Short-Term Load Interval Prediction Using a Deep Belief Network

Short-Term Load Interval Prediction Using a Deep Belief Network

In load predication, point-based forecasting methods have been widely applied. However, uncertainties arising in load predication bring significant challenges for such methods. This therefore drives the development of new methods amongst which interval predication is one of the most effective. In th...

Full description

Bibliographic Details
Main Authors: Xiaoyu Zhang, Zhe Shu, Rui Wang, Tao Zhang, Yabing Zha
Format: Article
Language:English
Published: MDPI AG 2018-10-01
Series:Energies
Subjects:
deep belief network
lower upper bound estimation method
short-term load prediction
interval predication
Online Access:http://www.mdpi.com/1996-1073/11/10/2744
id doaj-fc98fc09f34c4241884d27065fca1a01
record_format Article
spelling doaj-fc98fc09f34c4241884d27065fca1a012020-11-24T22:23:08ZengMDPI AGEnergies1996-10732018-10-011110274410.3390/en11102744en11102744Short-Term Load Interval Prediction Using a Deep Belief NetworkXiaoyu Zhang0Zhe Shu1Rui Wang2Tao Zhang3Yabing Zha4College of System Engineering, National University of Defense Technology, Changsha 410073, ChinaCollege of System Engineering, National University of Defense Technology, Changsha 410073, ChinaCollege of System Engineering, National University of Defense Technology, Changsha 410073, ChinaCollege of System Engineering, National University of Defense Technology, Changsha 410073, ChinaCollege of System Engineering, National University of Defense Technology, Changsha 410073, ChinaIn load predication, point-based forecasting methods have been widely applied. However, uncertainties arising in load predication bring significant challenges for such methods. This therefore drives the development of new methods amongst which interval predication is one of the most effective. In this study, a deep belief network-based lower–upper bound estimation (LUBE) approach is proposed, and a genetic algorithm is applied to reinforce the search ability of the LUBE method, instead of simulated an annealing algorithm. The approach is applied to the short-term load prediction on some realistic electricity load data. To demonstrate the effectiveness and efficiency of the proposed method, it is compared with three state-of-the-art methods. Experimental results show that the proposed approach can significantly improve the predication accuracy.http://www.mdpi.com/1996-1073/11/10/2744deep belief networklower upper bound estimation methodshort-term load predictioninterval predication
collection DOAJ
language English
format Article
sources DOAJ
author Xiaoyu Zhang
Zhe Shu
Rui Wang
Tao Zhang
Yabing Zha
spellingShingle Xiaoyu Zhang
Zhe Shu
Rui Wang
Tao Zhang
Yabing Zha
Short-Term Load Interval Prediction Using a Deep Belief Network
Energies
deep belief network
lower upper bound estimation method
short-term load prediction
interval predication
author_facet Xiaoyu Zhang
Zhe Shu
Rui Wang
Tao Zhang
Yabing Zha
author_sort Xiaoyu Zhang
title Short-Term Load Interval Prediction Using a Deep Belief Network
title_short Short-Term Load Interval Prediction Using a Deep Belief Network
title_full Short-Term Load Interval Prediction Using a Deep Belief Network
title_fullStr Short-Term Load Interval Prediction Using a Deep Belief Network
title_full_unstemmed Short-Term Load Interval Prediction Using a Deep Belief Network
title_sort short-term load interval prediction using a deep belief network
publisher MDPI AG
series Energies
issn 1996-1073
publishDate 2018-10-01
description In load predication, point-based forecasting methods have been widely applied. However, uncertainties arising in load predication bring significant challenges for such methods. This therefore drives the development of new methods amongst which interval predication is one of the most effective. In this study, a deep belief network-based lower–upper bound estimation (LUBE) approach is proposed, and a genetic algorithm is applied to reinforce the search ability of the LUBE method, instead of simulated an annealing algorithm. The approach is applied to the short-term load prediction on some realistic electricity load data. To demonstrate the effectiveness and efficiency of the proposed method, it is compared with three state-of-the-art methods. Experimental results show that the proposed approach can significantly improve the predication accuracy.
topic deep belief network
lower upper bound estimation method
short-term load prediction
interval predication
url http://www.mdpi.com/1996-1073/11/10/2744
work_keys_str_mv AT xiaoyuzhang shorttermloadintervalpredictionusingadeepbeliefnetwork
AT zheshu shorttermloadintervalpredictionusingadeepbeliefnetwork
AT ruiwang shorttermloadintervalpredictionusingadeepbeliefnetwork
AT taozhang shorttermloadintervalpredictionusingadeepbeliefnetwork
AT yabingzha shorttermloadintervalpredictionusingadeepbeliefnetwork
_version_ 1725765728270811136

Similar Items