Building Energy Load Forecasting using Deep Neural Networks

📝 Abstract

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 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 variants of the LSTM: 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 where 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.

💡 Analysis

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 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 variants of the LSTM: 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 where 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.

📄 Content

© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. Accepted version of the paper appearing in Proceedings of the 42nd Annual Conference of the IEEE Industrial Electronics Society (IECON), 2016

Building Energy Load Forecasting using Deep Neural Networks

Daniel L. Marino, Kasun Amarasinghe, Milos Manic Department of Computer Science Virginia Commonwealth University Richmond, Virginia marinodl@vcu.edu, amarasinghek@vcu.edu, misko@ieee.org

Abstract—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 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 variants of the LSTM: 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 where 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.
Keywords—Deep Learning; Deep Neural Networks; Long- Short-Term memory; LSTM; Energy; Building Energy; Energy Load forecasting I. INTRODUCTION
Buildings are identified as a major energy consumer worldwide, accounting for 20%-40% of the total energy production [1]-[3]. In addition to being a major energy consumer, buildings are shown to account for a significant portion of energy wastage as well [4]. As energy wastage poses a threat to sustainability, making buildings energy efficient is extremely crucial. Therefore, in making building energy consumption more efficient, it is necessary to have accurate predictions of its future energy consumption.
At the grid level, to minimizing the energy wastage and making the power generation and distribution more efficient, the future of the power grid is moving to a new paradigm of smart grids [5], [6]. Smart grids are promising, unprecedented flexibility in energy generation and distribution [7]. In order to provide that flexibility, the power grid has to be able to dynamically adapt to the changes in demand and efficiently distribute the generated energy from the various sources such as renewables [8]. Therefore, intelligent control decisions should be made continuously at aggregate level as well as modular level in the grid. In achieving that goal and ensuring the reliability of the grid, the ability of forecasting the future demands is important. [6], [9]. Further, demand or load forecasting is crucial for mitigating uncertainties of the future [6]. In that, individual building level demand forecasting is crucial as well as forecasting aggregate loads. In terms of demand response, building level forecasting helps carry out demand response locally since the smart grids incorporate distributed energy generation [6]. The advent of smart meters have made the acquisition of energy consumption data at building and individual site level feasible. Thus data driven and statistical forecasting models are made possible [7]. Aggregate level and building level load forecasting can be viewed in three different categories: 1) Short-term 2) Medium- term and 3) Long-term [6]. It has been determined that the load forecasting is a hard problem and in that, individual building level load forecasting is even harder than aggregate load forecasting [6], [10]. Thus, it has received increased attention from researchers. In literature, two main methods can be found for performing energy load forecasting: 1) Physics principles based models and 2) Statistical and machine learning based models. Focus of the presented work is on the second category of statistical load forecasting. In [7], the authors used Artificial Neural Network (ANN) ensembles to perform the building level load forecasting. ANNs have been explored

This content is AI-processed based on ArXiv data.