File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Improving Load Forecasting Performance via Sample Reweighting

TitleImproving Load Forecasting Performance via Sample Reweighting
Authors
KeywordsData inequality
influence function
load forecasting
sample reweighting
Issue Date1-Jul-2023
PublisherInstitute of Electrical and Electronics Engineers
Citation
IEEE Transactions on Smart Grid, 2023, v. 14, n. 4, p. 3317-3320 How to Cite?
AbstractLoad forecasting techniques have been well developed and almost all typical models are trying to minimize the error over all equally weighted data samples. However, treating all data equally in the training process is unreasonable and even harmful to model generalization. This letter investigates the issue of data inequality which is crucial but is overlooked in the field of load forecasting. To this end, we propose a novel load forecasting method via sample reweighting to tackle the data inequality issue. The proposed method determines the weights of samples using their gradients and influence functions in the process of training. Case studies on a real-world load dataset using linear models, artificial neural networks, and long short term memory in both deterministic and probabilistic situations all reveal that assigning different weights to various samples can improve forecasting accuracy.
Persistent Identifierhttp://hdl.handle.net/10722/339052
ISSN
2023 Impact Factor: 8.6
2023 SCImago Journal Rankings: 4.863
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorWang, CX-
dc.contributor.authorZhou, YZ-
dc.contributor.authorWen, QS-
dc.contributor.authorWang, Y-
dc.date.accessioned2024-03-11T10:33:30Z-
dc.date.available2024-03-11T10:33:30Z-
dc.date.issued2023-07-01-
dc.identifier.citationIEEE Transactions on Smart Grid, 2023, v. 14, n. 4, p. 3317-3320-
dc.identifier.issn1949-3053-
dc.identifier.urihttp://hdl.handle.net/10722/339052-
dc.description.abstractLoad forecasting techniques have been well developed and almost all typical models are trying to minimize the error over all equally weighted data samples. However, treating all data equally in the training process is unreasonable and even harmful to model generalization. This letter investigates the issue of data inequality which is crucial but is overlooked in the field of load forecasting. To this end, we propose a novel load forecasting method via sample reweighting to tackle the data inequality issue. The proposed method determines the weights of samples using their gradients and influence functions in the process of training. Case studies on a real-world load dataset using linear models, artificial neural networks, and long short term memory in both deterministic and probabilistic situations all reveal that assigning different weights to various samples can improve forecasting accuracy.-
dc.languageeng-
dc.publisherInstitute of Electrical and Electronics Engineers-
dc.relation.ispartofIEEE Transactions on Smart Grid-
dc.subjectData inequality-
dc.subjectinfluence function-
dc.subjectload forecasting-
dc.subjectsample reweighting-
dc.titleImproving Load Forecasting Performance via Sample Reweighting-
dc.typeArticle-
dc.identifier.doi10.1109/TSG.2023.3269205-
dc.identifier.scopuseid_2-s2.0-85153798025-
dc.identifier.volume14-
dc.identifier.issue4-
dc.identifier.spage3317-
dc.identifier.epage3320-
dc.identifier.eissn1949-3061-
dc.identifier.isiWOS:001017487000066-
dc.publisher.placePISCATAWAY-
dc.identifier.issnl1949-3053-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats