Probabilistic Load Forecasting Based on Adaptive Online Learning

Load forecasting is crucial for multiple energy management tasks such as scheduling generation capacity, planning supply and demand, minimizing trade costs. Such relevance has increased even more in recent years due to the integration of renewable energies, electric cars, microgrids. Conventional load techniques obtain single-value forecasts by exploiting consumption patterns past demand. However, cannot assess intrinsic uncertainties capture dynamic changes patterns. To address these problems, this paper presents a method probabilistic based on adaptive online learning hidden Markov models. We propose with theoretical guarantees, experimentally their performance scenarios. In particular, we develop that update model parameters recursively, sequential prediction using most parameters. The evaluated datasets corresponding regions have different sizes display assorted time-varying results show proposed can significantly improve existing wide range