Analytic Evaluation of Engine NVH Robustness Due to Manufacturing Variations

In an effort to improve quality, shorten engine development times, and reduce costly and time-consuming experimental work, analytic modeling is being used upstream in the product development process to evaluate engine robustness to noise factors. This paper describes a model-based method for evaluating engine NVH (Noise, Vibration, and Harshness) robustness due to manufacturing variations for a statistically significant engine population. A brief discussion of the cycle simulation model and its capabilities is included. The methodology consists of Monte Carlo simulations involving several noise factors to obtain the crank-angle resolved response of the combustion process and Fourier analysis of the resulting engine torque. Further analysis of the Fourier results leads to additional insights regarding the relative importance of and sensitivity to the individual noise factors. While the cost and resources required to experimentally evaluate a large engine population can be prohibitive, the analytical modeling proved to be a cost-effective way of analyzing the engine robustness taking into account manufacturing process capability.