This repository contains a Python-based motor simulator designed to aid in control algorithm development. The simulation allows fine-grained control over motor and application settings and supports real-time switching behavior and dead-time effects.

• Simulates motor electrical and mechanical dynamics for Synchronous motors (Asynchronous TBD)
• Takes into account mutual and self inductances, as well as inductance saturation and inductance time derivatives
• Accounts for dead-time and switching behavior
• Supports back-emf harmonics
• Allows the simulation of short-circuiting all phases
• Supports MTPA (Maximum torque per ampere) LUT (Lookup table) generation based on motor parameters
• Supports loading external MTPA LUTs
• Supports AFC (Adaptive feed forward cancellation) control for harmonic attenuation
• Supports decoupling
• Features a simple battery model
• Supports an input filter (RLC network) from the battery
• Supports unbalanced phases

The simulator integrates electrical and mechanical motor equations to simulate motor behavior. A current vector is commanded, and following a current loop, the FOC voltages Vd and Vq are calculated. These voltages are transformed to transistor states (Taking dead-time into account). The phase currents are found by solving the voltage equations. And the user can plot anything they wish:

• Phase currents
• Phase voltages
• Torque
• Speed
• Inductances (Self, mutual, dq)
• etc.

Modify parameters by directly adjusting the config class initialization in the script:

class Config:
    def __init__(self):
        '''
        Initializes all script parameters:

        Args:
        ...

Parameter units and explanations are in the script.

# Instantiate objects
config = Config()
motor = Motor(config)
sim = Simulation(config)
app = Application(config)
control = MotorControl(config)
lut = LUT(config)

# Uncomment to show closed loop bode plots of q and d axes:
# estimate_BW(control, app)

# If direct torque command is desired, a LUT is required. Generate or upload.
if (app.torque_command_flag):
    if (app.generate_lut):
        # Calculates this motor's MTPA LUT
        lut.mtpa_gen(motor, app)
    else:
        # Upload LUTs from text file
        iq_table, id_table = extract_iq_id_tables()
        lut.mtpa_lut = np.stack((iq_table, -id_table), axis=-1) / 10


# Run the simulation
simulate_motor(motor, sim, app, control, lut)

# Plot results
...

This project requires the following Python modules:

• control: https://python-control.readthedocs.io/en/0.10.1/index.html
• scipy
• matplotlib
• numpy
• tqdm: https://tqdm.github.io/
• re
• tkinter
• math

This project is licensed under the MIT License - see the LICENSE file for details.

• Inspired by Ben Katz's motor simulator: https://build-its-inprogress.blogspot.com/2016/12/some-motor-math-and-simulation.html. I saw he built a motor simulator in MATLAB which made me want to build a motor simulator in Python.