An Integrated Gaussian Process Modeling Framework for Residential Load Prediction

Guangrui Xie, Xi Chen, Yang Weng

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

While adding new capabilities, the distributed energy resource proliferation raises great concern about challenges such as dynamic fluctuations of voltages. For robust and efficient operational planning purposes, we propose an integrated Gaussian process (IGP) modeling framework for reliable hourly load prediction. The proposed IGP modeling framework has the following unique features. (1) IGP utilizes not only the data streams generated by the target customer but also those generated by relevant customers in the power system; an effective input space dimension reduction method is proposed to significantly improve the computational efficiency while maintaining the high predictive accuracy of IGP. (2) An adaptive data communication rate controlling scheme is proposed to further enhance the predictive performance of IGP, by optimally and dynamically adjusting the data communication rate used for each customer under the total data communication bandwidth constraint often imposed. Taking into account the highly uncertain load and generation behaviors of distributed energy resources (DERs), the proposed IGP framework is tested on various standard IEEE test cases with load and renewable generation data collected from real-world power systems with DERs. The superiority and efficacy of IGP are verified by our simulation results.

Original languageEnglish (US)
JournalIEEE Transactions on Power Systems
DOIs
StateAccepted/In press - Jun 28 2018

Fingerprint

Energy resources
Communication
Computational efficiency
Bandwidth
Planning
Electric potential

Keywords

  • adaptive sampling
  • Data communication
  • Gaussian processes
  • Kernel
  • Load forecasting
  • Load forecasting
  • Load modeling
  • Predictive models
  • renewable integration
  • Training

ASJC Scopus subject areas

  • Energy Engineering and Power Technology
  • Electrical and Electronic Engineering

Cite this

An Integrated Gaussian Process Modeling Framework for Residential Load Prediction. / Xie, Guangrui; Chen, Xi; Weng, Yang.

In: IEEE Transactions on Power Systems, 28.06.2018.

Research output: Contribution to journalArticle

@article{a2124d68d5604d6f888e0056709f143b,
title = "An Integrated Gaussian Process Modeling Framework for Residential Load Prediction",
abstract = "While adding new capabilities, the distributed energy resource proliferation raises great concern about challenges such as dynamic fluctuations of voltages. For robust and efficient operational planning purposes, we propose an integrated Gaussian process (IGP) modeling framework for reliable hourly load prediction. The proposed IGP modeling framework has the following unique features. (1) IGP utilizes not only the data streams generated by the target customer but also those generated by relevant customers in the power system; an effective input space dimension reduction method is proposed to significantly improve the computational efficiency while maintaining the high predictive accuracy of IGP. (2) An adaptive data communication rate controlling scheme is proposed to further enhance the predictive performance of IGP, by optimally and dynamically adjusting the data communication rate used for each customer under the total data communication bandwidth constraint often imposed. Taking into account the highly uncertain load and generation behaviors of distributed energy resources (DERs), the proposed IGP framework is tested on various standard IEEE test cases with load and renewable generation data collected from real-world power systems with DERs. The superiority and efficacy of IGP are verified by our simulation results.",
keywords = "adaptive sampling, Data communication, Gaussian processes, Kernel, Load forecasting, Load forecasting, Load modeling, Predictive models, renewable integration, Training",
author = "Guangrui Xie and Xi Chen and Yang Weng",
year = "2018",
month = "6",
day = "28",
doi = "10.1109/TPWRS.2018.2851929",
language = "English (US)",
journal = "IEEE Transactions on Power Systems",
issn = "0885-8950",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - An Integrated Gaussian Process Modeling Framework for Residential Load Prediction

AU - Xie, Guangrui

AU - Chen, Xi

AU - Weng, Yang

PY - 2018/6/28

Y1 - 2018/6/28

N2 - While adding new capabilities, the distributed energy resource proliferation raises great concern about challenges such as dynamic fluctuations of voltages. For robust and efficient operational planning purposes, we propose an integrated Gaussian process (IGP) modeling framework for reliable hourly load prediction. The proposed IGP modeling framework has the following unique features. (1) IGP utilizes not only the data streams generated by the target customer but also those generated by relevant customers in the power system; an effective input space dimension reduction method is proposed to significantly improve the computational efficiency while maintaining the high predictive accuracy of IGP. (2) An adaptive data communication rate controlling scheme is proposed to further enhance the predictive performance of IGP, by optimally and dynamically adjusting the data communication rate used for each customer under the total data communication bandwidth constraint often imposed. Taking into account the highly uncertain load and generation behaviors of distributed energy resources (DERs), the proposed IGP framework is tested on various standard IEEE test cases with load and renewable generation data collected from real-world power systems with DERs. The superiority and efficacy of IGP are verified by our simulation results.

AB - While adding new capabilities, the distributed energy resource proliferation raises great concern about challenges such as dynamic fluctuations of voltages. For robust and efficient operational planning purposes, we propose an integrated Gaussian process (IGP) modeling framework for reliable hourly load prediction. The proposed IGP modeling framework has the following unique features. (1) IGP utilizes not only the data streams generated by the target customer but also those generated by relevant customers in the power system; an effective input space dimension reduction method is proposed to significantly improve the computational efficiency while maintaining the high predictive accuracy of IGP. (2) An adaptive data communication rate controlling scheme is proposed to further enhance the predictive performance of IGP, by optimally and dynamically adjusting the data communication rate used for each customer under the total data communication bandwidth constraint often imposed. Taking into account the highly uncertain load and generation behaviors of distributed energy resources (DERs), the proposed IGP framework is tested on various standard IEEE test cases with load and renewable generation data collected from real-world power systems with DERs. The superiority and efficacy of IGP are verified by our simulation results.

KW - adaptive sampling

KW - Data communication

KW - Gaussian processes

KW - Kernel

KW - Load forecasting

KW - Load forecasting

KW - Load modeling

KW - Predictive models

KW - renewable integration

KW - Training

UR - http://www.scopus.com/inward/record.url?scp=85049339337&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85049339337&partnerID=8YFLogxK

U2 - 10.1109/TPWRS.2018.2851929

DO - 10.1109/TPWRS.2018.2851929

M3 - Article

AN - SCOPUS:85049339337

JO - IEEE Transactions on Power Systems

JF - IEEE Transactions on Power Systems

SN - 0885-8950

ER -