Time Series Forecasting with Multi-Headed Attention-Based Deep Learning for Residential Energy Consumption

Bu, Seok-Jun; Cho, Sung-Bae

doi:10.3390/en13184722

Cited by 44 publications

(28 citation statements)

References 41 publications

(51 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There exists ongoing research for integrating these methods for forecasting mainly energy consumption [30][31][32][33][63][64][65]. Such methods can be implemented in the suggested model and be used in the first step of the methodology for forecasting the urban district's energy consumption and weather data, as well as in the fourth step for energy generation from DG equipment and real-time energy management.…”

Section: Discussionmentioning

confidence: 99%

“…There exists ongoing research of different methods of forecasting energy data, such as the hour-energy, one-step-ahead, one-hour-ahead, one-day-ahead forecasting, and newer methods which combine artificial intelligence [30][31][32][33]. Nevertheless, hour-energy balance based on trends from local historical data (i.e., energy consumption and weather data of the urban district and its area) can increase the acceptance of the infrastructure design by decision makers and residents of the district.…”

Section: Design Algorithmmentioning

confidence: 99%

See 1 more Smart Citation

A Systematic Methodology for Design of Sustainable Urban Neighborhood Energy Infrastructure

2021

View full text Add to dashboard Cite

The growing share of global energy consumption by cities (currently over 65%) raises the requirements for a systematic holistic approach for designing urban energy infrastructure in order to ensure its sustainability. A literature review of state-of-the-art modeling of urban energy infrastructure design emphasized the incomprehensive sustainability of the performed evaluations, as they accounted for several aspects of sustainability but missed others. Omitting important aspects can have significant implications which can put the sustainability of the energy infrastructure at risk. In this study, we attempted to develop a comprehensive model for designing sustainable energy infrastructure for urban districts, which accounts for the four aspects of sustainability: social, technical, environmental, and economic. The model is based on a four-step methodology: district characterization, a technological survey for distributed generation and energy storage, selecting suitable technologies according to social and technical criteria, and simulations of different energy infrastructure configurations to find the most suitable configurations basing on economic and environmental criteria. The research includes a case study in which the model was implemented for the Technion campus in Haifa. The developed model proved to be a comprehensive, efficient, and versatile tool for designing urban energy infrastructure.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Design Algorithmmentioning

confidence: 99%

A Systematic Methodology for Design of Sustainable Urban Neighborhood Energy Infrastructure

2021

View full text Add to dashboard Cite

show abstract

“…To overcome and extract the irregular energy patterns and reduce the translational variance in the power forecasting for residential area consumption, Seok-Jun et al proposed multi-headed attention and Convolutional Recurrent Neural Network-based deep learning model [12]. To model the transient and impulsive nature of the power demand, it calculates the attention scores using softmax and the dot product in the present network.…”

Section: Related Workmentioning

confidence: 99%

Energy Load Forecasting Using a Dual-Stage Attention-Based Recurrent Neural Network

Özcan

Çatal

Kasif

2021

Sensors

View full text Add to dashboard Cite

Providing a stable, low-price, and safe supply of energy to end-users is a challenging task. The energy service providers are affected by several events such as weather, volatility, and special events. As such, the prediction of these events and having a time window for taking preventive measures are crucial for service providers. Electrical load forecasting can be modeled as a time series prediction problem. One solution is to capture spatial correlations, spatial-temporal relations, and time-dependency of such temporal networks in the time series. Previously, different machine learning methods have been used for time series prediction tasks; however, there is still a need for new research to improve the performance of short-term load forecasting models. In this article, we propose a novel deep learning model to predict electric load consumption using Dual-Stage Attention-Based Recurrent Neural Networks in which the attention mechanism is used in both encoder and decoder stages. The encoder attention layer identifies important features from the input vector, whereas the decoder attention layer is used to overcome the limitations of using a fixed context vector and provides a much longer memory capacity. The proposed model improves the performance for short-term load forecasting (STLF) in terms of the Mean Absolute Error (MAE) and Root Mean Squared Errors (RMSE) scores. To evaluate the predictive performance of the proposed model, the UCI household electric power consumption (HEPC) dataset has been used during the experiments. Experimental results demonstrate that the proposed approach outperforms the previously adopted techniques.

show abstract

“…To improve the LSTM capability to deal with the varying length of input features, the attention mechanism is integrated into the LSTM algorithm to improve the prediction performance [32]. The attention mechanism is also used to improve the prediction accuracy for a sudden increase in power usage [33]. Compared with the load prediction of a whole building or a whole floor, the electric load prediction at a single-unit level is more difficult because of the greater randomness.…”

Section: Introductionmentioning

confidence: 99%

A Novel Short-Term Residential Electric Load Forecasting Method Based on Adaptive Load Aggregation and Deep Learning Algorithms

et al. 2021

View full text Add to dashboard Cite

Short-term residential load forecasting is the precondition of the day-ahead and intra-day scheduling strategy of the household microgrid. Existing short-term electric load forecasting methods are mainly used to obtain regional power load for system-level power dispatch. Due to the high volatility, strong randomness, and weak regularity of the residential load of a single household, the mean absolute percentage error (MAPE) of the traditional methods forecasting results would be too big to be used for home energy management. With the increase in the total number of households, the aggregated load becomes more and more stable, and the cyclical pattern of the aggregated load becomes more and more distinct. In the meantime, the maximum daily load does not increase linearly with the increase in households in a small area. Therefore, in our proposed short-term residential load forecasting method, an optimal number of households would be selected adaptively, and the total aggregated residential load of the selected households is used for load prediction. In addition, ordering points to identify the clustering structure (OPTICS) algorithm are also selected to cluster households with similar power consumption patterns adaptively. It can be used to enhance the periodic regularity of the aggregated load in alternative. The aggregated residential load and encoded external factors are then used to predict the load in the next half an hour. The long short-term memory (LSTM) deep learning algorithm is used in the prediction because of its inherited ability to maintain historical data regularity in the forecasting process. The experimental data have verified the effectiveness and accuracy of our proposed method.

show abstract

Time Series Forecasting with Multi-Headed Attention-Based Deep Learning for Residential Energy Consumption

Cited by 44 publications

References 41 publications

A Systematic Methodology for Design of Sustainable Urban Neighborhood Energy Infrastructure

A Systematic Methodology for Design of Sustainable Urban Neighborhood Energy Infrastructure

Energy Load Forecasting Using a Dual-Stage Attention-Based Recurrent Neural Network

A Novel Short-Term Residential Electric Load Forecasting Method Based on Adaptive Load Aggregation and Deep Learning Algorithms

Contact Info

Product

Resources

About