Work

Cap-Aware Boosting for Green Day-Ahead Electricity Price Forecasting

Year: 2026
Type: preprint
Abstract: Forecasting 15-minute Market Clearing Prices in the Indian Energy Exchange Green Day-Ahead Market (GDAM) is complicated by pronounced diurnal renewable cycles and frequent upper price-cap events ($10,000\text{ Rs/MWh}$). We propose a leak-safe, cap-aware LightGBM architecture that blends an $L_1$ log-price regressor with an upper-price classifier alongside nine rolling-calibrated quantile heads. Across 257 delivery days (January–September 2026; 24,672 blocks), our baseline model achieves $397.9\text{ Rs/MWh}$ MAE and $171.6\text{ Rs/MWh}$ mean pinball loss, cutting MAE by $31.8\%$ relative to a zero-shot univariate Chronos-2 model ($583.3\text{ Rs/MWh}$) and $32.5\%$ over persistence ($589.1\text{ Rs/MWh}$). On a volatile October–November 2026 holdout, decoupling uncapped price regression from cap classification while adding pre-gate covariates—same-morning real-time prices, prior-day unmatched sell volumes, and capacity-weighted clear-sky solar irradiance—further reduces MAE to $716.0\text{ Rs/MWh}$ versus $2,000.4\text{ Rs/MWh}$ for Chronos-2. Domain-informed covariates within cap-aware boosting drive accuracy gains far more effectively than replacing tabular ensembles with univariate foundation models.
Language: English
Open Access status: green
OpenAlex ID: W7235661956
Cites: 8
Cited by:
Related to: 10