This assignment focuses on using Bayesian Optimization (BO) to maximize the yield strength of an alloy by optimizing its composition and processing conditions.
The yield strength of a metal determines the point at which it begins to plastically deform. Maximizing yield strength is crucial for:
- Aerospace applications
- Automotive industry
- Nautical engineering
A common strengthening mechanism is the formation of vanadium carbide precipitates, which:
- Inhibit atomic plane movement
- Provide thermal and chemical stability
- Form based on specific processing conditions
| Parameter | Range | Description |
|---|---|---|
| Vanadium Content | 1-5 wt% | Weight percentage of vanadium |
| Temperature | 500-1100°C | Aging temperature |
| Time | 0.5-24 hours | Aging duration |
| Process | CR/RX | Cold Rolling (CR) or Recrystallization (RX) |
t = 12 # hours
temperature = 800 # °C
v_prct = 3 # weight percentage of Vanadium
process = "RX" # recrystallization
ys = measure_yield_strength(t, temperature, v_prct, process)
print(ys)Use Honegumi to:
- Generate optimization template
- Configure parameter space
- Set up experiment with 25-trial budget
- Find optimal parameters
- Store results in
optimal_params - Record best yield strength in
optimal_yield_strength
- Use
get_feature_importances() - Analyze parameter significance
- Store results in
feature_importances
- Perform cross-validation using
cross_validate() - Calculate diagnostics with
compute_diagnostic() - Record:
- Correlation coefficient in
corr_coeff - Root mean squared error in
rmse
- Correlation coefficient in
- Analyze parameter perturbations (±3%)
- Generate stability heatmap
- Report minimum performance impact
- Python 3.8+
- Required packages listed in
requirements.txt
# Environment will be automatically set up in GitHub Codespace
# Manual setup if needed:
pip install -r requirements.txtpytest- The objective function
measure_yield_strength()is provided inutils.py - Limited to 25 experimental trials
- Consider practical parameter variations in production settings