https://research.birmingham.ac.uk/portal/files/53643809/APEN2017.pdf

Performance-based health monitoring, diagnostics and prognostics for condition-based maintenance of gas turbines: A review

Prognostics and Health Management (PHM): Where are we and where do we (need to) go in theory and practice

We are performing the digital transition of industry, living the 4th industrial revolution, building a new World in which the digital, physical and human dimensions are interrelated in complex socio-cyber-physical systems. For the sustainability of these transformations, knowledge, information and data must be integrated within model-based and data-driven approaches of Prognostics and Health Management (PHM) for the assessment and prediction of structures, systems and components (SSCs) evolutions and process behaviors, so as to allow anticipating failures and avoiding accidents, thus, aiming at improved safe and reliable design, operation and maintenance. There is already a plethora of methods available for many potential applications and more are being developed: yet, there are still a number of critical problems which impede full deployment of PHM and its benefits in practice. In this respect, this paper does not aim at providing a survey of existing works for an introduction to PHM nor at providing new tools or methods for its further development; rather, it aims at pointing out main challenges and directions of advancements, for full deployment of condition-based and predictive maintenance in practice. 

Remaining useful life prediction based on health index similarity

Prognostics and health management: A review from the perspectives of design, development and decision

When a robust mathematical model of a process equipment is available, model-based diagnostic methods can be used to identify the occurrence of faults in a system. However, these methods are less effective when the non-linearity, complexity, and modeling uncertainties of the system increase. In recent years, a new discipline, known as artificial intelligence-based methods, has emerged, which allows the behavior of the system to be studied using operational data. While single-learner artificial neural network (ANN)-based models demonstrate a satisfactory level of capability in assessing the health of gas turbines, this article investigated the application of a multiple networks artificial neural network (multi-nets ANN) model using a multiple-views multiple learners approach to provide a real-time performance-based automatic fault detection (AFD) system in gas turbine engines. Towards this end, a number of key performance variables, which are commonly measurable on most industrial gas turbine engines, were monitored, and their associated ANNs were trained for healthy conditions. Two back-propagation training algorithms, namely the Levenberg–Marquardt and Bayesian regularization algorithms, and the k-fold cross-validation technique, were employed to train the optimal networks using a training data set. Using the trained multi-nets ANN model, two case studies were conducted pertaining to the detection of drop in compressor flow capacity and compressor fouling in an industrial 18.7-MW twin-shaft gas turbine engine. The ability of all the trained networks of the multi-nets model to detect these faulty conditions was investigated. The results obtained showed that among the four trained networks in the multi-nets model, the associated network for gas generator rotational speed was able to track these incidents earlier. The findings demonstrated the capabilities and performance of the proposed multi-nets model with regard to AFD in industrial gas turbines.

A multi-nets ANN model for real-time performance-based automatic fault diagnosis of industrial gas turbine engines

This study aimed at quantifying the effect of mechanical load on the performance of an 18.7 MW offshore gas turbine engine. The targeted engine is of two-shaft free power turbine configuration that operates as a mechanical driver for a process compressor in the gas compression service. The study is a part of a comprehensive performance health monitoring program to address the diagnostic and prognostic requirements in oil and gas offshore platforms and is motivated by the need to provide in-depth knowledge of the gas turbine engine performance. In this work, only the context of some design point key performance parameters and a limited set of collected operational data from the gas turbine in the real plant are available. Therefore, three major tasks, namely design point calculation, characteristic map tuning and off-design performance adaption, were needed to be performed. In order to check the validity of the proposed model, the obtained simulation results were compared with the operational data. The results indicate the maximum inaccuracy of the proposed model is 3.04 %. Finally, by employing the developed model, the engine capability for power generation when exposed to various load speeds is investigated. The obtained result demonstrates at the maximum gas generator speed, every 3 % decrease in mechanical speed leads to 1 % decline in the gas turbine power output. Moreover, when the gas turbine operates under design power load and mechanical speed is lower than 80 % of design speed, every 1 % decrease in load speed results in 0.2 % loss in thermal efficiency. The established relationship will assist proper assessment of mechanical drive gas turbines for performance health monitoring.

Performance evaluation of a twin-shaft gas turbine engine in mechanical drive service

Recent advances in hydrogen compressors for use in large-scale renewable energy integration

The effect of rigid bed proximity on flow parameters and hydrodynamic loads in offshore pipelines exposed to turbulent flow is investigated numerically. The Galerkin finite volume method is employed to solve the unsteady incompressible 2D Navier–Stokes equations. The large eddy simulation turbulence model is solved using the artificial compressibility method and dual time-stepping approach. The proposed algorithm is developed for a wide range of turbulent flows with Reynolds numbers of 9500 to 1.5×104. Evaluation of the developed numerical model shows that the proposed technique is capable of properly predicting hydrodynamic forces and simulating the flow pattern. The obtained results show that the lift and drag coefficients are strongly affected by the gap ratio. The mean drag coefficient slightly increases as the gap ratio increases, although the mean lift coefficient rapidly decreases. The vortex shedding suppression happen at the gap ratio of less than 0.2.

Mohammadreza Tahan

Papers

Performance-based health monitoring, diagnostics and prognostics for condition-based maintenance of gas turbines: A review

A multi-nets ANN model for real-time performance-based automatic fault diagnosis of industrial gas turbine engines

Performance evaluation of a twin-shaft gas turbine engine in mechanical drive service

Recent advances in hydrogen compressors for use in large-scale renewable energy integration

Effect of bed vicinity on vortex shedding and force coefficients of fluid flow on an offshore pipeline