Short-Term Load Forecasting Using Time Pooling Deep Recurrent Neural\n Network

2021/09/26 by Elahe Khoshbakhti Vaygan, Vaygan, Elahe Khoshbakhti, Roozbeh Rajabi +3
Engineering · #Computer Vision and Pattern Recognition (cs.CV) #Energy Load and Power Forecasting #FOS: Computer and information sciences #Machine Learning (cs.LG) #Smart Grid Energy Management #Traffic Prediction and Management Techniques

paper · pdf · doi:10.48550/arxiv.2109.12498

openalex publication_date 2021/09/26 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28

Abstract

Integration of renewable energy sources and emerging loads like electric\nvehicles to smart grids brings more uncertainty to the distribution system\nmanagement. Demand Side Management (DSM) is one of the approaches to reduce the\nuncertainty. Some applications like Nonintrusive Load Monitoring (NILM) can\nsupport DSM, however they require accurate forecasting on high resolution data.\nThis is challenging when it comes to single loads like one residential\nhousehold due to its high volatility. In this paper, we review some of the\nexisting Deep Learning-based methods and present our solution using Time\nPooling Deep Recurrent Neural Network. The proposed method augments data using\ntime pooling strategy and can overcome overfitting problems and model\nuncertainties of data more efficiently. Simulation and implementation results\nshow that our method outperforms the existing algorithms in terms of RMSE and\nMAE metrics.\n

Related

• Appliance Level Short-term Load Forecasting via Recurrent Neural Network
• Short Term Load Forecasting Using Deep Neural Networks
• Deep Learning for Time Series Forecasting: The Electric Load Case
• Improving Power Generation Efficiency using Deep Neural Networks
• LoadCNN: A Low Training Cost Deep Learning Model for Day-Ahead Individual Residential Load Forecasting