Probabilistic Hourly Load Forecasting Using Additive Quantile Regression Models

被引:18
作者
Sigauke, Caston [1 ]
Nemukula, Murendeni Maurel [2 ]
Maposa, Daniel [2 ]
机构
[1] Univ Venda, Dept Stat, Private Bag X5050, ZA-0950 Thohoyandou, South Africa
[2] Univ Limpopo, Dept Stat & Operat Res, Private Bag X1106, ZA-0727 Sovenga, South Africa
基金
新加坡国家研究基金会;
关键词
additive quantile regression; Lasso; load forecasting; generalised additive models; PEAK ELECTRICITY DEMAND; SHORT-TERM;
D O I
10.3390/en11092208
中图分类号
TE [石油、天然气工业]; TK [能源与动力工程];
学科分类号
0807 ; 0820 ;
摘要
Short-term hourly load forecasting in South Africa using additive quantile regression (AQR) models is discussed in this study. The modelling approach allows for easy interpretability and accounting for residual autocorrelation in the joint modelling of hourly electricity data. A comparative analysis is done using generalised additive models (GAMs). In both modelling frameworks, variable selection is done using least absolute shrinkage and selection operator (Lasso) via hierarchical interactions. Four models considered are GAMs and AQR models with and without interactions, respectively. The AQR model with pairwise interactions was found to be the best fitting model. The forecasts from the four models were then combined using an algorithm based on the pinball loss (convex combination model) and also using quantile regression averaging (QRA). The AQR model with interactions was then compared with the convex combination and QRA models and the QRA model gave the most accurate forecasts. Except for the AQR model with interactions, the other two models (convex combination model and QRA model) gave prediction interval coverage probabilities that were valid for the 90%, 95% and the 99% prediction intervals. The QRA model had the smallest prediction interval normalised average width and prediction interval normalised average deviation. The modelling framework discussed in this paper has established that going beyond summary performance statistics in forecasting has merit as it gives more insight into the developed forecasting models.
引用
收藏
页数:21
相关论文
共 38 条
[1]  
Abuella M, 2017, NORTH AMER POW SYMP
[2]  
[Anonymous], MODELING FORECASTING
[3]  
[Anonymous], R PACKAGE HIERNET VE
[4]  
[Anonymous], 2017, DOING MAGIC ANAL SEA
[5]   A LASSO FOR HIERARCHICAL INTERACTIONS [J].
Bien, Jacob ;
Taylor, Jonathan ;
Tibshirani, Robert .
ANNALS OF STATISTICS, 2013, 41 (03) :1111-1141
[6]   A novel multi-time-scale modeling for electric power demand forecasting: From short-term to medium-term horizon [J].
Boroojeni, Kianoosh G. ;
Amini, M. Hadi ;
Bahrami, Shahab ;
Iyengar, S. S. ;
Sarwat, Arif I. ;
Karabasoglu, Orkun .
ELECTRIC POWER SYSTEMS RESEARCH, 2017, 142 :58-73
[7]   Performance Analysis of Short-Term Electricity Demand with Atmospheric Variables [J].
Chapagain, Kamal ;
Kittipiyakul, Somsak .
ENERGIES, 2018, 11 (04)
[8]  
Chikobvu D, 2012, J ENERGY SOUTH AFR, V23, P23
[9]   Stacking Ensemble Learning for Short-Term Electricity Consumption Forecasting [J].
Divina, Federico ;
Gilson, Aude ;
Gomez-Vela, Francisco ;
Torres, Miguel Garcia ;
Torres, Jose E. .
ENERGIES, 2018, 11 (04)
[10]   Dynamic factors in periodic time-varying regressions with an application to hourly electricity load modelling [J].
Dordonnat, Virginie ;
Koopman, Siem Jan ;
Ooms, Marius .
COMPUTATIONAL STATISTICS & DATA ANALYSIS, 2012, 56 (11) :3134-3152