Numerical and Experimental Investigation of a Low Order Radial Turbine Model for Engine-Level Optimization of Turbocharger Design

This paper presents the development and validation of a meanline model by means of numerical and experimental methods, to determine it's feasibility as an optimization tool for turbocharger matching. Using a parametric turbine model, numerical experiments were conducted accounting for variations of several key turbine design parameters and a wide operating range. The resulting dataset was used to test the accuracy of the meanline model when calibrated to a baseline design and thus evaluate it's ability of extrapolating to different designs. The loss models were examined in more detail, and a set of loss models which provided the most accurate results is presented. The meanline model was further validated experimentally using dynamometer test results of 6 turbine designs from the same parametric turbine model. The result showed that for design point and high power operation, an error of less than 3.1% and 2.0% was achieved for efficiency and mass flow parameters, respectively. This led to the conclusion that the model would be sufficiently accurate to represent design changes relevant to turbocharger matching.