On model identification of engine for unmanned aerial vehicle

In the light of knowledge of statistics theory, the multi-element regression forecast algorithm is researched. Based on the experiments data of a two-stroke gasoline engine for unmanned aerial vehicle (UAV), a steady-state engine model is established by using a mathematical tool of stepwise regression. The procedure of modeling is described in detail. By comparing with experimental data, this model can describe the loading characteristic of the engine, and can achieve high fidelity model. The established engine model provides a basis for research of drive-train system and automatic control of UAV.