We are proud to announce that JModelica.org release 1.7 is now available for download. The main highlights of this release are:
- Improved support for hybrid systems, including friction models and ideal diodes. The improvements include solution of mixed systems of equations, which significantly strengthens the FMI-based simulation capabilities of JModelica.org.
- Support for tearing of equation systems. A graph-theoretical algorithm for computation of tearing variables of equation systems has been implemented, which enables faster simulation times of FMUs.
- Support for external FORTRAN functions. This feature enables use of the LAPACK functions in the ModelicaStandard Library.
- Support for function inlining. In order to simplify models containing functions, typically used in media libraries, inlining of function calls has been implemented. In particular, this feature is useful for simplification of models used for optimization.
- Support for export of Modelica functions in stand-alone DLLs. This feature is useful for exporting Modelica functions, e.g., media functions into stand-alone DLLs (shared object files) for interfacing in third party tools.
- Refactorization of the JModelica.org Python code: a new stand-alone package, PyFMI, is offered. Scripting and simulation of Functional Mock-Up Units (FMUs) in Python is of interest for many FMI-compliant tools. Therefore, the PyFMI package is now provided both as part of JModelica.org and as a stand-alone package.
- A new dynamic optimization algorithm for DAEs implemented in Python based on collocation and CasADi has been implemented. The new algorithm provides significantly improved flexibility and performance.
With the 1.7 release, we introduce compliance reports, where we provide diagnostics on what models in the Modelica Standard Library are supported. The purpose is to assist users in assessing the level of Modelica compliance in JModelica.org.
For additional information, see the release notes. A binary installer for Windows is available at the download page.