A method for predicting carbon emissions from green transportation based on wavelet threshold denoising

The commonly used prediction methods nowadays find it difficult to balance the long-term and short-term operating conditions, resulting in low prediction accuracy. Therefore, a green traffic carbon emissions prediction method based on wavelet threshold denoising is proposed. Firstly, the LMDI algorithm is used to calculate the carbon emission impact factors and complete the collection of green transportation carbon emission data. Secondly, the wavelet threshold denoising method is used to denoise the green traffic carbon emission data. Finally, using the denoised carbon emission data as input and the carbon emission prediction results as output, an improved BP neural network is used to construct a traffic carbon emission prediction model. The test results indicate that the relative error of the studied prediction model is smaller, and the Pearson similarity coefficient is relatively larger, indicating that the prediction model has good predictive ability in both short and long term predictions.