Nonlinear forces in urban thermal environment using Bayesian optimization-based ensemble learning

Wú, Zhìqiáng; Qiao, Renlu; Zhao, Shuang; Liu, Xiaochang; Gao, Shuo; Liu, Zhi‐Yu; Ao, Xiang; Zhou, Shiqi; Wang, Zhensheng; Jiang, Qingrui

doi:10.1016/j.scitotenv.2022.156348

Cited by 33 publications

(3 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Even more and more research on the impact of land use on TOD ridership shifting to more sophisticated non-linear models (Cheng et al, 2020;Wu et al, 2022), findings could not avoid the diverse effects of mixed land use across the various catchment areas, causing the MAUP. The effect of a specifically built environment variable on ridership may saturate or reduce up to a certain scale, and figuring out its effective range might offer more precise guidelines for land use planning for TOD (Ding et al, 2019;van Wee & Handy, 2016).…”

Section: 2mentioning

confidence: 99%

Integrated impact of urban mixed land use on TOD ridership: A multi-radius comparative analysis

Gu,

Lin,

Wang

2024

JTLU

View full text Add to dashboard Cite

The global trend toward urbanization has spurred the widespread adoption of transit-oriented development (TOD). While previous research has extensively explored the relationship between land use and TOD ridership, much of it has focused on linear associations at a singular scale. Leveraging recent advancements in nonlinear modeling and the accessibility of open-source data, this study employs a comprehensive two-step methodology. Firstly, K-means clustering algorithm categorizes TOD sites in Shenzhen into three distinct clusters, providing a site-based understanding of their characteristics. Subsequently, a Light Gradient Boosting Machine (LightGBM) classification model, complemented by SHapley Additive exPlanations (SHAP) values for interpretation, quantitatively evaluates the influence of mixed land use on TOD ridership across various catchment areas. As for the findings, we discover that land-use factors have different effects on TOD site patronage at different buffer radii and delve into the intricacies of these effects. Further results reveal non-linear relationships with varying degrees of positivity and negativity. For instance, residents and health sites positively impact patronage across all buffer radii, while certain commercial land uses exhibit a negative influence. The study demonstrates how the importance of different land-use structures varies across these clusters, shedding light on the nuanced impacts of land use on TOD catchment areas. Our research optimizes land-use mixes based on predominant cluster characteristics by offering actionable recommendations for urban managers.

show abstract

Section: 2mentioning

confidence: 99%

Integrated impact of urban mixed land use on TOD ridership: A multi-radius comparative analysis

Gu,

Lin,

Wang

2024

JTLU

View full text Add to dashboard Cite

show abstract

“…Therefore, through the architecture of this new analytical process, we bring the analysis tools from small-scale experiments conducted in the laboratory to a more urban scale and urban design application scenarios. It also serves as a bridge between traditional physical space simulation analyses and statistical model construction under the new perspective of the era of big data [64,65]. Our method solves the current problem that most of the research mentioned above only remained at the level of analysis and evaluation and lacked clear design guidelines for interventions that can be implemented.…”

Section: Strengths and Weaknesses Of The Existing Toolsmentioning

confidence: 99%

A Survey of the Landscape Visibility Analysis Tools and Technical Improvements

Wú

Wang

Gan

et al. 2023

IJERPH

Self Cite

View full text Add to dashboard Cite

Visual perception of the urban landscape in a city is complex and dynamic, and it is largely influenced by human vision and the dynamic spatial layout of the attractions. In return, landscape visibility not only affects how people interact with the environment but also promotes regional values and urban resilience. The development of visibility has evolved, and the digital landscape visibility analysis method allows urban researchers to redefine visible space and better quantify human perceptions and observations of the landscape space. In this paper, we first reviewed and compared the theoretical results and measurement tools for spatial visual perception and compared the value of the analytical methods and tools for landscape visualization in multiple dimensions on the principal of urban planning (e.g., complex environment, computational scalability, and interactive intervention between computation and built environment). We found that most of the research was examined in a static environment using simple viewpoints, which can hardly explain the actual complexity and dynamic superposition of the landscape perceptual effect in an urban environment. Thus, those methods cannot effectively solve actual urban planning issues. Aiming at this demand, we proposed a workflow optimization and developed a responsive cross-scale and multilandscape object 3D visibility analysis method, forming our analysis model for testing on the study case. By combining the multilandscape batch scanning method with a refined voxel model, it can be adapted for large-scale complex dynamic urban visual problems. As a result, we obtained accurate spatial visibility calculations that can be conducted across scales from the macro to micro, with large external mountain landscapes and small internal open spaces. Our verified approach not only has a good performance in the analysis of complex visibility problems (e.g., we defined the two most influential spatial variables to maintain good street-based landscape visibility) but also the high efficiency of spatial interventions (e.g., where the four recommended interventions were the most valuable), realizing the improvement of intelligent landscape evaluations and interventions for urban spatial quality and resilience.

show abstract

“…However, current city-level assessments of building energy consumption present challenges. On the one hand, the top-down approach, which depends on monitoring and statistics Wu et al, 2022), is often lacking in smaller cities or cities with insufficient economic development. On the other hand, the resources and time required for assessments that rely on bottomup approaches with energy consumption simulation engines often prove prohibitive in the early stages of planning (W. Wang et al, 2021).…”

Section: Introductionmentioning

confidence: 99%

Predicting building energy consumption in urban neighborhoods using machine learning algorithms

Jiang,

Huang,

et al. 2024

FURP

Self Cite

View full text Add to dashboard Cite

Assessing building energy consumption in urban neighborhoods at the early stages of urban planning assists decision-makers in developing detailed urban renewal plans and sustainable development strategies. At the city-level, the use of physical simulation-based urban building energy modeling (UBEM) is too costly, and data-driven approaches often are hampered by a lack of available building energy monitoring data. This paper combines a simulation-based approach with a data-driven approach, using UBEM to provide a dataset for machine learning and deploying the trained model for large-scale urban building energy consumption prediction. Firstly, we collected 18,789 neighborhoods containing 248,938 buildings in the Shanghai central area, of which 2,702 neighborhoods were used for UBEM. Simultaneously, building functions were defined by POI data and land use data. We used 14 impact factors related to land use and building morphology to define each neighborhood. Next, we compared the performance of six ensemble learning methods modeling impact factors with building energy consumption and used SHAP to explain the best model; we also filtered out the features that contributed the most to the model output to reduce the model complexity. Finally, the balanced regressor that had the best prediction accuracy with the minimum number of features was used to predict the remaining urban neighborhoods in the Shanghai central area. The results show that XGBoost achieves the best performance. The balanced regressor, constructed with the 9 most contributing features, predicted the building rooftop photovoltaics potential, total load, cooling load, and heating load with test set accuracies of 0.956, 0.674, 0.608, and 0.762, respectively. Our method offers an 85.5%-time advantage over traditional methods, with only a maximum of 22.75% of error.

show abstract

Nonlinear forces in urban thermal environment using Bayesian optimization-based ensemble learning

Cited by 33 publications

References 42 publications

Integrated impact of urban mixed land use on TOD ridership: A multi-radius comparative analysis

Integrated impact of urban mixed land use on TOD ridership: A multi-radius comparative analysis

A Survey of the Landscape Visibility Analysis Tools and Technical Improvements

Predicting building energy consumption in urban neighborhoods using machine learning algorithms

Contact Info

Product

Resources

About