Does a Sequential Modular Approach (SMA) algorithm exist for solving component-based systems of non-linear equations


I am polling this board to see if anyone knows if an algorithm exists in any Julia package to perform the Sequential Modular Approach (SMA) technique for solving systems of connected components where each component is a compilation of non-linear algebraic equations.

This approach is commonly used in flowsheeting software, for example IDAES, which uses a Python pkg, called Pyomo, that uses SequentialDecomposition (this has many resembalnces to `ModelingToolkit in my opinion).

There is an argument in the flowsheeting software world for using either SMA vs Equation-Oriented Approach (EOA); they each have pros and cons. I have used MTK already to perform EOA successfully, although I am curious if the SMA algorithm exists or if it is being considered for development. I assume it could fit in nicely within the MTK framework.

Thanks, Chris

MTK generates something similar from its tearing results, especially with ODAEProblem. We plan to develop some more tools around this directly shortly as well.

Thank you for the status update. Looking forward to it! Please let me know if there is a best way to stay in the loop for when these developments are released.