MultiLoad-GAN: A GAN-Based Synthetic Load Group Generation Method Considering Spatial-Temporal Correlations

Hu, Yi; Li, Yiyan; Song, Lidong; Lee, Han Pyo; Rehm, P. J.; Makdad, Matthew; Miller, Edmond; Lu, Ning

doi:10.1109/tsg.2023.3302192

Title: MultiLoad-GAN: A GAN-Based Synthetic Load Group Generation Method Considering Spatial-Temporal Correlations

Journal Article · 04 August 2023 · IEEE Transactions on Smart Grid

DOI: https://doi.org/10.1109/tsg.2023.3302192 · OSTI ID:2329284

^[1];

^[2];

^[3];

^[3]; Rehm, P. J. ^[4]; Makdad, Matthew ^[5]; Miller, Edmond ^[5];

^[3]

North Carolina State University, Raleigh, NC (United States); North Carolina State University
North Carolina State University, Raleigh, NC (United States); Shanghai Jiao Tong University (China)
North Carolina State University, Raleigh, NC (United States)
ElectriCities, Raleigh, NC (United States)
New River Light and Power, Boone, NC (United States)

This paper presents a deep-learning framework, Multi-load Generative Adversarial Network (MultiLoad-GAN), for generating a group of synthetic load profiles (SLPs) simultaneously. The main contribution of MultiLoad-GAN is the capture of spatial-temporal correlations among a group of loads that are served by the same distribution transformer. This enables the generation of a large amount of correlated SLPs required for microgrid and distribution system studies. Here, the novelty and uniqueness of the MultiLoad-GAN framework are three-fold. First, to the best of our knowledge, this is the first method for generating a group of load profiles bearing realistic spatial- temporal correlations simultaneously. Second, two complementary realisticness metrics for evaluating generated load profiles are developed: computing statistics based on domain knowledge and comparing high-level features via a deep-learning classifier. Third, to tackle data scarcity, a novel iterative data augmentation mechanism is developed to generate training samples for enhancing the training of both the classifier and the MultiLoad-GAN model. Simulation results show that MultiLoad- GAN can generate more realistic load profiles than existing approaches, especially in group level characteristics. With little finetuning, MultiLoad-GAN can be readily extended to generate a group of load or PV profiles for a feeder or a service area.

View Accepted Manuscript (DOE)

Cite

Export

Save

Research Organization:: North Carolina State University, Raleigh, NC (United States)

Sponsoring Organization:: USDOE Office of Energy Efficiency and Renewable Energy (EERE), Renewable Power Office. Solar Energy Technologies Office

Grant/Contract Number:: EE0008770

OSTI ID:: 2329284

Journal Information:: IEEE Transactions on Smart Grid, Journal Name: IEEE Transactions on Smart Grid Journal Issue: 2 Vol. 15; ISSN 1949-3053

Publisher:: IEEECopyright Statement

Country of Publication:: United States

Language:: English

References (16)

Joint forecasting of multi-energy loads for a university based on copula theory and improved LSTM network Ren, Hongbo; Li, Qifen; Wu, Qiong Energy Reports, Vol. 8 https://doi.org/10.1016/j.egyr.2022.05.208	journal	November 2022
Generating realistic building electrical load profiles through the Generative Adversarial Network (GAN) Wang, Zhe; Hong, Tianzhen Energy and Buildings, Vol. 224 https://doi.org/10.1016/j.enbuild.2020.110299	journal	October 2020
Two approaches for synthesizing scalable residential energy consumption data Liu, Xiufeng; Iftikhar, Nadeem; Huo, Huan Future Generation Computer Systems, Vol. 95 https://doi.org/10.1016/j.future.2019.01.045	journal	June 2019
Generation of synthetic benchmark electrical load profiles using publicly available load and weather data Pillai, Gobind G.; Putrus, Ghanim A.; Pearsall, Nicola M. International Journal of Electrical Power & Energy Systems, Vol. 61 https://doi.org/10.1016/j.ijepes.2014.03.005	journal	October 2014
Short term load forecasting and the effect of temperature at the low voltage level Haben, Stephen; Giasemidis, Georgios; Ziel, Florian International Journal of Forecasting, Vol. 35, Issue 4 https://doi.org/10.1016/j.ijforecast.2018.10.007	journal	October 2019
Short-term load forecasting based on support vector regression considering cooling load in summer Hu, Li; Zhang, Lei; Wang, Tao 2020 Chinese Control And Decision Conference (CCDC) https://doi.org/10.1109/ccdc49329.2020.9164387	conference	August 2020
Residential Energy Load Profile Generation Using a Probabilistic Approach Gruber, Jorn K.; Prodanovic, Milan 2012 European Modelling Symposium (EMS), 2012 Sixth UKSim/AMSS European Symposium on Computer Modeling and Simulation https://doi.org/10.1109/ems.2012.30	conference	November 2012
Synthesizing Energy Consumption Data Using a Mixture Density Network Integrated with Long Short Term Memory Sarochar, Jonathan; Acharya, Ipsita; Riggs, Hugo 2019 IEEE Green Technologies Conference(GreenTech) https://doi.org/10.1109/greentech.2019.8767148	conference	April 2019
A study for clustering method to generate Typical Load Profiles for Smart Grid Kim, Young-Il; Kang, Shin-Jae; Ko, Jong-Min 8th International Conference on Power Electronics - ECCE Asia https://doi.org/10.1109/icpe.2011.5944675	conference	May 2011
GAN-based Model for Residential Load Generation Considering Typical Consumption Patterns Gu, Yuxuan; Chen, Qixin; Liu, Kai 2019 IEEE Power & Energy Society Innovative Smart Grid Technologies Conference (ISGT) https://doi.org/10.1109/isgt.2019.8791575	conference	February 2019
A Prediction-Based Smart Meter Data Generator Iftikhar, Nadeem; Liu, Xiufeng; Nordbjerg, Finn Ebertsen 2016 19th International Conference on Network-Based Information Systems (NBiS) https://doi.org/10.1109/nbis.2016.15	conference	September 2016
Data-Driven Generation of Synthetic Load Datasets Preserving Spatio-Temporal Features Pinceti, Andrea; Kosut, Oliver; Sankar, Lalitha 2019 IEEE Power & Energy Society General Meeting (PESGM) https://doi.org/10.1109/pesgm40551.2019.8973532	conference	August 2019
A time series probabilistic synthetic load curve model for residential customers Dickert, J.; Schegner, P. 2011 IEEE Trondheim PowerTech https://doi.org/10.1109/ptc.2011.6019365	conference	June 2011
Generative Adversarial Network for Synthetic Time Series Data Generation in Smart Grids Zhang, Chi; Kuppannagari, Sanmukh R.; Kannan, Rajgopal 2018 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids (SmartGridComm) https://doi.org/10.1109/smartgridcomm.2018.8587464	conference	October 2018
Generative adversarial networks Goodfellow, Ian; Pouget-Abadie, Jean; Mirza, Mehdi Communications of the ACM, Vol. 63, Issue 11 https://doi.org/10.1145/3422622	journal	October 2020
Multiple-Load Forecasting for Integrated Energy System Based on Copula-DBiLSTM Zheng, Jieyun; Zhang, Linyao; Chen, Jinpeng Energies, Vol. 14, Issue 8 https://doi.org/10.3390/en14082188	journal	April 2021

Similar Records

ProfileSR-GAN: A GAN Based Super-Resolution Method for Generating High-Resolution Load Profiles

Journal Article · 2022 · IEEE Transactions on Smart Grid · OSTI ID:2329473

Song, Lidong; Li, Yiyan; Lu, Ning

Generative adversarial networks (GAN) based efficient sampling of chemical composition space for inverse design of inorganic materials

Journal Article · 2020 · npj Computational Materials · OSTI ID:1647313

Dan, Yabo; Zhao, Yong; Li, Xiang; +3 more

Predicting CO₂ Plume Migration in Heterogeneous Formations Using Conditional Deep Convolutional Generative Adversarial Network

Journal Article · 2019 · Water Resources Research · OSTI ID:1614300

Zhong, Zhi; Sun, Alexander Y.; Jeong, Hoonyoung

Related Subjects

24 POWER TRANSMISSION AND DISTRIBUTION
Data augmentation
generative adversarial networks
load profile group generation
machine learning
negative sample generation
synthetic data

Title: MultiLoad-GAN: A GAN-Based Synthetic Load Group Generation Method Considering Spatial-Temporal Correlations

Citation Formats

References (16)

Similar Records

Related Subjects