Humboldt-Universität zu Berlin - High Dimensional Nonstationary Time Series

IRTG1792DP2021 020

Advanced Statistical Learning on Short Term Load Process Forecasting

Junjie Hu
Brenda López Cabrera
Awdesch Melzer

Short Term Load Forecast (STLF) is necessary for effective scheduling, operation
optimization trading, and decision-making for electricity consumers. Modern and
efficient machine learning methods are recalled nowadays to manage complicated
structural big datasets, which are characterized by having a nonlinear temporal
dependence structure. We propose different statistical nonlinear models to
manage these challenges of hard type datasets and forecast 15-min frequency
electricity load up to 2-days ahead. We show that the Long-short Term Memory
(LSTM) and the Gated Recurrent Unit (GRU) models applied to the production line
of a chemical production facility outperform several other predictive models in
terms of out-of-sample forecasting accuracy by the Diebold-Mariano (DM) test
with several metrics. The predictive information is fundamental for the risk
and production management of electricity consumers.

Short Term Load Forecast, Deep Neural Network, Hard Structure Load Process

JEL Classification:
C51, C52, C53, Q31, Q41