Smoothed Bernstein Online Aggregation for Short-Term Load Forecasting in IEEE DataPort Competition on Day-Ahead Electricity Demand Forecasting: Post-COVID Paradigm

We present a winning method of the IEEE DataPort Competition on Day-Ahead Electricity Demand Forecasting: Post-COVID Paradigm. The day-ahead load forecasting approach is based novel online forecast combination multiple point prediction models. It contains four steps: i) data cleaning and preprocessing, ii) new holiday adjustment procedure, iii) training individual models, iv) by smoothed Bernstein Online Aggregation (BOA). flexible can quickly adjust to energy system situations as they occurred during after COVID-19 shutdowns. ensemble models ranges from simple time series sophisticated like generalized additive (GAMs) high-dimensional linear estimated lasso. They incorporate autoregressive, calendar, weather effects efficiently. All steps contain concepts that contribute excellent performance proposed method. especially true for procedure fully adaptive BOA approach.