Autonomous identification of fuel parameters for adaptive model-based control of large gas engines

The operational efficiency and performance of gas engines, such as those manufactured by INNIO Jenbacher GmbH & Co OG in Jenbach, are affected by a multitude of factors, including engine type, fuel quality, geographic location, systems aging, and other boundary conditions. These engines are increasingly used in different environments and subjected to various gas conditions, necessitating a robustness to gas quality changes during operation. However, the large number of possible variations represents a significant challenge in the complete parameterization of the controller. This thesis presents the adaptation and implementation of an algorithm for the autonomous identification of parameters that characterize the gas type and quality of fuel, from now referred to as Fuel Parameters Estimator or FPE. The algorithm was developed using MATLAB and Simulink via a combination of theoretical modeling and simulation. It was then validated on a real engine in an INNIO Jenbacher GmbH & Co OG test cell, demonstrating a significant improvement in the controller's ability to adapt to fuel quality changes. This work is a new step in the ongoing efforts to improve the adaptability and efficiency of gas engines, offering an innovative approach to controller parameterization. The identification of relevant fuel parameters simplifies the controller's tuning, improves the robustness of gas engines, makes them more relocatable and better suited for a wider range of applications. The findings of this research are significant for the future design and operation of gas engines in diverse and changing environments.