Symbolic Simplification Framework in a Modelica Compiler

Abstract: The aim of the study is to develop a framework for symbolic simplification algorithms in the JModelica.org compiler. It should make the procedure of adding new algorithms to the compiler easier. Instead of the fixed order currently used in the compiler, algorithms will perform their simplifications to the model equations iteratively. This opens up the possibility for more simplifications being made, when all algorithms have access to the constantly updated, reduced and simplified model equations. We have implemented such a framework in the JModelica.org compiler for the algorithms Alias Elimination and Variability Propagation. Also, we have designed a canonical form for equations to alleviate the implementation of new algorithms to the framework. The framework improves simulation time at the cost of increased compile time, whereas the effects of the canonical form remain to be evaluated.