Promoting excellence in mobility engineering

  1. FISITA Store
  2. Technical Papers

Bond Graph Modeling and Forward Simulation of a Parallel Hybrid Electric Powertrain with Epicyclical Automatic Transmission
EAEC-07/PT-P10

Authors

Dragos N. Cruceru - University Politehnica of Bucharest
Ion Tabacu - University of Pitesti
Ioan-mircea Oprean - University Politehnica of Bucharest
Cistian N. Andreescu - University Politehnica of Bucharest

Abstract

Keywords: Hybrid Electric Vehicle, Automatic Transmission, Modeling and Simulation, Bond Graphs, Block Diagrams, Variable Dynamics - Systems.

The purpose of the paper is to develop a dynamic simulation model of a complex hybrid electric propulsion system. The powertrain system consists of an internal combustion engine, one or two electric motors and an epicyclical automatic transmission. All subsystems are modeled individually by means of the bond graph language and then linked together to form the complex powertrain system. Adequate, compatible powertrain subsystem models have been chosen (in terms of accuracy), in order to avoid algebraic loops or simulation blockings. Bond graph models of electric motors catch the basic physical phenomena of electric machines, and integrate well into the assembled system bond graph. Causality issues are of great importance when coupling the individual systems and components, as the authors emphasize throughout the paper.

The chosen modeling language is the bond graph, which is a link between physical (real) model and its mathematical (equations) description. Increased attention is given to methods of generalizing the powertrain system governing equations, as in the case of the automatic transmission model, which is designed to be flexible and easily changeable to other transmission models. The authors present three different original techniques for determining the governing equations of the automatic transmission. A computer code implements the algorithms and generates the motion equations, for all gears and shifts, automatically and in analytical way, before transferring them to the numerical simulation program. A 3D transmission matrix synthesizes all governing equations, thus the SimulinkTM model can be easily generalized for almost any kind of epicyclical AT.

Four hypotheses regarding drivetrain (rigid or elastic drive shafts, tire model) are taken into account, and a unique, generic SimulinkTM drivetrain model is given, valid for all drivetrain hypotheses. A series of breakaway acceleration simulations are performed with two simulation models. Conclusions are presented and discussed.

Add to basket