Dynamic Modeling of Gas Turbine Rotor Shaft Faults

Most imminent faults in gas turbines often emanate from the rotor shaft of the engine. Some of these faults that could lead to catastrophe include misalignment, imbalance, crack, and eccentricity. These defects are equally likely to lead to unscheduled downtime resulting in large economic losses to equipment owners. It is against this backdrop that the rotor shaft of a gas turbine system was isolated and used for this dynamic model to reduce downtime. A method of dynamic modeling was used to consider how the aforementioned faults could be addressed at the design stage of the gas turbine engine. Modeling and simulation of the faults were carried out, and the obtained results compared favorably with what theory suggests. It was observed that cracking, as the most prominent rotor shaft fault, could manifest even at a turbine speed of 7,264 rpm (0.18 m for 0.8776 mm/s vibration velocity amplitude). Artificial neural networks (ANN) were then used to validate and link the results together, which also confirmed the authenticity of the work. Also, a Visual Basic program was used in the course of the various simulations adopted for the modeling, with faults being randomized every 3000 ms, and outputs were easily displayed on desktop computer screen. The work therefore showed how an ANN could be integrated into the monitoring of gas turbine rotor shaft defects. In its totality, the monitoring technique metamorphosed into the development of a software code-named "The MICE" for monitoring essential performance parameters in gas turbine operations.