Jodie Herbert completed Energy and Renewable: Machine Learning for the Power Grid

Built an end-to-end turbine yield predictor using leakage-free sensor inputs and a genuine time-based holdout. Compared a LinearRegression baseline with a tuned RandomForest, then evaluated the final model on held-out data with MAE, RMSE, R2, and a residual read. Also explained the model in plain language for a plant manager, including its main failure mode when inputs drift outside the trained range.

• Leakage-Free Feature Selection
• Time-Based Holdout Evaluation
• Baseline-to-Ensemble Comparison
• Plain-Language Model Explanation