Effects of green quantity and structure on thermal comfort and air quality of urban residential areas based on ENVl-met model

Air quality in urban residential areas depends on the vehicles, celebrations, carbon emissions, etc. experienced around the day/ year. The effects of carbon emissions on residents’ thermal comfort are more important than control measures, for which measurements are mandatory. This article, therefore, introduces an ENVi-Met-based thermal structure assessment model to identify the air quality affected by carbon emissions. In the thermal structure assessment, the three-dimensional space of the urban Area, including the emitting and emission-conducted regions, is modeled to compute the air quality index (AQI). A stabilized assessment of the mean AQI that distinguishes between the worst and better thermal structures is estimated from continuous assessments. In this process, the change in AQI from the lowest to the highest is the boundary for the structural evaluation. The primary objective of this study is to develop an integrated framework that explains how carbon emissions affect the thermal structure and air quality in urban residential areas. The work employs the ENVI-met microclimate simulation system to model the spatial and temporal distribution of emissions, thereby facilitating a complex representation of pollutant dynamics within three-dimensional urban environments. It subsequently sets high and low AQI limits that reflect differing thermal conditions, making it easier to distinguish areas that are thermally strained from those that are well-ventilated. The model employs convergence-based transfer learning to maintain stable AQI forecasts over time, ensuring predictions remain consistent even as environmental variables change. Lastly, the framework examines thermal comfort by considering the combined effects of emission intensity, vegetation absorption, and microclimatic interactions. The boundary-based change differentiation is validated using converged transfer learning to identify the maximum changes in thermal structures. Learning converges for the AQI differentiation value for stabilization detection. Therefore, this stabilization value is used to train the learning network to maintain boundary consistency across different structural changes. The proposed model achieves an 11.39% high AQI analysis with a maximum Kappa of 12.58% between the convergence and stability for the time/day and stable variants under hot climatic conditions.