Fixpoint Controller Simulation

Question

We already have a working and well tested plecs standalone simulation model. As a next step we would like to transform the control logic from plecs model into VHDL code for hardware implementation. There is no floating point unit on the FPGA available - the final controller must be of fixedpoint form. Is there any way to simulate a fixed point control logic?

Our first idea:

• Write some script that implements the controller. For example a python script relying on https://pypi.org/project/pyPhix/ for fixed point operations.
• Run this python script using the "DLL" block that can be found under Library browser -> Control -> Functions & Tables.
  Unfortunately we did not get this DLL block to execute any python code at all. The manual only explains the inclusion of C/C++. We could not find a MWE in python so far.
• Last step would be to transfer from python script to VHDL. This is fairly easy once all fixedpoint number format are known.

Any suggestions are warmly welcome!

 

Answer 1

Hi,

Unfortunately only DLLs using C and C++ are officially supported in PLECS. Googling gives a few option that you could try, but again this is fully experimental and not officially supported/tested by the Plexim developers.

1. Implement the fixed-point control algorithm in C using the FixMath library (http://www.nongnu.org/fixmath/doc/index.html)
2. Embed python into C, see e.g. https://docs.python.org/2/extending/embedding.html or https://www6.software.ibm.com/developerworks/education/l-pythonscript/l-pythonscript-ltr.pdf
3. It might be possible to create a dll from Python using Cython (see https://cython.org/#about or https://docs.python.org/3/faq/windows.html#is-a-pyd-file-the-same-as-a-dll)

Best regards,
NF