The coronavirus disease 2019 (COVID-19) pandemic has urged many governments in the world to enforce a strict lockdown where all nonessential businesses are closed and citizens are ordered to stay at home. One of the consequences of this policy is a significant change in electricity consumption patterns. Since load forecasting models rely on calendar or meteorological information and are trained on historical data, they fail to capture the significant break caused by the lockdown and have exhibited poor performances since the beginning of the pandemic. In this paper we introduce two methods to adapt generalized additive models, alleviating the aforementioned issue. Using Kalman filters and fine-tuning allows to adapt quickly to new electricity consumption patterns without requiring exogenous information. The proposed methods are applied to forecast the electricity demand during the French lockdown period, where they demonstrate their ability to significantly reduce prediction errors compared to traditional models. Finally, expert aggregation is used to leverage the specificities of each predictions and enhance results even further.

We present the winning strategy for the IEEE DataPort Competition on Day-Ahead Electricity Load Forecasting: Post-Covid Paradigm. This competition was organized to design new forecasting methods for unstable periods such as the one starting in Spring 2020. First, we pre-process the data with a statistical correction of the meteorological variables. Second, we apply standard statistical and machine learning models. Third, we rely on state-space models to adapt the aforementioned forecasters. It achieves the right compromise between two extremes. Indeed, machine learning methods allow to learn complex dependence to explanatory variables on a historical data set but fail to forecast non-stationary data accurately. Conversely, purely time-series models such as autoregressives are adaptive in essence but fail to capture dependence to exogenous variables. Finally, we use aggregation of experts, and we leverage the diversity of the set of obtained forecasters to improve our final predictions. The evaluation period of the competition was the occasion of trial and error and we put the focus on the final procedure.