Search

A deep learning approach is developed to swiftly evaluate phase change in computational fluid dynamics (CFD) simulations of a multi-component, liquid–gas two-phase injection flow. This method significantly improves computational efficiency by using a deep feedforward neural network (DFNN) to replace the complex iterative solution of multi-species vapor–liquid equilibrium (VLE). The DFNN takes instantaneous pressure, temperature, and system composition as input and predicts the corresponding phase equilibrium state. A parametric study was conducted to optimize the neural network's hyperparameters, including the activation function, number of hidden layers, and neurons per hidden layer. The rate of phase change is then calculated as a linear relaxation toward phase equilibrium, guiding subsequent computational steps in the CFD solver. A case study was performed to test the proposed methodology, involving the injection of a superheated liquid ethanol–water mixture into a gaseous nitrogen environment. The simulation results and computational cost were examined. It is found that the DFNN model, while accurately representing the non-ideal non-equilibrium phase change of a multi-component injection flow, speeds up the VLE solution by four orders of magnitude, leading to a 30%–40% reduction in overall flow simulation time. This model shows promise for injection flow simulations, especially for systems with a large number of compositions, such as sustainable aviation fuels. [ABSTRACT FROM AUTHOR]

Accurate predictions of fuel spray behavior and mixture formation in simulations of direct-injection spark-ignition (DISI) engines are fundamental to ensure proper description of all subsequent processes including ignition, combustion, and emissions. In this work, the spray evolution in a single-cylinder optical DISI engine was studied experimentally and numerically with the goal of enabling predictive computational fluid dynamics (CFD) modeling of in-cylinder sprays. The authors explored a wide range of operating conditions characterized by several fuel injection temperatures and engine speeds, using a well-characterized nine-component gasoline surrogate known as PACE-20. The effect of flash boiling and intake crossflow on the spray is discussed, with a focus on evaluating the ability of the spray models to capture highly transient spray behavior. In the experiments, the fuel temperature was varied between 20°C and 80°C, allowing for non-flash- to flash-boiling transition to emerge with enhanced flashing intensity at the highest temperatures. Spray collapse resulted in vapor-rich regions, owing to the locally lower inertia of the fluid. Varying the engine speed from 650 to 1950 rpm promoted increasingly more turbulent in-cylinder crossflow which interacted with the spray during the injection event and resulted in enhanced spray dispersion. The CFD model was able to capture the spray morphology transition at different fuel temperatures and engine speeds adequately. It is shown that the spray breakup model could capture the transitional spray behavior induced by flash boiling atomization and intake flow via proper initialization of the spray cone angle and calibration of the spray models' constants. [ABSTRACT FROM AUTHOR]