Data-driven prediction and optimization toward net-zero and positive-energy buildings: A systematic review

Building and Environment, Journal Year: 2023, Volume and Issue: 242, P. 110578 - 110578

Published: July 4, 2023

Language: Английский

---

A predictive model for daylight performance based on multimodal generative adversarial networks at the early design stage

Energy and Buildings, Journal Year: 2024, Volume and Issue: 305, P. 113876 - 113876

Published: Jan. 3, 2024

Language: Английский

---

Advanced Deep Learning Algorithms for Energy Optimization of Smart Cities

Energies, Journal Year: 2025, Volume and Issue: 18(2), P. 407 - 407

Published: Jan. 18, 2025

Advanced deep learning algorithms play a key role in optimizing energy usage smart cities, leveraging massive datasets to increase efficiency and sustainability. These analyze real-time data from sensors IoT devices predict demand, enabling dynamic load balancing reducing waste. Reinforcement models optimize power distribution by historical patterns adapting changes real time. Convolutional neural networks (CNNs) recurrent (RNNs) facilitate detailed analysis of spatial temporal better usage. Generative adversarial (GANs) are used simulate scenarios, supporting strategic planning anomaly detection. Federated ensures privacy-preserving sharing distributed systems, promoting collaboration without compromising security. technologies driving the transformation towards sustainable energy-efficient urban environments, meeting growing demands modern cities. However, there is view that if pace development maintained with large amounts data, computational/energy costs may exceed benefits. The article aims conduct comparative assess potential this group technologies, taking into account efficiency.

Language: Английский

---

Architectural spatial layout planning using artificial intelligence

Automation in Construction, Journal Year: 2023, Volume and Issue: 154, P. 105019 - 105019

Published: July 27, 2023

Language: Английский

---

Concepts of user-centred lighting controls for office applications: A systematic literature review

Building and Environment, Journal Year: 2024, Volume and Issue: 254, P. 111321 - 111321

Published: Feb. 29, 2024

Due to the very limited information available during building design phase, lighting automation systems tend follow pre-defined control curves and use normative default values. However, these are unlikely reflect wide variation in daily working hours individual preferences. Inadequate user modelling can result missed energy comfort targets, as well insufficient light doses. Given that sector accounts for around one third of world's demand, is main consumers, a higher level representation essential meet climate environmental targets. The general practical applicability user-centred concepts usually fails due implementation, inadequate mapping objectives required parameters, availability systems. This comprehensive literature review 160 articles evaluates potential relation target criteria identifies necessary technical system components greater applicability. focus on daylight artificial their application office environments. From results obtained, key elements better implementation were derived. These include zoned lighting, human-in-the-loop approaches, sensor fusion. Post-occupancy evaluation, supported by social science methods, help capture relevant physiological psychological parameters. concludes post-occupancy optimisation applications offers great overcoming previous limitations subsequently reducing performance gaps.

Language: Английский

---

Spatial efficiency: An outset of lighting application efficacy for indoor lighting

Building and Environment, Journal Year: 2024, Volume and Issue: 255, P. 111409 - 111409

Published: March 22, 2024

Electric lighting plays a critical role in buildings, not only impacting the well-being, satisfaction, and performance of building occupants but also accounting for significant portion energy consumption. The commonly used efficiency metrics lighting, such as luminous efficacy or power density, fall short quantifying effective light architectural spaces. To address shortcomings existing measures, application that characterizes efficient delivery from source to target should be utilized. Lighting can account efficiencies temporal, electrical, visual, spatial dimensions. This study outlines method quantify electric buildings. As proof concept, simulated data are analyzed two targets (horizontal work plane level occupant field view) based on primary characteristics built environment systems. findings indicate design variables (e.g., room size, luminaire distribution type, reflectance surfaces, significantly affect specified settings. Therefore, values exhibit considerable variation among distinct systems spaces customized suit unique requirements each setting. Future research will investigate establish complementary components framework.

Language: Английский

---

Machine-learned kinetic Façade: Construction and artificial intelligence enabled predictive control for visual comfort

Automation in Construction, Journal Year: 2023, Volume and Issue: 156, P. 105093 - 105093

Published: Sept. 16, 2023

Language: Английский

---

A review and guide on selecting and optimizing machine learning algorithms for daylight prediction

Building and Environment, Journal Year: 2023, Volume and Issue: 244, P. 110822 - 110822

Published: Sept. 10, 2023

Language: Английский

---

The impacts of urban canyons morphology on daylight availability and energy consumption of buildings in a hot-summer Mediterranean climate

Solar Energy, Journal Year: 2023, Volume and Issue: 266, P. 112181 - 112181

Published: Nov. 20, 2023

Language: Английский

---

An interactive assessment framework for residential space layouts using pix2pix predictive model at the early-stage building design

Smart and Sustainable Built Environment, Journal Year: 2022, Volume and Issue: 13(4), P. 809 - 827

Published: Dec. 5, 2022

Purpose In this study, a novel framework based on deep learning models is presented to assess energy and environmental performance of given building space layout, facilitating the decision-making process at early-stage design. Design/methodology/approach A methodology using an image-based model called pix2pix proposed predict overall daylight, ventilation residential layout. The then evaluated by being applied 300 sample apartment units in Tehran, Iran. Four were trained illuminance, spatial daylight autonomy (sDA), primary intensity maps. simulation results considered ground truth. Findings showed average structural similarity index measure (SSIM) 0.86 0.81 for predicted illuminance sDA maps, respectively, score 88% representative each which outputted within three seconds. Originality/value study helps upskilling design professionals involved with architecture, engineering construction (AEC) industry through engaging artificial intelligence human–computer interactions. specific novelties research are: first, evaluating indoor metrics (daylight ventilation) alongside layouts model, second, widening assessment scope group spaces forming layout five different floors third, incorporating impact context intended objectives.

Language: Английский

---