Predictive methods for CO 2 emissions and energy use in vehicles at intersections

doi:10.21203/rs.3.rs-5157930/v1

Download PDF

Article

Predictive methods for CO 2 emissions and energy use in vehicles at intersections

https://doi.org/10.21203/rs.3.rs-5157930/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

This study examines CO₂ emissions and vehicle energy consumption at high-traffic intersections in urban areas. Existing emission models at the macro, meso, and microscales often fail to accurately represent real traffic conditions, especially at intersections with frequent stop-and-go maneuvers. New predictive models were developed using methods such as linear regression, least absolute shrinkage and selection operator (LASSO), Ridge regression, Random Forest, and Extreme Gradient Boosting (XGBoost), with XGBoost providing the highest accuracy. The density-based spatial clustering of applications with noise (DBSCAN) algorithm was used to group data specific to intersection areas, enabling targeted analysis. Real-world driving data were collected using portable emissions measurement systems and the Hioki 3390 power analyzer. The developed models were validated and applied in simulations, including Vissim software, to improve road infrastructure planning and traffic management. These methods offer a refined approach to reducing emissions and optimizing energy use in urban transportation networks.

emission modeling

vehicle emissions

exhaust measurements

portable emissions measurement systems

energy modeling

electric vehicles

Air pollution and energy efficiency in urban transportation are critical issues in the context of sustainable development. With 2023 recorded as the warmest year in Earth's history [1, 2], cities with heavy traffic, signalized intersections, and roundabouts are hotspots for emissions and delays [3, 4]. These areas, characterized by frequent stops, accelerations, and decelerations, contribute significantly to CO₂ emissions and energy consumption [5, 6].

This study addresses gaps in predictive emission and energy consumption models, focusing on the microscale dynamics of urban intersections. Traffic flow models are typically divided into macro, meso, and micro scales [7]. Macroscale models analyze large regions to inform broad transportation policies [8–10], while mesoscale models focus on specific urban zones, evaluating the effects of infrastructure elements like intersections on emissions and energy [11–13]. Microscale models, such as car-following and point-queue models, track individual vehicles at specific locations like intersections [14], offering critical insights into traffic dynamics and emissions [15, 16]. However, existing microscale models often rely on limited speed ranges, reducing accuracy in predicting emissions at intersections where vehicle conditions vary significantly [17–20].

Given these limitations, there is a need for improved models to analyze emissions and energy consumption, particularly in traffic management scenarios like adaptive signal control, which can reduce stops and emissions [21, 22]. Current models, especially for electric and hybrid vehicles, require updates to improve CO₂ emission estimates [23–26]. Emerging models, including car-following, point-queue, and shockwave models, offer new approaches but still have limitations in practical applications.

This study aims to develop new models for CO₂ emissions and energy consumption at intersections using real-world data from the Portable Emissions Measurement System (PEMS) and the Hioki 3390 power analyzer. Techniques like DBSCAN clustering and machine learning models (e.g., XGBoost, Random Forest) were applied, with XGBoost providing the best predictive performance. These models, integrated with simulations like Vissim, offer insights for future traffic management and infrastructure planning, including for electric vehicles.

The study framework, shown in Fig. 1, consists of four main parts. First, 12 vehicles (EURO2 to EURO6) and one electric vehicle (EV) were selected to test the proposed methodology across different emission profiles. Each vehicle was prepared for the installation of the Portable Emissions Measurement System (PEMS) and a power analyzer. A specific urban route with eight intersections was chosen to capture the dynamics of high-traffic areas, characterized by frequent stops and accelerations that contribute to elevated emissions and energy consumption.

In the second phase, CO₂ emissions and energy consumption data were collected using PEMS, which recorded vehicle speed, position (via GPS), and additional OBD II data. In the third phase, the data underwent quality checks and processing. The DBSCAN algorithm was used to isolate intersection-specific data for further analysis. The data were then stored in .csv format and used to develop machine learning models predicting CO₂ emissions and energy consumption based on vehicle speed and acceleration, with the goal of microscale applications.

Finally, predictive models were developed using techniques such as linear regression, LASSO, Ridge, Random Forest, and XGBoost. Model accuracy was validated through metrics like mean squared error (MSE) and R². XGBoost outperformed other models, highlighting its potential to improve traffic management and reduce emissions in urban intersections, where emissions are typically highest.

2.1. Vehicles used for the test

For the tests, 12 vehicles were used and their selected parameters are presented in Table 1. Executing emission and energy consumption models for this fleet provides information on the scalability of the developed method for larger computational models. The choice of this set of vehicles was also intended to diversify the types of power source. The selected vehicles include those powered by gasoline, diesel, LPG, CNG, as well as electric vehicles. All vehicles were tested at a service station before conducting road tests, in order to detect potential faults, following the standard procedure for approved vehicles for road use.

Table 1

Selected technical parameters of the tested vehicles
No.	The emission standard	Production year	Engine capacity [cm³]	Engine type	Fuel type	Max. power [kW]/with speed [rpm]	Max. torque [Nm]/with speed [rpm]	Gearbox / number of gears	Power system	Aftertreatment system	Weight [kg]
Vehicle 1	Euro 2	1998	1598	Spark ignition	Gasoline	88/6300	144/4500	Manual / 5	MPI	TWC	1230
Vehicle 2	Euro 3 (1)	2001	1991	Spark ignition	Gasoline/LPG	90/5800	175/4500	Manual / 5	MPI	TWC	1600
Vehicle 3	Euro 3 (2)	2001	2435	Spark ignition	Gasoline/CNG	103/4500	220/3750	Manual / 5	MPI	TWC	1660
Vehicle 4	Euro 4 (1)	2003	1199	Spark ignition	Gasoline	55/5600	110/4000	Manual / 5	MPI	TWC	1040
Vehicle 5	Euro 4 (2)	2004	1998	Spark ignition	Gasoline	115/6000	190/4500	Manual / 5	MPI	TWC	1430
Vehicle 6	Euro 5 (1)	2011	1591	Spark ignition	Gasoline	99/6300	164/4850	Manual / 6	GDI	TWC	1305
Vehicle 7	Euro 5 (2)	2012	1397	Spark ignition	Gasoline	96/5500	190/2250	Manual / 6	MPI	TWC	1280
Vehicle 8	Euro 6 (1)	2014	1149	Spark ignition	Gasoline/LPG	55/5500	105/4250	Manual / 5	MPI	TWC	980
Vehicle 9	Euro 6 (2)	2018	1560	Diesel	Diesel oil	88/3500	300/1750	Manual / 6	CR	DPF + SCR + DOC	1429
Vehicle 10	Euro 6 (3)	2017	1197	Spark ignition	Gasoline	74/4500	175/1500	Manual / 6	MPI	TWC	1205
Vehicle 11	Euro 6 (4)	2021	1497	Spark ignition / Electric (59 kW)	Gasoline/ Hybrid	67/4500	169/4000	Manual / 6	MPI	TWC	1095
Vehicle 12	-	2019	-	Electric motor	Electricity	125/4775 kW	-	Single-speed automatic transmission	EV	-	1390

Data for vehicles meeting the same emission standards and powered by the same fuels were aggregated into a single group for further analysis. An overview of the vehicles used in the study is presented in Fig. 2. The vehicles were tested during the period from May to July of 2018–2023. The broad range of different vehicles examined allows for the collection of a substantial amount of data, which will subsequently be processed to develop emission models for conventional vehicles and energy consumption models for electric vehicles, specifically for intersection areas.

2.2. Testing equipment

The research utilized the PEMS system from the Rzeszów University of Technology's Department of Automotive and Transport Engineering. This system measures exhaust emissions using sensors like a flame ionization detector (FID) for hydrocarbons (HC), a non-dispersive infrared (NDIR) spectrometer for CO and CO₂, and a chemiluminescence detector (CLD) for nitrogen oxides (NO and NO₂) [27]. European Commission regulations allow a maximum 10% deviation between PEMS and stationary analyzer (chassis dynamometer) measurements for CO₂ [28, 29]. PEMS is placed in the vehicle trunk, with sensors connected to the exhaust pipe, which must be heated to 190°C to prevent hydrocarbon condensation [30]. The system also includes sensors for air temperature, humidity, and a GPS transmitter. An OBDII interface was connected to monitor engine performance.

Energy consumption was measured using the Hioki PW3390 power analyzer, which accurately measures electrical power over a wide range, from direct current to inverter frequencies, with ± 0.04% accuracy [31]. The device has four input channels and can handle currents up to 4000 A AC/DC, making it suitable for high-frequency and low-power factor analyses [32]. Its advanced features, like harmonic analysis and transient power calculations, are valuable for studying power conversion efficiency and inverter motor performance in electric and hybrid vehicles. Figure 3 shows the system setup.

Table 2 shows selected technical parameters for measurements using the PEMS system and power analyzer.

Table 2

Selected technical parameters of measuring devices
Parameter	PEMS Horiba OBS-2200	Hioki PW3390 Power Analyzer
Measurement Method	NDIR (CO, CO2), FID (THC), CLD (NOx)	Voltage and current sensors, broad frequency range up to 200 kHz
Accuracy	± 2.5% (CO, CO2, THC, NOx)	Voltage: ±0.04% rdg. ±0.05% f.s.
Accuracy	± 2.5% (CO, CO2, THC, NOx)	Current: ±0.04% rdg. ±0.05% f.s.
Measurement Range	CO: 0–10%	Voltage range: 15 to 1500 V
	CO2: 0–10%	Current range: 0.1 A to 20 kA
	THC: 0-10000 ppm
	NOx: 0-3000 ppm
Sampling Frequency	1 Hz	50 ms
Power Measurement Range	-	0.0150 W to 39.600 MW
Harmonic Measurement	-	Up to 100th order
Noise Measurement	-	Max. frequency: 200 kHz
Synchronization Frequency Range	-	0.5 Hz to 5 kHz
External Interfaces	LAN, USB	LAN, USB, RS-232C, CF card
Power Supply	-	100 240 V AC, 50/60 Hz

2.3. Description of the research route

The route investigated included urban segments within the city of Rzeszów. As a dynamically developing city in southeastern Poland, Rzeszów has a well-developed road infrastructure, covering both national and local roads. The city is accessed by the S19 expressway, which is part of the international route Carpathia, making Rzeszów a major transportation hub in the region. The city center and its suburbs experience high traffic volumes, particularly during peak hours, often resulting in congestion. The public transport system is mainly based on buses, supported by a well-developed network of connections; however, private vehicle traffic remains predominant. Rzeszów continuously modernizes its streets by introducing new solutions, such as turbo roundabouts and intelligent traffic management systems, to improve traffic flow and safety. An important aspect of the city's infrastructure is its numerous bike lanes and pedestrian areas, which promote alternative transportation modes.

The route is depicted in Fig. 4. It covered a stretch of 16 kilometers, including a passage through eight intersections that were analyzed and used for further data processing. Intersections 1, 2, 4, 5, and 7 were signalized intersections, while intersections 6 and 8 were two-lane roundabouts. The traversal of these intersections involved vehicle queueing and frequent stop and go operations, leading to increased emissions. Each vehicle completed the route twice, traveling to the end at intersection 8 and then returning.

2.4. Methods of analysis and data processing obtained

Data analysis was conducted using Google Colab, a cloud-based platform that enables code execution in a Jupyter Notebook environment, offering access to powerful resources like GPUs and TPUs [33]. Vehicle data were saved in a .csv file and stored in Google Drive. Python, along with libraries like NumPy, Pandas, Matplotlib, Seaborn, and scikit-learn, was used for data analysis and model development.

The data for each vehicle group were imported into Colab, where quality checks ensured completeness and numerical accuracy. A key step was developing algorithms to detect vehicle movement near intersections, crucial for modeling start-stop behaviors. The k-means and DBSCAN clustering methods were analyzed for this purpose. DBSCAN, a density-based clustering algorithm, effectively identifies clusters without requiring a predefined number of clusters, making it ideal for detecting irregular shapes like intersection areas [34–36]. This method helped isolate dense traffic zones while classifying free-flowing areas as noise.

The analysis then evaluated various regression models for predicting CO₂ emissions and energy consumption using two explanatory variables: speed (V) and acceleration (a). After data cleaning, models were trained and tested using five algorithms: linear regression, LASSO, Ridge, Random Forest, and XGBoost. Model performance was assessed using mean squared error (MSE) and the coefficient of determination (R²), with the best-performing model selected for further analysis.

TThis chapter presents the results from data collected exclusively for intersection areas using DBSCAN algorithms to capture vehicle approaches, traversals, and exits. Based on this data, detailed CO₂ emissions and energy consumption models were developed and validated for passenger vehicles.

Figure 5 outlines the data processing steps. For model development, vehicles were grouped by emission standards (e.g., EURO6), and traffic data was clustered using DBSCAN to focus on intersection areas with frequent start-stop operations. This method enables accurate predictions of emissions and energy consumption in these critical traffic zones.

3.1. Data selection for intersection areas

In the process of selecting data for intersection areas, the first step involved applying the DBSCAN algorithm to cluster data points based on their geographical locations. Prior to clustering, geographic coordinate data (latitude and longitude) were standardized using the StandardScaler method. This standardization normalized the data, eliminating the effects of scale differences among the coordinates. Subsequently, experiments were conducted with various parameter values for the DBSCAN algorithm, including eps (neighborhood radius) and min_samples (minimum number of points required to form a cluster). The testing of these parameters aimed to identify the optimal settings for extracting significant clusters corresponding to intersections within the study area. The use of scaled coordinates in combination with different parameters facilitated the visualization and analysis of results, which is crucial for the subsequent stage of data selection and analysis for intersection areas. The results of the analyzes for different values of the neighborhood radius and minimum number of points for the route studied are presented in Fig. 6.

Figure 7 illustrates the optimal predictions for the data groups that denote intersections along the study road segment. This chart uses the analysis parameters eps = 0.1 and min_samples = 20, which accurately identified 8 intersections, corroborating real-world data. The cluster centroids were determined on the basis of the average geographical coordinates (latitude and longitude) of the points within each identified cluster by the DBSCAN algorithm. Initially, all points assigned to a specific cluster by the algorithm were gathered, and then the average values of their geographical coordinates were computed. These resultant values represent the centroids of the clusters, marking the central locations of the intersections in the analyzed area. The centroids obtained, depicted on the graph as red crosses, serve as reference points for further analyses related to vehicle emissions and energy consumption within these designated areas.

On the chart, the points labeled 0, -1 and other numbers represent different assignment of groups made by the DBSCAN algorithm. The points labeled 0 belong to the first identified cluster, indicating that DBSCAN considered them part of a dense data region forming a cluster. The algorithm sequentially assigns numbers to clusters, starting from 0, indicating the first group of points that meets the density criteria. Points labeled as -1 are classified as noise or outliers. These points were not assigned to any cluster because they did not have a sufficient number of neighboring points within the specified eps distance to be considered part of a cluster. The − 1 label indicates that these points are isolated or do not fit into any of the dense regions identified. This classification allows for visualizing how DBSCAN differentiates between clustered points and those deemed as noise.

The centers of the intersection areas analyzed are the points of maximum accumulation of start-stop operations for passing vehicles. At these locations, internal combustion vehicles generate the highest levels of exhaust emissions, contributing significantly to the environmental pollution around these road arteries. Pedestrians in these areas, especially near crosswalks, are particularly exposed to exhaust emissions.

3.2. Creation and validation of CO₂ emission and energy consumption models for motor vehicles

The developed method for predicting intersection locations enables rapid data segregation, which can then be used, for example, to create more accurate models to predict CO₂ emissions from internal combustion engine vehicles and energy consumption for electric vehicles.

To predict CO₂ emissions and energy consumption based on variables V and a, a comprehensive process of model development and validation was carried out. Initially, data uploaded to Google Colab were processed and saved as a separate CSV file, including information on emissions or energy consumption depending on the type of vehicle powertrain, as well as vehicle speed and acceleration data. A sample view of the data and its columns used for model training is presented in Fig. 8.

Initially, the data were cleaned to remove invalid and missing values, and then converted to a numerical format. After data preparation, the data set was split into training and test sets in a 70:30 ratio. Standardization of input variables was applied to ensure comparability in scale. Five regression models were trained: Linear regression, LASSO, Ridge, Random Forest, and XGBoost. Each model was evaluated using the mean squared error (MSE) and the coefficient of determination (R²). The coefficient of determination measures how well a regression model fits the data. It is defined as (1):

$$\:{R}^{2}=1-\frac{\sum\:_{t=1}^{n}\left({y}_{t}-\widehat{y}{\:}_{t}\right){\:}^{2}}{\sum\:_{t=1}^{n}\left({y}_{t}-\stackrel{-}{y}\right){\:}^{2}}$$

where:

y_t – the observed actual values,

$\:\widehat{y}{\:}_{t}$ – are the values predicted by the model,

$\:\stackrel{-}{y}$ – is the mean of the actual values,

n – the number of observations.

The Mean Squared Error (MSE) measures the average of the squared differences between the actual and predicted values. It is defined as (2):

$$\:MSE=\frac{1}{n}\sum\:_{t=1}^{n}\left({y}_{t}-\widehat{y}{\:}_{t}\right){\:}^{2}$$

where:

y_t – the observed actual values,

$\:\widehat{y}{\:}_{t}$ – are the values predicted by the model,

n – the number of observations.

MSE measures how large the prediction errors are. Smaller MSE values indicate a better fit of the model as they represent smaller differences between actual and predicted values.

The data for vehicles were aggregated into larger groups according to the EURO emission standard and further categorized according to the type of fuel used. Consequently, 10 groups were created, for which emission and energy consumption models for intersection areas were developed and subsequently validated. The validation results are presented in Table 3.

Table 3

Validation results of the obtained models by vehicle emission class
Vehicle Type	Validation Metric	Linear Regression	LASSO	Ridge	Random Forest	XGBoost
EURO2 (gasoline)	MSE	0.29	0.22	0.2	0.16	0.18
EURO2 (gasoline)	R²	0.62	0.55	0.62	0.71	0.67
EURO3 (gasoline)	MSE	0.22	0.13	0.12	0.09	0.08
EURO3 (gasoline)	R²	0.75	0.72	0.74	0.82	0.85
EURO3 (LPG)	MSE	0.21	0.12	0.11	0.08	0.07
EURO3 (LPG)	R²	0.78	0.76	0.77	0.82	0.84
EURO3 (CNG)	MSE	0.21	0.12	0.11	0.08	0.07
EURO3 (CNG)	R²	0.8	0.77	0.78	0.82	0.85
EURO4 (gasoline)	MSE	0.11	0.11	0.11	0.07	0.06
EURO4 (gasoline)	R²	0.82	0.8	0.81	0.85	0.88
EURO5 (gasoline)	MSE	0.11	0.11	0.1	0.07	0.05
EURO5 (gasoline)	R²	0.83	0.81	0.82	0.86	0.9
EURO6 (gasoline)	MSE	0.22	0.13	0.13	0.08	0.06
EURO6 (gasoline)	R²	0.8	0.78	0.79	0.85	0.9
EURO6 (Diesel)	MSE	0.24	0.15	0.14	0.09	0.07
EURO6 (Diesel)	R²	0.78	0.76	0.77	0.8	0.83
EURO6 (Hybrid)	MSE	0.13	0.14	0.13	0.09	0.06
EURO6 (Hybrid)	R²	0.79	0.77	0.78	0.82	0.88
EV (energy)	MSE	7.33	7.43	7.34	0.94	0.4
EV (energy)	R²	0.74	0.73	0.74	0.97	0.99

Table 3, which presents the validation results of the regression models for different types of vehicles, including electric vehicles (EV), reveals significant differences in the effectiveness of predicting CO₂ emissions and energy consumption. The regression models analyzed include linear regression, LASSO, Ridge, Random Forest, and XGBoost. The results indicate that XGBoost achieves the best performance across both quality metrics: Mean square error (MSE) and coefficient of determination (R²). For electric vehicles, the XGBoost model achieved the lowest MSE of 0.40 and the highest R² of 0.99, demonstrating exceptional precision and explanatory power for data variability. In comparison, other models, such as linear regression and LASSO, showed considerably lower performance in terms of both MSE and R². Therefore, XGBoost is recommended to predict CO₂ emissions and energy consumption for electric vehicles, and this technique was selected for further analysis. However, consideration should also be given to optimizing other models, such as Random Forest, which also yielded promising results.

3.3. Use of developed models for prediction of CO₂ and energy consumption

This chapter details the use of models developed to predict CO₂ emissions and energy consumption for electric vehicles (EVs). These models, grouped by emission class, were applied to verify their accuracy for infrastructure assessments. Comparative CO₂ emission maps and cumulative values were generated. Additionally, a general microscale CO₂ emission model was created using data from the entire road segment to serve as a reference for validating the intersection-specific model.

This approach enables comparison between real-world data and both the general and intersection models, helping to identify potential deviations and assess the applicability of general models to the unique traffic conditions at intersections.

3.3.1. Generation of CO₂ emission maps and comparison of predictive capabilities of models

The developed models for CO₂ emissions and energy consumption allow not only the prediction of CO₂ values and vehicle energy consumption, but also the generation of maps that indicate their points of origin. This is particularly important in traffic management strategies, such as at intersections with traffic lights. In this way, potential areas with the highest accumulation of emissions can be identified, which directly impacts pedestrians near these road arteries. An example visualization of emissions for a group of vehicles that meet the EURO 5 standard is shown in Fig. 9.

To compare the methods of creating emission models and further validate the obtained results, an additional model was created based on all aggregated data from the road tests, referred to as the "city micro model." The model based solely on data from intersections was named the "intersection micro model."

Figure 9 identifies the areas with the highest CO₂ emissions, with Intersection 1 and the approach to Intersection 2 showing the most emissions due to frequent start-stop operations. Both microscale models generally align with the actual emission hotspots, but underestimations occur in areas with the highest emissions. Cumulative emissions for the road section reveal a 20% difference between modeled and actual values, which could scale in larger studies. Converting these results to emission factors in g/km would yield similar discrepancies.

Table 4 compares actual emissions with results from the microscale intersection model, the full city route model, and the COPERT program. COPERT, widely used for environmental analyses in Europe, estimates vehicle emissions based on factors like vehicle type, fuel, and engine technology [37, 38]. For this study, COPERT used the average speed over a 1.05 km section for comparison.

Table 4

Comparison of sum of real CO₂ emissions and energy consumption, values for intersection micro, city micro and COPERT program models
Vehicle Type	Real world	Intersection micro model	City micro model	COPERT
EURO2 (gasoline)	220 g	218 g	170 g	201 g
EURO3 (gasoline)	200 g	198 g	160 g	185 g
EURO3 (LPG)	185 g	183 g	150 g	170 g
EURO3 (CNG)	180 g	178 g	145 g	165 g
EURO4 (gasoline)	190 g	188 g	155 g	175 g
EURO5 (gasoline)	183 g	182 g	138 g	168 g
EURO6 (gasoline)	160 g	159 g	120 g	150 g
EURO6 (Diesel)	150 g	149 g	121 g	142 g
EURO6 (Hybrid)	110 g	109 g	85 g	100 g
EV (energy)	210 Wh	195 Wh	210 Wh	-

Table 4 presents a comparison of carbon dioxide (CO₂) emissions and energy consumption for various types of vehicles under different scenarios, including real-world driving conditions, the micro-scale intersection model, the micro-scale city model and COPERT projections.

The data from the microscale intersection model yield almost identical emission results to the actual data. In this case, road test data were not used to train the intersection micro-model. Models specifically developed for comparative purposes for urban areas show discrepancies of approximately 20–23% compared to real-world data. The values obtained from the COPERT model also tend to underestimate CO₂ emissions.

3.3.2. Use for Vissim simulation - prediction for electrification of vehicles of the future fleet

Since speed and acceleration were chosen as explanatory variables, the developed models exhibit a high degree of versatility in their applications. These models can be utilized with any new real-world data, but they can also be applied in simulation scenarios. One such software that enables traffic modeling is Vissim. Vissim is advanced software for microscopic traffic modeling that allows for the simulation and analysis of driver behavior and traffic flow under various road conditions [39, 40]. It is used to model traffic in cities, on highways, at intersections, roundabouts, and to analyze transportation infrastructure, including public transport systems, pedestrians, and cyclists.

Vissim enables a highly accurate representation of real traffic by modeling individual vehicles that behave similarly to actual drivers. The software is particularly useful for evaluating the effectiveness of different engineering solutions and forecasting the impact of new infrastructure investments on traffic flow and pollutant emissions. Figure 10 illustrates an example of using the energy consumption model developed for electric vehicles in a simulation context.

The use of intersection models in Vissim software, along with the incorporation of various types of vehicles, such as those powered by electric engines, allows a more optimal planning of future road infrastructure and facilitates the analysis of traffic control strategies at such intersections. For this purpose, the CO₂ emission models developed and the energy consumption model for electric vehicles can be used effectively. This approach allows an accurate estimation of vehicle emissions generation and energy consumption, as well as the recovery of energy from regenerative braking at and around intersections.

This study presents a methodology for developing precise predictive models for CO₂ emissions and energy consumption at road intersections, using algorithms that automatically detect data specific to intersections. Since intersections experience frequent start-stop operations, leading to higher emissions, the DBSCAN algorithm was used to cluster data from these areas. Validation results show that the "intersection micro" model accurately predicts CO₂ emissions and energy consumption in high-traffic areas like intersections. Comparisons with the COPERT model and a general urban microscale model reveal a 10–25% underestimation of emissions in non-intersection models.

The novelty of this approach lies in the first-time use of algorithms to automatically identify intersection areas for precise emission modeling. Related studies include [41], which focuses on congestion and pollution at urban intersections in Hong Kong, using dynamic traffic models like the point-queue and shockwave models to estimate delays and emissions. Another relevant study, [42], examines India's transport sector's impact on air pollution, using VISSIM for traffic modeling and showing how vehicle composition and signal timings influence emissions at a Vadodara intersection. Additionally, [43] investigates PM2.5 levels at a busy intersection in Xi'an, China, using the WNN model to predict particulate matter concentrations with high accuracy under varying conditions.

In terms of CO₂ prediction techniques, [44] uses the XGBoost technique to model CNG vehicle emissions, achieving high precision. Similarly, [45] applies machine learning methods like random forest and gradient boosting to model THC and NOx emissions for vehicles with start-stop technology, showing R² values of 0.9 for both pollutants.

While previous research has focused on modifying traffic patterns to assess emissions or using average emission models at intersections, none specifically address the localized emissions modeling at intersections. This study emphasizes the importance of such models for urban planners, as they allow for precise emission forecasting at key points in road networks. Integration with traffic simulation software like VISSIM also opens opportunities for optimizing future road infrastructure, especially with increasing vehicle electrification.

In this study, dynamic models for estimating CO₂ emissions and energy consumption in urban road networks were developed and evaluated, with a particular focus on intersections. Models such as the intersection micro model, the city micro model, and the widely used COPERT software were validated against real-world data, demonstrating their ability to accurately represent areas with the highest concentration of emissions and total energy consumption. However, the analysis revealed that while the intersection micromodel is closely aligned with actual emission data, both the city micro model and COPERT tend to underestimate emissions by approximately 20–23%. These discrepancies highlight the challenges of scaling emission forecasts from localized areas to broader urban environments.

Furthermore, the application of these models within Vissim simulation software underscores their potential for planning and optimizing urban infrastructure for future vehicle fleets, including electric vehicles (EVs). By providing precise forecasts of energy consumption and emissions at intersections, these models offer valuable information to support pollution reduction efforts in cities and improve traffic management strategies. The study emphasizes the importance of integrating detailed traffic and emission modeling to advance sustainable urban transportation systems. A key limitation is the need to scale models to a broader range of vehicles; future research should focus on extending these models to include diverse vehicle technologies and improving their scalability.

Acknowledgement

This work was supported by The Ministry of Infrastructure and Development under the Eastern Poland Development Operational Program including European Regional Development Fund, which financed the research instruments.

Data Availability Statement: The original contributions presented in the study are included in the article, further inquiries can be directed to the corresponding author

Competing interests

The authors declare no competing interests.

Confict of interest

On behalf of all authors, the corresponding author states that there is no confict of interest.

Oh, S. H., Choe, S., Song, M., Yu, G. H., Schauer, J. J., Shin, S. A., Bae, M. S., 2024. Effects of long-range transport on carboxylic acids, chlorinated VOCs, and oxidative potential in air pollution events. Environmental Pollution 347, 123666.
Cubells, J., Miralles-Guasch, C., Marquet, O., 2024. Traffic pollution as a privilege: An intersectional approach to environmental justice and transport emissions. Transportation Research Part D: Transport and Environment 126, 104032.
Coelho, M. C., Farias, T. L., Rouphail, N. M., 2006. Effect of roundabout operations on pollutant emissions. Transportation Research Part D: Transport and Environment 11 (5), 333-343.
Meneguzzer, C., Gastaldi, M., Rossi, R., Gecchele, G., Prati, M. V., 2017. Comparison of exhaust emissions at intersections under traffic signal versus roundabout control using an instrumented vehicle. Transportation Research Procedia 25, 1597-1609.
Josey, K. P., Delaney, S. W., Wu, X., Nethery, R. C., DeSouza, P., Braun, D., Dominici, F., 2023. Air pollution and mortality at the intersection of race and social class. New England Journal of Medicine 388 (15), 1396-1404.
Mądziel, M., 2024. Energy modeling for electric vehicles based on real driving cycles: An artificial intelligence approach for microscale analyses. Energies 17 (5), 1148.
Wu, Y., Zhang, S., Hao, J., Liu, H., Wu, X., Hu, J., Stevanovic, S., 2017. On-road vehicle emissions and their control in China: A review and outlook. Science of the Total Environment 574, 332-349.
Nocera, S., Basso, M., Cavallaro, F., 2017. Micro and macro modelling approaches for the evaluation of the carbon impacts of transportation. Transportation Research Procedia 24, 146-154.
Shamsi, H., Tran, M. K., Akbarpour, S., Maroufmashat, A., Fowler, M., 2021. Macro-Level optimization of hydrogen infrastructure and supply chain for zero-emission vehicles on a Canadian corridor. Journal of Cleaner Production 289, 125163.
Xu, Z., Kang, Y., Cao, Y., Li, Z., 2021. Deep amended COPERT model for regional vehicle emission prediction. Science China. Information Sciences 64 (3), 139202.
Shang, W. L., Zhang, M., Wu, G., Yang, L., Fang, S., Ochieng, W., 2023. Estimation of traffic energy consumption based on macro-micro modelling with sparse data from Connected and Automated Vehicles. Applied Energy 351, 121916.
Mądziel, M., 2023a. Future cities carbon emission models: Hybrid vehicle emission modelling for low-emission zones. Energies 16 (19), 6928.
Rakha, H., Yue, H., Dion, F., 2011. VT-Meso model framework for estimating hot-stabilized light-duty vehicle fuel consumption and emission rates. Canadian Journal of Civil Engineering 38 (11), 1274-1286.
Quaassdorff, C., Borge, R., Pérez, J., Lumbreras, J., de la Paz, D., de Andrés, J. M., 2016. Microscale traffic simulation and emission estimation in a heavily trafficked roundabout in Madrid (Spain). Science of the Total Environment 566, 416-427.
Jamshidnejad, A., Papamichail, I., Papageorgiou, M., De Schutter, B., 2017. A mesoscopic integrated urban traffic flow-emission model. Transportation Research Part C: Emerging Technologies 75, 45-83.
Jin, W. L., 2015. Point queue models: A unified approach. Transportation Research Part B: Methodological 77, 1-16.
Chen, C. Y., Lu, T. H., Wang, W. M., Liao, C. M., 2023. Assessing regional emissions of vehicle-based tire wear particle from macro-to micro/nano-scales with pandemic lockdowns and electromobility scenarios implications. Chemosphere 311, 137209.
Mądziel, M., 2023b. Liquified Petroleum Gas-Fuelled Vehicle CO2 emission modelling based on portable emission measurement system, On-Board diagnostics data, and Gradient-Boosting machine learning. Energies 16 (6), 2754.
Khreis, H., 2016. Critical issues in estimating human exposure to traffic-related air pollution: Advancing the assessment of road vehicle emissions estimates. In World Conference on Transport Research, Transportation Research Procedia, WCTR.
Ghermandi, G., Fabbi, S., Bigi, A., Veratti, G., Despini, F., Teggi, S., Torreggiani, L., 2019. Impact assessment of vehicular exhaust emissions by microscale simulation using automatic traffic flow measurements. Atmospheric Pollution Research 10 (5), 1473-1481.
Osorio, C., Nanduri, K., 2015. Urban transportation emissions mitigation: Coupling high-resolution vehicular emissions and traffic models for traffic signal optimization. Transportation Research Part B: Methodological 81, 520-538.
Eom, M., Kim, B. I., 2020. The traffic signal control problem for intersections: A review. European Transport Research Review 12, 1-20.
Mądziel, M., 2023c. Vehicle emission models and traffic simulators: A review. Energies 16 (9), 3941.
Lyu, P., Wang, P. S., Liu, Y., Wang, Y., 2021. Review of the studies on emission evaluation approaches for operating vehicles. Journal of Traffic and Transportation Engineering (English Edition) 8 (4), 493-509.
Deng, F., Lv, Z., Qi, L., Wang, X., Shi, M., Liu, H., 2020. A big data approach to improving the vehicle emission inventory in China. Nature Communications 11 (1), 2801.
Agarwal, A. K., Mustafi, N. N., 2021. Real-world automotive emissions: Monitoring methodologies, and control measures. Renewable and Sustainable Energy Reviews 137, 110624.
Gallus, J., Kirchner, U., Vogt, R., Benter, T., 2017. Impact of driving style and road grade on gaseous exhaust emissions of passenger vehicles measured by a Portable Emission Measurement System (PEMS). Transportation Research Part D: Transport and Environment 52, 215-226.
Ziółkowski, A., Fuć, P., Jagielski, A., Bednarek, M., Konieczka, S., 2023. Comparison of the energy consumption and exhaust emissions between hybrid and conventional vehicles, as well as electric vehicles fitted with a range extender. Energies 16 (12), 4669.
European Commission, 2017. Commission Regulation (EU) 2017/1151 of 1 June 2017 Supplementing Regulation (EC) No 715/2007 of the European Parliament and of the Council on Type-Approval of Motor Vehicles with Respect to Emissions from Light Passenger and Commercial Vehicles (Euro 5 a). Official Journal of the European Union (692).
Pielecha, J., Kurtyka, K., 2023. Exhaust emissions from Euro 6 vehicles in WLTC and RDE—Part 2: Verification by experimental measurement. Energies 16 (22), 7533.
Kondo, H., Yamaura, C., Saito, Y., Kobayashi, H., 2017. Effectiveness of phase correction when evaluating the efficiency of high-efficiency motor drives. HIOKI EE Corporation, Ueda, Nagano, Technical Notes.
Vrana, M., Moravek, J., Mastny, P., 2015. Photovoltaic power plant inspection and diagnostic. Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska (3), 55-58.
Bisong, E., Bisong, E., 2019. Google colaboratory: Building machine learning and deep learning models on Google Cloud Platform: A comprehensive guide for beginners. Building Machine Learning and Deep Learning Models on Google Cloud Platform: A Comprehensive Guide for Beginners, 59-64.
Civera, M., Sibille, L., Fragonara, L. Z., Ceravolo, R., 2023. A DBSCAN-based automated operational modal analysis algorithm for bridge monitoring. Measurement 208, 112451.
Lu, Z., Zhu, Z., Bi, J., Xiong, K., Wang, J., Lu, C., Yan, W., 2021, October. Bolt 3D point cloud segmentation and measurement based on DBSCAN clustering. In 2021 China Automation Congress (CAC) (pp. 420-425). IEEE.
Rui, Y., Zhou, Z., Cai, X., Dong, L., 2022. A novel robust method for acoustic emission source location using DBSCAN principle. Measurement 191, 110812.
O'Driscoll, R., ApSimon, H. M., Oxley, T., Molden, N., Stettler, M. E., Thiyagarajah, A., 2016. A Portable Emissions Measurement System (PEMS) study of NOx and primary NO2 emissions from Euro 6 diesel passenger cars and comparison with COPERT emission factors. Atmospheric Environment 145, 81-91.
Lejri, D., Can, A., Schiper, N., Leclercq, L., 2018. Accounting for traffic speed dynamics when calculating COPERT and PHEM pollutant emissions at the urban scale. Transportation Research Part D: Transport and Environment 63, 588-603.
Leyn, U., Vortisch, P., 2015. Calibrating VISSIM for the German highway capacity manual. Transportation Research Record 2483 (1), 74-79.
Tumminello, M. L., Macioszek, E., Granà, A., Giuffrè, T., 2023. A methodological framework to assess road infrastructure safety and performance efficiency in the transition toward cooperative driving. Sustainability 15 (12), 9345.
Zhu, F., Lo, H. K., Lin, H. Z., 2013. Delay and emissions modelling for signalised intersections. Transportmetrica B: Transport Dynamics 1 (2), 111-135.
Chauhan, B. P., Joshi, G. J., Parida, P., 2019. Car following model for urban signalised intersection to estimate speed-based vehicle exhaust emissions. Urban Climate 29, 100480.
Song, J., Qiu, Z., Ren, G., Li, X., 2020. Prediction of pedestrian exposure to traffic particulate matters (PMs) at urban signalized intersection. Sustainable Cities and Society 60, 102153.
Mądziel, M., 2024. Modelling CO2 emissions from vehicles fuelled with compressed natural gas based on on-road and chassis dynamometer tests. Energies 17 (8), 1850.
Mądziel, M., 2024c. Quantifying emissions in vehicles equipped with energy-saving start–stop technology: THC and NOx modeling insights. Energies 17 (12), 2815.

No competing interests reported.

Download PDF

Reviewers agreed at journal
16 Oct, 2024
Reviewers agreed at journal
14 Oct, 2024
Reviewers invited by journal
14 Oct, 2024
Editor assigned by journal
14 Oct, 2024
Editor invited by journal
30 Sep, 2024
Submission checks completed at journal
26 Sep, 2024
First submitted to journal
26 Sep, 2024

You are reading this latest preprint version

Predictive methods for CO 2 emissions and energy use in vehicles at intersections

Status:

Version 1

Abstract

Figures

1. Introduction

2. Material and methods

2.1. Vehicles used for the test

2.2. Testing equipment

2.3. Description of the research route

2.4. Methods of analysis and data processing obtained

3. Results

3.1. Data selection for intersection areas

3.2. Creation and validation of CO₂ emission and energy consumption models for motor vehicles

3.3. Use of developed models for prediction of CO₂ and energy consumption

3.3.1. Generation of CO₂ emission maps and comparison of predictive capabilities of models

3.3.2. Use for Vissim simulation - prediction for electrification of vehicles of the future fleet

4. Discussion

5. Conclusions

Declarations

References

Additional Declarations

Status:

Version 1

Predictive methods for CO 2 emissions and energy use in vehicles at intersections

Status:

Version 1

Abstract

Figures

1. Introduction

2. Material and methods

2.1. Vehicles used for the test

2.2. Testing equipment

2.3. Description of the research route

2.4. Methods of analysis and data processing obtained

3. Results

3.1. Data selection for intersection areas

3.2. Creation and validation of CO2 emission and energy consumption models for motor vehicles

3.3. Use of developed models for prediction of CO2 and energy consumption

3.3.1. Generation of CO2 emission maps and comparison of predictive capabilities of models

3.3.2. Use for Vissim simulation - prediction for electrification of vehicles of the future fleet

4. Discussion

5. Conclusions

Declarations

References

Additional Declarations

Status:

Version 1

3.2. Creation and validation of CO₂ emission and energy consumption models for motor vehicles

3.3. Use of developed models for prediction of CO₂ and energy consumption

3.3.1. Generation of CO₂ emission maps and comparison of predictive capabilities of models