Quantile-Mixer: A Novel Deep Learning Approach for Probabilistic Short-Term Load Forecasting

被引:5
作者
Ryu, Seunghyoung [1 ,2 ]
Yu, Yonggyun [1 ]
机构
[1] Korea Atom Energy Res Inst, Appl Artificial Intelligence Sect, Daejeon 34057, South Korea
[2] Sejong Univ, Dept Intelligent Mechatron Engn, Seoul 05006, South Korea
基金
新加坡国家研究基金会;
关键词
Deep learning; short-term load forecasting; MLP-mixer; probabilistic forecasting; quantile regression; uncertainty; energy AI; ENERGY-STORAGE SYSTEMS; ADAPTIVE DROOP CONTROL; STATE; MICROGRIDS; ARCHITECTURES; BATTERIES; SOC;
D O I
10.1109/TSG.2023.3290180
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
As the power grid becomes more complex and dynamic, accurate short-term load forecasting (STLF) with probabilistic information is a prerequisite for various smart grid applications. For doing this, various deep learning models have been proposed, and recent models increase model size and complexity to achieve better accuracy which could also increase the burden on model design, computation time, and resources. To this end, we propose a novel deep learning model for accurate and efficient probabilistic STLF (PSTLF). First, we develop an STLF model utilizing the multi-layer perceptron (MLP)-mixer structure, i.e., MLP-mixer for STLF (MM-STLF), that has an advantage in forecasting accuracy and efficiency compared to the other deep learning models. Then, we propose a random quantile regression (RQR) method that takes a cumulative probability tau as an input to the model and is trained on random tau s. By combining MM-STLF and RQR, we develop a novel deep-PSTLF model, namely quantile-mixer (Q-mixer). We evaluate the overall performance of the proposed model with seven load datasets in terms of prediction error, model size, and inference time, respectively. Through experiments, various STLF models and probabilistic forecasting methods are compared, and the experimental results demonstrate the effectiveness of Q-mixer in load forecasting.
引用
收藏
页码:2237 / 2250
页数:14
相关论文
共 50 条
[31]   An ensemble approach for short-term load forecasting by extreme learning machine [J].
Li, Song ;
Goel, Lalit ;
Wang, Peng .
APPLIED ENERGY, 2016, 170 :22-29
[32]   Short-term Load Forecasting with LSTM based Ensemble Learning [J].
Wang, Lingxiao ;
Mao, Shiwen ;
Wilamowski, Bogdan .
2019 INTERNATIONAL CONFERENCE ON INTERNET OF THINGS (ITHINGS) AND IEEE GREEN COMPUTING AND COMMUNICATIONS (GREENCOM) AND IEEE CYBER, PHYSICAL AND SOCIAL COMPUTING (CPSCOM) AND IEEE SMART DATA (SMARTDATA), 2019, :793-800
[33]   Machine Learning for Short-Term Load Forecasting in Smart Grids [J].
Ibrahim, Bibi ;
Rabelo, Luis ;
Gutierrez-Franco, Edgar ;
Clavijo-Buritica, Nicolas .
ENERGIES, 2022, 15 (21)
[34]   A Hybrid Deep Learning Model with Evolutionary Algorithm for Short-Term Load Forecasting [J].
Al Mamun, Abdullah ;
Hoq, Muntasir ;
Hossain, Eklas ;
Bayindir, Ramazan .
2019 8TH INTERNATIONAL CONFERENCE ON RENEWABLE ENERGY RESEARCH AND APPLICATIONS (ICRERA 2019), 2019, :886-891
[35]   Short-Term Load Forecasting of Integrated Energy Systems Based on Deep Learning [J].
Huan, Jiajia ;
Hong, Haifeng ;
Pan, Xianxian ;
Sui, Yu ;
Zhang, Xiaohui ;
Jiang, Xuedong ;
Wang, Chaoqun .
2020 5TH ASIA CONFERENCE ON POWER AND ELECTRICAL ENGINEERING (ACPEE 2020), 2020, :16-20
[36]   A deep learning model for short-term power load and probability density forecasting [J].
Guo, Zhifeng ;
Zhou, Kaile ;
Zhang, Xiaoling ;
Yang, Shanlin .
ENERGY, 2018, 160 :1186-1200
[37]   An effective deep learning neural network model for short-term load forecasting [J].
Li, Ning ;
Wang, Lu ;
Li, Xinquan ;
Zhu, Qing .
CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE, 2020, 32 (07)
[38]   Short-Term Load Forecasting Based on VMD and Combined Deep Learning Model [J].
Wang, Nier ;
Xue, Sheng ;
Li, Zhanming .
IEEJ TRANSACTIONS ON ELECTRICAL AND ELECTRONIC ENGINEERING, 2023, 18 (07) :1067-1075
[39]   Multiscale-integrated deep learning approaches for short-term load forecasting [J].
Yang, Yang ;
Gao, Yuchao ;
Wang, Zijin ;
Li, Xi'an ;
Zhou, Hu ;
Wu, Jinran .
INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS, 2024, 15 (12) :6061-6076
[40]   Deep Learning Based on Multi-Decomposition for Short-Term Load Forecasting [J].
Kim, Seon Hyeog ;
Lee, Gyul ;
Kwon, Gu-Young ;
Kim, Do-In ;
Shin, Yong-June .
ENERGIES, 2018, 11 (12)