https://orcid.org/0000-0002-9151-5290
Figures
Abstract
This study aimed to investigate the impact of altitude, cold start, ambient temperatures, and driving behaviors on the Real Driving Emissions (RDE) of China VI standard light-duty gasoline vehicles. Tests were conducted on actual roads and in vehicle emission environment simulation laboratories at altitudes of 700 m, 1,300 m, 1,900 m, and 2,400 m in Yunnan. The results showed that: (1) as altitude increased, the CO emission factor exhibited a positive correlation trend, peaking at 2400 m with a 2.56-fold increase compared to 700 m. The NOX emission factor displayed an “N” distribution, with aggressive driving at 1900 m yielding 1.89 times higher emissions than normal driving and 3.02 times higher than low-temperature driving; (2) under low-temperature rotating wheel conditions, PN emission factors were 7.55 times higher than normal driving and 71.71 times higher than aggressive driving, indicating that driving behavior significantly influenced NOX emissions, while low-temperature environments had a greater impact on PN emissions; (3) compared to non-cold-start conditions, the cold-start phase increased urban CO, NOX, and PN emission factors by 4.72% to 225.00%, 0.19% to 15.38%, and 6.45% to 430.36%, respectively, with total emission factors increasing by 0.07% to 0.55%, 0.00% to 6.00%, and 1.03% to 242.64%.
Citation: Wang J, Wang L, Li J, Li J, Xu F, Han F, et al. (2025) RDE & dynamometer analysis of light-duty vehicle emissions across altitudes, temperatures, and driving styles. PLoS ONE 20(3): e0318298. https://doi.org/10.1371/journal.pone.0318298
Editor: Tien Anh Tran, Vietnam Maritime University, VIET NAM
Received: August 8, 2024; Accepted: January 13, 2025; Published: March 19, 2025
Copyright: © 2025 Wang et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting information files.
Funding: The National Natural Science Foundation of China (NSFC) (No.51968065) The National Natural Science Foundation of China (NSFC) (Grant No. 51968065), as the funder of this research, has provided necessary support and contributions in terms of validation, visualization, supervision, project management, and funding acquisition. However, despite the funder's contributions in these areas, the design of the study, data collection and analysis, as well as the final decision to publish, were entirely carried out autonomously by the research team, ensuring the independence and academic integrity of the research.
Competing interests: The authors have declared that no competing interests exist.
Introduction
As the road vehicle with the largest production, sales, and ownership of motor vehicles in China, light-duty vehicles have the characteristics of large total amounts of pollutants and carbon emissions and wide operating areas. They have become the focus of China’s control in the fields of motor vehicle pollution reduction and carbon reduction. However, some studies have shown that laboratory test cycles cannot fully reflect the actual driving conditions [1–4], resulting in large differences between laboratory emission results and actual emissions [5–8].
In order to accurately test and evaluate the actual road pollutant emission characteristics of vehicles, the China Ⅵ emission standard for light-duty vehicles (GB18352.6-2016) includes RDE into the type approval content for the first time [9]. Many scholars use the Portable Emission Measurement System (PEMS) to conduct actual road emission tests on traditional light-duty gasoline and diesel vehicles and hybrid vehicles [10–16]. Different from the strict boundary condition control in laboratory test, the RDE test allows expansion in the ambient temperature range of −7 ~ 35 °C and the altitude range of 0 ~ 2400 m and conducts actual road emissions on actual roads composed of urban, suburban, and highway, and use the CO2 moving average window method (MAW) to calculate emissions [17]. With the implementation of China VI standards, a lot of scholars and institutions have focused on the boundary conditions that may affect RDE test results, including stroke dynamics parameters [18–20], altitude [21–23], cold start [24,25], data treatment methods [26–28] and road gradient conditions [29] have been studied. Related factors have an impact on the actual road emissions of light-duty vehicles. Roberto Aliandro Varella et al. [30,31] research team conducted extensive RDE tests on light-duty vehicles, employing MAW, PB, and VSP methodologies to process the data. They observed that while overall CO2 and NOX emissions were comparable among the methods, notable differences emerged in NOX emissions specifically during urban driving. The PB and VSP methods, due to their similar theoretical foundations, exhibited similar trends in CO2 and NOX emissions, with a strong correlation. Furthermore, the study introduced the concept of emissions index to analyze pollutant emissions during cold starts, revealing that lower ambient temperatures led to increased NOX emissions, with a more pronounced effect on gasoline-powered vehicles.
Separately, Kwon S et al. [32,33] and his team utilized PEMS equipment to test light-duty diesel vehicles under RDE conditions, uncovering that NOX emission factors varied significantly (−4% to 60%) depending on the driving route. Notably, the PB method yielded lower NOX emission factors compared to the MAW method. Jens Gallus et al. [34,35] compared PN emissions between laboratory bench tests and real-world driving conditions, finding a high degree of consistency but also significant variations attributed to changing driving conditions. Additionally, they observed a sharp increase in PN emissions during cold starts under low temperatures. Research into driving behavior indicated that aggressive driving significantly elevated CO2 and NOX emissions, with CO2 increasing by 20–40% and NOX by 50–255%, while CO and HC emissions showed lesser variation. The Polish BOSMAL Automotive R & D Institute, led by Piotr Bielaczyc et al. [36] investigated the emission characteristics of gasoline and diesel vehicles under different test cycles (NEDC, WLTC, FTP-75). They found that NOX, PM, and PN emissions were significantly higher under WLTC conditions. While NEDC and WLTC showed relatively similar emission compliance factors, NEDC and FTP-75 displayed greater discrepancies. Moreover, Piotr Bielaczyc’s et al. [37] research emphasized that RDE conditions offered a broader load distribution compared to NEDC. WLTP conditions produced CO2 emission factors that were on average 10% higher than NEDC but closer to RDE. However, WLTP’s NOX emission factors were 2–5 times higher than NEDC, and RDE’s NOX emissions were approximately three times higher than WLTP. CADC’s NOX emission factors were similar to RDE, while DYN’s were 2–8 times higher. Collectively, these studies underscore the complexity and variability of vehicle emissions, emphasizing the crucial role of testing methodologies and driving conditions in accurate emission assessments.
However, due to the poor reproducibility of RDE testing on actual roads conditions, there are still considerable differences in conclusions. In addition, the China Ⅵ standard stipulates that the RDE test only records cold-start emission data, which needs to be excluded from the final urban and total emission calculations, resulting in the calculation results not fully reflecting the actual road emission situation.
As the major source of air pollutants. light-duty vehicles contribute more to the world’s air pollution than any other human activity, especially in high-traffic areas [38]. Light-duty gasoline vehicles are responsible for almost all of the carbon monoxide and lead in urban air, as well as most of the nitrogen oxides, volatile organic compounds (VOCs), particulates and toxic chemicals [39]. As the world’s leading consumers of oil, vehicles also emit large amounts of carbon dioxide and other gases that contribute to global warming. Vehicle also play a major role in stratospheric ozone depletion due to the widespread use of chlorofluorocarbons (CFCs) in automobile air conditioning. The damage caused by vehicle pollutants has become unavoidable. Although almost 50% of all new vehicles produced each year are equipped with emission control devices, the number of vehicles and the number of vehicle miles travelled are growing much faster than the emission reductions that have been achieved to date [40]. Research on light-duty vehicle emission characteristics is critical to saving energy and reducing carbon emissions.
Cold-start emissions account for a large proportion of urban trip emissions, and the effect of driving behavior on cold-start emissions is more significant and sensitive than the effect on non-cold-start emissions in RDE tests [41]. Based on RDE tests conducted at different ambient temperatures, it was also shown that the effect of ambient temperature on cold-start emissions is only a minor factor of driving behavior [42]. To address the effect of high altitude on vehicle emissions, the China VI emission standards introduced the RDE test program. The results showed that carbon monoxide emissions increased with altitude. PN and NOX emissions also increased with altitude increasing, while NOX emissions at 2990 m showed a decreasing trend. In general, the higher altitude, the lower the vehicle CO2 emissions. The CO2 emissions of the naturally-aspirated test vehicles correlate linearly with altitude: for every 1,000 m of altitude gain, the WLTC CO2 emissions decrease by 5.31% [43]. Not only does the intensity of the driving maneuver have an effect on the emission factors, but also the duration and frequency of individual operating states [44].
This article selects a light-duty gasoline vehicle that meets the China VI Standards Stage B as the research object and uses the PEMS to test actual road emissions at altitudes of 700 m, 1300 m, 1900 m, and 2400 m. At the same time, CATARC Automotive Test Center (Kunming) Co., Ltd. (referred to as Kunming Test Center) have built a light-duty vehicle emission environment simulation laboratory at the actual altitude of 1900 m in Kunming City to carry out Type I te0sts (exhaust pollution after cold start at normal temperature emissions test) and RDE tests based on vehicle dynamometers, study and analyze the influence of cold start, altitude, ambient temperature and driving behavior (dynamic parameters) on RDE emissions. The aim of this paper is to investigate the emission characteristics of China VI light-duty gasoline vehicles under road conditions at different altitudes in Yunnan Province. In this paper, we conducted RDE tests in four cities with different altitudes in Yunnan Province. We used the moving average window method to analyze the RDE test data to compare the effects of altitude, cold start, ambient temperature, and driving behavior on the RDE test. The conclusions of this paper can shed light on the emission characteristics of China VI light-duty gasoline vehicles under different conditions, provide useful information for policy formulation, and provide a reference for the improvement of RDE regulations.
This study conducted RDE tests on light-duty gasoline vehicles complying with China’s National VI standards in different altitude regions of Yunnan Province, China, at altitudes of 700 m, 1300 m, 1900 m, and 2400 m. The tests were carried out on actual roads and in vehicle emission simulation laboratories. This study reveals new insights into the emission characteristics of these vehicles under various environmental conditions, with its main novelty lying in the comprehensive investigation of the combined effects of altitude, cold start, ambient temperature, and driving behavior on vehicle emissions.
The main objectives of this study are as follows:
- (1). To assess the specific impact of altitude changes on the emission factors of CO, NOX, and PN;
- (2). To thoroughly analyze the impact of different driving behaviors (including aggressive driving, normal driving, and low-temperature driving conditions) on NOX and PN emissions;
- (3). To compare vehicle emissions during the cold start phase with those under urban driving conditions, in order to fully understand the combined effects of different driving phases and environmental conditions on vehicle emissions.
Materials and methods
Test vehicle
A light-duty gasoline truck, which meets the China VI Standard Stage B, was selected as the research object. It is equipped with a 1.5 L naturally aspirated engine, and the after-treatment system used a three-way catalytic converter. During the testing process, the test vehicle used the same batch of commercially available #92 gasoline that meets China VI standards, driven by a professional RDE driver and accompanied by a researcher who monitored the data in real-time. The main technical parameters of the test vehicle are shown in Table 1.
Fig 1 shows the installation diagram of the vehicle emission testing equipment. The OBS-ONE [45] manufactured by HORIBA, Japan, was selected, which consists of two main modules, i.e., the gas module, the particulate module. The measurement parameters of the main pollutants in OBS-ONE are shown in Table 2. The gas module is used to test the concentration of gaseous pollutants in the vehicle exhaust and consists of a gaseous pollutant analyzer (OBS-ONE GS), which uses non-dispersive infrared (NDIR) to measure CO and CO2, and chemiluminescent detector (CLD) to measure NO and NO2. In the particulate module, the on-board measurement system (OBD) is used to analyze the gaseous pollutant concentration in the vehicle exhaust by condensation particle CPC (Condensation Particle Counter) method to analyze the number concentration of solid particles in a specified particle size range. The test system is also equipped with a Global Positioning System (GPS) receiver that calculates vehicle speed based on the vehicle’s longitude, latitude and altitude. In addition, PEMS is equipped with an atmospheric pressure sensor (measuring range 0 to 115 kPa with an accuracy of ± 2% of full scale), an ambient temperature sensor (measuring range −40 °C to 60 °C with an accuracy of ± 0.5 °C) and a humidity sensor (measuring range 0 to 100 RH with an accuracy of ± 1.5% of full scale). Information about the environmental conditions during the test is obtained from these sensors. The test system also collects engine operating data through OBD, including engine speed, engine coolant temperature, throttle position, fuel consumption rate, TWC converter temperature, intake mass flow, air pressure and other data, and exhaust mass flow. The test system is powered by an external power supply at 24 V. Before use, the PEMS unit is warmed up and calibrated according to RDE test regulations.
Experiment with test boundary conditions
Based on the Yunnan Plateau mountainous three-dimensional road, an actual road condition emission test was carried out. During the test, the vehicle air conditioner was turned on, and the interior temperature was maintained within the range of 23 ± 2 °C. The influence of altitude, ambient temperature, and driving behavior on RDE emission results and control the errors of variables such as ambient temperature and road conditions during the test were studied. Ambient temperature can affect the performance of a vehicle’s engine [46]. Low temperatures can lead to decreased engine efficiency due to thicker oil and slower chemical reactions, while high temperatures may cause the engine to overheat, increasing wear on engine components [47]. These factors, in turn, can impact the vehicle’s ability to maintain a certain speed or achieve optimal fuel efficiency. Secondly, ambient temperature also affects the interior comfort of the vehicle. Extreme temperatures, both hot and cold, can make it difficult for passengers to maintain a comfortable body temperature. This can affect their comfort level and overall satisfaction with the vehicle’s performance [48]. In our experiments, we recognize that ambient temperature was not directly controlled or recorded. However, we took measures to ensure that the experiments were conducted under similar conditions, such as scheduling the tests during similar times of day or using temperature-controlled environments. While this may not fully eliminate the effects of ambient temperature, it helps minimize potential variations and ensures the consistency of our results. In future work, we plan to consider ambient temperature as a variable in our experiments. This will allow us to more thoroughly evaluate its effects on vehicle performance and interior comfort, and provide more comprehensive recommendations and insights.
The influence of humidity on engine characteristics is a complex and multifaceted phenomenon, particularly during road testing, where the effects become more pronounced due to varying weather conditions such as rain, snow, or fog. As humidity increases, the water vapor content in the air also rises, leading to a decrease in air density [49]. For engines, this reduced air density means a reduced volume of air entering the cylinders per unit time, affecting the fuel-to-air mixture ratio and subsequent combustion process. Specifically, increased humidity may result in condensation of water within the engine’s intake system, forming droplets or mist [50]. This can obstruct the intake passages, reduce airflow, and potentially cause corrosion in the combustion chamber. Additionally, the water content in the mixture absorbs heat, lowering the combustion temperature and reducing the combustion speed, which affects the engine’s power output and fuel economy [51]. During rainy, snowy, or foggy weather conditions, engines face even greater challenges during road testing. Rainwater not only increases humidity but can also splash directly into the engine’s intake or exhaust passages, causing malfunctions or damage. Snowy and foggy days may also impact the engine’s heat dissipation performance, leading to overheating or performance degradation [52].
The test is conducted in strict accordance with the requirements of the RDE regulations, the duration of the RDE test is between 90 min–120 min, the driving route includes 34% of the urban road, 33% of the suburban road, 33% of the highway road, and the each road section proportion of the driving error is controlled within 10%, the minimum distance of each section of the driving is greater than 16 km, and the difference in altitude between the test start point and the end point is less than 100 m. Fig 2 shows the vehicle speed cycle of RDE, which reflects the characteristics of actual road conditions in Yunnan province and meets the GB18352.6-2016 standard requirement. In order to reduce the error of test results, two tests were conducted under each altitude condition, and the moving average window method was used for calculation. After the pollutant deviation of the two tests was ≤10%, the arithmetic mean of the two test results was taken as the final test. The results are analyzed; otherwise, additional tests are required for testing. Table 3 shows the RDE test location and altitude. The altitude covers ordinary altitude conditions, extended altitude conditions and further extended altitude conditions.
Analysis of test results
RDE test condition judgment and verification
Judgment of window integrity and normality.
The RDE test standard stipulates that the normality and integrity of the window should be verified when calculating using the moving average window method. The integrity verification requires that the number of windows in urban, suburban, and highway sections be greater than 15% of the total number of windows, and the integrity of the six RDE experiments all passed the requirements [53].
In the calculation of the CO2 MAW method, the “vehicle - CO2 characteristic curve” needs be used to evaluate the normality of the CO2 window. The vehicle CO2 characteristic curve is shown in Fig 3. The irregular curves in the figure are obtained from the RDE test and correspond to the CO2 window on the axes. The parameters at points P1, P2, and P3 consist of dots, which are determined by the average speeds and CO2 emission factors in the low-speed, high-speed, and ultra-high-speed segments of the vehicle’s WLTC cycle. The three points are connected to form the CO2 characteristic curve of the vehicle. The window average speed is bound by 45 and 80 km/h and is divided into urban, suburban, and highway. The basic tolerance and extended tolerance of the vehicle CO2 characteristic curve are defined as tol1 = 25% and tol2 = 50%, respectively.
The RDE test procedures require that when more than 50% of urban, suburban, and highway windows fall within the basic tolerance range defined by the characteristic curve, the results can be judged to be normal. If the above minimum requirement of 50% is not met, the range of the upper limit tol1 can be increased in steps of 1% until the 50% window requirement is met [36]. Table 4 shows the normality and completeness of the CO2 window of the test prototype, both of which meet the requirements of the GB18352.6-2016.
Verification of stroke dynamics parameters.
The GB18352.6-2016 standard stipulates that RDE testing requires verification of stroke dynamic characteristics to determine whether all dynamic characteristics are excessive or insufficient while driving in urban, suburban, and highway sections [54]. The core of stroke dynamics verification lies in two parameters: RPA (relative positive acceleration) and v.apos-[95] (95th percentile of the product of vehicle speed and positive acceleration greater than 0.1 m/s2), divided according to instantaneous speed per second. The three-speed sets (urban, suburban and highway) all need to meet the verification of two parameters, respectively, and the RDE test trip is considered valid. RPA is used to characterize the actual road driving acceleration and intensity of vehicle [55]. The RDE test requires verification of the RPA of each speed group to deter-mine whether the driving stroke is too gentle. The purpose of verifying the RPA is to ensure that the driving is not too gentle.
In the formula, ai + is the positive acceleration of the test vehicle in i second, m/s2;Vi is the speed of the test vehicle in i second, m/s; X is the displacement of the test vehicle within the time interval T.
In the formula: ai is the acceleration of the test vehicle in i second, m/s2; Δt is the time interval between i second and Δi + t, where Δt = 1, s.
v·apos-[95] represents the 95th percentile value of vehicle acceleration within a specific time period or on a particular road section. It signifies that 95% of all the collected acceleration data are less than or equal to this specific value. This parameter can intuitively reflect the intensity of acceleration behavior during the vehicle’s actual driving process [54]. Specifically, a higher v·apos-[95] value usually indicates more aggressive driving behavior, such as frequent acceleration or hard acceleration. The purpose of verifying v.apos-[95] is to limit driving from being too aggressive. v.apos-[95] in each speed group should be verified during RDE testing.
The RDE standard stipulates that the stroke dynamic parameters such as acceleration, apos (positive acceleration greater than 0.1 m/s2), and RPA are calculated based on the speed signal with a vehicle speed greater than 3 km/h, an accuracy of 0.1%, and a sampling frequency of more than 1 Hz. The v.apos-[95] and RPA verification results of the test vehicle under different altitude conditions are shown in Table 5.
It can be seen from Table 5 that the actual values of v.apos-[95] of each road section of the test sample under different altitude conditions are less than the reference values in the brackets, and the actual RPA values of each road section are greater than the reference values in the brackets. RDE test both v.apos-[95], and RPA have passed the verification. In the stages of urban, suburban, and highway, the v.apos-[95] and RPA of test condition D are significantly higher than other normal driving test conditions. From this, it can be seen that the intensity of aggressive driving in test condition D is more obvious. In addition, under test conditions A to F, RPA gradually decreases on urban, suburban, and highway roads, which shows that the actual intensity of vehicle driving on roads gradually decreases and is also consistent with the vehicle driving speed characteristics of different roads.
Fig 3 shows the acceleration characteristic distribution during aggressive driving and normal driving during the RDE test at an altitude of 1,900 m. It can be seen from Fig 2 that the acceleration distribution of normal driving is mainly concentrated in the range of −1.5~1 m/s2, while the acceleration of aggressive driving is mainly concentrated in the range of −3~2 m/s2, which is significantly higher than normal driving. At the same time, aggressive driving conditions are mainly reflected in urban and suburban road sections. The acceleration distribution range is large, especially since the deceleration process is more aggressive. The deceleration under extreme driving conditions reaches −4 m/s2, while in high-speed sections, it is controlled by the speed range of the highway. At limits such as 80~120 km/h, the difference in acceleration distribution characteristics between normal driving and aggressive driving is very small. From this, it can be inferred that aggressive driving conditions mainly have a greater impact on emissions in urban and suburban stages [56].
The statistical analysis of RDE tests and bench tests.
In order to further analyze the difference between the emission results of the dynamometer and actual road conditions, the test vehicle in the study was carried out the WLTC emission test by dynamometer under the 23 °C normal temperature and −7 °C low temperature environment and the altitude of 1914 meters, respectively. The results were compared with the test results of No. C condition and No. E condition respectively, as shown in Table 6.
The results show that the pollutant emission results of CO, NOX and PN in No. C condition are about 1.25 times, 1.35 times and 1.15 times of that in WLTC condition at 23 °C normal temperature, respectively. The same pollutant emissions in No. E condition are about 1.35 times, 1.32 times and 1.25 times of that in WLTC condition, respectively. There are two main reasons, First, the large difference between the WLTC condition and the actual road condition, especially the WLTC condition is only a “Speed-Time” relationship curve, while the actual road slope and other road factors have a significant impact on emissions. Second, the actual road test results are affected by driving behavior, environmental temperature and humidity [46], and the test conditions are more complex, resulting in the actual road emissions are significantly higher than the dynamometer test results [48].
Analysis of emission factors of test vehicles at different altitudes and under road conditions
In order to study the pollutant emission characteristics of different test roads in the RDE test, the pollutant emission analysis is divided into four stages: urban, suburban, high-speed, and total. Under comprehensive test conditions such as different altitudes, temperatures, and driving behaviors, the emission factors of three pollutants, CO, NOX, and PN, of the test vehicle in urban areas, suburbs, highways, and the entire actual road are shown in Fig 4, and Table 7 shows the explanation of A-F. Overall, as the altitude increases, there are certain differences in the emission trends of CO, NOX, and PN pollutants from test vehicles in various road sections [57].
Fig 4(a) shows that the CO emission factor is positively correlated with the in-crease in altitude in urban and suburban areas and total travel stages. The CO emission factor is the largest when the altitude reaches 2400 m. Compared with the altitude of 700 m, the CO emission has increased about 2.56 times. Analyzing the main reasons, this paper studies the test vehicle equipped with a 1.5 L naturally aspirated engine. Com-pared with the plain environment, light-duty vehicle engines with naturally aspirated technology have reduced intake pressure, airflow, and insufficient engine power in the plateau region, resulting in increased injection penetration distance and weakened engine combustion performance, especially in urban areas under low-load conditions [58]. Therefore, enrichment will occur under medium and large loads in suburban areas and highways. At this time, the altitude factor has weakened; the CO emissions at an altitude of 1900 m are only about two times higher than those at an altitude of 100 m. During the urban start-up stage, in order to ensure the normal start of the vehicle, the fuel injection volume is relatively large, but the reduction in air intake volume causes the mixture to be (locally) too rich, exacerbating CO emissions. CO emissions in the high-speed stage first decrease and then increase with increasing altitude. Especially under aggressive driving behavior, the CO emissions in the high-speed stage rise instantly. At this time, the engine load is large, the cylinder fuel injection volume in-creases, and the probability of insufficient fuel combustion increases, making it difficult for the fuel in the cylinder to reach the optimal air-fuel ratio in a short period of time, resulting in insufficient combustion of the mixture and causing an increase in CO emissions from vehicles in the high-speed segment at high engine speeds and high loads.
NOX emissions mainly come from transportation (especially road transportation and shipping), oil and gas activities, and land-based industries. NOX refers to com-pounds composed of only two elements, nitrogen and oxygen [59]. Fig 4(b) shows that the full NOX emission factor shows an “N” type change trend as the altitude increases from 700 m to 2400 m. At an altitude of 2400 m, the NOX emission factor is the highest, reaching 16.30 mg/km. However, it is worth noting that at an altitude of 1900 m, com-pared with normal driving (C) and low-temperature hub (E) conditions, the NOX emission factor under aggressive driving (D) conditions is significantly higher, which is approximately 1.89 times of the C driving conditions, and 3.02 times that of driving conditions E. It can be seen that the intensity of driving behavior has a greater impact on NOX emissions. Analyze the reasons: NOX generation conditions are high-temperature and oxygen enrichment, drastic load changes during aggressive driving behavior, a continuous high-temperature environment, and obvious increase in transition conditions, and other factors that lead to an increase in engine fuel injection and a significant increase in NOX. Under various altitude conditions, the NOX emission factor in the urban stage is higher than that in the suburban and highway stages, mainly because NOX is generated under high temperatures and oxygen-rich conditions. In the urban stage, fuel injection is enriched to ensure vehicle power under medium and high loads, making it difficult to generate NOX; fuel injection and enrichment are not performed under low-load conditions in urban areas, so NOX emissions are relatively small. The overall change in NOX increases with the increase in altitude, and the NOX emission shows a trend of first decreasing and then increasing.
Fig 4(c) shows that the PN emission factor shows a “V” shaped trend with the increase in altitude from 700 m to 2400 m. However, compared to the normal driving (C) and aggressive driving (D) conditions at an altitude of 1900 m, the PN emission factor was the highest in the low-temperature rotating hub (E) condition at 1.85 × 10 + 11#/km, which is about 7.55 times of the driving condition C, and 71.71 times of driving conditions D. It can be seen that the low-temperature environment has a greater impact on PN emissions. Analyze the reasons: First, the engine cylinder wall temperature is low in low-temperature environments, and the mixture is not fully burned, which leads to an increase in soot, soluble organic matter, and other particles under low load. At this time, the TWC has not reached its ignition temperature, and the soluble organic matter has not been released in time. Oxidation leads to an increase in particulate matter emissions; second, PN emissions are related to the fuel oil film participating in combustion in the engine cylinder. The greater the quality of the oil film, the greater the PN emissions. The main reasons for the increase in oil film quality are: the higher the altitude, the lower the air density, the atmospheric pressure increases, the fuel spray penetration distance in-creases, and the oil film will be generated on the combustion chamber wall due to the spray hitting the wall. The higher the altitude, the greater the fuel penetration distance, the greater the quality of the coanda oil film, and the greater the impact on PN emissions. In addition, under different altitude conditions, PN emissions are mainly in the high-speed stage. The main reason for this is that the increase in fuel injection following the increased power output demand of the engine in the high-speed phase even causes the emission control strategy to open the cycle, but the oxygen content in the air decreases, leading to a sharp increase in particulate matter emissions.
Analysis of the impact of cold start on pollutant emissions of light-duty vehicles
Effect of the cold start phase on pollutant emissions.
The GB18352.6-2016 standard stipulates that two conditions, time and engine coolant temperature, are used to determine whether the vehicle is in the cold start stage, that is, within 300 seconds after the vehicle starts the engine for the first time or the coolant temperature reaches 70 °C as a sign of the end of the cold start, but the premise is that the time for the coolant temperature to reach 70 °C does not exceed 300 seconds after the engine is firstly started [60].
The RDE emission test results are mainly calculated by multiplying the pollutant emission factors in three stages: urban, suburban, and highway, respectively, by weighting factors of 0.34, 0.33, and 0.33. The average coolant temperature during the cold start phase of the vehicle is mainly related to the initial coolant temperature (usually consistent with the ambient temperature) and the operating conditions of the engine. The low-temperature environment will affect the average coolant temperature and the ignition time of the exhaust after-treatment system and prolong the cold start [61]. Du-ration and thus have an impact on cold start emissions. However, cold start pollutant emissions are only generated during the starting process of the test vehicle, which only affects pollutant emissions in the urban phase and has no impact on the suburban and high-speed phases. Therefore, the study mainly analyzes the impact of a cold start on the urban phase and highway phase of the RDE test. The impact of pollutant emissions throughout the process [62].
Fig 5 shows a comparison of pollutant emission factors with and without the cold start phase for the city and full-scale tests, assessing and analyzing the impact of the cold start phase data on city and total pollutant emissions. It can be seen from Fig 4 that under various altitude test conditions, the pollutant emission factors in the urban and total stages with cold start data are higher than the results without cold start data, and the pollutant emission factors in the urban stage are higher than those without cold start data. The main reason is that the ignition time of the TWC of the test vehicle is about 100 seconds. The TWC of the test vehicle did not ignite after a cold start, and the catalytic conversion efficiency was low [63]. It was unable to efficiently reduce the emissions of various pollutants, resulting in an increase in pollutant emissions. Compared with the data without the cold start stage, the urban CO, NOX, and PN emission factors with the cold start stage increased by 4.72% ~ 225.00%, 0.19% ~ 15.38%, 0.19%~15.38% and 6.45%~430.36% respectively. The whole-process emission factors increased by 0.07% ~ 0.55%, 0.00% ~ 6.00%, and 1.03% ~ 242.64%, respectively. It can be seen from this that the cold start mainly has a greater impact on CO and PN emissions in the urban stage and a smaller impact on NOX emissions, but has a greater impact on PN emissions in the entire stage.
“Deviation” refers to the difference in experimental results between the conditions where cold start is considered and where it is not.
Study on the transient emission characteristics of cold-start pollutants.
The instantaneous emission data of CO, NOX and PN within 300 seconds of engine cold start under different altitude conditions from A to F were selected to study and analyze the pollutant emission characteristics and change patterns of the test vehicle during the cold start stage [64]. As shown in Fig 6, there are certain differences between the instantaneous emissions of CO, NOX, and PN during the cold start stage of the test vehicle and factors such as altitude, driving behavior, and ambient temperature.
It can be seen from Fig 6(a) that as the altitude increases, the transient concentration of CO emissions shows an upward trend. Especially at the altitude of 2400 m, CO emissions increase sharply, and the peak emission concentration reaches 9%. The main reason is that as the altitude increasing, the atmospheric pressure, air density, and ambient temperature all decrease to varying degrees [65]. In order to start and run the engine smoothly, the ECU control system adopts an injection strategy of increasing the fuel injection volume, but the air intake volume of the low oxygen content leads to insufficient combustion in the engine cylinder and a significant increase in CO emissions [66].
It can be seen from Fig 6(b) that within 200 seconds after the engine is started, the transient concentration of NOX emissions increases with the increase in altitude, especially when the peak value reaches 0.04 ppm at an altitude of 700 m. Since the air-fuel ratio in low-altitude areas is higher than that in high-altitude areas, the oxygen content in the mixture is sufficient, making it easier to generate NOX. In high altitude areas, the engine must operate under a larger load to obtain normal power output, which in turn leads to an increase in cylinder temperature and NOX emissions. For aggressive driving condition D, the transient concentration of NOX emissions suddenly increased sharply at 180 ~ 200 s during a cold start. The main reason was that the test vehicle accelerated sharply at this moment after the cold start, and the engine load increased significantly, resulting in a rapid increase in NOX emissions. However, The TWC has not yet reached its ignition temperature and cannot effectively reduce NOX, resulting in an instant in-crease in the NOX emission concentration in the exhaust gas [67].
It can be seen from Fig 6(c) that during the cold start phase, PN emissions are mainly generated in the first 100 s. The reason is that during the cold start phase of the test vehicle, the engine cylinder wall temperature is low, and the mixture is not fully burned. This leads to an increase in particles such as soot and soluble organic matter [68]. At this time, the TWC has not reached its ignition temperature, and the soluble organic matter is not oxidized in time, resulting in an increase in particulate matter emissions. In particular, the PN concentration was more obvious under the two test conditions of the aggressive driving test (D) at an altitude of 1,900 m and the low-temperature rotating hub test (E), with the peak concentration reaching 7 × 10 + 11#/cm3. The longer warm-up time of the engine under high temperature causes an increase in particulate matter emissions, while under aggressive driving conditions, the increase in the fuel injection volume of the test vehicle and insufficient in-cylinder fuel mixing and combustion leads to an increase in particulate matter emissions [69].
Analysis of the impact of driving behavior on RDE emissions
In the study, two dynamic parameters, RPA and v.apos-[95], were used to describe the driving behavior characteristics and the correlation between the dynamic parameters RPA and v.apos-[95], and various pollutant emissions under different driving behaviors was analyzed. Fig 7 shows the comparison of CO2 emission characteristic curves under normal driving and aggressive driving conditions at an altitude of 1900 m. The thick black line represents the CO2 characteristic curve, which is part of the RDE window method verification.
It can be seen from Fig 6 that under normal driving conditions, the window CO2 emission factor is almost within the basic tolerance line (±25%) and is closer to the CO2 characteristic curve; under aggressive driving conditions, the window CO2 emission factor is Within the range of 25% ~ +50%, especially in the low and medium speed range where the vehicle speed is ≤ 80 km/h, the CO2 emission factor is within the range of the CO2 emission characteristic curve and the + 50% range. Comparing Fig 7a and Fig 7b, it can be seen that the change in CO2 emission factor is most obvious in the high-speed section (80 km/h < V < 120 km/h) window. Therefore, it can be inferred that urban and suburban roads have frequent acceleration and deceleration due to frequent vehicle acceleration and deceleration. The working conditions are more aggressive, and the aggressive driving operating point has a greater impact on the change range of the window CO2 emission factor.
Analysis of the relationship between v.a pos- [95] and pollutant emission factors
Fig 8 shows the relationship between v.apos-[95] and pollutant emission factors under normal driving and aggressive driving conditions at an altitude of 1900 m.
It can be seen from Fig 7 that the CO and PN emission factors have a strong positive correlation with the intensity of actual road driving. Under the same altitude conditions, compared with normal driving, aggressive driving behavior causes CO emissions to increase by 33% to 235% and PN emissions to increase by 35% to 273%. The main reason is that drivers frequently drive violently during aggressive driving. When the accelerator pedal is stepped on, and the throttle is suddenly opened wide, the mixture may be momentarily lean due to the inertia of the mixture flow, the combustion temperature may be too low, and the mixture may be incompletely burned, etc., which may lead to increased emissions of CO and PN [70].
However, NOX emissions do not show a good correlation with driving intensity, and there is no obvious pattern. When v.apos-[95] is small, the more aggressive the driving, the higher the NOX emission factor. The main reason is that during the vehicle’s starting stage, the internal temperature is low, and the TWC converter has not reached its light-off temperature, resulting in poor NOX emissions. As v.apos-[95] increases, the NOX emission factors under normal driving and aggressive driving conditions gradually approach the same level. The reason for the above phenomenon is that when v.apos-[95] has a large value, the vehicle is in a high-speed and high-load state. At this time, the gasoline engine valve opening is higher than about 85%. Due to the power mixture requirements, the engine needs to emit maximum power to overcome the larger External resistance, which requires the supply of a richer mixed gas. The excess air coefficient is generally 0.85 ~ 0.9. The mixed gas is incompletely burned, NOX is easily generated under high temperature and oxygen-rich conditions, and its stoichiometric ratio deviates from the optimal range of TWC.
Analysis of the relationship between RPA and pollutant emission factors
Table 8 shows the relationship between RPA and pollutant emission factors under normal driving and aggressive driving conditions at an altitude of 1900 m.
It can be seen from Table 6 that RPA has no obvious correlation with CO, NOX, and PN emissions. Under aggressive driving conditions, NOX emissions have a strong positive correlation with RPA. As RPA increases, NOX and PN emission factors show a significant increasing trend, reaching 23.5 g/km and 6.6 E + 09#/km, respectively, in urban areas, but the CO emission factor shows a downward trend with the increase of RPA. Under normal driving conditions, the CO emission factor also shows a downward trend with the increase of RPA, but the NOX and PN emissions change irregularly with the increase of RPA. This also shows that compared with the dynamic parameters RPA, the correlation between v.apos-[95] and various pollutant emissions is greater than RPA; that is, v.apos-[95] evaluates the intensity of driving behavior better than RPA.
Conclusion
This study conducted a thorough analysis of the key influencing factors in Real Driving Emissions (RDE) tests, with the main findings encompassing the following aspects:
The Prominent Role of Driving Behavior on Emissions: The study revealed the direct impact of driving behavior on pollutant emissions, particularly that aggressive driving significantly elevates NOX emissions, while low-temperature conditions markedly increase PN emissions. This discovery underscores the importance of optimizing driving behavior to reduce emissions.
The Complex Influence of Altitude on Emissions: As altitude increases, the CO emission factor notably rises, whereas the NOX and PN emission factors exhibit “N”-shaped and “V”-shaped trends, respectively. This indicates that emission control strategies must be tailored to specific altitudes.
The Significance of the Cold Start Phase: The cold start phase significantly impacts pollutant emission factors during urban driving, particularly with substantial increases in CO and PN emissions. Consequently, improving emission performance during the cold start phase is crucial for reducing overall emissions.
The Correlation between Driving Behavior Intensity and Emissions: By comparing emission data under different driving conditions, the study found a strong positive correlation between driving behavior intensity and CO and PN emission factors. This suggests that evaluating and adjusting driving behavior intensity can effectively control emission levels.
Exploration of Dynamic Parameters and Their Relationship with Emissions: v.apos-[95], as an effective parameter for assessing driving behavior intensity, exhibits a higher correlation with pollutant emissions than other dynamic parameters like RPA, providing a more accurate basis for optimizing driving behavior.
Planned Research Directions:
- (1). Further investigation of the relationship between driving behavior and emission factors, including the impact of different acceleration and deceleration patterns on various pollutants.
- (2). Examination of the combined effects of altitude and temperature on emission factors, particularly in high-altitude and cold regions.
- (3). Study of advanced emission control technologies and their effectiveness in reducing emissions under aggressive driving conditions and in low-temperature environments.
- (4). Development of driver training programs and incentives to promote more environmentally friendly driving practices, especially in urban and suburban areas.
- (5). Investigation of the impact of traffic congestion and road design on driving behavior and subsequent emissions, with a focus on improving road infrastructure to reduce emissions.
Supporting information
S4 Data. Analysis the impact of cold start on pollutant emissions.
https://doi.org/10.1371/journal.pone.0318298.s004
(XLSX)
References
- 1. Zhu QG, Yang ZJ, Wen Y, et al. A comparative research on the pollutant emissions of light-duty vehicles under real road driving and laboratory condition. Automo Eng. 2017;10(39):1125–113.
- 2. García-Contreras R, Soriano JA, Fernández-Yáñez P, Sánchez-Rodríguez L, Mata C, Gómez A, et al. Impact of regulated pollutant emissions of Euro 6d-Temp light-duty diesel vehicles under real driving conditions. J Clean Prod. 2021;286:124927.
- 3. Al-rifai J. Correlation analysis of driving conditions and on-road emissions trends for vehicles. JUEE. 2017;11(1):63–72.
- 4. Khemri N, Supina J, Syed F, et al. Developing a real-world, second-by-second driving cycle database through public vehicle trip surveys. SAE Paper. 2019;1:5074.
- 5. Mei H, Wang L, Wang M, Zhu R, Wang Y, Li Y, et al. Characterization of exhaust CO, HC and NOx emissions from light-duty vehicles under real driving conditions. Atmosphere. 2021;12(9):1125.
- 6. Khan T, Frey HC. Comparison of real-world and certification emission rates for light duty gasoline vehicles. Sci Total Environ. 2018;622–623:790–800. pmid:29223906
- 7. Song B, Ge Y, Yin H, Yang Z, Wang X, Tan J. A study on the effects of driving dynamics parameters on the results of RDE test. Automot Eng. 2018;40(4):389–95.
- 8. Kwon S, Park Y, Park J, Kim J, Choi K-H, Cha J-S. Characteristics of on-road NO x emissions from Euro 6 light-duty diesel vehicles using a portable emissions measurement system. Sc Total Environ. 2017;576:70–7.
- 9. Ministry of Ecology and Environment. Limits and measurement methods of emissions from light-duty vehicles (China 6) Ministry of Ecology and Environment of the People’s Republic of China 2016. GB18352.6-2016. Available from: https://www.mee.gov.cn/ywgz/fgbz/bz/bzwb/dqhjbh/dqydywrwpfbz/201612/t20161223_369476.shtml
- 10. Mahesh S, Ramadurai G, Shiva Nagendra SM. Real-world emissions of gaseous pollutants from diesel passenger cars using portable emission measurement systems. Cities Soc. 2018;41:104–13.
- 11. Fu B, Yang Z, Yin H, Zhou X, Tan J, Ge Y. A research on the real driving emission characteristics of light-duty gasoline vehicles. Automot Eng. 2017;39(4):376–80.
- 12. Wang Y, Hao C, Ge Y, Hao L, Tan J, Wang X, et al. Fuel consumption and emission performance from light-duty conventional/hybrid-electric vehicles over different cycles and real driving tests. Fuel. 2020;278:118340.
- 13. Huang Y, Surawski NC, Organ B, Zhou JL, Tang OHH, Chan EFC. Fuel consumption and emissions performance under real driving: comparison between hybrid and conventional vehicles. Sci Total Environ. 2019;659:275–82. pmid:30599346
- 14. McCaffery C, Zhu H, Li C, Durbin TD, Johnson KC, Jung H, et al. On-road gaseous and particulate emissions from GDI vehicles with and without gasoline particulate filters (GPFs) using portable emissions measurement systems (PEMS). Sci Total Environ. 2020;710:136366. pmid:31923692
- 15. Luján JM, Bermúdez V, Dolz V, Monsalve-Serrano J. An assessment of the real-world driving gaseous emissions from a Euro 6 light-duty diesel vehicle using a portable emissions measurement system (PEMS). Atmos Environ. 2018;174:112–21.
- 16. Yang ZJ, Fu BZ, Yin H, et al. Research on actual driving emission characteristics of light-duty diesel vehicles. Automot Eng. 2017;39(5):497–502.
- 17. Ge YS, Wang CY, Dong HL, et al. The adaptability of the RDE window method for actual China road. Transactions of Beijing Institute of Technology. 2020;40(9):924–8.
- 18. Available from:
- 19. Song B, Ge YS, Yin H, et al. A study on the effects of driving dynamics parameters on the results of RDE test. Automot Eng. 2018;40(4):389–395,442.
- 20. Zhang YJ, Dong HL, Li YB, et al. Driving behavior correction of RDE test results based on window dynamics factor. Automotive Engineering. 2020;42(10):1354–63.
- 21. Yang CZ, Du BC, Li YB, et al. Correlation analysis of driving dynamic parameters and RDE characteristics. Journal of Chongqing University. 2021;44(7):14–25.
- 22. Zheng F, Zhao HG, Ji Jl, et al. Real driving emission characteristics of China Ⅵ hybrid electric vehicle based on altitude conditions in Beijing and Kunming. Research of Environmental Sciences. 2022;35(7):1581–8.
- 23. Ni HZ, Wang LL, et al. A research of the effects of altitude on the emissions and fuel consumption of a gasoline vehicle. Automot Eng. 2014;36(10):1205–9.
- 24. Bishop GA, Morris JA, Stedman DH, Cohen LH, Countess RJ, Countess SJ, et al. The effects of altitude on heavy-duty diesel truck on-road emissions. Environ Sci Technol. 2001;35(8):1574–8. pmid:11329704
- 25. Wang ZH, Wu PH, Liu E. Influence of cold start on CO and NOx emissions in real driving. Journal of East China Jiaotong University. 2020;37(6):96–102.
- 26. Seo J, Yun B, Kim J, Shin M, Park S. Development of a cold-start emission model for diesel vehicles using an artificial neural network trained with real-world driving data. Sci Total Environ. 2022;806(Pt 3):151347. pmid:34728203
- 27. Zheng SK, Ge YS, Chen WC, et al. A research on the effects of different data processing methods on the results of RDE test. Automot Eng. 2018;40(6):646–51, 665.
- 28. Xu H, Lei Y, Liu M, Ge Y, Hao L, Wang X, et al. Research on the CO2 emission characteristics of a light-vehicle real driving emission experiment based on vehicle-specific power distribution. Atmosphere. 2023;14(9):1467.
- 29. Varella RA, Faria MV, Mendoza-Villafuerte P, Baptista PC, Sousa L, Duarte GO. Assessing the influence of boundary conditions, driving behavior and data analysis methods on real driving CO2 and NOx emissions. Sci Total Environ. 2019;658:879–94. pmid:30583183
- 30. Varella R, Duarte G, Baptista P, et al. Comparison of data analysis methods for European real driving emissions regulation. SAE Technical Paper. 2017;2017-01–0997.
- 31. Varella R, Duarte G, Baptista P, et al. Analysis of the influence of outdoor temperature in vehicle cold-start operation following EU real driving emission test procedure. SAE Int J Commer Veh. 2017;10(2).
- 32. Kwon S, Park Y, Park J, Kim J, Choi K-H, Cha J-S. Characteristics of on-road NOx emissions from Euro 6 light-duty diesel vehicles using a portable emissions measurement system. Sci Total Environ. 2017;576:70–7. pmid:27780101
- 33. Kwon S, Kwon S, Lee J, Oak S, Seo Y, Park S, et al. Data Evaluation Methods for Real Driving Emissions using Portable Emissions Measurement System(PEMS). Transactions of the Korean Society of Mechanical Engineers B. 2015;39(12):965–73.
- 34. Gallus J, Kirchner U, Vogt R, Börensen C, Benter T. On-road particle number measurements using a portable emission measurement system (PEMS). Atmospheric Environment. 2016;124:37–45.
- 35. Gallus J, Kirchner U, Vogt R, Benter T. 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. 2017;52:215–26.
- 36. Bielaczyc P, Woodburn J, Szczotka A. Exhaust emissions of gaseous and solid pollutants measured over the NEDC, FTP-75 and WLTC chassis dynamometer driving cycles. SAE Technical Paper. 2016;2016-01–1008.
- 37. Titus F, Berlet P, Sobek F, et al. Emission reduction during cold start by combustion controlled increase of in-cylinder temperatures. SAE Technical Paper. 2018.;01–1740:114.
- 38. Gallus J, Kirchner U, Vogt R, Benter T. 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. 2017;52:215–26.
- 39. Breuer JL, Samsun RC, Stolten D, Peters R. How to reduce the greenhouse gas emissions and air pollution caused by light and heavy duty vehicles with battery-electric, fuel cell-electric and catenary trucks. Environ Int. 2021;152:106474. pmid:33711760
- 40. Claßen J, Krysmon S, Dorscheidt F, Sterlepper S, Pischinger S. Real driving emission calibration—review of current validation methods against the background of future emission legislation. Applied Sciences. 2021;11(12):5429.
- 41. Prevedouros P, Mitropoulos L. Life cycle emissions and cost study of light duty vehicles. Transportation Research Procedia. 2016;15:749–60.
- 42. Du B, Zhang L, Geng Y. Testing and evaluation of cold-start emissions in a real driving emissions test. Transportation Research Part D. 2020;86.
- 43. Wang R, Zhang L, Peng J, Han J, Du B, Gong B, et al. Effect of ambient high-temperature on real driving emission tests of a light-duty gasoline vehicle. Atmospheric Pollution Research. 2024;15(2):101998.
- 44. Wang H, Ge Y, Hao L, Xu X, Tan J, Li J, et al. The real driving emission characteristics of light-duty diesel vehicle at various altitudes. Atmospheric Environment. 2018;191:126–31.
- 45. Zhao T, Zhang Y, Song Q, Meng Q, Zhou S, Cong J. Tire and road wear particles in the aquatic organisms - A review of source, properties, exposure routes, and biological effects. Aquat Toxicol. 2024;273:107010. pmid:38917645
- 46. Wang R, Zhang L, Peng J, Han J, Du B, Gong B, et al. Effect of ambient high-temperature on real driving emission tests of a light-duty gasoline vehicle. Atmospheric Pollution Research. 2024;15(2):101998.
- 47. Wu X, Zhao H, He L, Yang X, Jiang H, Fu M, et al. Impacts on real-world extra cold start emissions: fuel injection, powertrain, aftertreatment and ambient temperature. Environ Pollut. 2023;324:121339. pmid:36863441
- 48. Zhu R, Wei Y, He L, Wang M, Hu J, Li Z, et al. Particulate matter emissions from light-duty gasoline vehicles under different ambient temperatures: physical properties and chemical compositions. Sci Total Environ. 2024;926:171791. pmid:38508249
- 49. Serrano JR, Martín J, Piqueras P, Tabet R, Gómez J. Effect of natural and forced charge air humidity on the performance and emissions of a compression-ignition engine operating at high warm altitude. Energy. 2023;266:126409.
- 50. Wang X, Liang Y, Tang X, Jiang X. A multi-compartment electric vehicle routing problem with time windows and temperature and humidity settings for perishable product delivery. Expert Systems with Applications. 2023;233:120974.
- 51. Zhang J, Sui C, Zhang B, Li J. Effects of swirl intensity on flame stability and NO emission in swirl-stabilized ammonia/methane combustion. Applications in Energy and Combustion Science. 2023;14:100138.
- 52. Benajes J, García A, Monsalve-Serrano J, Martínez-Boggio S. Optimization of the parallel and mild hybrid vehicle platforms operating under conventional and advanced combustion modes. Energy Conversion and Management. 2019;190:73–90.
- 53. Wang Y, Wang J, Hao C, Wang X, Li Q, Zhai J, et al. Characteristics of instantaneous particle number (PN) emissions from hybrid electric vehicles under the real-world driving conditions. Fuel. 2021;286:119466.
- 54. Huang R, Ni J, Shi X, Wang Q, Cheng Z. Assessing and characterizing the effect of altitude on fuel economy, particle number and gaseous emissions performance of gasoline vehicles under real driving. SAE Technical Paper Series. 2023;
- 55. Qiao P, Ni J, Huang R, Cheng Z. Prediction of instantaneous particle number for light-duty gasoline vehicles under real driving conditions based on ensemble learning. Journal of Cleaner Production. 2024;434:139859.
- 56. Xia Y, Liao C, Chen X, Zhu Z, Chen X, Wang L, et al. Future reductions of China’s transport emissions impacted by changing driving behaviour. Nat Sustain. 2023;6(10):1228–36.
- 57. Jiang Z, Wu L, Niu H, Jia Z, Qi Z, Liu Y, et al. Investigating the impact of high-altitude on vehicle carbon emissions: a comprehensive on-road driving study. Sci Total Environ. 2024;918:170671. pmid:38316305
- 58. Serrano JR, Martín J, Piqueras P, Tabet R, Gómez J. Effect of natural and forced charge air humidity on the performance and emissions of a compression-ignition engine operating at high warm altitude. Energy. 2023;266:126409.
- 59. Wang Z, Luo K, Yu H, Feng K, Ding H. NOx Emission prediction of heavy-duty diesel vehicles based on Bayesian optimization -Gated Recurrent Unit algorithm. Energy. 2024;292:130559.
- 60. Wu X, Zhao H, He L, Yang X, Jiang H, Fu M, et al. Impacts on real-world extra cold start emissions: fuel injection, powertrain, aftertreatment and ambient temperature. Environ Pollut. 2023;324:121339. pmid:36863441
- 61. Zhai Z, Xu J, Zhang M, Wang A, Hatzopoulou M. Quantifying start emissions and impact of reducing cold and warm starts for gasoline and hybrid vehicles. Atmospheric Pollution Research. 2023;14(1):101646.
- 62. Zare A, Stevanovic S, Jafari M, Verma P, Babaie M, Yang L, et al. Analysis of cold-start NO2 and NOx emissions, and the NO2/NOx ratio in a diesel engine powered with different diesel-biodiesel blends. Environ Pollut. 2021;290:118052. pmid:34479164
- 63. Bae S-H, Chae E, Park Y-S, Lee S-W, Yun J-H, Choi S-S. Characteristics of tire-road wear particles (TRWPs) and road pavement wear particles (RPWPs) generated through a novel tire abrasion simulator based on real road pavement conditions. Sci Total Environ. 2024;944:173948. pmid:38880134
- 64. Jeong S, Ryu H, Shin H, Lee MG, Hong J, Kim H, et al. Quantification of tire wear particles in road dust based on synthetic/natural rubber ratio using pyrolysis-gas chromatography-mass spectrometry across diverse tire types. Sci Total Environ. 2024;942:173796. pmid:38851327
- 65. Khardi S. Emission factors of tyre wear particles emitted by light road vehicles in real driving conditions: a new challenge for clean road transport to improve urban air quality. Atmosphere. 2024;15(6):665.
- 66. Medina R, Sánchez R-V, Cabrera D, Cerrada M, Estupiñan E, Ao W, et al. Gearbox faults severity classification using Poincaré plots of acoustic emission signals. Applied Acoustics. 2024;219:109918.
- 67. Hongsheng S, Lixia D, Xiaoying Y, et al. Research on carbon emission for preventive maintenance of wind turbine gearbox based on stochastic differential equation. School of Automation and Electrical Engineering. 2024,121(4):973–86.
- 68. Eckert JJ, da Silva SF, Santiciolli FM, de Carvalho ÁC, Dedini FG. Multi-speed gearbox design and shifting control optimization to minimize fuel consumption, exhaust emissions and drivetrain mechanical losses. Mechanism and Machine Theory. 2022;169:104644.
- 69. Mon Phyo MZ, Phairote W, Srilomsak M, Charoenphonphanich C, Masomtob M, Chollacoop N, et al. Effect of a retrofitted metallic microfiber partial flow diesel particulate filter on a light duty diesel vehicle particle emission characteristics. Journal of the Energy Institute. 2024;112:101491.
- 70. Wang X, Ge YS, Jiangping , et al. Emission characteristics of light duty vehicle in RDE tests under different environmental conditions. China Environmental Science. 2022;42(6):2561–8.