A method, equipment, and medium for predicting carbon emission factors per passenger in public transportation.

By constructing a multi-level influencing factor model and introducing a sparse activation strategy, the dynamic change problem of public transport carbon emission prediction is solved, the prediction accuracy and interpretability are improved, and more scientific emission reduction strategies are supported.

CN120013301BActive Publication Date: 2026-05-26ZHEJIANG UNIV
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
ZHEJIANG UNIV
Filing Date
2025-04-22
Publication Date
2026-05-26

AI Technical Summary

Technical Problem

Existing methods for predicting carbon emissions from public transport rely on fixed emission factors and static traffic data, making it difficult to depict the dynamic changes in public transport operations. They also ignore the influence of multiple levels of factors, such as stations, links, and routes, resulting in low model prediction accuracy and poor interpretability, which affects the scientific validity and effectiveness of emission reduction strategies.

Method used

We construct influencing factors at the bus stop, link, and route levels, integrate multi-level factors through routing mechanisms and sparse activation strategies, predict the carbon emissions per passenger, introduce expert equilibrium loss to prevent overload or idleness of expert sub-models, and use the MST-Boost model for prediction.

Benefits of technology

It improves the accuracy and interpretability of carbon emission prediction for public transportation, provides a more accurate basis for evaluating emission reduction strategies, and enhances the robustness and generalization ability of the model.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120013301B_ABST
    Figure CN120013301B_ABST
Patent Text Reader

Abstract

This invention discloses a method, device, and medium for predicting public transport carbon emission factors per passenger. The method includes: constructing bus stop-level, bus link-level, and bus route-level influencing factors; using these factors as input factors, processing them through a routing mechanism, calculating the routing score for each input factor, and then assigning each input factor to a corresponding expert sub-model to predict the corresponding public transport carbon emission per passenger. The method also involves identifying key factors for public transport carbon emissions and / or determining the coupling effect between pairs of input factors. Context-sensitive frequency and call balance are introduced to evaluate the local and global load of the expert sub-models, and the impact of expert switching on the prediction output is quantified through expert perturbation sensitivity.
Need to check novelty before this filing date? Find Prior Art