Next Article in Journal
Electricity Cost Savings in Energy-Intensive Companies: Optimization Framework and Case Study
Previous Article in Journal
Current Challenges in Operation, Performance, and Maintenance of Photovoltaic Panels
Previous Article in Special Issue
Traffic Signal Optimization to Improve Sustainability: A Literature Review
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Energies
Volume 17
Issue 6
10.3390/en17061308
energies-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Comparison of the Real-Driving Emissions (RDE) of a Gasoline Direct Injection (GDI) Vehicle at Different Routes in Europe

by
Barouch Giechaskiel
1,*,
Victor Valverde
2,
Anastasios Melas
1,
Michaël Clairotte
1,
Pierre Bonnel
1 and
Panagiota Dilara
3
1
Joint Research Centre (JRC), European Commission, 21027 Ispra, Italy
2
Unisystem S.A., 20145 Milan, Italy
3
Directorate-General for Environment (DG-ENV), European Commission, 1160 Brussels, Belgium
*
Author to whom correspondence should be addressed.
Energies 2024, 17(6), 1308; https://doi.org/10.3390/en17061308
Submission received: 17 February 2024 / Revised: 25 February 2024 / Accepted: 26 February 2024 / Published: 8 March 2024
(This article belongs to the Special Issue Vehicle Energy Consumption and Emissions in Intelligent and Sustainable Transportation Systems)
Browse Figures
Versions Notes

Abstract

:
On-road real-driving emissions (RDE) tests with portable emissions measurement systems (PEMS) are part of the vehicle emissions regulations in the European Union (EU). For a given vehicle, the final emission results depend on the influence of the ambient conditions and the trip characteristics (including the driver’s behaviour) on the vehicle performance and the instrument measurement uncertainty. However, there are not many studies that have examined the emissions variability of a single vehicle following different routes. In this study, a 1.2 L gasoline direct injection (GDI) Euro 5b passenger car without a particulate filter and a PEMS was circulated in seven European laboratories. At their premises, the laboratories performed two to five repetitions of on-road trips compliant with the EU RDE regulation. The ambient temperature ranged between 7 °C and 23 °C. The average emission levels of the vehicle were 135 g/km for CO2, 77 mg/km for CO, 55 mg/km for NOx, and 9.2 × 1011 #/km for particle number. The coefficient of variance in the emissions following the same route was 2.9% for CO2, 23.8% for CO, 23.0% for NOx, and 5.8% for particle number. The coefficient of variance in the emissions following different routes in Europe was 6.9% for CO2, 9.1% for CO, 0.0% for NOx, and 9.1% for particle number. The previous values include the specific vehicle emissions variability under the narrow test conditions of this study, but only partly the PEMS measurement uncertainty because the same instrument was used in all the trips. The results of this study can be used by laboratories conducting RDE tests to assess their uncertainty budget when testing or comparing vehicles of similar technology.

1. Introduction

Measuring vehicle emissions on the road is becoming more and more common [1,2,3,4,5,6,7,8,9,10]. On the one hand, a lot of research is conducted to determine the realistic behaviour or emission factors of light-duty and heavy-duty vehicles [11,12,13,14,15,16,17]. On the other hand, most regulations worldwide require emissions compliance to the limits not only in the laboratory but also on the road under normal conditions of use [18]. The on-road testing for regulatory purposes started in the United States of America (USA) in 2007 for heavy-duty vehicles. Europe introduced Portable Emissions Measurement Systems (PEMS) testing a few years later (in 2013) for heavy-duty vehicles with Regulation (EU) 582/2011 [19] and for light-duty vehicles (in 2017) with Real-Driving Emissions (RDE) Regulation (EU) 2016/427 [20]. Since then, various regulations have improved the procedures [21,22,23,24] and were included in the Euro 6 regulation [25]. Table 1 summarises some of the key additions in the light-duty RDE regulation, sometimes called first to fifth RDE packages. The first two packages were consolidated in the worldwide harmonised light-vehicle test procedure (WLTP) Regulation (EU) 2017/1151 [25], which was further developed by the third and fourth packages. Euro 6d-t entered into force in September 2017, Euro 6d in September 2020, and Euro 6e in September 2023 (January 2025 for Euro 6ebis). In Asia, China introduced on-road provisions with China 6b in 2023 and India with Bharat Stage 6 Phase II in 2023 [18,26].
Although on-road testing is more representative of the reality than laboratory testing, the results might exhibit higher scatter due to the variable vehicle response and usage and the wider ranges of possible ambient and driving conditions, even when keeping all possible influencing parameters constant (e.g., fuel, pre-conditioning, weight (payload), use of air-conditioning, and other auxiliaries, etc.) [27]. The variability of emissions measured over different routes is caused by a composite of (i) the stability of the engine and the aftertreatment devices under identical conditions, (ii) the influence of the ambient conditions on the vehicle emissions performance (including parameters such as temperature, wind, road quality), (iii) the influence of the trip characteristics (e.g., driving style, route composition, traffic conditions, etc.) on the vehicle emissions performance, and (iv) the instrument accuracy and stability. The European regulation requires that light-duty vehicles respect the NOx and particle number (PN) Euro 6 emission limits under the normal driving conditions defined by the RDE regulation; i.e., the potential variability of the emissions caused by points (i) to (iii) are included. The intention of the regulators is to avoid poor performance of the emissions control technology (e.g., efficient only under certain conditions) and to capture the exceedance of the regulated limits if and when this happens. Within the regulated conditions, voluntary strategies (Auxiliary Emissions Strategies) are permitted. Outside the regulated conditions, auxiliary strategies may be permitted under certain conditions aiming to protect the engine and the aftertreatment against damage or accident and for the safe operation of the vehicle [28]. These conditions are assessed by the responsible type approval authority, in particular, if their effect on the emissions can be justified and quantified [29,30].
It is important to avoid cases in which a vehicle would be falsely considered to exceed the emission limits due to the instrument measurement inaccuracy; i.e., point (iv) should be taken into account in the emission compliance assessment. For this reason, the maximum PEMS measurement uncertainty that can occur during all potential RDE testing conditions is taken into account in the EU regulation as a “margin” on top of the regulated emission limits (NOx and PN), applicable to all vehicles (with direct injection engines for PN) and regulated testing conditions [31,32]. A novelty of the Euro 6e is that the “margin” is now included in the final RDE results and not on the limit (Appendix 11, Annex III of Regulation (EU) 2023/443 [24]). Consequently, the same limit applies to testing in the laboratory and on the road. The term “conformity factor” (1+margin) referring to permissible exceedance of the laboratory limits on the road is not used anymore, but only “margin”. The approach of Euro 6e will also be followed in Euro 7 [33]. In the USA, both the measurement uncertainty and the differences between laboratory engine and on-road vehicle testing are taken into account in the not-to-exceed limit [34].
On the one hand, a wide range of conditions (driving style, ambient temperature, altitude, vehicle payload, i.e., “boundary conditions” according to the regulation) is desirable to assess the compliance of a car. On the other hand, similar conditions are desirable when the aim is to compare the emissions of different vehicles or have an indication of the laboratory’s repeatability for certification purposes. Repeatability is the precision (i.e., closeness) of measurements of the same or similar objects at the same location with the same measurement procedure, system, operator, and operating conditions over a short period of time. Reproducibility, on the other hand, is the precision of measurements of the same or similar objects but at different locations, with different measuring systems and operators and over a longer period of time. While the definitions are straightforward for laboratory testing, for on-road RDE testing, the terms cannot be strictly used because it is almost impossible to have identical conditions on two trips at the same or at different locations (local traffic and weather conditions cannot be controlled, for instance) [35]. Furthermore, the inherent variability of the vehicle response is included and cannot be avoided. Under strict laboratory conditions, this variability can be small, but during the on-road testing, the environmental conditions and the trip characteristics are not well-controlled and might result in high vehicle emissions variability. Only a few studies have examined the variability of results following the same route and trying to keep all parameters similar [36,37]. A few researchers examined different routes with different trip characteristics [38,39,40]. To our knowledge, there are no studies that have compared the emissions and the characteristics of different RDE trips for the same vehicle and the same PEMS at different locations. As the studies regarding the influence of the trip on the results are limited, there is a need to better understand the variability of the emissions due to different trip characteristics, in particular, those compliant with the EU RDE regulation, which, in principle, is the same with China’s, India’s, and United Nations’ Regulation 168 [41].
This study addresses the trip-to-trip variability over different RDE trips compliant with the EU requirements. In particular, the variability of the RDE testing procedure is assessed under similar (i.e., narrow range of) environmental and driving conditions using the same vehicle and the same measurement equipment at different locations in Europe. According to our knowledge, no other similar studies have been conducted before. The results of this study can be a first input of expected differences due to different trip characteristics for both regulators and researchers. It has to be highlighted that this study does not assess the possible variation in the PEMS response (measurement uncertainty), which has been thoroughly assessed in other studies [32,42,43]. As many changes took place in the RDE regulation over the years, the text also gives some historical information regarding the evolution of the RDE technical specifications.

2. Materials and Methods

The objective of the original inter-laboratory comparison exercise was to assess the measurement performance of particle number (PN) PEMS. This exercise directly involved the regulators, the industry, and the technical services. For this reason, a PEMS (i.e., gas analyzers, PN analyzer, exhaust flow meter, GPS, and weather station and batteries) and two additional PN-PEMS were installed in a reference (“Golden”) vehicle and were tested in different laboratories and their premises across Europe [44]. The original report assessed the measurement performance of the two additional PN-PEMS by comparing them to the laboratory reference systems on chassis dynamometers and to each other.
In this study, we evaluate the PEMS data (gases and particles) that were not analyzed in the previous report, as they were outside of the scope of the exercise. Furthermore, the approach is different. Here, we focus on the variability of the on-road tests for gases and particles, while in the original report, only relative differences in PN-PEMS to the reference laboratory systems were assessed. Using the results from the same PEMS applied in different laboratories allows for expressing the variability of the emissions on the same or different routes without taking into account the bias and uncertainty related to the use of different PEMS. The results of the two additional PN-PEMS and their measurement uncertainty can be found in the original report [44]. The details of the exercise can be found in the report as well, so here, only the most relevant information for understanding this study is summarised. The on-road tests took place at the end of 2015 at the premises of the following European laboratories in this chronological order:
  • Lab #1: JRC (Joint Research Centre), Ispra, Italy (mid-September);
  • Lab #2: Volkswagen, Wolfsburg, Germany (end of September);
  • Lab #3: Bosmal (Automotive R&D institute), Bielsko-Biala, Poland (beginning of October);
  • Lab #4: Honda R&D Europe, Offenbach am Main, Germany (end of October);
  • Lab #5: Audi, Neckarsulm, Germany (beginning of November);
  • Lab #6: Volvo, Gothenburg, Sweden (end of November);
  • Lab #7: TÜV Nord, Essen, Germany (beginning of December);
  • Lab #1 repeated at the end of the exercise (end of December) in order to confirm the stability and the absence of drift of the emissions.
Although the tests were performed 2 years before RDE’s introduction to the regulation and 8 years before the last package, they still have high interest and value as no such study has been conducted since then. Furthermore, the next Euro 7 step has kept the same (or similar) testing methodology (RDE) and technical specifications for the instruments.
The vehicle was a C-segment Volkswagen Golf, with 1.2 L gasoline direct injection (GDI) with a three-way catalyst (but no particulate filter), with maximum power of 63 kW, manual gearbox, kerb weight of 1205 kg, maximum laden mass of 1720 kg, and 9800 km on the odometer. The equipment installed in the car was a PEMS (for gases and PN) and two additional PN-PEMS. It was type-approved as a Euro 5b vehicle (model year 2015), and thus, no emission limits were applicable for RDE testing, as the RDE procedure was not in the regulation at that time.
The PEMS was the Semtech LDV from Sensors (Saline, MI, USA) measuring CO and CO2 with non-dispersive infrared detectors, NO and NO2 with non-dispersive ultraviolet detectors, and (solid) PN with condensation particle counter with thermal pre-treatment [31]. A 1.5-inch Exhaust Flow Meter (EFM) was used (maximum exhaust flow around 250 kg/h). It also logged the position coordinates, altitude, as well as ambient temperature, pressure, and relative humidity. The PEMS was installed on the tow bar of the car. The two additional PN-PEMS (prototype NPET from Horiba (Kyoto, Japan) and Nanomet 3 from Testo (Lenzkirch, Germany)) were installed in the cabin of the vehicle. More details about the PN-PEMS can be found elsewhere [31], and their results will not be presented here. At some laboratories, one of the two PN-PEMS was not installed because it was not functioning, and at some other laboratories, a second person was inside the car, in addition to the driver. The additional weight associated with the instrumentation and one driver was typically around 125 kg, 24% of the maximum payload, but it varied between 20% and 40%, well below the 90% of the maximum payload, which is the maximum applicable payload limit defined by the RDE regulation.
The testing protocol included 3 days in the laboratory and 2 days on the road (with morning and afternoon tests). The laboratory tests were the applicable type approval cycle of the specific vehicle NEDC (new European driving cycle) and the current type approval cycle WLTC (worldwide harmonised light vehicles test cycle). The vehicles were soaked in closed garages with temperatures around 20 °C; thus, the influence of the external ambient temperature during cold start was small. The soak duration did not fully respect the Euro 6d and 6e RDE requirements (minimum 6 h), as tests were run in the morning and afternoon. The drivers were instructed to drive normally according to their laboratory procedures for RDE tests. They were allowed to turn on the heater or air-conditioning of the car. All laboratories used reference E5 fuel from the same batch with 5% ethanol, 31.7% aromatics, and 3 mg/kg sulphur content.
Before each trip, the analyzers were zeroed with nitrogen, and their maximum range was adjusted using span gases with 1% accuracy (2% for NO2 in the first RDE package [20]). The span concentration range was different at each laboratory but fulfilled the requirement that the maximum concentration during a test should be lower than two times the span value, and all exceedances should be <1% of the total trip duration. At the end of the trip, the span ranges were controlled, and they were within 2%. The zero levels were also within 0.2% for CO2, 10 ppm for NOx, and 100 ppm for CO. Note that Euro 6d RDE specified a maximum drift of 5 ppm for NOx and 75 ppm for CO. With Euro 6e, the maximum NOx drift is 3 ppm. Each laboratory used its own personnel to operate the PEMS and drive the vehicle.
The data were analyzed in accordance with the rules of Regulation (EU) 2017/1151 [25] using EMROAD 6.04. A short description follows; the details can be found in the regulation. Instantaneous data were binned in urban (≤60 km/h), rural, or motorway (>90 km/h) parts, depending on the vehicle speed. The assessment was performed for the total trip and each sub-part. Initially, it was confirmed that the operation of PEMS was error-free, and the concentrations were within their calibrated range. The next step was the evaluation of the trip characteristics, such as urban, rural, motorway shares, altitude, temperature range, vehicle conditioning at cold start, cold start requirements, sub-part trip requirements (e.g., average speed, stop time), and cumulative altitude gain. No correction was needed and/or applied for extended ambient temperature or altitude conditions, as the conditions were within the moderate range. Then, the 95th cumulative percentile of speed × acceleration and the relative positive acceleration for each sub-part were calculated and compared with the respective limits. Finally, applying the moving average window, the CO2 of each window was compared with the reference values from the type approval cycle. The calculation of the emissions (i.e., from ppm to mg/km) for the total trip and each sub-part was the next step. No drift correction was applied to the PEMS concentrations. The final emissions were not corrected with the PEMS margin (1 + 0.1 for NOx given in Appendix 11 of Regulation (EU) 2023/443 for Euro 6e [24]). The analysis was performed without correcting for the RDE evaluation factor (as per Appendix 6 of Regulation (EU) 2017/1151 for Euro 6d [25] or Appendix 11 of Regulation (EU) 2023/443 for Euro 6e [24]). This factor is calculated as the ratio of the test (trip) CO2 and the type approval cycle CO2 and takes into account the severity of the trip. For ratios <1.3, no correction is applied, while for ratios >1.5, the inverse of the ratio is applied to the final emission result. For ratios in between, a linear function is used to calculate the final correction value. The effect of this correction will also be presented because it is included in the latest RDE regulation. For this correction, as the CO2 value was not available at the certificate of conformity of the vehicle, we used the average cold start WLTC CO2 value determined at the first laboratory.
Finally, the integrated results for the urban or the total trips were assessed following ISO 5725-2 [45] to determine the variance in the emissions following the same route (equations for within labs variance) and the variance in the emissions following different routes (equations for between labs variance) (see also [46] for simplified equations). The first is a weighted average of the laboratories’ same route’s “repeatability”. The second is the weighted variability of the laboratories’ averages. The analysis was carried out for each pollutant and CO2. The analysis was also repeated, removing the first 300 s (cold start) in order to remove as much as possible the effect of the ambient conditions on the emissions associated with the cold start, such effect being substantial for gasoline vehicles [47]. Variance is a measure of variability, but in this text, we use the terms interchangeably. The final statistical analysis is not part of the regulation but is typical for the inter-laboratory correlation exercises.

3. Results

This section will present the three assessment steps: trip validity; emissions calculations; and statistical analysis. A total of 24 RDE road tests were performed at the premises of the seven participating laboratories, plus two more repetitions at the end of the campaign from the first laboratory. The trips were compliant with the first package RDE, and any deviations from the subsequent RDEs will be discussed.

3.1. Trip Characteristics

Table 2 gives the average ambient conditions during the various on-road trips. It should be remembered that the vehicle was soaked in temperature-controlled areas with an ambient temperature of around 20 °C. The ambient temperature during the RDE tests ranged between 3 °C and 25 °C (averages between 7 °C and 23 °C) with an overall average value of 14.2 °C. The ambient pressure of the trips ranged between 99 kPa and 103.9 kPa, with an overall average value of 102 kPa. The average relative humidity was around 60%. According to the regulation, all trips were carried out at “moderate” conditions.
Table 3 summarises the average trip characteristics and the respective limits set in the legislation. In general, the trips fulfilled the requirements with the exception of the cold start provisions. Most of the trips had an initial idling duration of >15 s and a cold start–stop time of >90 s, thus not fulfilling Euro 6d and Euro 6e RDE regulations. These requirements were not part of the draft requirements when the tests were conducted in 2015. They were added with the third part of the RDE legislation in 2017 with Regulation (EU) 2017/1154 [22], and, therefore, not fulfilling them does not explicitly violate the RDE requirements. In general, the average values were within 10% of each other. All laboratories followed the urban, rural, and motorway, in this order, except laboratory #7, which followed the urban, motorway, and then the rural part. It should be remembered that the order urban–rural–motorway became obligatory with Regulation (EU) 2016/427 [20], applicable from 2017.
The average trip dynamics are presented in Figure 1. The clouds of points at speeds 29, 72, and 106 km/h represent the urban, rural, and motorway parts, respectively. All trips respected the lower limits of relative positive acceleration (RPA) and the upper limits of the product of velocity and acceleration (v × a), with some of the trips close to the limits, although they were introduced as requirements later [21].
The average exhaust flow rate of the total trip was 35 kg/h, and for the urban part, it was 25 kg/h. The peaks at the various trips reached 160–210 kg/h, thus covering 64–84% of the flow meter range. Amendment of the RDE regulation in Regulation (EU) 2018/1832 [23] recommends having the maximum expected flow rate during the test covering at least 75% of the EFM full range.

3.2. On-Road Emission Results

Real-time examples of emissions of the specific vehicle can be found in Figure A1 in Appendix A. In general, the pollutant concentrations were high at the cold start and at the motorway part.
Table 4 summarises the maximum concentrations and the 90th percentile of the analyzers’ signals that were measured during the campaign. The span gas concentrations that were used for the calibration of the analyzers were in accordance with the regulation. They covered at least 90% of the concentration values obtained from 99% of the measurement of the valid parts of the emissions test. The RDE concentrations were not more than double the span gas ranges. What is impressive is that while the peaks of CO exceeded 12,400 ppm, the majority of the concentrations during the trips were <255 ppm. To a smaller degree, this happened also to NO. These results show the wide range that the analyzers must cover and the importance of accurate linearization, even at low concentrations. It also indicates that any drift might have an impact on the final result [32].
Figure 2 presents the average results of the trips for each laboratory for CO2, NOx, and PN for the urban part (left panels) or the total trip (right panels), including the cold start. The tests of the first laboratory at the start and end of the inter-comparison exercise were quite close to each other. The PN and NOx were around 15% lower at the end of the testing, and CO2 was 9% higher. The CO (not shown) was around 30% higher. However, the tests in the second period were carried out at 7 °C, while in the first period, at 23 °C of the ambient temperature, which can explain such differences.
Even though it is not meaningful to perform a statistical analysis with only two repetitions, the differences between the urban part and the total trip were not statistically significant, except for NOx for the total trip. Nevertheless, based on the laboratory means, there was no obvious drift for the combination vehicle and instrumentation.
Regarding the tests at all laboratories for CO2, two distinct levels could be seen for the total trip: around 145 g/km (Labs #1, #3, and #5); and around 125 g/km (Labs #2, #4, #6, and #7). The high CO2 group had an additional passenger (Lab #1), lower ambient temperature, use of a heater or air conditioning and/or higher v × a. The variability (i.e., error bars) of the total trip was almost half (1–6 g/km) of the urban part variability (2–13 g/km).
NOx ranged between 45 and 65 mg/km for the total trip without any particular trend. NOx was ~100 mg/km in the urban part and had a rather high scatter. NO2 was negligible, with values from 0 up to 9 mg/km. The trips with higher NO2 emissions were those that had an increasing trend of NO2 over time, which was due to the drift of the analyzer (see example Figure A1 in Appendix A). The NOx variability (error bars) was around 5–16 mg/km for the total trip and 23–48 mg/km for the urban part.
PN varied in a narrow range of around 9 × 1011 #/km without any particular trend. The variability (error bars) was 0.1–1 × 1011 #/km for the total trip and 0.1–2 × 1011 #/km for the urban part.

3.3. Statistical Analysis

The results of the statistical analysis according to ISO 5725-2 for the variance in the emissions following the same or different routes are given in Table 5. Only the first tests of Lab #1 were considered in the analysis in order to avoid having two data sets from the same laboratory and bias the results. No laboratory was removed from the analysis based on the Grubbs and Cochran tests. The variance in the same route (within routes variability) for the total trips was 2.9% for CO2, around 24% for CO and NOx, and 6% for PN. The variance in different routes (between routes variability) was 7% for CO2, 9% for CO, negligible for NOx, and 9% for PN. The very low value for NOx has to do with the high scatter of the repeats at the laboratories, making the contribution of the variability of the means at all laboratories negligible. While the means at each laboratory were within 20% of the mean value, the variability at the labs was 18–41%. One interesting finding is that the variance in the emissions following the same route was higher than following different routes for CO and NOx, while the opposite applied to CO2. This will be discussed in the Discussion section, but it indicates that CO2 is sensitive to conditions such as weight and driving dynamics, while CO and NOx depend more on the cold start part of a trip. The last column, “Combined”, combines within and between variances and practically gives the expected variance in the vehicle as measured at different locations. It is calculated as the square root of the sum of the squares of the two values. It is the equivalent of “reproducibility” of laboratory tests, although this term cannot be used for on-road tests (see Section 1).
The results did not change significantly, correcting with the CO2 a ratio of a trip to laboratory certification cycle. Excluding the first 300 s of the trips and repeating the analysis (i.e., trips with hot start) did not improve the coefficient of variances. However, the absolute emissions were lower, especially for CO, which is usually mostly associated with the cold start emission, and thus, the absolute variances were lower (but not the relative variance).

4. Discussion

This is the first study to compare RDE-compliant routes at different locations in Europe using the same car and PEMS. Even though the original objective was to assess the measurement uncertainty of PN-PEMS (see [44]), the data were proven to be useful for the assessment of the trip variance following the RDE procedure, the variance following the same route and the variance following different routes at different locations. In general, the results of an RDE test depend on the vehicle and its stability, the influence of the operating conditions on the vehicle emissions, and the measurement accuracy of the measurement equipment. In this inter-laboratory exercise, the circulation of a “Golden” (reference) PEMS made it possible to exclude (or minimize) the uncertainty related to the use of different equipment. This does not mean that the results were necessarily the “true” emissions, but any bias of the instrument should be similar at all locations. Testing at the chassis dynamometers of the laboratories with two cold start NEDCs (new European driving cycle) and five hot start WLTCs (worldwide harmonised light vehicle test cycle) showed that the average deviations of PEMS to laboratory (bag) results were +9.4% (with one standard deviation ±2.9%) and +7.5% (±3.6%) for CO2 for the NEDCs and WLTCs, respectively, and +17.3% (±25.0%) and +2.0% (±17.5%) for NOx. Nevertheless, the differences were in most cases within the permissible tolerances prescribed in the RDE regulation for the CO2 (maximum difference between PEMS and laboratory 10 g/km or 10%, which was reduced to 7.5% with Regulation (EU) 2023/443 [24]) and NOx (15 mg/km or 15%, which was reduced to 10 mg/km or 12.5% with Regulation (EU) 2023/443). Similar differences for the same PEMS model were found by others [48]. While the PEMS vs. laboratory differences for NOx were not statistically significant, for CO2, they were. Typically, CO2 analyzers are accurate within 2%; thus, the difference indicates that the specific flow meter was overestimating the flow. Indeed, the CO2 deviation was similar to the difference in the PEMS flow to the exhaust flow calculated by the laboratory (total dilution tunnel flow minus dilution air) [44]. Although the CO2 is not included in the RDE pass/fail criteria, it can invalidate a test based on the overall trip dynamics requirement with the moving average window (MAW) (as defined in Appendix 5 of RDE regulation Annex IIIA [25]), and it can indirectly affect the results by the correction with the CO2 ratio of RDE to laboratory WLTC (Appendix 6 of the same RDE regulation). Other studies showed that the PEMS uncertainty can be as high as 43% for NOx and 50% for PN, but in most cases, it should be less [31,32]. Inter-laboratory PEMS exercises gave differences of <4% for CO2, 10–20% for NOx, and 35–40% for PN (the assessment was performed in the laboratory) [46,49].
The emissions of the “Golden” circulating vehicle with a gasoline direct injection engine and a three-way catalyst remained relatively stable over the exercise. The tests of the first laboratory at the start and end of the exercise had some differences (PN and NOx were 15% lower, CO 30% higher, and CO2 was 9% higher at the end of the testing). The lower ambient temperature, the use of the heater or air-conditioning, and the slightly higher v × a of the trips at the end of testing can partly explain this trend for CO2 [50].
With the exception of the cold start, three-way catalysts work efficiently, and emissions of CO, HCs (that were not measured in this study), and NOx are generally low. Higher emissions can be seen at deviations of stoichiometric mixture [51,52], e.g., dynamic driving with hard accelerations [53] or high engine loads, which was not the case of the conducted trips. The direct injection engine was proven to be a stable and substantial source of particles as well [54]. The urban and total trip emission levels of NOx (113 mg/km and 55 mg/km) and PN (10.2 × 1011 and 9.2 × 1011 #/km) were close to the limit or slightly higher than the limits for Euro 6d gasoline vehicles (60 × 1.43 = 86 mg/km for NOx and 1.5 × 6 × 1011 = 9 × 1011 #/km for PN), although not applicable to this car (type approved as Euro 5). RDE studies with Euro 5 and Euro 6b/c GDI vehicles without gasoline particulate filter (GPF) found emission levels between 20 and 83 mg/km for NOx and 0.6 and 1.7 × 1012 #/km for PN [55,56].
The environmental and driving conditions (Table 2), in general, varied in a narrow range, well below the temperature boundaries of the extended conditions (0 °C and 30 °C). Regulation (EU) 2023/443 increased the maximum extended ambient temperature from 35 °C to 38 °C, starting with Euro 6ebis (in force 01/01/2025). It also softened this requirement by invalidating a test only in case the emissions were above the limits. Even though the ambient temperature varied in the range of 3 °C to 25 °C (mean temperatures between 7 °C and 23 °C), all laboratories soaked the vehicles in a temperature-controlled area with ambient temperature around 20 °C, minimizing the effect of external temperature on the cold start emissions and the impact of using auxiliaries (e.g., air conditioning). It is highly likely that the scatter would be higher if the vehicle was soaked outside. Since the third package (2017/1154) [22], if the vehicle is conditioned for the last three hours prior to the test at an average temperature that falls within the extended range, then the 1.6 extended factor applies to the cold start period, even if the running conditions are not within the extended temperature range. The rest of the trip characteristics, such as distances, shares, etc., also varied in a relatively narrow range (typically within ±10%). Bigger variability of the trip characteristics was noticed at the cold start part. The reason was that the testing was conducted before the introduction of RDE in the regulation, and in particular, the third package that introduced the new provisions regarding idling duration and cold start average speed; so, in some cases, the results were not compliant to the Euro 6d or later RDE requirements. Note that with Regulation EU 2023/443 [24], the test is invalid only if the final emissions exceed the limit. Nevertheless, repeating the statistical analysis excluding the first 300 s (cold start), no improvement in the (relative) variance values was noted (partly explained by the lower absolute emission factors). The removal of cold start reduced the absolute emissions by 3% for CO2, 84% for CO, 7% for NOx, and 8% for PN for the urban part. The contribution of cold start was higher [56] or similar [4,57] to what has been reported by others for RDE trips. The relative contribution depends on the absolute emissions during the cold start period but also on the absolute emissions at the rest of the test and the trip duration. For pollutants such as CO, where the majority of the emissions occur at the cold start, the inclusion or not plays an important role.
Repeating the analysis using the pollutant emissions corrected with the RDE evaluation factor (see Section 2: CO2 ratio of the trip to the certification cycle; practically normalizing the emissions to the same CO2 emissions), did not change the variances. This implies that the dependence of CO, NOx, and PN on the CO2 or power of the vehicle is small for properly operating after-treatment devices and the small range of CO2 of this study.
In one laboratory, the order of testing was not urban, rural, and motorway, but urban, motorway, and rural (Lab #7). No differences were seen compared to the rest of the laboratory trips for the rural and motorway parts. Later amendment with Regulation (EU) 2018/1832 [23] defined that the order has to be urban, rural, motorway.
A few labs (#1, #5) had higher driving dynamics than some others (#6, #2, #7), but still well within the RDE boundaries (see Figure 1). This was also evident at the CO2 values (Figure 2a,b) but not at the other pollutants (Figure 2c–f). This indicates that the vehicle could control the emissions within the boundaries of the regulation well.
The lack of NO2 zero checks and/or the 10 ppm permitted drift at that time (5 ppm for NO and 5 ppm for NO2) resulted in high NO2 measurements scatter (0–9 mg/km). This drift, however, did not impact the NOx variance as NO2 was <15% of NOx. Euro 6e RDE regulation [24] allows for a maximum (total) NOx drift of 3 ppm and drift correction on the instantaneous values.
In addition to the differences in routes, the different driving dynamics, the lack of weight control, and air conditioning use influenced the CO2 variance [50,58]. The big variability of CO2 under real conditions is well known. A study that monitored the fuel consumption (almost proportional to CO2 emissions) of a single vehicle throughout one year with twenty different persons using and driving the vehicle found deviations of ±14% (one standard deviation) [59]. Our results also showed a scatter of 8-14% for urban or total trips with cold or hot engine start, highlighting the difficulties in assessing CO2 (and fuel consumption) with on-road trips. Nevertheless, the results give an indication of the variance in the same or different routes under normal RDE conditions; thus, they could be useful for researchers who declare or interpret RDE results with gasoline-fueled vehicles. It should be emphasized that the results include the variability of the vehicle. In our case, it was proven to be relatively constant after >4000 km of driving; thus, they give a good indication of the RDE testing procedure. However, for other vehicles and different environmental or trip conditions, the variability of the vehicle emissions could be much higher. This depends on the calibration and strategy of the engine and/or aftertreatment devices. For example, NOx emissions of diesel vehicles are more difficult to control and could depend on the ambient temperature [38,39,48,60], the exhaust gas temperature [61,62], the use of the exhaust gas recirculation (EGR) [63], or regeneration events [64]. It is also well known that more dynamic driving (with harder accelerations) (or expressed as higher v × a or RPA) will increase the emissions and, consequently, the variance in the results [12,61,65,66,67,68,69]. The traffic, the road grade, the altitude, and the wind also have an effect on CO2 and pollutants [60,70,71,72,73,74]. In general, vehicle-specific power (VSP) is correlated with emissions [75], and the use of auxiliaries, e.g., air conditioning, can also influence the results [38,39,50]. For PN, the existence of a particulate filter and vehicle hybridization can also increase the variability [76]. Thus, more complex aftertreatment devices (such as those installed in Euro 6d vehicles and later), wider ambient temperatures, more dynamic driving and wider changes in altitude will increase the scatter (and variance) due to their greater influence on the vehicle emissions.
Due to the wide range of environmental and trip conditions, RDE regulation was designed as a pass–fail test and not for comparisons of vehicles. Here, we gave the variability ranges of the same or different routes under moderate RDE conditions, including cold start. If, however, the objective is the comparability of vehicle emissions, then the guidelines of the European Committee for Standardization (CEN) Workshop Agreement (CWA) 17379:2019 should be followed [77]. The standard was built upon the RDE methods and defined a process to test the emissions during urban driving, with the aim of comparing the on-road emissions performance of different vehicles. Among others, some requirements include trips of at least 10 km long, with a hot engine, at ambient temperatures between 10 °C and 20 °C, and average speed between 20 km/h and 40 km/h. According to this CEN, to compare emissions between vehicle models, it is necessary to collect at least five valid trips from at least two different vehicles of the specific model, fulfilling the above-mentioned requirements.
One final note is that the PEMS measurement uncertainty is different from the “route” variability values that were discussed here. Two trips at two locations with two different PEMS could have differences that combine the two factors (trip influence and PEMS influence).
In closing, it should be emphasized that the variance following the same route is at acceptable levels, comparable to chassis dynamometer laboratory repeatability for the gaseous pollutants (but not for CO2). For example, reported repeatability for the NEDC cycle can be <0.5% for CO2 (at 200 g/km), but for NOx (at 20 mg/km), it can be 15% or higher, especially for PN [78,79,80].
Similarly, the laboratory reproducibility based on inter-laboratory correlation exercises compared to RDE is lower for CO2 (2%) and lower-to-similar for the rest of the pollutants (10–30%) [46,81]. The variance in the CO2, NOx, and PN for the cold start NEDC and hot WLTC of the laboratories that participated in our study was 2.1–2.3%, 20–37%, and 20–22%, respectively [44].

5. Conclusions

A Euro 5b gasoline direct injection vehicle with a portable emissions measurement system (PEMS) was circulated at seven European laboratories. Trips compliant with the European Union’s (EU) real-driving emissions (RDE) regulation were conducted at the premises of the laboratories. The variance in different routes around Europe was 7% for CO2, negligible for NOx and 9% for PN and CO. The variance following the same route was 3% for CO2, around 24% for CO and NOx, and 6% for PN. The data showed that the variance in CO and NOx repeating the same route (24%) could be higher than following different routes (<9%). The opposite was noticed for CO2 (7% vs. 3%). For PN, both the same and different route variances contributed to the final result. The results of this study indicate that the NOx and CO emissions at different locations could be comparable under similar environmental (temperature, altitude) conditions and trip characteristics (dynamics, shares of urban/motorway parts). For CO2, though, stricter protocols and control of conditions are necessary if results need to be compared. When trip emissions are compared, though, in addition to the trip influence that was discussed in this study, the PEMS uncertainty should be taken into account. Finally, the variance values derived from the Euro 5b gasoline direct injection vehicle with the three-way catalyst of this study do not necessarily apply to more advanced or different combustion and aftertreatment technologies (e.g., hybrids or diesel with selective catalytic reduction for NOx).

Author Contributions

Conceptualization, B.G.; methodology, B.G.; writing—original draft preparation, B.G.; writing—review and editing, A.M., M.C., V.V., P.B. and P.D. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

Data is contained within the article.

Acknowledgments

The authors would like to acknowledge all the laboratories that participated in this inter-laboratory comparison.

Conflicts of Interest

The opinions expressed in this manuscript are those of the author and should in no way be considered to represent an official opinion of the European Commission. Mention of trade names or commercial products does not constitute endorsement or recommendation by the author or the European Commission. Author Victor Valverde was employed by the company Unisystem S.A. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Appendix A

Figure A1 gives an example of real-time signals of the wet concentration of CO, PN, NO, NO2, and CO2. As the speed profile reveals (bottom of the figure), the first 3100 s are the urban part; then the rural part continues until 5000 s and then the motorway part. Gasoline engines work stoichiometrically, so the CO2 is relatively constant, around 13.8%, with lower values at fuel cut-offs. The NO2 is very low, while NO is emitted during a cold start (when the catalyst temperature is low) and decelerations (lean conditions and lack of oxygen reduce the conversion efficiency). The gradual increase in NO2 up to 5 ppm for this particular example is an indication of the analyzer’s drift. PN is high during the whole cycle, particularly at cold start and accelerations (note that this vehicle is not equipped with a particle filter). CO is very high during the cold start but is practically at zero when the catalyst reaches its light-off temperature; then, there are a few spikes during accelerations and some emissions at high speeds, probably because the exhaust flow is higher than the oxidation capacity of the catalyst and/or poor lambda management.
Energies 17 01308 g0a1aEnergies 17 01308 g0a1b
Figure A1. Real-time example of emission profiles during an RDE trip of laboratory #5: (a) CO; (b) PN; (c) NO; (d) NO2; (e) CO2; (f) exhaust flow rate; (g) speed.
Figure A1. Real-time example of emission profiles during an RDE trip of laboratory #5: (a) CO; (b) PN; (c) NO; (d) NO2; (e) CO2; (f) exhaust flow rate; (g) speed.
Energies 17 01308 g0a1aEnergies 17 01308 g0a1b

References

  1. May, A.A.; Nguyen, N.T.; Presto, A.A.; Gordon, T.D.; Lipsky, E.M.; Karve, M.; Gutierrez, A.; Robertson, W.H.; Zhang, M.; Brandow, C.; et al. Gas- and Particle-Phase Primary Emissions from in-Use, on-Road Gasoline and Diesel Vehicles. Atmos. Environ. 2014, 88, 247–260. [Google Scholar] [CrossRef]
  2. Andersson, J.; May, J.; Favre, C.; Bosteels, D.; de Vries, S.; Heaney, M.; Keenan, M.; Mansell, J. On-Road and Chassis Dynamometer Evaluations of Emissions from Two Euro 6 Diesel Vehicles. SAE Int. J. Fuels Lubr. 2014, 7, 919–934. [Google Scholar] [CrossRef]
  3. Pielecha, J.; Merkisz, J.; Markowski, J.; Jasiński, R. Analysis of Passenger Car Emission Factors in RDE Tests. E3S Web Conf. 2016, 10, 00073. [Google Scholar] [CrossRef]
  4. Luján, J.M.; 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, 112–121. [Google Scholar] [CrossRef]
  5. Lozhkina, O.V.; Lozhkin, V.N. Estimation of Nitrogen Oxides Emissions from Petrol and Diesel Passenger Cars by Means of On-Board Monitoring: Effect of Vehicle Speed, Vehicle Technology, Engine Type on Emission Rates. Transp. Res. Part D Transp. Environ. 2016, 47, 251–264. [Google Scholar] [CrossRef]
  6. Czerrwiński, J.; Comte, P. Testing Emissions of Passenger Cars in Laboratory and On-Road (PEMS, RDE). Combust. Engines 2016, 166, 17–23. [Google Scholar] [CrossRef]
  7. Merkisz, J.; Pielecha, J. Selected Remarks about RDE Test. Combust. Engines 2016, 166, 54–61. [Google Scholar] [CrossRef]
  8. McCaffery, C.; Zhu, H.; Li, C.; Durbin, T.D.; Johnson, K.C.; Jung, H.; Brezny, R.; Geller, M.; Karavalakis, G. 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. [Google Scholar] [CrossRef]
  9. Gis, W.; Gis, M.; Wiśniowski, P.; Bednarski, M. Comparative Tests of a Passenger Car with Compression Ignition Engine on Chassis Dynamometer during NEDC and WLTC Tests and during RDE Road Test. Combust. Engines 2019, 178, 228–234. [Google Scholar] [CrossRef]
  10. Gis, W.; Gis, M.; Pielecha, J.; Skobiej, K. Alternative Exhaust Emission Factors from Vehicles in On-Road Driving Tests. Energies 2021, 14, 3487. [Google Scholar] [CrossRef]
  11. Pelkmans, L.; Debal, P. Comparison of On-Road Emissions with Emissions Measured on Chassis Dynamometer Test Cycles. Transp. Res. Part D Transp. Environ. 2006, 11, 233–241. [Google Scholar] [CrossRef]
  12. Kumar Pathak, S.; Sood, V.; Singh, Y.; Channiwala, S.A. Real World Vehicle Emissions: Their Correlation with Driving Parameters. Transp. Res. Part D Transp. Environ. 2016, 44, 157–176. [Google Scholar] [CrossRef]
  13. Rašić, D.; Rodman Oprešnik, S.; Seljak, T.; Vihar, R.; Baškovič, U.Ž.; Wechtersbach, T.; Katrašnik, T. RDE-Based Assessment of a Factory Bi-Fuel CNG/Gasoline Light-Duty Vehicle. Atmos. Environ. 2017, 167, 523–541. [Google Scholar] [CrossRef]
  14. Triantafyllopoulos, G.; Dimaratos, A.; Ntziachristos, L.; Bernard, Y.; Dornoff, J.; Samaras, Z. A Study on the CO2 and NOx Emissions Performance of Euro 6 Diesel Vehicles under Various Chassis Dynamometer and On-Road Conditions Including Latest Regulatory Provisions. Sci. Total Environ. 2019, 666, 337–346. [Google Scholar] [CrossRef] [PubMed]
  15. Kontses, A.; Triantafyllopoulos, G.; Ntziachristos, L.; Samaras, Z. Particle Number (PN) Emissions from Gasoline, Diesel, LPG, CNG and Hybrid-Electric Light-Duty Vehicles under Real-World Driving Conditions. Atmos. Environ. 2020, 222, 117126. [Google Scholar] [CrossRef]
  16. Vojtíšek-Lom, M.; Beránek, V.; Klír, V.; Jindra, P.; Pechout, M.; Voříšek, T. On-Road and Laboratory Emissions of NO, NO2, NH3, N2O and CH4 from Late-Model EU Light Utility Vehicles: Comparison of Diesel and CNG. Sci. Total Environ. 2018, 616–617, 774–784. [Google Scholar] [CrossRef] [PubMed]
  17. Tu, R.; Xu, J.; Wang, A.; Zhai, Z.; Hatzopoulou, M. Effects of Ambient Temperature and Cold Starts on Excess NOx Emissions in a Gasoline Direct Injection Vehicle. Sci. Total Environ. 2020, 760, 143402. [Google Scholar] [CrossRef] [PubMed]
  18. Giechaskiel, B.; Melas, A.; Martini, G.; Dilara, P. Overview of Vehicle Exhaust Particle Number Regulations. Processes 2021, 9, 2216. [Google Scholar] [CrossRef]
  19. European Commission Commission Regulation (EU) No 582/2011 of 25 May 2011 Implementing and Amending Regulation (EC) No 595/2009 of the European Parliament and of the Council with Respect to Emissions from Heavy Duty Vehicles (Euro VI) and Amending Annexes I and III to Directive 2007/46/EC of the European Parliament and of the Council. Off. J. Eur. Union 2011, L167, 1–168.
  20. European Commission Commission Regulation (EU) 2016/427 of 10 March 2016 Amending Regulation (EC) No 692/2008 as Regards Emissions from Light Passenger and Commercial Vehicles (Euro 6). Off. J. Eur. Union 2016, L82, 1–98.
  21. European Commission Commission Regulation (EU) 2016/646 of 20 April 2016 Amending Regulation (EC) No 692/2008 as Regards Emissions from Light Passenger and Commercial Vehicles (Euro 6). Off. J. Eur. Union 2016, L109, 1–22.
  22. European Commission Commission Regulation (EU) 2017/1154 of 7 June 2017 Amending Regulation (EU) 2017/1151 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 and Euro 6) and on Access to Vehicle Repair and Maintenance Information, Amending Directive 2007/46/EC of the European Parliament and of the Council, Commission Regulation (EC) No 692/2008 and Commission Regulation (EU) No 1230/2012 and Repealing Regulation (EC) No 692/2008 and Directive 2007/46/EC of the European Parliament and of the Council as Regards Real-Driving Emissions from Light Passenger and Commercial Vehicles (Euro 6) (Text with EEA Relevance). Off. J. Eur. Union 2017, L175, 708–732.
  23. European Commission Commission Regulation (EU) 2018/1832 of 5 November 2018 Amending Directive 2007/46/EC of the European Parliament and of the Council, Commission Regulation (EC) No 692/2008 and Commission Regulation (EU) 2017/1151 for the Purpose of Improving the Emission Type Approval Tests and Procedures for Light Passenger and Commercial Vehicles, Including Those for in-Service Conformity and Real-Driving Emissions and Introducing Devices for Monitoring the Consumption of Fuel and Electric Energy. Off. J. Eur. Union 2018, L301, 1–314.
  24. European Commission Commission Regulation (EU) 2023/443 of 8 February 2023 Amending Regulation (EU) 2017/1151 as Regards the Emission Type Approval Procedures for Light Passenger and Commercial Vehicles. Off. J. Eur. Union 2023, L66, 1–237.
  25. European Commission Commission Regulation (EU) 2017/1151 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 and Euro 6) and on Access to Vehicle Repair and Maintenance Information, Amending Directive 2007/46/EC of the European Parliament and of the Council, Commission Regulation (EC) No 692/2008 and Commission Regulation (EU) No 1230/2012 and Repealing Commission Regulation (EC) No 692/2008. Off. J. Eur. Union 2017, L175, 1–643. Available online: https://eur-lex.europa.eu/eli/Reg/2017/1151/2023-09-01 (accessed on 17 February 2024).
  26. The Gazette of India. Ministry of Road Transport and Highways; Part II, Section 3(i), Regd. No. D. L.-33004/99; Government of India: New Delhi, India, 17 February 2023.
  27. Joumard, R.; Laurikko, J.; Han, T.L.; Geivanidis, S.; Samaras, Z.; Merétei, T.; Devaux, P.; André, J.-M.; Cornelis, E.; Lacour, S.; et al. Accuracy of Exhaust Emission Factor Measurements on Chassis Dynamometer. J. Air Waste Manag. Assoc. 2009, 59, 695–703. [Google Scholar] [CrossRef] [PubMed]
  28. European Commission Regulation (EC) No 715/2007 of the European Parliament and of the Council of 20 June 2007 on Type Approval of Motor Vehicles with Respect to Emissions from Light Passenger and Commercial Vehicles (Euro 5 and Euro 6) and on Access to Vehicle Repair and Maintenance Information. Off. J. Eur. Union 2007, L171, 1–16.
  29. European Commission Regulation (EU) 2018/858 of the European Parliament and of the Council of 30 May 2018 on the Approval and Market Surveillance of Motor Vehicles and Their Trailers, and of Systems, Components and Separate Technical Units Intended for Such Vehicles, Amending Regulations (EC) No 715/2007 and (EC) No 595/2009 and Repealing Directive 2007/46/EC. The Latest Consolidated Version Can Be Found at: Http://Data.Europa.Eu/Eli/Reg/2018/858/2023-07-30. Off. J. Eur. Union 2018, L151, 1–218.
  30. European Commission Commission Notice Guidance on the Identification of the Presence of Defeat Devices with Regards to Emissions of Light-Duty Vehicles Approved with Real Driving Emissions (RDE), Heavy-Duty Vehicles and on Anti-Tampering Protection 2023/C 68/01. Off. J. Eur. Union 2023, L68, 1–20.
  31. Giechaskiel, B.; Bonnel, P.; Perujo, A.; Dilara, P. Solid Particle Number (SPN) Portable Emissions Measurement Systems (PEMS) in the European Legislation: A Review. Int. J. Environ. Res. Public Health 2019, 16, 4819. [Google Scholar] [CrossRef]
  32. Giechaskiel, B.; Clairotte, M.; Valverde-Morales, V.; Bonnel, P.; Kregar, Z.; Franco, V.; Dilara, P. Framework for the Assessment of PEMS (Portable Emissions Measurement Systems) Uncertainty. Environ. Res. 2018, 166, 251–260. [Google Scholar] [CrossRef] [PubMed]
  33. General Secretariat of the Council Agreed Document ST_5142_2024_INIT: Regulation on Type-Approval of Motor Vehicles and Engines and of Systems, Components and Separate Technical Units Intended for Such Vehicles, with Respect to Their Emissions and Battery Durability (Euro 7)—Letter to the Chair of the European Parliament ENVI Committee. 2024. Available online: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=consil%3AST_5142_2024_INIT (accessed on 15 February 2024).
  34. Feist, M.D.; Sharp, C.A.; Spears, M.W. Determination of Pems Measurement Allowances for Gaseous Emissions Regulated under the Heavy-Duty Diesel Engine in-Use Testing Program: Part 1 – Project Overview and PEMS Evaluation Procedures. SAE Int. J. Fuels Lubr. 2009, 2, 435–454. [Google Scholar] [CrossRef]
  35. Matla, J.; Sitnik, L.J.; Sroka, Z.J.; Włostowski, R.M. Aspect of Repeatability in Real Driving Emissions Tests Regarding to Velocity Profile. IOP Conf. Ser. Mater. Sci. Eng. 2020, 977, 012006. [Google Scholar] [CrossRef]
  36. Bodisco, T.; Zare, A. Practicalities and Driving Dynamics of a Real Driving Emissions (RDE) Euro 6 Regulation Homologation Test. Energies 2019, 12, 2306. [Google Scholar] [CrossRef]
  37. Pielecha, J.; Jasiński, R.; Magdziak, A. Practicability of Passenger Vehicle Driving Emission Tests According to New European Union Procedures. MATEC Web Conf. 2017, 118, 00021. [Google Scholar] [CrossRef]
  38. 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, 70–77. [Google Scholar] [CrossRef]
  39. Donateo, T.; Giovinazzi, M. Some Repeatability and Reproducibility Issues in Real Driving Emission Tests; SAE Technical Paper 2018-01-5020; SAE International: Warrendale, PA, USA, 2018. [Google Scholar] [CrossRef]
  40. Merkisz, J.; Pielecha, J. Comparison of Real Driving Emissions Tests. IOP Conf. Ser. Mater. Sci. Eng. 2018, 421, 042055. [Google Scholar] [CrossRef]
  41. European Commission UN Regulation No 168—Uniform Provisions Concerning the Approval of Light Duty Passenger and Commercial Vehicles with Regards to Real Driving Emissions (RDE). Off. J. Eur. Union 2024, L, 1–98.
  42. Su, S.; Hou, P.; Wang, X.; Lyu, L.; Ge, Y.; Lyu, T.; Lai, Y.; Luo, W.; Wang, Y. Evaluating the Measurement Uncertainty of On-Road NOx Using a Portable Emission Measurement System (PEMS) Based on Real Testing Data in China. Atmosphere 2023, 14, 702. [Google Scholar] [CrossRef]
  43. Adamiak, B.; Szczotka, A.; Woodburn, J.; Merkisz, J. Comparison of Exhaust Emission Results Obtained from Portable Emissions Measurement System (PEMS) and a Laboratory System. Combust. Engines 2023, 195, 128–135. [Google Scholar] [CrossRef]
  44. Riccobono, F.; Giechaskiel, B.; Mendoza Villafuerte, P. Particle Number PEMS Inter-Laboratory Comparison Exercise; Publications Office: Luxembourg, 2016. [Google Scholar]
  45. ISO 5725–2:1994; Accuracy (Trueness and Precision) of Measurement Methods and Results—Part 2: Basic Method for the Determination of Repeatability and Reproducibility of a Standard Measurement Method. International Organization for Standardization: Geneva, Switzerland, 1994.
  46. Giechaskiel, B.; Casadei, S.; Mazzini, M.; Sammarco, M.; Montabone, G.; Tonelli, R.; Deana, M.; Costi, G.; Di Tanno, F.; Prati, M.; et al. Inter-Laboratory Correlation Exercise with Portable Emissions Measurement Systems (PEMS) on Chassis Dynamometers. Appl. Sci. 2018, 8, 2275. [Google Scholar] [CrossRef]
  47. Weilenmann, M.; Favez, J.-Y.; Alvarez, R. Cold-Start Emissions of Modern Passenger Cars at Different Low Ambient Temperatures and Their Evolution over Vehicle Legislation Categories. Atmos. Environ. 2009, 43, 2419–2429. [Google Scholar] [CrossRef]
  48. Cha, J.; Lee, J.; Chon, M.S. Evaluation of Real Driving Emissions for Euro 6 Light-Duty Diesel Vehicles Equipped with LNT and SCR on Domestic Sales in Korea. Atmos. Environ. 2019, 196, 133–142. [Google Scholar] [CrossRef]
  49. Giechaskiel, B.; Casadei, S.; Rossi, T.; Forloni, F.; Di Domenico, A. Measurements of the Emissions of a “Golden” Vehicle at Seven Laboratories with Portable Emission Measurement Systems (PEMS). Sustainability 2021, 13, 8762. [Google Scholar] [CrossRef]
  50. Fontaras, G.; Ciuffo, B.; Zacharof, N.; Tsiakmakis, S.; Marotta, A.; Pavlovic, J.; Anagnostopoulos, K. The Difference between Reported and Real-World CO2 Emissions: How Much Improvement Can Be Expected by WLTP Introduction? Transp. Res. Procedia 2017, 25, 3933–3943. [Google Scholar] [CrossRef]
  51. Brandt, E.P.; Wang, Y.; Grizzle, J.W. Dynamic Modeling of a Three-Way Catalyst for SI Engine Exhaust Emission Control. IEEE Trans. Contr. Syst. Technol. 2000, 8, 767–776. [Google Scholar] [CrossRef]
  52. Kim, J.; Myung, C.; Lee, K. Exhaust Emissions and Conversion Efficiency of Catalytic Converter for an Ethanol-fueled Spark Ignition Engine. Biofuels Bioprod. Bioref. 2019, 13, 1211–1223. [Google Scholar] [CrossRef]
  53. Satlawa, M.; Pajdowski, P.; Pietras, D. The Impact of the Driver’s Driving Style on the EXHAUST Emissions of a Passenger Car during a Real Road Cycle; SAE Technical Paper 2020-01-2214; SAE International: Warrendale, PA, USA, 2020. [Google Scholar]
  54. Raza, M.; Chen, L.; Leach, F.; Ding, S. A Review of Particulate Number (PN) Emissions from Gasoline Direct Injection (GDI) Engines and Their Control Techniques. Energies 2018, 11, 1417. [Google Scholar] [CrossRef]
  55. Demuynck, J.; Favre, C.; Bosteels, D.; Hamje, H.; Andersson, J. Real-World Emissions Measurements of a Gasoline Direct Injection Vehicle without and with a Gasoline Particulate Filter; SAE Technical Paper 2017-01-0985; SAE International: Warrendale, PA, USA, 2017. [Google Scholar]
  56. Merkisz, J.; Bielaczyc, P.; Pielecha, J.; Woodburn, J. RDE Testing of Passenger Cars: The Effect of the Cold Start on the Emissions Results; SAE Technical Paper 2019-01-0747; SAE International: Warrendale, PA, USA, 2019. [Google Scholar]
  57. Wang, X.; Jing, X.; Li, J. Cold Start Emission Characteristics and Its Effects on Real Driving Emission Test. In Application of Intelligent Systems in Multi-modal Information Analytics; Sugumaran, V., Xu, Z., Shankar, P., Zhou, H., Eds.; Advances in Intelligent Systems and Computing; Springer International Publishing: Cham, Swizterland, 2019; Volume 929, pp. 873–882. ISBN 978-3-030-15738-8. [Google Scholar]
  58. Kim, K.; Chung, W.; Kim, M.; Kim, C.; Myung, C.-L.; Park, S. Inspection of PN, CO2, and Regulated Gaseous Emissions Characteristics from a GDI Vehicle under Various Real-World Vehicle Test Modes. Energies 2020, 13, 2581. [Google Scholar] [CrossRef]
  59. Pavlovic, J.; Fontaras, G.; Ktistakis, M.; Anagnostopoulos, K.; Komnos, D.; Ciuffo, B.; Clairotte, M.; Valverde, V. Understanding the Origins and Variability of the Fuel Consumption Gap: Lessons Learned from Laboratory Tests and a Real-Driving Campaign. Environ. Sci. Eur. 2020, 32, 53. [Google Scholar] [CrossRef]
  60. Ko, J.; Myung, C.-L.; Park, S. Impacts of Ambient Temperature, DPF Regeneration, and Traffic Congestion on NOx Emissions from a Euro 6-Compliant Diesel Vehicle Equipped with an LNT under Real-World Driving Conditions. Atmos. Environ. 2019, 200, 1–14. [Google Scholar] [CrossRef]
  61. Mera, Z.; Fonseca, N.; López, J.-M.; Casanova, J. Analysis of the High Instantaneous NOx Emissions from Euro 6 Diesel Passenger Cars under Real Driving Conditions. Appl. Energy 2019, 242, 1074–1089. [Google Scholar] [CrossRef]
  62. Mera, Z.; Fonseca, N.; Casanova, J.; López, J.-M. Influence of Exhaust Gas Temperature and Air-Fuel Ratio on NOx Aftertreatment Performance of Five Large Passenger Cars. Atmos. Environ. 2021, 244, 117878. [Google Scholar] [CrossRef]
  63. Prati, M.V.; Costagliola, M.A.; Zuccheroso, A.; Napolitano, P. Assessment of Euro 5 Diesel Vehicle NOx Emissions by Laboratory and Track Testing. Environ. Sci. Pollut. Res. 2019, 26, 10576–10586. [Google Scholar] [CrossRef]
  64. Giechaskiel, B. Particle Number Emissions of a Diesel Vehicle during and between Regeneration Events. Catalysts 2020, 10, 587. [Google Scholar] [CrossRef]
  65. Clenci, A.; Sălan, V.; Niculescu, R.; Iorga-Simăn, V.; Zaharia, C. Assessment of Real Driving Emissions via Portable Emission Measurement System. IOP Conf. Ser. Mater. Sci. Eng. 2017, 252, 012084. [Google Scholar] [CrossRef]
  66. Pielecha, J.; Kurtyka, K.; Skobiej, K. The Impact of Vehicle Dynamic Parameters on the Exhaust Emissions in RDE Tests. Combust. Engines 2018. [Google Scholar] [CrossRef]
  67. Kurtyka, K.; Pielecha, J. The Evaluation of Exhaust Emission in RDE Tests Including Dynamic Driving Conditions. Transp. Res. Procedia 2019, 40, 338–345. [Google Scholar] [CrossRef]
  68. Varella, R.A.; Faria, M.V.; Mendoza-Villafuerte, P.; Baptista, P.C.; Sousa, L.; Duarte, G.O. Assessing the Influence of Boundary Conditions, Driving Behavior and Data Analysis Methods on Real Driving CO2 and NOx Emissions. Sci. Total Environ. 2019, 658, 879–894. [Google Scholar] [CrossRef] [PubMed]
  69. Luján, J.M.; Guardiola, C.; Pla, B.; Pandey, V. Impact of Driving Dynamics in RDE Test on NOx Emissions Dispersion. Proc. Inst. Mech. Eng. Part D J. Automob. Eng. 2020, 234, 1770–1778. [Google Scholar] [CrossRef]
  70. Wyatt, D.W.; Li, H.; Tate, J.E. The Impact of Road Grade on Carbon Dioxide (CO2) Emission of a Passenger Vehicle in Real-World Driving. Transp. Res. Part D Transp. Environ. 2014, 32, 160–170. [Google Scholar] [CrossRef]
  71. 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). Transp. Res. Part D Transp. Environ. 2017, 52, 215–226. [Google Scholar] [CrossRef]
  72. Costagliola, M.A.; Costabile, M.; Prati, M.V. Impact of Road Grade on Real Driving Emissions from Two Euro 5 Diesel Vehicles. Appl. Energy 2018, 231, 586–593. [Google Scholar] [CrossRef]
  73. Wang, Y.; Ge, Y.; Wang, J.; Wang, X.; Yin, H.; Hao, L.; Tan, J. Impact of Altitude on the Real Driving Emission (RDE) Results Calculated in Accordance to Moving Averaging Window (MAW) Method. Fuel 2020, 277, 117929. [Google Scholar] [CrossRef]
  74. Giechaskiel, B.; Komnos, D.; Fontaras, G. Impacts of Extreme Ambient Temperatures and Road Gradient on Energy Consumption and CO2 Emissions of a Euro 6d-Temp Gasoline Vehicle. Energies 2021, 14, 6195. [Google Scholar] [CrossRef]
  75. Rodríguez, R.A.; Virguez, E.A.; Rodríguez, P.A.; Behrentz, E. Influence of Driving Patterns on Vehicle Emissions: A Case Study for Latin American Cities. Transp. Res. Part D Transp. Environ. 2016, 43, 192–206. [Google Scholar] [CrossRef]
  76. Melas, A.; Selleri, T.; Franzetti, J.; Ferrarese, C.; Suarez-Bertoa, R.; Giechaskiel, B. On-Road and Laboratory Emissions from Three Gasoline Plug-In Hybrid Vehicles-Part 2: Solid Particle Number Emissions. Energies 2022, 15, 5266. [Google Scholar] [CrossRef]
  77. CWA 17379:2019; General Guideline on Real Drive Test Methodology for Compiling Comparable Emission Data. CEN: Brussels, Belgium, 2019.
  78. Jaworski, A.; Kuszewski, H.; Ustrzycki, A.; Balawender, K.; Lejda, K.; Woś, P. Analysis of the Repeatability of the Exhaust Pollutants Emission Research Results for Cold and Hot Starts under Controlled Driving Cycle Conditions. Environ. Sci. Pollut. Res. 2018, 25, 17862–17877. [Google Scholar] [CrossRef]
  79. Chłopek, Z. Testing of Non-Repeatability of Pollution Emission from Motor-Car Engines. Combust. Engines 2004, 118, 40–50. [Google Scholar] [CrossRef]
  80. Andrych-Zalewska, M.; Chlopek, Z.; Merkisz, J.; Pielecha, J. Comparison of Gasoline Engine Exhaust Emissions of a Passenger Car through the WLTC and RDE Type Approval Tests. Energies 2022, 15, 8157. [Google Scholar] [CrossRef]
  81. Pavlovic, J.; Marotta, A.; Ciuffo, B.; Serra, S.; Fontaras, G.; Anagnostopoulos, K.; Tsiakmakis, S.; Arcidiacono, V.; Hausberger, S.; Silberholz, G. Correction of Test Cycle Tolerances: Evaluating the Impact on CO2 Results. Transp. Res. Procedia 2016, 14, 3099–3108. [Google Scholar] [CrossRef]
Energies 17 01308 g001
Figure 1. Average trip dynamics for the different laboratories: (a) relative positive acceleration (RPA); (b) 95th percentile of velocity per positive acceleration. Both in function of the average speed of urban, rural, and motorway parts. The dashed lines correspond to the RDE thresholds.
Figure 1. Average trip dynamics for the different laboratories: (a) relative positive acceleration (RPA); (b) 95th percentile of velocity per positive acceleration. Both in function of the average speed of urban, rural, and motorway parts. The dashed lines correspond to the RDE thresholds.
Energies 17 01308 g001
Energies 17 01308 g002aEnergies 17 01308 g002b
Figure 2. Average trip results for each laboratory for urban parts (left panels) and total trips (right panels): (a) CO2 urban; (b) CO2 total; (c) NOx (NO2 red, NO blue bars) urban; (d) NOx (NO2 red, NO blue bars) total; (e) PN urban ×1011 #/km; (f) PN total ×1011 #/km. Cold start is included in urban part. Error bars show one standard deviation of two to five repetitions. No PN measurements available for laboratory #3.
Figure 2. Average trip results for each laboratory for urban parts (left panels) and total trips (right panels): (a) CO2 urban; (b) CO2 total; (c) NOx (NO2 red, NO blue bars) urban; (d) NOx (NO2 red, NO blue bars) total; (e) PN urban ×1011 #/km; (f) PN total ×1011 #/km. Cold start is included in urban part. Error bars show one standard deviation of two to five repetitions. No PN measurements available for laboratory #3.
Energies 17 01308 g002aEnergies 17 01308 g002b
Table 1. Overview of key RDE EU regulations additions.
Table 1. Overview of key RDE EU regulations additions.
RegulationEuroMNOxMPNDriftDriving DynamicsOther Changes
2016/427
[20]		6c--NO ≤ 5 ppm
NO2 ≤ 5 ppm		-RDE for type approval
2016/646
[21]		6d-t
6d		1.10
0.50		-
-		SameRPA, v × a
CPEG
2017/1154
[22]		6d-tsame0.50NOx ≤ 5 ppmsameCold start inclusion; order urban, rural, motorway; hybrids; regeneration
2018/1832
[23]		6d0.43sameSamesameRDE also for ISC and surveillance; simplified evaluation method; EFM should cover 75% FS
2023/443
[24]		6e0.100.34NOx ≤ 3 ppmsameTmax,extended 38 °C (Euro 6ebis) (from 35 °C); stricter vehicle conditioning test for cold start test; soaking up to 72 h (from 56), drift correction
CPEG = cumulative positive elevation gain; EFM = exhaust flow meter; EU = European Union; M = margin; FS = full scale; RDE = real driving emissions; RPA = relative positive acceleration.
Table 2. Average ambient conditions of the various trips.
Table 2. Average ambient conditions of the various trips.
ParameterUnits#1a#2#3#4#5#6#7#1b
Temperature°C23.117.311.615.516.98.313.97.0
PressurekPa99.0103.9101.3103.0102.1102.2103.6102.5
Relative humidity-61%61%57%56%60%63%59%45%
Table 3. Average trip characteristics per laboratory. In italics, values that are outside of the Euro 6d RDE 4th package (Regulation (EU) 2018/1832) [23] specifications.
Table 3. Average trip characteristics per laboratory. In italics, values that are outside of the Euro 6d RDE 4th package (Regulation (EU) 2018/1832) [23] specifications.
Parameter (RDE Thresholds)Units#1a#2#3#4#5#6#7#1b
Number of valid trips-23344352
Total trip distancekm7874828882978379
Total trip duration (90–120)min928810910510711310793
Urban distance (>16)km2927303134332931
Rural distance (>16)km2524273027292623
Motorway distance (>16)km2423262822352825
Urban distance share (29–44)%3736363541343539
Rural distance share (23–43)%3233323432293230
Motorway distance share (23–43)%3131313127363431
Urban average speed (15–40) 1,2km/h2931242828292530
Rural average speedkm/h7675767473727577
Motorway average speedkm/h114105114114109106103113
Total trip average speedkm/h5152465046524751
Motorway speed >145 km/h (<3)%00000000
Motorway speed >100 km/h (≥5)min10812128181010
Urban stop time (6–30) 1,3%2416232117172720
Initial altitudem2107532513415075104208
Altitude difference (≤100)m301883710049737
Total CPEG (<1200 m/100 km) 1m615452872262413338801645
Urban CPEG (<1200 m/100 km) 1m628455944246410447840767
Idling events >300 s 4Y/NNNNNNNYN
Initial idling duration (≤15) 5s1981034793181911
Cold start average speed (15–40) 5km/h915182410181622
Cold start max speed (<60) 5km/h4752485654387456
Cold start–stop time (≤90) 5s17574121691173813947
1 added to second package. This requirement has been softened in Regulation (EU) 2023/443 [24], and the test is invalid only if the final RDE emissions are above the limits; 2 it was 15–30 at first package; 3 it was >10; 4 at second package long idling was excluded from analysis. With third package, the trip is void. This requirement has been softened in Regulation (EU) 2023/443 [24], and the test is invalid only if the final RDE emissions are above the limits; 5 added with third package. CPEG = cumulative positive elevation gain.
Table 4. Maximum and 90th percentile of the analyzers’ RDE (wet) concentration signals and span gas ranges. Range gives the maximum measured concentrations at all laboratories.
Table 4. Maximum and 90th percentile of the analyzers’ RDE (wet) concentration signals and span gas ranges. Range gives the maximum measured concentrations at all laboratories.
ParameterPN × 106
(#/cm3)		CO2
(%)		CO
(ppm)		NO
(ppm)		NO2
(ppm)
Maximum3.3–1713.7–14.112,400–26,600830–23005–50
A 90th percentile1.2–3.113.4–13.775–25538–1700–17
Span gas range-14.8–18.07600–16,000870–1940210–510
Table 5. Statistical analysis of the results. “CO2 corr.” refers to correction of Appendix 6 of Regulation (EU) 2017/1151 [25] or Appendix 11 of Regulation (EU) 2023/443 [24]. “Hot” refers to test removing the initial 300 s.
Table 5. Statistical analysis of the results. “CO2 corr.” refers to correction of Appendix 6 of Regulation (EU) 2017/1151 [25] or Appendix 11 of Regulation (EU) 2023/443 [24]. “Hot” refers to test removing the initial 300 s.
PollutantTrip PartMeanSame RouteDifferent RoutesCombined
CO2Urban (cold)164.7 g/km4.9%7.4%8.8%
Urban (hot)160.0 g/km7.1%12.0%13.9%
Total (cold)134.6 g/km2.9%6.9%7.5%
Total (hot)131.9 g/km3.3%7.4%8.1%
COUrban (cold) CO2 corr.81 mg/km34.1%0.0%34.1%
Urban (cold)113 mg/km29.0%4.9%29.4%
Urban (hot)18 mg/km32.4%18.4%37.2%
Total (cold) CO2 corr.70 mg/km22.6%0.0%22.6%
Total (cold)77 mg/km23.8%9.1%25.5%
Total (hot)43 mg/km26.5%4.3%26.9%
NOxUrban (cold) CO2 corr.80 mg/km29.4%1.0%29.4%
Urban (cold)113 mg/km30.6%0.0%30.6%
Urban (hot)104 mg/km26.6%3.0%26.8%
Total (cold) CO2 corr.50 mg/km22.9%0.0%22.9%
Total (cold)55 mg/km23.0%0.0%23.0%
Total (hot)51 mg/km20.0%9.8%22.3%
PN × 1011Urban (cold) CO2 corr.7.3 #/km14.2%1.7%14.3%
Urban (cold)10.2 #/km11.6%7.2%13.7%
Urban (hot)9.3 #/km11.2%6.2%12.8%
Total (cold) CO2 corr.8.5 #/km6.0%4.7%7.6%
Total (cold)9.2 #/km5.8%9.1%10.8%
Total (hot)8.9 #/km6.0%9.2%10.9%
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Giechaskiel, B.; Valverde, V.; Melas, A.; Clairotte, M.; Bonnel, P.; Dilara, P. Comparison of the Real-Driving Emissions (RDE) of a Gasoline Direct Injection (GDI) Vehicle at Different Routes in Europe. Energies 2024, 17, 1308. https://doi.org/10.3390/en17061308

AMA Style

Giechaskiel B, Valverde V, Melas A, Clairotte M, Bonnel P, Dilara P. Comparison of the Real-Driving Emissions (RDE) of a Gasoline Direct Injection (GDI) Vehicle at Different Routes in Europe. Energies. 2024; 17(6):1308. https://doi.org/10.3390/en17061308

Chicago/Turabian Style

Giechaskiel, Barouch, Victor Valverde, Anastasios Melas, Michaël Clairotte, Pierre Bonnel, and Panagiota Dilara. 2024. "Comparison of the Real-Driving Emissions (RDE) of a Gasoline Direct Injection (GDI) Vehicle at Different Routes in Europe" Energies 17, no. 6: 1308. https://doi.org/10.3390/en17061308

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop