No abstract

This paper presents an autoignition combustion model, which can be used in conjunction with a one-dimensional gas dynamics and zero-dimensional thermodynamics engine simulation code for homogeneous charge compression ignition (HCCI) engine cycle simulations. The model consists of two parts: the first predicts autoignition timing, based on the integral of the knock delay time, and the second predicts the heat release rate, based on the Watson correlation for diesel engines. The equations for the model were fitted with empirical coefficients derived from experimental results. The experimental data were measured under a number of engine operating conditions for different loads, speeds, and equivalence ratios. The experiments were performed on a Jaguar V6 HCCI engine employing internal exhaust gas residual trapping through cam profile switching and variable valve timing control. It was found that the location of 50 per cent mass fraction burned (MFB) and the peak rate of pressure rise can be chosen as the key parameters charactering the HCCI combustion in engine simulations. The fitted combustion model was then implemented into the V6 engine Ricardo WAVE TM model and validation of its predictive calculations was obtained using a different set of experimental data. The results show that the predicted 50 per cent MFB position is within an average error of 1.84u crank angle (CA) and the predicted peak rate of pressure rise is within an average error of 0.53 bar/deg CA (7.2 per cent).

The wave rotor (pressure exchanger) is a device working based on a relatively simple idea of operation, but is challenging in its technical realization and difficult to simulate numerically. It has been common practice to create and use specialized codes for simulating the wave rotor operation. The current work presents an attempt of developing 2D and 3D models of radial and axial wave rotors using the commercial software package FLUENT. In this study geometrical models are used for the device casing and rotor cells. The application of carefully chosen initial and boundary conditions enabled the realization of relative motion of the rotor model. The vast information about the unsteady processes occurring during simulation are visualized. It occurs that such type of models are useful for the final test of devices, after the geometry was optimized by the use of specialized but much simpler 1D codes.

No abstract

No abstract