Symbolic manipulation techniques for model simplification in object-oriented modelling of large scale continuous systems

In the present work, techniques for the symbolic manipulation of general no nlinear differential algebraic equation (DAE) systems are presented, and us ed for model simplification purposes, to support efficient simulation of la rge scale continuous systems in an object-oriented modelling environment. T he specific problems addressed are efficient elimination of trivial equatio ns by means of substitution, system block lower triangular (BLT) partitioni ng, and tearing, i.e. hiding of algebraic variables. Moreover, the weakenin g heuristic criterion for decoupling of large systems, via dynamic approxim ation, is studied. All these techniques have been successfully implemented and tested in MOSES (modular object-oriented software environment for simul ation), in order to define a complete model simplification process. The res ults achieved by applying the discussed algorithms and criteria on serial m ultibody systems are illustrated. A brief overview on further known symboli c manipulation techniques is also given, comparing them with the proposed o nes, throughout the paper. (C) 1998 IMACS/Elsevier Science B.V.