Univ logo Fuel composition and engine control using state-space and neural network By KinYip Chan Supervised by Prof. Andrzej Ordys, Dr. Olga Duran, Dr. Olga Duran, Dr. Konstantin Volkov Dr. Konstantin Volkov Dr. Jiamei Deng Dr. Jiamei Deng UKACC PhD Presentation Showcase
Univ logo UKACC PhD Presentation Showcase Slide 2 Introduction Fuel composition is supply by suppliers Named Gasoline(Petrol) or Diesel Contains hydrocarbon atoms Mix with Methanol, Ethanol (Alcohol) Composition varies Fuel identification Further Engine Control
Univ logo UKACC PhD Presentation Showcase Slide 3 Motivation and methodology Background and motivation for research Lower emissions Better fuel economy Improve engine performance Research methodology Emission – non-linear behaviour (CO2, CO, O2, NOx) Computer based engine simulation Fuel estimation C H O System identification – state-space, LQR, neural network Tune controller Current status System identified Different speed and torque Real engine simulationReal engine simulation
Univ logo UKACC PhD Presentation Showcase Slide 4 Engine Simulation Model
Univ logo UKACC PhD Presentation Showcase Slide 5 Engine Simulation Model
Univ logo UKACC PhD Presentation Showcase Slide 6 Result of SS, LQR and NN Emissions CO2, O2, CO, NOx
Univ logo UKACC PhD Presentation Showcase Slide 7 Different fuel mixture Isooctane C8H18 Methanol C1H4O1 Ethanol C2H6O1
Univ logo UKACC PhD Presentation Showcase Slide 8 Conclusion or summary or future work Engine system identification Non-linear behaviours (emissions) State space Controller stabilise by adding neural network Fuel estimation Hydrocarbon atoms Future work Different speed Real engine data
Univ logo End of Presentation Thank you for your attention UKACC PhD Presentation Showcase Slide 9