A review of data-driven approaches for prediction and classification of building energy consumption

Wei, Yixuan; Zhang, Xingxing; Shi, Yong; Liang, Xin; Pan, Song; Wu, Jinshun; Han, Mengjie; Zhao, Xiaoyun

doi:10.1016/j.rser.2017.09.108

Cited by 533 publications

(252 citation statements)

References 90 publications

Supporting

Mentioning

237

Contrasting

Unclassified

Order By: Relevance

“…Data-driven approaches have addressed variety of energy prediction and load forecasting tasks and thus has attracted significant research attention [39]. Common data driven approaches for time series data includes linear regression that fits the best straight line through the training data and using ordinary least square method to estimate the parameters by minimizing the sum of the squared vertical distances.…”

Section: Alternative Modeling Approachesmentioning

confidence: 99%

Optimal Deep Learning LSTM Model for Electric Load Forecasting using Feature Selection and Genetic Algorithm: Comparison with Machine Learning Approaches †

et al. 2018

View full text Add to dashboard Cite

Abstract:Background: With the development of smart grids, accurate electric load forecasting has become increasingly important as it can help power companies in better load scheduling and reduce excessive electricity production. However, developing and selecting accurate time series models is a challenging task as this requires training several different models for selecting the best amongst them along with substantial feature engineering to derive informative features and finding optimal time lags, a commonly used input features for time series models. Methods: Our approach uses machine learning and a long short-term memory (LSTM)-based neural network with various configurations to construct forecasting models for short to medium term aggregate load forecasting. The research solves above mentioned problems by training several linear and non-linear machine learning algorithms and picking the best as baseline, choosing best features using wrapper and embedded feature selection methods and finally using genetic algorithm (GA) to find optimal time lags and number of layers for LSTM model predictive performance optimization. Results: Using France metropolitan's electricity consumption data as a case study, obtained results show that LSTM based model has shown high accuracy then machine learning model that is optimized with hyperparameter tuning. Using the best features, optimal lags, layers and training various LSTM configurations further improved forecasting accuracy. Conclusions: A LSTM model using only optimally selected time lagged features captured all the characteristics of complex time series and showed decreased Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) for medium to long range forecasting for a wider metropolitan area.

show abstract

Section: Alternative Modeling Approachesmentioning

confidence: 99%

Optimal Deep Learning LSTM Model for Electric Load Forecasting using Feature Selection and Genetic Algorithm: Comparison with Machine Learning Approaches †

et al. 2018

View full text Add to dashboard Cite

show abstract

“…The enormous progress of the economy and the rapid rising of human living standards have motivated the demand for indoor thermal comfort (22°C to 28°C), which leads to the increase in energy consumption . It is estimated that more than 40% of the entire energy consumption in the world results from building energy consumption, especially from ventilation, air‐conditioning, and heating consumptions closely related to indoor thermal comfort . Accordingly, to reduce the energy consumption of maintaining indoor thermal comfort, the building energy efficiency needs to be maximized.…”

Section: Introductionmentioning

confidence: 99%

Form‐stable Na ₂ SO ₄ ·10H ₂ O‐Na ₂ HPO ₄ ·12H ₂ O eutectic/hydrophilic fumed silica composite phase change material with low supercooling and low thermal conductivity for indoor thermal comfort improvement

Fang

Tang

et al. 2020

Int J Energy Res

View full text Add to dashboard Cite

Summary In this research, a novel form‐stable composite phase change material (CPCM) based on Na2SO4·10H2O‐Na2HPO4·12H2O binary eutectic hydrated salt (EHS) as phase change material (PCM) and porous hydrophilic fumed silica (SiO2) as the carrier was prepared via the impregnation method, which is aimed at being integrated into building envelopes for the improvement of indoor thermal comfort and the reduction of building energy consumption. Thereinto, Na2SiO3·9H2O utilized as the nucleating agent suppressed the supercooling of the EHS enormously. Differential scanning calorimetry (DSC) result and the crystalline phase obtained by X‐ray diffraction (XRD) revealed that the mixture consisting of 20‐wt% Na2SO4·10H2O and 80‐wt% Na2HPO4·12H2O exhibited a eutectic melting behavior. Depending on DSC and leakage tests, the optimum amount of SiO2 in CPCM was determined as 30 wt%, which successfully stabilized the shape of the EHS to avoid leaking, simultaneously reduced thermal conductivity, and further facilitated its crystallization. Additionally, scanning electron microscope (SEM) manifested that the EHS was well adsorbed into the mesopores of SiO2. The obtained CPCM revealed an applicable phase transition temperature (25.16°C), moderate melting enthalpy (142.9 kJ·kg−1), negligible supercooling degree (0.24°C), and low thermal conductivity. More importantly, after 200 melting‐solidifying cycles, it displayed an excellent thermal reliability. All these results showed that the CPCM possessed desirable properties for applications.

show abstract

“…In recent decades, the energy demand of our society is continuously increasing, especially from the building sector . As a major energy consuming sector (about 40% of world total primary energy), buildings are always required to provide comfortable indoor environment for the occupants, and this has led to great challenges to the application of energy . Therefore, energy saving of building and improving the efficiency of their energy systems become necessary for sustainable development.…”

Section: Introductionmentioning

confidence: 99%

Climate applicability study of building envelopes containing phase change materials

Zheng

Zou

et al. 2019

Int J Energy Res

View full text Add to dashboard Cite

Summary Building envelopes incorporating phase change materials (PCMs) can contribute to reducing the energy consumption of buildings and enhance indoor thermal environment comfort. In this study, two building models were developed in EnergyPlus to explore the applicability of using PCMs in different climate zones. Six significantly different cities from five climate zones in China have been investigated. The simulation results reflected that for climates with small fluctuation weather conditions, PCMs with appreciate transition temperature TR can contribute to positive influences on energy saving. For example, in Guangzhou, PCMs with TR of near 26oC and 22oC contribute to the highest energy‐saving rates of 12.0% and 12.4%, for external and internal PCMs addition. And the ideal transition temperature for each climates is subjected to its own outdoor comprehensive temperatures and its indoor set temperature, for external and internal PCMs addition, respectively. For other climates with the big fluctuation weather condition, PCMs' latent heat function to the energy savings is not significant.

show abstract

A review of data-driven approaches for prediction and classification of building energy consumption

Cited by 533 publications

References 90 publications

Optimal Deep Learning LSTM Model for Electric Load Forecasting using Feature Selection and Genetic Algorithm: Comparison with Machine Learning Approaches †

Optimal Deep Learning LSTM Model for Electric Load Forecasting using Feature Selection and Genetic Algorithm: Comparison with Machine Learning Approaches †

Form‐stable Na ₂ SO ₄ ·10H ₂ O‐Na ₂ HPO ₄ ·12H ₂ O eutectic/hydrophilic fumed silica composite phase change material with low supercooling and low thermal conductivity for indoor thermal comfort improvement

Climate applicability study of building envelopes containing phase change materials

Contact Info

Product

Resources

About

A review of data-driven approaches for prediction and classification of building energy consumption

Cited by 533 publications

References 90 publications

Optimal Deep Learning LSTM Model for Electric Load Forecasting using Feature Selection and Genetic Algorithm: Comparison with Machine Learning Approaches †

Optimal Deep Learning LSTM Model for Electric Load Forecasting using Feature Selection and Genetic Algorithm: Comparison with Machine Learning Approaches †

Form‐stable Na 2 SO 4 ·10H 2 O‐Na 2 HPO 4 ·12H 2 O eutectic/hydrophilic fumed silica composite phase change material with low supercooling and low thermal conductivity for indoor thermal comfort improvement

Climate applicability study of building envelopes containing phase change materials

Contact Info

Product

Resources

About

Form‐stable Na ₂ SO ₄ ·10H ₂ O‐Na ₂ HPO ₄ ·12H ₂ O eutectic/hydrophilic fumed silica composite phase change material with low supercooling and low thermal conductivity for indoor thermal comfort improvement