Building energy load forecasting using Deep Neural Networks

By Daniel L. Marino, Kasun Amarasinghe, Milos Manic. Annual Conference of the IEEE Industrial Electronics Society, 2016

570 citations recorded by the source corpus.

Ensuring sustainability demands more efficient energy management with minimized energy wastage. Therefore, the power grid of the future should provide an unprecedented level of flexibility in energy management. To that end, intelligent decision making requires accurate predictions of future energy demand/load, both at aggregate and individual site level. Thus, energy load forecasting have received increased attention in the recent past. However, it has proven to be a difficult problem. This paper presents a novel energy load forecasting methodology based on Deep Neural Networks, specifically, Long Short Term Memory (LSTM) algorithms. The presented work investigates two LSTM based architectures: 1) standard LSTM and 2) LSTM-based Sequence to Sequence (S2S) architecture. Both methods were implemented on a benchmark data set of electricity consumption data from one residential customer. Both architectures were trained and tested on one hour and one-minute time-step resolution datasets. Experimental results showed that the standard LSTM failed at one-minute resolution data while performing well in one-hour resolution data. It was shown that S2S architecture performed well on both datasets. Further, it was shown that the presented methods produced comparable results with the other deep learning methods for energy forecasting in literature.

Subfields: Computer Vision and Pattern Recognition, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/addba7f62763606216dd1d9b34908b3c6f9158db

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Electrical and Electronic Engineering

Citations in the corpus

Short-Term Residential Load Forecasting Based on LSTM Recurrent Neural Network (2019, background): “Also, the similar work of [23] lacked the justifiable discussion of the reason”
Optimal energy management strategies for energy Internet via deep reinforcement learning approach (2019, methodology): “To illustrate, neural networks are used for the power modeling of PVs and loads in [28] and [29], respectively.”
Deep neural networks for energy load forecasting (2017, methodology): “In [2] and [24], first three years were used for training and the last year data were used for testing.”
Predicting residential energy consumption using CNN-LSTM neural networks (2019, methodology): “analyzed a new energy load prediction methodology based on deep neural networks and an LSTM-based Sequence-to-Sequence (Seq2Seq) [46].”
A survey on long short-term memory networks for time series prediction (2021, methodology): “Seq2Seq network with LSTM cells for the prediction of output sequences of variable time horizons according to [28].”
On-Line Building Energy Optimization Using Deep Reinforcement Learning (2017, methodology): “[29], which is followed by the work of Manic et al.”
Optimal Deep Learning LSTM Model for Electric Load Forecasting using Feature Selection and Genetic Algorithm: Comparison with Machine Learning Approaches † (2018, methodology): “[33] presented a novel energy load forecasting methodology using two different deep architectures namely, a standard LSTM and an LSTM with sequence to Sequence architecture that produced promising results.”
Intelligent Buildings of the Future: Cyberaware, Deep Learning Powered, and Human Interacting (2016, methodology): “FIGUre 10 – The prediction results for the standard LSTm [39].”
Deep Learning for Time Series Forecasting: The Electric Load Case (2019, methodology): “In [33] a LSTM based encoderdecoder model is shown to produce superior performance compared to standard LSTM.”
Review of Smart Meter Data Analytics: Applications, Methodologies, and Challenges (2018, methodology): “The recently proposed clustering method in [193] has been used in [29]; the progress in deep learning in [194] has been used in [195].”
A Review on Digital Twin Technology in Smart Grid, Transportation System and Smart City: Challenges and Future (2023, unknown): “A stacked denoising auto-encoders (SDAs) model is proposed for electricity load forecasting [24].”
A Survey of Deep Learning for Scientific Discovery (2020, background): “This paradigm has also found some scientiﬁc applications as varied as biology [23] and energy forcasting [145].”