Title : ( Bootstrap Aggregating as an ensembled machine learning algorithm for power consumption prediction under asymmetric loss with linear model-base learners )
Authors: Mahdi Doostparast , Mohammad Pouyani , Mohammad Hossein Yaghmaee Moghaddam ,
دانلود فایل برای اعضای دانشگاهAccess to full-text not allowed by authors
Abstract
It is necessary to know the amount of electric energy demand in each period in order to plan accurately and adopt appropriate policies. Therefore, the prediction of power consumption and electric energy demand is important for different economic sectors. This paper proposes two critical concepts for power consumption predictions in electricity distribution systems. The former deals with an asymmetric measure of inaccuracy, and the later provides an ensemble approach to wrap up the existing predictors. These two milestones are explored in details. For illustration purposes, a real data set on power consumption from 38 distribution transformers during the last two years at MEEDC is analyzed.
Keywords
Consumption Pattern; Ensemble Learning; LINEX Loss; Machine Learning; Prediction; Squared Error Loss
@inproceedings{paperid:1094396,
author = {Doostparast, Mahdi and محمد پویانی and Yaghmaee Moghaddam, Mohammad Hossein},
title = {Bootstrap Aggregating as an ensembled machine learning algorithm for power consumption prediction under asymmetric loss with linear model-base learners},
booktitle = {بیست و هفتمین کنفرانس بین المللی شبکه های توزیع نیروی برق},
year = {2023},
location = {مشهد, IRAN},
keywords = {Consumption Pattern; Ensemble Learning; LINEX Loss; Machine Learning; Prediction; Squared Error Loss},
}
%0 Conference Proceedings
%T Bootstrap Aggregating as an ensembled machine learning algorithm for power consumption prediction under asymmetric loss with linear model-base learners
%A Doostparast, Mahdi
%A محمد پویانی
%A Yaghmaee Moghaddam, Mohammad Hossein
%J بیست و هفتمین کنفرانس بین المللی شبکه های توزیع نیروی برق
%D 2023
