Review of electric vehicles integration impacts in distribution networks: Placement, charging/discharging strategies, objectives and optimisation models

Journal of Energy Storage, Journal Year: 2023, Volume and Issue: 72, P. 108672 - 108672

Published: Aug. 16, 2023

Language: Английский

---

Automated machine learning-based framework of heating and cooling load prediction for quick residential building design

Energy, Journal Year: 2023, Volume and Issue: 274, P. 127334 - 127334

Published: March 25, 2023

Language: Английский

---

A comparison between grey-box models and neural networks for indoor air temperature prediction in buildings

Journal of Building Engineering, Journal Year: 2024, Volume and Issue: 84, P. 108583 - 108583

Published: Jan. 21, 2024

Language: Английский

---

Statistical and machine learning approaches for energy efficient buildings

Energy and Buildings, Journal Year: 2025, Volume and Issue: unknown, P. 115309 - 115309

Published: Jan. 1, 2025

Language: Английский

---

Modeling energy-efficient building loads using machine-learning algorithms for the design phase

Energy and Buildings, Journal Year: 2023, Volume and Issue: 283, P. 112807 - 112807

Published: Jan. 20, 2023

Language: Английский

---

A multivariate time series graph neural network for district heat load forecasting

Energy, Journal Year: 2023, Volume and Issue: 278, P. 127911 - 127911

Published: May 25, 2023

Heat load prediction is essential for energy efficiency and carbon reduction in district heating systems. However, heat influenced by many factors, such as building characteristics, consumption behavior, climate, making its challenging. Traditional methods based on physical models are complex insufficiently accurate, whereas most data-driven statistical ignore customer behaviors their correlation, do not account the temporal inertia of consumption. This paper proposes a graph ambient intelligence (GAIN) method prediction, which classifies customers profiles uses collaborative attention graphs to capture associations weather impact loads. GAIN also incorporates recursive autoregressive model The proposed evaluated four metrics compared with fifteen baseline methods. results show that achieves lowest daily forecasting errors terms RMSE, MAE, CV-RMSE, values 6.972, 4.442, 0.191, respectively. Besides, maximum 25%, 29%, 25% respectively, other when taking meteorological factors into account.

Language: Английский

---

Batteries temperature prediction and thermal management using machine learning: An overview

Energy Reports, Journal Year: 2023, Volume and Issue: 10, P. 2277 - 2305

Published: Sept. 15, 2023

Batteries, particularly lithium-ion batteries, play an important role in powering our modern world, from portable devices to electric vehicles and renewable energy storage. However, during charging discharging, they generate heat due chemical reactions within them. This can lead reduced performance, shortened lifespan, even safety risks if not properly managed. To address this problem, Machine learning has been emerged as a changing tool battery technology its ability analyze large datasets that be used predicting temperatures enhancing their thermal management. In work, we machine features along with look at various categories, frameworks, applications. comprehensive study, methods neural networks temperature prediction management are analyzed discussed training algorithms. Moreover, the paper reviews summarizes research publications examining using As result, there is no superior algorithm for management, performance of model may vary depending on data set, algorithm, other parameters. among these algorithms researchers preferring use artificial accuracy complexity. particular, network integrated proper cooling reduce by more than 25%.

Language: Английский

---

Attention mechanism-based transfer learning model for day-ahead energy demand forecasting of shopping mall buildings

Energy, Journal Year: 2023, Volume and Issue: 270, P. 126878 - 126878

Published: Feb. 7, 2023

Language: Английский

---

Modeling and forecasting electricity consumption amid the COVID-19 pandemic: Machine learning vs. nonlinear econometric time series models

Sustainable Cities and Society, Journal Year: 2023, Volume and Issue: 98, P. 104860 - 104860

Published: Aug. 15, 2023

Accurately modelling and forecasting electricity consumption remains a challenging task due to the large number of statistical properties that characterize this time series such as seasonality, trend, sudden changes, slow decay autocrrelation function, among many others. This study contributes literature by using comparing four advanced econometrics models, machine learning deep models1 analyze forecast during COVID-19 pre-lockdown, lockdown, releasing-lockdown, post-lockdown phases. Monthly data on Qatar's total has been used from January 2010 December 2021. The empirical findings demonstrate both econometric models are able capture most important features characterizing consumption. In particular, it is found climate change based factors, e.g temperature, rainfall, mean sea-level pressure wind speed, key determinants terms forecasting, results indicate autoregressive fractionally integrated moving average three state markov switching with exogenous variables outperform all other models. Policy implications energy-environmental recommendations proposed discussed.

Language: Английский

---

An improved attention-based deep learning approach for robust cooling load prediction: Public building cases under diverse occupancy schedules

Sustainable Cities and Society, Journal Year: 2023, Volume and Issue: 96, P. 104679 - 104679

Published: May 28, 2023

Space cooling in buildings is responsible for massive energy consumption and carbon emissions. Accurate load prediction can facilitate the implementation of energy-efficiency control strategies practice. In this paper, an improved attention-based deep learning approach proposed robust ultra-short-term prediction. First, a novel time representation introduced to extract periodicity non-periodicity loads efficiently. Then, long short-term memory with attention mechanism extracts properly steps by identifying relevant hidden states learns high-level temporal dependency. The additionally incorporates extreme gradient boosting through error reciprocal method, enhancing elimination errors improving robustness. study takes Guangzhou as example generates using diverse occupancy schedules five building types based on Chinese National Standard Typical Meteorological Year data. evaluated datasets comprising loads, meteorological data, contextual information. Through results analysis, outperforms other models terms accuracy robustness across all types. Additionally, model interpretation provided regarding feature importance matrixes, which enhances understanding transparency final from approach.

Language: Английский

---