Abstract:In urban areas, traffic-related contamination is one of the main contributors to environmental deterioration, and the pollution from public transit buses is a major component. To mitigate these impacts, it is essential to estimate bus emissions and analyze their characteristics. This paper proposes a hybrid model based on gated recurrent unit (GRU) and extreme gradient boosting (XGBoost), termed GRU-XGB, to predict gaseous pollutants from bus emissions (CO, CO2, HC, NOX) under real conditions. On-road experimental data collected from CNG-fueled and diesel-powered buses in Zhenjiang was used as a case study to verify the model's effectiveness. A comparison between the proposed and other state-of-the-art models reveals that GRU-XGB performs best for all evaluation metrics on both microscopic and aggregative levels, with an average correlation coefficient above 0.98 and an average MAPE lower than 9%. Moreover, the results of estimation errors analysis suggest that the real conditions of bus stations are more complicated than those of intersections and road sections. In most cases, however, the emission factors produced from intersections are proven to be the highest. Furthermore, operating patterns are shown to be the most significant factors, with relative importance equal to 45.09% and 71.68% for CNG and diesel buses, respectively. Besides, the results also indicate that humidity has little impact on this issue, while the influence of temperature is obvious, with relative importance equal to 17.56% and 9.41% for CNG and diesel buses, separately. Such findings can provide theoretical guidance for both emission estimation and environmental protection. Also, it is applicable for the management of accurate monitoring from an urban-level and can be integrated into emission simulation tools.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the impact of gaseous pollutants (such as carbon monoxide (CO), carbon dioxide (CO₂), hydrocarbons (HC) and nitrogen oxides (NOₓ)) emitted by buses on environmental quality in urban traffic. In order to mitigate these impacts, it is necessary to accurately estimate the emissions of buses under actual conditions and analyze their characteristics. Specifically, the paper proposes a hybrid model (called GRU - XGB) based on Gated Recurrent Unit (GRU) and eXtreme Gradient Boosting (XGBoost) for predicting gaseous pollutants emitted by buses. This model aims to solve the following key problems: 1. **Time - dependence**: Existing studies usually use current or past driving states (such as speed, acceleration, number of passengers, etc.) to estimate bus emissions on the road, while ignoring the influence of driving patterns. The driving pattern is a composite variable that may cover the influence of multiple factors and is directly related to the emission rate. 2. **Model comparison**: There is a lack of comprehensive comparison of various models in solving this problem. 3. **Impact of weather conditions**: Although weather conditions (such as temperature, humidity, etc.) may affect emissions, most existing studies do not consider this when estimating bus emissions. ### Solution To overcome the above problems, the paper proposes a hybrid model named GRU - XGB, whose main features are as follows: - **GRU part**: Use a two - layer GRU network to process historical emission data and generate new features that can represent the operation mode. GRU is an improved Recurrent Neural Network (RNN), which is especially suitable for processing time - series data and can effectively capture long - term dependencies. - **XGBoost part**: Combine the features generated by GRU and other external factors (such as driving state, road condition, weather condition, etc.), and perform the final emission prediction through XGBoost. XGBoost is an efficient tree - boosting system and can perform excellently when dealing with high - dimensional non - linear data. ### Main contributions 1. **New framework**: Developed the GRU - XGB model for estimating CO, CO₂, HC and NOₓ emitted by buses. 2. **Model performance comparison**: Implemented several state - of - the - art models and compared their performance in predicting bus emissions. 3. **Comprehensive consideration of factors**: Considered the influence of operation mode and weather conditions in the study and quantified their relative importance. ### Experimental verification The paper uses actual data collected in Zhenjiang, China, including data of compressed natural gas (CNG) buses and diesel buses, to carry out model verification and analysis. The experimental results show that the GRU - XGB model performs best on micro - and macro - level evaluation indicators, with an average correlation coefficient exceeding 0.98 and an average absolute percentage error (MAPE) lower than 9%. ### Conclusion This study provides a theoretical guidance, which is helpful for emission estimation and environmental protection. At the same time, this model is suitable for accurate monitoring at the city level and can be integrated into emission simulation tools.

Estimating gaseous pollutants from bus emissions: A hybrid model based on GRU and XGBoost

Estimation of bus emission models for different fuel types of buses under real conditions

Understanding and estimating the carbon dioxide emissions for urban buses at different road locations: A comparison between new-energy buses and conventional diesel buses

Unraveling nonlinear and interaction effects of various determinants on bus gaseous emissions

Exploring the influence of contributing factors and impact degree on bus emissions in real-world conditions

Roadside Assessment of a Modern City Bus Fleet: Gaseous and Particle Emissions

Supplementary Material to "A Novel Method for Quantifying the Contribution of Regional Transport to PM2.5 in Beijing (2013–2020): Combining Machine Learning with Concentration-Weighted Trajectory Analysis"

Application of the XGBoost Machine Learning Method in PM2.5 Prediction: A Case Study of Shanghai

On-road Remote Sensing Measurements and Fuel-Based Motor Vehicle Emission Inventory in Hangzhou, China

NOx Emission prediction of heavy-duty diesel vehicles based on Bayesian optimization -Gated Recurrent Unit algorithm

Vehicle mix evaluation in Beijing's passenger-car sector: From air pollution control perspective

Evaluating real-world CO2 and NOX emissions for public transit buses using a remote wireless on-board diagnostic (OBD) approach.

Integrated MOVES model and machine learning method for prediction of CO 2 and NO from light-duty gasoline vehicle

Features Importance Analysis of Diesel Vehicles' NOx and CO2 Emission Predictions in Real Road Driving Based on Gradient Boosting Regression Model

Air quality and climate impacts of alternative bus technologies in Greater London

Study on the Emission Characteristics of Typical City Buses under Actual Road Conditions

Real-world Fuel Consumption and CO2 Emissions of Urban Public Buses in Beijing

Online characterization of primary and secondary emissions of particulate matter and acidic molecules from a modern fleet of city buses

Spatiotemporal Analysis of CO2 Emissions and Emission Reduction Potential of Beijing Buses Using Smart Card Data

Assessing on-road emissions from urban buses in different traffic congestion scenarios by integrating real-world driving, traffic, and emissions data

Models for predicting vehicle emissions: A comprehensive review