The electricity sector presents new challenges in the operation and planning of power systems, such as the forecast of power demand. This paper proposes a comprehensive approach for evaluating statistical methods and techniques of electric demand forecast. The proposed approach is based on smoothing methods, simple and multiple regressions, and ARIMA models, applied to two real university buildings from Ecuador and Spain. The results are analyzed by statistical metrics to assess their predictive capacity, and they indicate that the Holt-Winter and ARIMA methods have the best performance to forecast the electricity demand (ED).
|Title of host publication||Advances in Emerging Trends and Technologies - Volume 2|
|Editors||Miguel Botto-Tobar, Joffre León-Acurio, Angela Díaz Cadena, Práxedes Montiel Díaz|
|Number of pages||12|
|State||Published - 1 Jan 2020|
|Event||1st International Conference on Advances in Emerging Trends and Technologies, ICAETT 2019 - quito, Ecuador|
Duration: 29 May 2019 → 31 May 2019
|Name||Advances in Intelligent Systems and Computing|
|Conference||1st International Conference on Advances in Emerging Trends and Technologies, ICAETT 2019|
|Period||29/05/19 → 31/05/19|
Bibliographical notePublisher Copyright:
© 2020, Springer Nature Switzerland AG.
- Electric demand
- Load Uncertainties
- Statistical methods