Extreme Learning Machine-Based Diagnostics for Component Degradation in a Microturbine

Micro turbojets are used for propelling radio-controlled aircraft, aerial targets, and personal air vehicles. When compared to full-scale engines, they characterized by relatively low efficiency durability. In this context, the degraded performance of gas path components could lead an unacceptable reduction in overall engine performance. work, a data-driven model based on conventional artificial neural network (ANN) extreme learning machine (ELM) was estimating degradation micro turbojet. The training datasets containing data with were generated using validated GSP Monte Carlo approach. particular, compressor turbine simulated three different flight regimes. It confirmed that component had similar impact than at sea level. Finally, testing process ELM algorithm four input vectors. Two vectors extensive number virtual sensors, other two reduced just fuel flow exhaust temperature. Even small high prediction accuracy maintained takeoff cruise but slightly worse variable conditions.