Energy mapping of existing building stock in Cambridge using energy performance certificates and thermal infrared imagery DOI Creative Commons
Yinglong He, Jia-Yu Pan, Ramit Debnath

и другие.

Environmental Data Science, Год журнала: 2024, Номер 3

Опубликована: Янв. 1, 2024

Abstract Both energy performance certificates (EPCs) and thermal infrared (TIR) images play key roles in mapping the of urban building stock. In this paper, we developed parametric archetypes using an EPC database conducted temperature clustering on TIR acquired from drones satellite datasets. We evaluated 1,725 EPCs existing stock Cambridge, UK, to generate consumption profiles. Drone-based individual buildings two Cambridge University colleges were processed a machine learning pipeline for anomaly detection investigated influence specific factors that affect reliability management applications: ground sample distance (GSD) angle view (AOV). The results suggest construction year influences their consumption. For example, modern over 30% more energy-efficient than older ones. parallel, found show almost double savings potential through retrofitting compared newly constructed buildings. imaging showed anomalies can only be properly identified with GSD 1 m/pixel or less. A 1-6 detect hot areas surfaces. > 6 cannot characterize but does help identify heat island effects. Additional sensitivity analysis is sensitive AOV GSD. Our study informs newer approaches diagnostics thermography supports decision-making large-scale retrofitting.

Язык: Английский

Reflecting City Digital Twins (CDTs) for sustainable urban development: Roles, challenges and directions DOI Creative Commons
Qian-Cheng Wang, Maoran Sun, Xuan Liu

и другие.

Digital engineering., Год журнала: 2025, Номер unknown, С. 100035 - 100035

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

3
Towards carbon neutrality: mapping mass retrofit opportunities in Cambridge, UK DOI Creative Commons

Humberto Mora,

Ronita Bardhan

Royal Society Open Science, Год журнала: 2025, Номер 12(1)

Опубликована: Янв. 1, 2025

This study proposes a methodology and proof of concept to target prioritize mass retrofitting residential buildings in the UK using open building datasets that combine fabric energy efficiency fuel poverty meet net-zero targets. The methodological framework uses series multi-variate statistical geospatial methods consider urban, socio-economic physical attributes. In addition, thermal imaging is implemented provide insights at scale. We define hard-to-decarbonize (HtD) metric enable clustering different types establish priorities. Using Cambridge, UK, as case study, five neighbourhoods were identified characterized help determine decarbonization intervention found one clusters HtD requires more policy support from government for implementation retrofit strategies. achieved has potential inform decision making. Of relevance, it applicable urban contexts.

Язык: Английский

Процитировано

0
Estimating the Time Constant Using Smart Thermostat Data Acquisition and Manipulation: A Whole Building Experimental Study DOI
Danlin Hou,

Lukas Allan,

Hadia Awad

и другие.

Journal of Building Engineering, Год журнала: 2025, Номер unknown, С. 112485 - 112485

Опубликована: Март 1, 2025

Язык: Английский

Процитировано

0
Comprehensive street built environmental recognizabililty evaluation by integrating visual and spatial structural data DOI Creative Commons
Yi Liu, Yang Yang, Qi Dong

и другие.

Journal of Urban Management, Год журнала: 2024, Номер 13(4), С. 772 - 786

Опубликована: Авг. 24, 2024

Evaluating the recognizability of street built environments provides crucial support for urban planning, security monitoring and navigation. Although view images (SVIs) are widely used in studies, it overlooks interconnection among different locations, which can also affect perceptions about environmental recognizability. To address this issue, study proposes a deep learning-based model called RB-Node, comprehensively integrates spatial structural features road network visual from SVIs, achieving 82.56% accuracy. It appears that image information dominates Additionally, contributes significantly to accurate classification nodes waterfront promenade areas. Moreover, scene-text information, subset features, helps classify commercial historical Furthermore, 1056 samples were collected through an eye-tracking experiment validate evaluation results, as well compare decision-making process between humans RB-Node. According RB-Node behaviour human observed behavior follow similar patterns, although tend be more holistic than RB-Node's. This better understanding targeted recommendations renewal.

Язык: Английский

Процитировано

0
Measuring shaded bike lanes for heat stress mitigation with deep learning: A case study in Amsterdam, Netherlands DOI Creative Commons

Biru Cao,

Maoran Sun, Ronita Bardhan

и другие.

Urban Climate, Год журнала: 2024, Номер 57, С. 102126 - 102126

Опубликована: Сен. 1, 2024

Язык: Английский

Процитировано

0
Energy mapping of existing building stock in Cambridge using energy performance certificates and thermal infrared imagery DOI Creative Commons
Yinglong He, Jia-Yu Pan, Ramit Debnath

и другие.

Environmental Data Science, Год журнала: 2024, Номер 3

Опубликована: Янв. 1, 2024

Abstract Both energy performance certificates (EPCs) and thermal infrared (TIR) images play key roles in mapping the of urban building stock. In this paper, we developed parametric archetypes using an EPC database conducted temperature clustering on TIR acquired from drones satellite datasets. We evaluated 1,725 EPCs existing stock Cambridge, UK, to generate consumption profiles. Drone-based individual buildings two Cambridge University colleges were processed a machine learning pipeline for anomaly detection investigated influence specific factors that affect reliability management applications: ground sample distance (GSD) angle view (AOV). The results suggest construction year influences their consumption. For example, modern over 30% more energy-efficient than older ones. parallel, found show almost double savings potential through retrofitting compared newly constructed buildings. imaging showed anomalies can only be properly identified with GSD 1 m/pixel or less. A 1-6 detect hot areas surfaces. > 6 cannot characterize but does help identify heat island effects. Additional sensitivity analysis is sensitive AOV GSD. Our study informs newer approaches diagnostics thermography supports decision-making large-scale retrofitting.

Язык: Английский

Процитировано

0