Search Results

The phenomenon of fuel-film dynamics for four-stroke small-scale spark-ignition engines is investigated in this paper. A first-order fuel-film model, so-called tau-x model, is used to represent the fuel dynamics. The parameters of fuel-film model, which consists of the portion of fuel that deposited on the manifold wall and the time constant of the fuel evaporation process, are identified using the recursive least squared technique. Performances of the proposed algorithm are evaluated using a nonlinear engine model in Matlab/Simulink. The preliminary simulation results show that the proposed algorithm can accurately be used to identify the parameters of fuel-film model. The experimental data are then utilized to study the characteristics of fuel-film dynamics, and show that the fuel-film dynamics is significantly affected by engine speed, throttle opening, injection timing, and intake temperature.

For a conventional scooter engine, not only the crankshaft position estimation is insufficient based on the one-tooth crankshaft wheel, but also the speed measurement might be contaminated by sensor noise easily. The authors propose a technique using Kalman filter to estimate crankshaft position and engine speed for digital ignition control of a scooter engine. A 125cc engine model, which is verified by the experimental data of the target engine, will be used to design the Kalman filter. A simulation model, which consists of nonlinear engine dynamics, powertrain dynamics, tire dynamics, and pitch plane dynamics, is used to evaluate the performance of proposed estimation algorithm with different tooth numbers of crankshaft wheel and various noise conditions.

In order to study the noise effect of the crank angle sensor on electronic fuel injection (EFI) control system, a Kalman filter with stroke identification is employed to estimate the crankshaft rotational dynamics. Estimated crank angle and speed are then used for EFI system. A 125 c.c. scooter engine verified by the experimental data is used to design the Kalman filter. A simulation model, which consists of nonlinear engine dynamics, powertrain dynamics, tire dynamics, and pitch-plane motorcycle dynamics, is established in Matlab/Simulink to evaluate the performance of the Kalman filter at various noise conditions.

This paper develops an engine model for the model-in-the-loop (MIL) and hardware-in-the-loop (HIL) application to shorten the time duration and reduce the costs of developing and verifying the engine management system (EMS). The target engine is a 1.0L V-type two cylinder water-cooled spark-ignition engine. The engine model is developed using a so-called modulization method, which includes to: (1) separate the sub-models according to the different physical phenomena; (2) collect the sub-models to establish a library; (3) execute the component modules based on a pre-determined sequence by a more flexible way. The engine model is then applied in MIL structure for testing and verifying the control strategies in the developed EMS. After all strategies are verified, the HIL structure is constructed by a hardware controller and a virtual engine in the xPC target. The execution time-step of engine model is analyzed to keep enough accuracy and numeric stability for real-time simulation.