Much of the research on climate change has focused on carbon reduction in cities or countries. However, more attention needs to be paid to how to achieve carbon neutrality in the urban design and planning stage, and the lack of quantitative analysis of carbon related to urban space makes it difficult to locate urban space and provide direct guidance for urban planning and design. This study proposed three optimization paths to achieve carbon neutrality in multi-scale urban building clusters. Firstly, we reconstructed the quantitative calculation system of urban building communities with the goal of carbon neutrality; secondly, we screened the carbon source reduction and carbon sink interventions that are suitable for multi-scale urban building communities; finally, we constructed a carbon emission and carbon sink calculation system of planning and design schemes based on the layout of relevant elements of planning and design schemes with a grid cell of 100 × 100 m. In practice, there was a gap of about 115,000 tons of CO₂ from the carbon-neutral target and 26% of carbon emission was distributed in the Xiajiabian Station TOD. In this study, nine types of carbon reduction measures were adopted to achieve carbon neutrality in the region, among which the highest carbon reduction was achieved by biomass energy measures, accounting for 29% of the total carbon reduction of 33,745.27 T. The objective of this study is to accurately and quantitatively assess the carbon targets of urban spaces at different scales and adopt effective measures to achieve carbon neutrality.

In this study, we proposed a novel traffic flow prediction model aimed at providing more accurate and effective traffic information. The proposed model utilized a multi-scale convolutional neural network in conjunction with a long short-term memory (LSTM) neural network to efficiently extract spatial and temporal features from the data. The network model was trained using the L1 loss function to optimize its performance.To evaluate the performance of the proposed model, we conducted experiments on two public datasets, PeMSD4 and PeMSD8. The results demonstrated the model's strong competitiveness in the field of traffic flow prediction. Specifically, on the PeMSD4 dataset, our model achieved a mean absolute error (MAE) of 20.94, a mean absolute percentage error (MAPE) of 13.32%, and a root mean square error (RMSE) of 30.96. These findings suggested that the proposed model hold significant promise in the realm of traffic flow forecasting.