A Dynamic Method to Analyze Cold-Start First Cycles Engine-Out Emissions at Elevated Cranking Speed Conditions of a Hybrid Electric Vehicle Including a Gasoline Direct Injection Engine

The cold crank-start stage, including the first three engine cycles, is responsible for a significant amount of the cold-start phase emissions in a Gasoline Direct Injection (GDI) engine. The engine crank-start is highly transient due to substantial engine speed changes, Manifold Absolute Pressure (MAP) dynamics, and in-cylinder temperatures. Combustion characteristics change depending on control inputs variations, including throttle angle and spark timing. Fuel injection strategy, timing, and vaporization dynamics are other parameters causing cold-start first cycles analysis to be more complex. Hybrid Electric Vehicles (HEVs) provide elevated cranking speed, enabling technologies such as cam phasing to adjust the valve timing and throttling, and increased fuel injection pressure from the first firings. To analyze the engine-out emissions, including unburnt Hydrocarbon (HC), Nitrogen Oxides (NOx), Carbon monoxide (CO), and Carbon dioxide (CO₂), the measured emissions in mole fraction need to be quantified in mass per cycle per cylinder considering all dynamics mentioned above. This study proposes a new method to quantify individual-cylinder engine-out emissions event by event dynamically. The method consists of the individual-cylinder GT-Power Three Pressure Analysis (TPA), in-cylinder parameters estimation, fuel vaporization Computational Fluid Dynamics (CFD) analysis, and exhaust gas dynamics in the exhaust manifold. Experimental MAP, cylinder and exhaust pressures, injection pulse width, GT-Power estimated parameters, and air mass flow meter data are used for the new method calibration and validation. The estimated trapped air charge and the equivalent combusted fuel masses are the most critical parameters affecting the precision of calculating engine-out emissions on a mass basis. The results show that the trapped air charge is estimated with a 2.7 mg average error. In addition, the simulated Indicated Mean Effective Pressure (IMEP) as representative of the mass of fuel contributed to the combustion during the same event was validated with a 0.06 bar average error. Furthermore, the fuel path analysis is carried out to validate the post-oxidization coefficient and lost fuel portion calibrated values, showing 75.3% and 15.8% post-oxidization rate of unburnt HC and 18.5% and 20% lost fuel portion for high cranking speed/highly retarded and low cranking speed/advanced spark timing conditions, respectively.