Dependence between urban morphology and outdoor air temperature: A tropical campus study using random forests algorithm

Yu, Zhuoping; Chen, Shisheng; Wong, Nyuk Hien; Ignatius, Marcel; Deng, Jiyu; He, Yueer; Hii, Daniel Jun Chung

doi:10.1016/j.scs.2020.102200

Cited by 40 publications

(10 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From the experimental results, the correlation between near-surface air temperature and SVF in the city during daytime is roughly positive. In other words, the denser the urban building area, the higher the probability of high temperature around the building area, and these results show a similar trend to the research results of Hai et al [38,39]. In addition, the correlation between daytime SVF and air temperature showed a weak trend of increasing and then decreasing with time series, which may be related to the timing of exposure to direct sunlight and the amount of solar radiation absorbed and released near the buildings during different time periods [40,41].…”

Section: Relationship Between Svf and Air Temperaturesupporting

confidence: 88%

Effects of Sky View Factor on Thermal Environment in Different Local Climate Zoning Building Scenarios—A Case Study of Beijing, China

Wang

Chen

et al. 2023

Buildings

View full text Add to dashboard Cite

As an essential structural indicator of buildings, sky view factor (SVF) is one of the most critical factors affecting the urban thermal environment. However, the relationship between SVF and the thermal environment at the neighborhood scale has not been adequately studied. Therefore, this paper investigates the relationship between SVF and air temperature in different building scenarios based on the Local Climate Zone (LCZ) classification framework. Firstly, the study is based on multi-source urban data and the Open Street Map (OSM) to map the local climate zones in Beijing. Then, a simulation model with different LCZs was constructed based on realistic scenarios using the microclimate simulation software ENVI-met, and the thermal environment was simulated in 24 h on a single day in summer. Finally, the SVF and air temperature relationship under different LCZ scenarios was calculated and analyzed. The results show that (1) the SVF values of LCZ 1, LCZ 2, and LCZ 5 show a more apparent positive correlation with air temperature than other categories, and the SVF values of LCZ 6–9 show a negative and then positive correlation with air temperature; (2) in the morning, the dense building areas show a weak correlation with air temperature, and the differences in air temperature corresponding to the SVF values in different zones are greater; (3) in the morning, the air temperature in the dense building areas showed a weak correlation, the difference between the SVF values and the air temperature in different intervals was different, and when the SVF was larger or smaller, the air temperature change was smaller and concentrated, and the correlation between the air temperature and the SVF in the open building areas was not obvious; (4) with 12:00 as the dividing line, the SVF and the air temperature in all categories showed a weak positive correlation after this time. This study can provide guidance on optimizing building layouts and mitigating the impacts of urban heat on human health.

show abstract

Section: Relationship Between Svf and Air Temperaturesupporting

confidence: 88%

Effects of Sky View Factor on Thermal Environment in Different Local Climate Zoning Building Scenarios—A Case Study of Beijing, China

Wang

Chen

et al. 2023

Buildings

View full text Add to dashboard Cite

show abstract

“…It is necessary to investigate whether commonly used parameter regression and other methods can accurately capture the correlation between the two variables. Exploring machine learning algorithms, which do not rely on predetermined relationship models, can provide valuable insights for optimizing correlation studies [67][68][69][70][71].…”

Section: The Blue-green Model In Land Spatial Planningmentioning

confidence: 99%

Evolution Patterns of Cooling Island Effect in Blue–Green Space under Different Shared Socioeconomic Pathways Scenarios

et al. 2023

View full text Add to dashboard Cite

Blue–green space refers to blue space (rivers and lakes) and green space (lawns and trees), which have the cooling island effect and are increasingly acknowledged as a potential and effective way to help alleviate the urban heat island effect. Scientific and flexible blue–green space planning is required, especially for medium- and large-scale urban agglomerations in the face of climate change. However, the temporal evolution and spatial patterns of the cooling island effect in the blue–green space under different future scenarios of climate change have not been fully investigated. This would impede long-term urban strategies for climate change adaptation and resilience. Here we studied the relationship between future climate change and blue–green spatial layout with Weather Research and Forecasting (WRF), based on the numerical simulation data of 15 global climate models under different extreme Shared Socioeconomic Pathway (SSP) scenarios. As a result, future changes in urban cooling island (UCI) magnitudes were estimated between historical (2015–2020) and future timelines: 2030s (2021–2040), 2050s (2041–2060), 2070s (2061–2080), and 2090s (2081–2100). Our results showed different land use types in blue and green space across the study area were predicted to present various changes in the next 80 years, with forest, grassland, and arable land experiencing the most significant land use transfer. The future UCI intensity of cities under SPP5-8.5 (12) was found to be lower than that under SPP2-4.5 (15), indicating that cities may be expected to experience decreases in UCI magnitudes in the future under SSP5-8.5. When there is no expansion of urban development land, we found that the conversion of different land use types into blue and green space leads to little change in future UCI intensity. While the area growth of forests and water bodies is proportional to the increase in UCI, the increase of farmland was observed to have the most significant impact on reducing the amplitude of urban UCI. Given that Huai’an City, Yancheng City, and Yangzhou City have abundant blue–green space, the urban cooling island effect was projected to be more significant than that of other cities in the study area under different SSP scenarios. The simulation results of the WRF model indicate that optimizing the layout of urban blue–green space plays an important role in modulating the urban thermal environment.

show abstract

“…90% of the complete data set was utilized as the training set, and the rest was used as the test set. The training data set was trained via DT, 51 RF, 52 KNN, 53 XGB, 54 and LGB models, respectively, and we have adjusted the hyperparameters of these five models, as summarized in Table S6. The 10-fold crossvalidation method was chosen to evaluate the models.…”

Section: Model Evaluation Andmentioning

confidence: 99%

Large Data Set-Driven Machine Learning Models for Accurate Prediction of the Thermoelectric Figure of Merit

Zhang

et al. 2022

ACS Appl. Mater. Interfaces

View full text Add to dashboard Cite

The figure of merit (zT) is a key parameter to measure the performance of thermoelectric materials. At present, the prediction of zT values via machine leaning has emerged as a promising method for exploring high-performance materials. However, the machine learning-based predictions still suffer from unsatisfactory accuracy, and this is related to the size of the data set, the hyperparameters of models, and the quality of the data. In this work, 5038 pieces of data of thermoelectric materials were selected, and several regression models were generated to predict zT values. This large data set-driven light gradient boosting (LGB) model with 57 features performed with an excellent accuracy, achieving a coefficient of determination (R 2) value of 0.959, a root mean squared error (RMSE) of 0.094, a mean absolute error (MAE) of 0.057, and a correlation coefficient (R) of 0.979. Owing to the large size of the data set, the prediction accuracy exceeds that of most reported zT predictions via machine learning. The “ME Lattice Parameter” was verified as the most important feature in the zT prediction. Furthermore, nine potential candidates were screened out from among one million pieces of data. This study solves the problem of the data set size, adjusts the hyperparameters of the models, uses feature engineering to improve data quality, and provides an efficient strategy to perform wide-ranging screening for promising materials.

show abstract

Dependence between urban morphology and outdoor air temperature: A tropical campus study using random forests algorithm

Cited by 40 publications

References 37 publications

Effects of Sky View Factor on Thermal Environment in Different Local Climate Zoning Building Scenarios—A Case Study of Beijing, China

Effects of Sky View Factor on Thermal Environment in Different Local Climate Zoning Building Scenarios—A Case Study of Beijing, China

Evolution Patterns of Cooling Island Effect in Blue–Green Space under Different Shared Socioeconomic Pathways Scenarios

Large Data Set-Driven Machine Learning Models for Accurate Prediction of the Thermoelectric Figure of Merit

Contact Info

Product

Resources

About