Uncertainty and Sensitivity Analysis of Gas Pulsations in the Suction Manifold of a Multi-Cylinder Automotive Compressor

All compressor simulation models have uncertainties in model parameters and variability in operational parameters. The thermodynamic and noise performance of a multi-cylinder, automotive compressor suction process are investigated in this study. The main focus is on uncertainty and variability in compressor models to improve model accuracy and identify feasible design solutions that are robust to sources of variability. Uncertainty Analysis and Sensitivity Analysis tools are used to assess and quantify the effects of variability and uncertainty in compressor simulation models. Based on initial sensitivity analysis of a few model input parameters, it is shown that the output response is most sensitive to the valve model parameters. The valve model is the key factor in determining the mass flow rate that is subsequently used in determining gas pulsations in the suction manifold of a compressor. The compressor simulation model is used to identify the combination of input variables that produce the maximum and minimum levels of gas pulsations and also to select the design and operation parameters which result in the lowest levels.