Modeling, Identification and Prediction of Inherent quasi-stationary Pressure Dynamics of a Common-Rail System using Neuro-Fuzzy Structures with Local Linear ARX models

Abstract

In the opening there is a short overview for the current approaches found in the literature regarding the methods and models used for the Common Rail diesel high pressure dynamics identification. The local linear Neuro-Fuzzy models are proposed as an alternative to the conventional analytical and empirical models. In the following, the diesel Common Rail system structure is presented along with the basic physical equations governing the process. A short analysis reveals the main factors influencing the fuel pressure behavior inside the Common Rail high pressure system. Finally, it is analyzed the feasibility for the high pressure system identification and modeling using a particular Neuro-Fuzzy network structure with local linear dynamic models by training its parameters with engine measured data.