| Abstract |
In the construction of smart city, the carbon emission reduction problem of road traffic needs to be solved urgently. It is of great significance to introduce reasonable low-carbon policies. Based on urban private cars trajectory data, this study, respectively, establishes the genetic algorithm-back propagation neural network model (GA-BP) and back propagation-adaptive boosting algorithm neural network model (BP-AdaBoost) to predict the carbon emissions of private cars. By comparing the two neural network models, the GA-BP neural network model has better prediction results. Next, this study establishes the cost-benefit model for consumers and compares consumers' participation willingness, emission reduction effect, and social benefits of consumers from the perspective of six kinds of low-carbon policies. The results show that the overall effect of the low-carbon policy mix of free quota is better than that of paid quota. In addition, different low-carbon policy mixes innovations have different policy implementation effects under different indicators. Overall, the low-carbon policy mix of carbon trading and emission reduction subsidy is better in the short term, and the low-carbon policy mix of carbon tax and emission reduction subsidy is better in the long term. © 2021 Wenjie Chen et al. |
