A Mathematical Model of a UAV Spark-Ignition Engine for the Design of Control Methods

This paper presents a real-time, nonlinear, control-oriented model for a two-stroke, spark-ignition aircraft engine. The safety and reliability of unmanned aerial vehicles (UAVs) are vital for their large-scale usage. Therefore, the design of control systems for normal as well as abnormal operation of UAVs is very essential. Timely detection and isolation of faults in an engine can save the aircraft from catastrophic consequences. Modeling is the first stage in the majority of control methods. This model is designed to be able to accurately and in real-time predict the output of an aircraft engine. Using existing modeling knowledge, a mean-value engine model is developed in this paper. The engine model consists of five submodels named the throttle body model, air dynamics model, fuel dynamics model, rotational dynamics model, and atmospheric model. The first four submodels are responsible for an accurate description of engine dynamics, while the atmospheric model covers the variation in the aerial environment. The simulation model is constructed using the MATLAB Simulink environment. The proposed engine model is validated with available data of a real UAV engine. The model is as generic as possible and can be employed as a simulation tool for control analysis and the implementation of model-base fault diagnosis schemes for UAV engines.