Processforge
Plan / Apply Workflow
Validate flowsheets, preview structural diffs, and apply changes with warm-start and automatic homotopy fallback.
Model, plan, and solve process flowsheets in Python.
Processforge brings an equation-oriented solver, dynamic simulation, and a plan/apply workflow to steady-state and transient process engineering — with versioned snapshots and multiple backends.
processforge — terminal
Core capabilities
Built for process engineering workflows
Equation-Oriented Solver
Global Newton-Raphson with Armijo line search. Pluggable backends: SciPy, Pyomo/IPOPT, and CasADi.
Convergence Guardrails
Automatic rollback to the last good snapshot and divergence reports when solves fail.
Dynamic Simulation
ODE time-marching for tank-based flowsheets with sequential-modular propagation and Wegstein convergence.
Modelica & FMU Export
Transpile flowsheets to native Modelica source and compile to Model Exchange or co-simulation FMUs.
JSON Flowsheet DSL
Declarative configuration with schema validation, unit connectivity checks, and automatic recycle detection.
Who it's for
Built for the people who design, simulate, and ship process systems
Who it's for
- Process engineers building steady-state and dynamic flowsheet models.
- Simulation developers who need a Python-native, scriptable solver backend.
- Researchers coupling thermal, hydraulic, and custom unit models in one framework.
What it can do for you
- Ship faster with a declarative JSON DSL and
pf init → plan → applyworkflow. - Trust convergence via snapshot versioning, warm starts, and automatic rollback.
- Pick your backend — SciPy, Pyomo/IPOPT, or CasADi — without rewriting the model.
Get started
Install from PyPI
pip install processforge