Abstract
AbstractProbabilistic forecasting of power consumption in a middle-term horizon (few months to a year) is a main challenge in the energy sector. It plays a key role in planning future generation plants and transmission grid. This paper proposes a novel model that (i) incorporates seasonality and autoregressive features in a traditional time-series analysis and (ii) includes weather conditions in a parsimonious machine learning approach, known as Gaussian Process. Applying to a daily power consumption dataset in North East England, provided by one of the largest energy suppliers, we obtain promising results in Out-of-Sample density forecasts up to one year, even using a small dataset, with only a two-year calibration set. For the evaluation of the achieved probabilistic forecasts, we consider the pinball loss—a metric common in the energy sector—and we assess the coverage—a procedure standard in the banking sector after the introduction of Basel II Accords—also running the conditional and unconditional tests for probability intervals. Results show that the proposed model outperforms benchmarks in terms of both accuracy and reliability.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have