7xx - Scripting & Automation¶

Programmatic control and external tool integration.

Learning Objectives¶

• Automate simulations with GeckoSCRIPT
• Integrate with MATLAB for parameter sweeps
• Co-simulate with Simulink models
• Create custom Java control blocks
• Use the REST API for remote control
• Control simulations from Python scripts

Integration Options Overview¶

Contents¶

701 - GeckoSCRIPT Basics¶

• GeckoSCRIPT.pdf - Complete scripting guide
• GeckoSCRIPT.ipes - Example script-controlled circuit
• buck_control.ipes / buck_control_tuning.m - Control tuning example

702 - MATLAB Integration¶

• GeckoSCRIPT_ example_matlab/ - MATLAB integration examples
• Loss calculation scripts (.scl)
• RMI interface for bidirectional communication

703 - Simulink Co-Simulation¶

• GeckoCIRCUITS_simulink_tutorial.pdf - Tutorial guide
• Gecko_VR1.mdl - Simulink model
• s_GeckoCIRCUITS.c - S-function source
• MEX files for various platforms

704 - Java Blocks¶

• demo_JAVA_Block.ipes - Java block introduction
• JavaBlockPMSM.ipes - PMSM control with Java
• Custom component development guide

705 - API Integration¶

• SIGNAL_ANALYSIS_API.md - API documentation
• websocket-client.html - WebSocket test client
• External signal source examples

706 - Python Integration (New)¶

• Parameter sweep automation
• Optimization with SciPy
• Data analysis with NumPy/Pandas
• Parallel simulation execution

Quick Start: Automation¶

GeckoSCRIPT (Built-in)¶

// Simple parameter sweep
for (D = 0.2; D <= 0.8; D += 0.1) {
    setParameter("PWM.1", "dutyCycle", D);
    runSimulation();
    Vout = getMeasurement("SCOPE.1", "ch1_avg");
    print("D=" + D + ", Vout=" + Vout);
}

MATLAB (RMI Interface)¶

% Connect to GeckoCIRCUITS
gecko = GeckoRemoteInterface('localhost', 1099);

% Run parameter sweep
D_values = 0.2:0.1:0.8;
for i = 1:length(D_values)
    gecko.setParameter('PWM.1', 'dutyCycle', D_values(i));
    gecko.runSimulation();
    Vout(i) = gecko.getMeasurement('SCOPE.1', 'ch1_avg');
end

Python (File-based or API)¶

import subprocess
import numpy as np

# Parameter sweep
D_values = np.linspace(0.2, 0.8, 7)
for D in D_values:
    # Run simulation
    subprocess.run(['java', '-jar', 'gecko.jar',
                   '--param', f'D={D}', 'buck.ipes'])
    # Analyze results
    results = np.loadtxt('output.csv')
    print(f"D={D:.1f}, Vout={results[:,1].mean():.2f}")

Use Case Selection Guide¶

Related Tutorials¶

• 104 - Running Simulations - Batch mode basics
• 105 - Analysis Tools - Automated analysis