Supported Analyses
eMachineSim is exposed publicly through the eMachineSim Python API. The table below summarizes the user-facing workflows currently documented for public use.
| Analysis area | Python entry point | Typical samples | Status |
|---|---|---|---|
| Steady thermal conduction | eMachineSim.thermal.run_file, eMachineSim.Session.solve | Examples/thermal/bdf, Examples/thermal/realistic_motor_steady | Main documented workflow. |
| Thermal network / FEMH coupling | eMachineSim.thermal.run_file | realistic_motor_steady | Supports external cooling, thermal equivalent circuits, and heat-balance diagnostics. |
| BDF/QVOL thermal import | eMachineSim.thermal.run_bdf, eMachineSim.run_file | QVOL_loss.bdf, QVOL_loss_mix.bdf | Supports selected Nastran bulk cards for thermal preprocessing. |
| Surface/gap thermal coupling | eMachineSim.thermal.run_file | realistic_motor_steady | Includes face pairing diagnostics and node-pair lumped coupling. |
| Structural static analysis | eMachineSim.structural.run_static | Examples/structural/rotor | Includes 2D rotor structural examples and contact-oriented diagnostics. |
| Modal/NVH analysis | eMachineSim.modal.run | Examples/structural/modal_stator | Includes modal, force mapping, frequency response, circumferential order, and simplified acoustic indicators. |
| Electrostatic analysis | eMachineSim.run_file | Examples/electrostatic/ring, Examples/electrostatic/3by3 | Smaller legacy/sample workflow; kept visible for future electric/electrostatic expansion. |
Recommended Tool for Each Workflow
Use eMachineSim.run_file() when you already have a JSON input file:
import eMachineSim
result = eMachineSim.run_file("input.json", r"path\to\run_directory")
Use eMachineSim.Session when a script changes parameters and solves repeatedly:
session = eMachineSim.Session()
session.initialize(input_json, run_directory)
session.update_input({"some": "patch"})
result = session.solve()
session.finalize()
Using SKILLS / MCP + Coding AI
For motor thermal analysis and NVH workflows, users do not need to memorize every JSON key or diagnostic CSV. eMachineSim SKILLS can guide a coding AI through example selection, input.json preparation, analysis execution, and CSV result checks.
The eMachineSim MCP server makes this workflow easier for AI agents by exposing structured tools for Examples, input JSON inspection, thermal heat-balance summaries, surface assignment checks, gap coupling, interface resistance, and cooling path summaries. Detailed setup belongs to the public repository; this documentation describes the benefits and workflow intent.
Current Scope Notes
- The Python API is JSON-first; it does not expose every internal assembly step as a separate public function.
- Electrostatic examples are documented as supported samples, not as the main product focus today.
- Motor NVH output is a simplified relative indicator workflow, not a calibrated absolute sound-pressure solver.