Short-Term Electricity Load Forecasting with Machine Learning
Open accessThe DOI matches its Crossref record; 33 of the 33 references it cites were found in a registry or an index.
Abstract
An accurate short-term load forecasting (STLF) is one of the most critical inputs for power plant units’ planning commitment. STLF reduces the overall planning uncertainty added by the intermittent production of renewable sources; thus, it helps to minimize the hydrothermal electricity production costs in a power grid. Although there is some research in the field and even several research applications, there is a continual need to improve forecasts. This research proposes a set of machine learning (ML) models to improve the accuracy of 168 h forecasts. The developed models employ features from multiple sources, such as historical load, weather, and holidays. Of the five ML models developed and tested in various load profile contexts, the Extreme Gradient Boosting Regressor (XGBoost) algorithm showed the best results, surpassing previous historical weekly predictions based on neural networks. Additionally, because XGBoost models are based on an ensemble of decision trees, it facilitated the model’s interpretation, which provided a relevant additional result, the features’ importance in the forecasting.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
A tutorial on support vector regression
Smola, A. J., & Schölkopf, B. (2004) · Statistics and Computing · doi:10.1023/b:stco.0000035301.49549.88 · Cited by 13,387 · cited 2×
“SVR is a regression version of the support vector machine (SVM), which was initially designed for classification problems [65].”
MethodAlgorithm aversion: People erroneously avoid algorithms after seeing them err.
Dietvorst, B. J., Simmons, J. P., & Massey, C. (2015) · Journal of Experimental Psychology: General · doi:10.1037/xge0000033 · Cited by 3,232 · cited 1×
“Studies have shown that many decision-makers exhibit an inherent distrust of automated predictive models, even if they are proven to be more accurate than human forecasters [60].”
Backgroundxarray: N-D labeled Arrays and Datasets in Python
Hoyer, S., & Hamman, J. (2017) · Journal of Open Research Software · doi:10.5334/jors.148 · Cited by 1,856 · cited 1×
“Weather data is available on daily NetCDF files, which can be treated with netCDF [53] and xarray [54] to select the desired variables and subsequently merge all the datasets into a unique modeling dataset.”
MethodZheng, H., Yuan, J., & Chen, L. (2017) · Energies · doi:10.3390/en10081168 · Cited by 635 · cited 1×
“Despite the good XGB performance, some authors recommend training the model based on similar days to enhance the forecast [34,35].”
MethodAn optimized model using LSTM network for demand forecasting
Abbasimehr, H., Shabani, M., & Yousefi, M. (2020) · Computers & Industrial Engineering · doi:10.1016/j.cie.2020.106435 · Cited by 538 · cited 1×
“An example of forecasting the next 24 h load from a smart grid, comparing LSTM results with a back-propagation ANN and SVR, demonstrates that LSTM can offer a MAPE of 1.9% against 3.3% from ANN and 4.8% of SVR [41].”
MethodInfluentialShort-term hourly load forecasting using time-series modeling with peak load estimation capability
Amjady, N. (2001) · IEEE Transactions on Power Systems · doi:10.1109/59.962429 · Cited by 515 · cited 4×
“New variables related to the date–time index are created to feed the ML models with this extra information about time, with this being one of the most critical steps for STLF [22,48].”
MethodInfluentialMassaoudi, M., Refaat, S. S., Chihi, I., Trabelsi, M., Oueslati, F. S., & Abu-Rub, H. (2021) · Energy · doi:10.1016/j.energy.2020.118874 · Cited by 400 · cited 1×
“Another approach to enhance the holidays’ forecast is to develop a specific model for holidays and merge the predictions with a regular days model [16] and use stacking techniques to enhance the forecast accuracy, even though it increases the training and predicting time [73].”
MethodAl-Musaylh, M. S., Deo, R. C., Adamowski, J. F., & Li, Y. (2018) · Advanced Engineering Informatics · doi:10.1016/j.aei.2017.11.002 · Cited by 300 · cited 2×
“Based on the results from References [10,36], ANN for STLF can outperform other forecasting methods if a robust hyperparameter optimization is performed to avoid the issues related to ANN tuning.”
ResultApplied Predictive Analytics : Principles and Techniques for the Professional Data Analyst
Abbott, D. (2014) · Cited by 283
Singh, P., & Dwivedi, P. (2018) · Applied Energy · doi:10.1016/j.apenergy.2018.02.131 · Cited by 236 · cited 1×
“Another solution joining ANN with ensemble approaches is presented in Reference [39], where the authors seek to improve ANN generalization ability using bagging-boosting.”
MethodTight and Compact MILP Formulation of Start-Up and Shut-Down Ramping in Unit Commitment
Morales-Espana, G., Latorre, J. M., & Ramos, A. (2013) · IEEE Transactions on Power Systems · doi:10.1109/tpwrs.2012.2222938 · Cited by 235 · cited 1×
“Consequently, an efficient thermal power plant dispatch can be achieved by minimizing the unit commitment production-transmission costs for the power system [7,8].”
BackgroundXue, P., Jiang, Y., Zhou, Z., Chen, X., Fang, X., & Liu, J. (2019) · Energy · doi:10.1016/j.energy.2019.116085 · Cited by 196 · cited 2×
“Based on the results from References [10,36], ANN for STLF can outperform other forecasting methods if a robust hyperparameter optimization is performed to avoid the issues related to ANN tuning.”
BackgroundMethodRelative evaluation of regression tools for urban area electrical energy demand forecasting
Johannesen, N. J., Kolhe, M., & Goodwin, M. (2019) · Journal of Cleaner Production · doi:10.1016/j.jclepro.2019.01.108 · Cited by 174 · cited 1×
“Nevertheless, their results can be interpreted, and some researchers used it as a baseline model [64].”
ResultKhwaja, A. S., Anpalagan, A., Naeem, M., & Venkatesh, B. (2020) · Electric Power Systems Research · doi:10.1016/j.epsr.2019.106080 · Cited by 163 · cited 1×
“A study to forecast the 10-day streamflow for a hydroelectric dam used a decomposition-based methodology to compare XGB and SVR [38].”
MethodYu, X. J., Wang, Y., Wu, L., Chen, G., Wang, L., & Qin, H. (2019) · Journal of Hydrology · doi:10.1016/j.jhydrol.2019.124293 · Cited by 152 · cited 1×
“The authors of Reference [37] proposed a multi-step-ahead forecasting methodology using XGB and SVR to forecast hourly heat load, where a “direct” and “recursive” forecasting strategy are compared.”
MethodEseye, A. T., Lehtonen, M., Tukia, T., Uimonen, S., & Millar, R. J. (2019) · IEEE Access · doi:10.1109/access.2019.2924685 · Cited by 131
Ensemble learning for electricity consumption forecasting in office buildings
Pinto, T. D. O., Praça, I., Vale, Z., & Silva, J. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.02.124 · Cited by 119 · cited 2×
“Nevertheless, researchers are starting to use it, showing outstanding performances against traditional methods [29,30].”
BackgroundMethodDeep learning versus traditional machine learning methods for aggregated energy demand prediction
Paterakis, N. G., Mocanu, E., Gibescu, M., Stappers, B., & van Alst, W. (2017) · 2017 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe) · doi:10.1109/isgteurope.2017.8260289 · Cited by 118 · cited 1×
“From the most straightforward persistence method, proposed in Reference [19], which follows the basic rule of “today equals tomorrow”, until the most recent deep learning algorithms, as exposed in the review article of Reference [20].”
MethodLebotsa, M. E., Sigauke, C., Bere, A., Fildes, R. A., & Boylan, J. E. (2018) · Applied Energy · doi:10.1016/j.apenergy.2018.03.155 · Cited by 109 · cited 1×
“Varying from very short-term applications, like forecasting the next 900 s for machine tools [14], moving to a few hours [15], predicting the day-ahead, which is the most common [16], and 48 h ahead [17], to weekly forecasts [18].”
BackgroundAtef, S., & Eltawil, A. (2020) · Electric Power Systems Research · doi:10.1016/j.epsr.2020.106489 · Cited by 108 · cited 2×
“The KNN method searches for the k most similar instances; when the k most similar samples are found, the target is obtained by local interpolation of the targets associated with the k found instances [42].”
MethodMachine learning based very short term load forecasting of machine tools
Dietrich, B., Walther, J., Weigold, M., & Abele, E. (2020) · Applied Energy · doi:10.1016/j.apenergy.2020.115440 · Cited by 107 · cited 3×
“This two-step approach [14] attempts to find a global optimum without digging through all the hyperparameter grids in the first step, reducing computational time, and refining the result by searching the second step.”
BackgroundResearch on Short-Term Load Forecasting Using XGBoost Based on Similar Days
Liao, X., Cao, N., Ma, L., & Kang, X. (2019) · International Conference on Intelligent Transportation, Big Data & Smart City (ICITBS) · doi:10.1109/icitbs.2019.00167 · Cited by 100 · cited 1×
“The authors of Reference [33] suggest using XGB, including weather variables and historical load, to forecast the hourly weekly load of a power plant.”
MethodDesigning a short-term load forecasting model in the urban smart grid system
Li, C. (2020) · Applied Energy · doi:10.1016/j.apenergy.2020.114850 · Cited by 98 · cited 3×
“For example, in the 2006 study of Reference [10], an ANN is proposed with a Levenberg-Marquardt training algorithm to forecast hourly, daily, and weekly load in Ontario, Canada, presenting good results without comparing with other algorithms.”
BackgroundMethodElectrical load prediction of healthcare buildings through single and ensemble learning
Cao, L., Li, Y., Zhang, J., Jiang, Y., Han, Y., & Wei, J. (2020) · Energy Reports · doi:10.1016/j.egyr.2020.10.005 · Cited by 87 · cited 1×
“SVR is frequently used for STLF with the linear [25] or radial basis function (RBF) kernel [66], identifying load patterns better than other linear models [67].”
MethodLoad and Renewable Energy Forecasting for a Microgrid using Persistence Technique
Dutta, S., Li, Y., Venkataraman, A., Costa, L. M., Jiang, T., Plana, R., Tordjman, P., Choo, F. H., Foo, C. F., & Puttgen, H. B. (2017) · Energy Procedia · doi:10.1016/j.egypro.2017.12.736 · Cited by 84 · cited 1×
“From the most straightforward persistence method, proposed in Reference [19], which follows the basic rule of “today equals tomorrow”, until the most recent deep learning algorithms, as exposed in the review article of Reference [20].”
MethodComparison of three short-term load forecast models in Southern California
Zhang, N., Li, Z., Zou, X., & Quiring, S. M. (2019) · Energy · doi:10.1016/j.energy.2019.116358 · Cited by 83 · cited 1×
“Weather data is available on daily NetCDF files, which can be treated with netCDF [53] and xarray [54] to select the desired variables and subsequently merge all the datasets into a unique modeling dataset.”
MethodLiu, X., Zhang, Z., & Song, Z. (2019) · Renewable and Sustainable Energy Reviews · doi:10.1016/j.rser.2019.109632 · Cited by 83
A Comparative Study of Predictive Approaches for Load Forecasting in Smart Buildings
Hadri, S., Naitmalek, Y., Najib, M., Bakhouya, M., Fakhri, Y., & Elaroussi, M. (2019) · Procedia Computer Science · doi:10.1016/j.procs.2019.09.458 · Cited by 61 · cited 3×
“Nevertheless, researchers are starting to use it, showing outstanding performances against traditional methods [29,30].”
BackgroundThe Price of InterpretabilityPreprint
Bertsimas, D., Delarue, A., Jaillet, P., & Martin, S. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1907.03419 · Cited by 54 · cited 1×
“One way to overcome “algorithm aversion” is to provide them with interpretability [61].”
BackgroundWeather Based Day-Ahead and Week-Ahead Load Forecasting using Deep Recurrent Neural Network
Zou, M., Fang, D., Harrison, G. P., & Djokic, S. (2019) · IEEE International Forum on Research and Technology for Society and Industry (RTSI) · doi:10.1109/rtsi.2019.8895580 · Cited by 28 · cited 1×
“Varying from very short-term applications, like forecasting the next 900 s for machine tools [14], moving to a few hours [15], predicting the day-ahead, which is the most common [16], and 48 h ahead [17], to weekly forecasts [18].”
BackgroundLong-term projections of the hourly electricity consumption in Danish municipalities
Andersen, F. M., Henningsen, G., Møller, N., & Larsen, H. V. (2019) · Energy · doi:10.1016/j.energy.2019.115890 · Cited by 20 · cited 1×
“However, other trending consumption patterns [77], like recharging more electric vehicles and having more solar production behind-the-meter, will produce residential and commercial hourly consumption changes.”
BackgroundForecasting the Portuguese Electricity Consumption using Least-Squares Support Vector Machines
Ferreira, P., Cuambe, I. D., Ruano, A. E., & Pestana, R. (2013) · IFAC Proceedings Volumes · doi:10.3182/20130902-3-cn-3020.00138 · Cited by 7 · cited 2×
“Similarly, the authors of Reference [17] proposed forecasting the 48 h of Portuguese electricity consumption by using SVR as a better alternative after previously submitting the ANN’s use for the same task in Reference [28].”
BackgroundMethodBoya‐Lara, C. (2019) · International Engineering, Sciences and Technology Conference (IESTEC) · doi:10.1109/iestec46403.2019.00-88 · Cited by 4 · cited 2×
“…_2 m _ in _ Panama _ city h , with temperature being the most important weather variable due to its positive relationship with electricity load [58], as illustrated in Figure 1 obtain good results. remains in the hourly granularity, the best regressor is expressed as: 𝐿…”
Background
Metadata
- Authors
- Aguilar Madrid, E., & Antonio, N.
- Venue
- Information
- Publisher
- MDPI AG
- Published
- 2021-01-22
- Volume
- 12
- Issue
- 2
- Pages
- 50
- ISSN
- 2078-2489 (ISSN-L)
- OpenAlex
- W3122860865
- Cited by
- 224
- References
- 33
- FWCI
- 8.21
- Type
- journal article
- Language
- en
- Open access
- gold
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.