An Analysis of the Effects of Torque, Engine Geometry, and Speed on Choosing an Engine Inertia Model to Minimize Prediction Errors

Abstract

Constant inertia engine models have been used for years by control engineers to attain real-time engine simulations. The accuracy of the estimation largely depends on the selection of the constant engine inertia value used in these engine models. A methodology that uses the least squares method to calculate the engine inertia value for constant inertia engine model is developed. This methodology enables the constant inertia engine model simulations to have minimum acceleration estimation errors. The effects of net external torque, engine geometry, and speed on the inertia value computation are also investigated. Since most of the constant inertia engine models use the averaged engine inertia as the constant inertia value, comparisons of these two inertia values in engine operation simulations are presented. Discussions of the advantages and disadvantages of both methods are also included. Precautions and restrictions of using the methodology are discussed as well.