Michaelle Mariam Ferdinand completed Energy and Renewable: Machine Learning for the Power Grid

Built a leakage-aware gas-turbine TEY predictor end to end: framed the target correctly, excluded downstream CO/NOx sensors, and validated on a locked later-year holdout rather than a random split. Compared a LinearRegression baseline with a tuned RandomForest, then documented held-out MAE, RMSE, R², residual behavior, limitations, and a practical monitoring plan for drift and retraining.

• Leakage-Free Feature Selection
• Time-Based Holdout Validation
• Baseline-to-Ensemble Comparison
• Honest Held-Out Evaluation