied to the fault classification of a model two-spool turbofan gas turbine engine implemented with single and dual component faults to test the effectiveness of the frameworks. It has been demonstrated in the application that all three framework configurations can provide satisfactory fault classification and that Framework 1 has higher fault classification success rate than that of Frameworks 2 and 3. In addition, Frameworks 2 and 3 have better performance in identifying fault types than Framework 1.

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.

At off-design conditions, engine performance model prediction accuracy depends largely on its component characteristic maps. With the absence of actual characteristic maps, performance adaptation needs to be done for good imitations of actual engine performance. A non-linear multiple point Genetic Algorithm based performance adaptation developed earlier by the authors using a set of non-linear scaling factor functions has been proven capable of making accurate performance prediction over a wide range of operating conditions. However, the success depends on searching the right range of scaling factor coefficients heuristically, in order to obtain optimum scaling factor functions. Such search ranges may be difficult to obtain and in many off-design adaption cases, it may be very time consuming due to the nature of trial and error process. In this paper, an improvement on the present adaptation method is presented using a Least Square method where the search range can be selected deterministically. In the new method, off-design adaptation is applied to individual off-design point first to obtain individual off-design point scaling factors. Then plots of the scaling factors against the off-design conditions are generated. Using the Least Square method, the relationship between each scaling factor and the off-design operating condition is generated. The regression coefficients are then used to determine the search range of the scaling factor coefficients before multiple off-design points performance adaptation is finally applied. The developed adaptation approach has been applied to a model single-spool turboshaft engine and demonstrated a simpler and faster way of obtaining the optimal scaling factor coefficients compared with the original off-design adaptation method.