Performance assessment of classical and fractional controllers for transient operation of gas turbine engines

Abstract

The nonlinear behaviour of gas turbine engines has motivated the development of advanced controllers for ensuring their safe and reliable operation. In this paper, the problem of controller design for a two-shaft industrial gas turbine is addressed. Specifically, a transient dynamic engine model has been developed in MATLAB/Simulink for assessing the performance behaviour of the engine. Observed engine behaviour during transient manoeuvres has enabled the development of a PI controller capable of ensuring a smooth gas turbine operation. The performance of the gas turbine engine implementing the developed PI controller has been also compared to a fractional PI controller. Results demonstrate and illustrate the remarkable impact that transient engine simulation has in the development of robust controllers.

Transient Behavior in Variable Geometry Industrial Gas Turbines: A Comprehensive Overview of Pertinent Modeling Techniques

“…Similarly, Park [ 24 ] has also developed a hybrid dynamic model of a distributed energy system having small gas turbine, diesel engine, fuel cell, solar source and synchronous machine. Several other researchers have established dynamic models for design of controller and effective control strategy for actuation of VIGVs and fuel valves [ 25 , 26 , 27 , 28 , 29 , 30 , 31 ]. Similarly, Kong and Kim [ 32 ] have focused on performance optimization and controller design of a turbojet engine.…”

Robust controller design for a three-shaft industrial gas turbine in the infinite grid power generation

“…In the field of the gas turbine, the provided controllers can be classified into robust and non-robust controllers. Several studies have been conducted in the field of robust controller design for gas turbine in order to investigate optimal (Watts et al, 1992), adaptive (Comporeale et al, 2002), predictive (Jurado and Capiro, 2006, Ghorbani et al, 2008; Jadhav et al, 2014; Tsoutsanis and Meskin, 2018), Nero-Fuzzy (Mohamed Iqbal and Joseph Xavier, 2017), and classical Proportional–Integral (PI) and Proportional–Integral–Derivative (PID) controller design (Balamurugan et al, 2013; Hajagos and Berube, 2001; Kim and Kim, 2003; Kim, 2004; Montazeri-Gh and Ilkhani, 2012; Selvakumar et al, 2013). Dynamic models in different controller designs were often presented by the Rowen and IEEE working group (deMello, 1994; Rowen 1983, 1992).…”