Adaptive Observer for Joint Estimation of Oxygen Fractions and Blend Level in Biodiesel Fueled Engines

In this paper, a dynamic oxygen fraction model for a biodiesel-compatible engine with a dual-loop exhaust gas recirculation (EGR) system is developed. When oxygenated fuel is applied in an engine, the intake manifold oxygen fraction, which is an important factor for both combustion and emissions, can be chosen as a new reference for evaluating the equivalent EGR level instead of EGR ratio. Based on this model, an adaptive observer is designed, and it is able to simultaneously estimate the oxygen fraction states and unknown fuel blend level. The adaptive observer introduced here is advantageous for its simple convergence condition and its efficient implementation. The analysis of the observer´s convergence and robustness is detailed. The performance of the observer is validated by the simulation and experimental results. The difference of intake oxygen fractions between pure diesel (B0) and pure soybean biodiesel (B100) are studied. The adaptive observer is expected to be valuable for adaptive control of the combustion and emissions of biodiesel-compatible engines.