When is Enough? Minimum Sample Sizes for On-Road Measurements of Car Emissions

Assessment of identification performance for high emission heavy-duty diesel vehicles by means of remote sensing

To improve the accuracy of detecting high NO (nitric oxide) emissions from heavy-duty diesel vehicles (HDDV) by remote sensing (RS), the emissions of one HDDV complied with China V regulation and one HDDV complied with China VI regulation at constant speeds, with and without after-treatment devices, are tested by a portable emission measurement system (PEMS) and RS. The optimized measurement procedures for detecting high NO emissions from China V and China VI HDDVs by RS are summarized. The correlation of RS and PEMS data shows that the ratio of NO to CO2 (carbon dioxide) is a more appropriate RS measurement than NO concentration alone for identifying high emitters, although NO concentrations of 600 ppm and 100 ppm can be used as a basis for distinguishing between China V and China VI HDDVs, respectively. When the NO/CO2 ratio is >200 × 10-4 and 25 × 10-4, identifying China V and China VI HDDV high emitters, respectively, is possible. Additionally considering the vehicle speed can reduce the high emitter identification error rate, and excluding data where vehicle acceleration is less than -0.1 m/s2 can further improve identification accuracy. Four new high-emitter identification methods based on different combinations of measurements are shown to improve identification efficiency with only small increases in identification error. This study provides evidence to support the future development of high-precision RS methodologies for identifying high-emission vehicles.

Detecting high emitting vehicle subsets using emission remote sensing systems

It is often assumed that a small proportion of a given vehicle fleet produces a disproportionate amount of air pollution emissions. If true, policy actions to target the highly polluting section of the fleet could lead to significant improvements in air quality. In this paper, high-emitter vehicle subsets are defined and their contributions to the total fleet emission are assessed. A new approach, using enrichment factor in cumulative Pareto analysis is proposed for detecting high emitter vehicle subsets within the vehicle fleet. A large dataset (over 94,000 remote-sensing measurements) from five UK-based EDAR (emission detecting and reporting system) field campaigns for the years 2016-17 is used as the test data. In addition to discussions about the high emitter screening criteria, the data analysis procedure and future issues of implementation are discussed. The results show different high emitter trends dependent on the pollutant investigated, and the vehicle type investigated. For example, the analysis indicates that 23 % and 51 % of petrol and diesel cars were responsible for 80 % of NO emissions within that subset of the fleet, respectively. Overall, the contributions of vehicles that account for 80 % of total fleet emissions usually reduce with EURO class improvement, with the subset fleet emissions becoming more homogenous. The high emitter constituent was more noticeable for pollutant PM compared with the other gaseous pollutants, and it was also more prominent for petrol cars when compared to diesel ones.

Data quality evaluation and calibration of on-road remote sensing systems based on exhaust plumes

In recent years, with rapid increases in the number of vehicles in China, the contribution of vehicle exhaust emissions to air pollution has become increasingly prominent. To achieve the precise control of emissions, on-road remote sensing (RS) technology has been developed and applied for law enforcement and supervision. However, data quality is still an existing issue affecting the development and application of RS. In this study, the RS data from a cross-road RS system used at a single site (from 2012 to 2015) were collected, the data screening process was reviewed, the issues with data quality were summarized, a new method of data screening and calibration was proposed, and the effectiveness of the improved data quality control methods was finally evaluated. The results showed that this method reduces the skewness and kurtosis of the data distribution by up to nearly 67%, which restores the actual characteristics of exhaust diffusion and is conducive to the identification of actual clean and high-emission vehicles. The annual variability of emission factors of nitric oxide decreases by 60% - on average - eliminating the annual drift of fleet emissions and improving data reliability.

Quantifying on-road vehicle emissions during traffic congestion using updated emission factors of light-duty gasoline vehicles and real-world traffic monitoring big data

Light-duty gasoline vehicles (LDGVs) have made up >90 % of vehicle fleets in China since 2019, moreover, with a high annual growth rate (> 10 %) since 2017. Hence, accurate estimates of air pollutant emissions of these fast-changing LDGVs are vital for air quality management, human healthcare, and ecological protection. However, this issue is poorly quantified due to insufficient reserves of timely updated LDGV emission factors, which are dependent on real-world activity levels. Here we constructed a big dataset of explicit emission profiles (e.g., emission factors and accumulated mileages) for 159,051 LDGVs based on an official I/M database by matching real-time traffic dynamics via real-world traffic monitoring (e.g., traffic volumes and speeds). Consequently, we provide robust evidence that the emission factors of these LDGVs follow a clear heavy-tailed distribution. The top 10 % emitters contributed >60 % to the total fleet emissions, while the bottom 50 % contributed <10 %. Such emission factors were effectively reduced by 75.7-86.2 % as official emission standards upgraded gradually (i.e., from China 2 to China 5) within 13 years from 2004 to 2017. Nevertheless, such achievements would be offset once traffic congestion occurred. In the real world, the typical traffic congestions (i.e., vehicle speed <5 km/h) can lead to emissions 5- 9 times higher than those on non-congested roads (i.e., vehicle speed >50 km/h). These empirical analyses enabled us to propose future traffic scenarios that could harmonize emission standards and traffic congestion. Practical approaches on vehicle emission controls under realistic conditions are proposed, which would provide new insights for future urban vehicle emission management.

Detecting candidate high NOx emitting light commercial vehicles using vehicle emission remote sensing

Vehicle emission remote sensing devices have been widely used for monitoring and assessing the real-world emission performance of vehicles. They are also well-suited to identify candidate high emitting vehicles as remote sensing surveys measure the on-road, real-driving emissions (RDE) of a high proportion of the operational vehicle fleet passing through a testing site. This study uses the Gumbel distribution to characterize the fuel-specific NO_x emission rates (g·kg^-1) from diesel vans (formally referred to as light commercial vehicles or LCVs) and screen candidate high emitting vehicles. Van emission trends of four European countries (Belgium, Sweden, Switzerland and the UK) from Euro 3 to Euro 6a/b have been studied, and the impact of road grade on candidate Euro 6a/b high-emitters is also evaluated. The measurements of Euro 6a/b fleets from four countries are pooled together, and a consistent 4% of candidate high-emitters are found in both class II and class III Euro 6a/b vans, accounting for an estimated 24% and 21% total NO_x emissions respectively. The pooled four country data is differentiated by vehicle models and manufacture groups. Engine downsizing of Euro 6a/b class II vans is suspected to worsen the emission performance when vehicles are driven under high engine load. The VW Group is found to be the manufacture with cleanest NO_x emission performance in the Euro 6a/b fleets. By distinguishing high-emitters from normally behaving vehicles, a more robust description of fleet behaviour can be provided and high-emitting vehicles targeted for further testing by plume chasing or in an inspection garage. If the vehicle is found to have a faulty, deteriorated or tampered emission after-treatment system, the periodic vehicle inspection safety and environmental performance certificate could be revoked.

Rapid assessments of light-duty gasoline vehicle emissions using on-road remote sensing and machine learning

In-time and accurate assessments of on-road vehicle emissions play a central role in urban air quality and health policymaking. However, official insight is hampered by the Inspection/Maintenance (I/M) procedure conducted in the laboratory annually. It not only has a large gap to real-world situations (e.g., meteorological conditions) but also is incapable of regular supervision. Here we build a unique dataset including 103,831 light-duty gasoline vehicles, in which on-road remote sensing (ORRS) measurements are linked to the I/M records based on the vehicle identification numbers and license plates. On this basis, we develop an ensemble model framework that integrates three machining learning algorithms, including neural network (NN), extreme gradient boosting (XGBoost), and random forest (RF). We demonstrate that this ensemble model could rapidly assess the vehicle-specific emissions (i.e., CO, HC, and NO). In particular, the model performs quite well for the passing vehicles under normal conditions (i.e., lower VSP (<18 kw/t), temperature (6-32 °C), relative humidity (<80%), and wind speed (<5 m/s)). Together with the current emission standard, we identify a large number of the 'dirty' (2.33%) or 'clean' (74.92%) vehicles in the real world. Our results show that the ORRS measurements, assisted by the machine-learning-based ensemble model developed here, can realize day-to-day supervision of on-road vehicle-specific emissions. This approach framework provides a valuable opportunity to reform the I/M procedures globally and mitigate urban air pollution deeply.

Rapid detection of high-emitting vehicles by on-road remote sensing technology improves urban air quality

Vehicle emissions are the most important source of air pollution in the urban environment worldwide, and their detection and control are critical for protecting public health. Here, we report the use of on-road remote sensing (RS) technology for fast, accurate, and cost-effective identification of high-emitting vehicles as an enforcement program for improving urban air quality. Using large emission datasets from chassis dynamometer testing, RS, and air quality monitoring, we found that significant percentages of in-use petrol and LPG vehicles failed the emission standards, particularly the high-mileage fleets. The RS enforcement program greatly cleaned these fleets, in terms of high-emitter percentages, fleet average emissions, roadside and ambient pollutant concentrations, and emission inventory. The challenges of the current enforcement program are conservative setting of cut points, single-lane measurement sites, and lack of application experience in diesel vehicles. Developing more accurate and vertical RS systems will improve and extend their applications.

A novel method for comparing passenger car fleets and identifying high-chance gross emitting vehicles using kerbside remote sensing data

The quantification and comparison of NO_X emission from in-situ car fleets, and identification of the highest emitters is an ongoing challenge. This challenge will become more important as new and increasingly complex emissions removal systems penetrate the market. We combine real-world data with new-to-the-field statistical methods to describe fleet-scale emissions behaviours and identify candidate gross-emitter vehicles. 19,605 passenger cars were observed using a Remote Sensing Device across Aberdeen in 2015. Of these, 736 were Euro 6 Passenger Cars. The distribution of observed pollutant per unit of fuel burnt ratios for most fuel type and Euro standards followed an asymmetrical shape best characterised by the Gumbel distribution. The Gumbel distribution approach was not able to fully replicate the distribution of measurements of petrol or Euro 6 diesel cars due to the presence of a subset of high-emitting outliers, ranging from the 13^th percentile for Euro 3 petrol to the 2^nd percentile for Euro 6 petrol, with Euro 6 diesel having a 5^th percentile outlier value. No outlier fraction was observed for pre-Euro 6 diesels. The off-model fractions resembled Gumbel distributed data and in some cases could be modelled as a separate distribution with the fleet behaving as a superposition of them. It is shown that VSP was not directly linked to this behaviour and it is hypothesised that it is caused by the emissions control systems operating sub-optimally. The reasons for sub-optimal operation are beyond the scope of this paper but may be linked to air-fuel mixture sensors, cold-start running and deterioration of the catalytic converter. Larger data-sets with more Euro 6 passenger cars are required to fully test this. Application of this methodology to larger data sets from more widely deployed remote sensing devices will allow observers to identify potentially problematic vehicles for further investigation into their emission control systems.

Statistical evaluation of on-road vehicle emissions measurement using a dual remote sensing technique

On-road remote sensing (RS) is a rapid, non-intrusive and economical tool to monitor and control the emissions of in-use vehicles, and currently is gaining popularity globally. However, a majority of studies used a single RS technique, which may bias the measurements since RS only captures a snapshot of vehicle emissions. This study aimed to use a unique dual RS technique to assess the characteristics of on-road vehicle emissions. The results show that instantaneous vehicle emissions are highly dynamic under real-world driving conditions. The two emission factors measured by the dual RS technique show little correlation, even under the same driving condition. This indicates that using the single RS technique may be insufficient to accurately represent the emission level of a vehicle based on one measurement. To increase the accuracy of identifying high-emitting vehicles, using the dual RS technique is essential. Despite little correlation, the dual RS technique measures the same average emission factors as the single RS technique does when a large number of measurements are available. Statistical analysis shows that both RS systems demonstrate the same Gamma distribution with ≥200 measurements, leading to converged mean emission factors for a given vehicle group. These findings point to the need for a minimum sample size of 200 RS measurements in order to generate reliable emission factors for on-road vehicles. In summary, this study suggests that using the single or dual RS technique will depend on the purpose of applications. Both techniques have the same accuracy in calculating average emission factors when sufficient measurements are available, while the dual RS technique is more accurate in identifying high-emitters based on one measurement only.

On-Road NOx and Smoke Emissions of Diesel Light Commercial Vehicles-Combining Remote Sensing Measurements from across Europe

Light commercial vehicles (LCVs) account for about 10-15% of road traffic in Europe. There have only been few investigations on their on-road emission performance. Here, on-road remote sensing vehicle emission measurements from 18 locations across four European countries are combined for a comprehensive analysis of NO_x and smoke emission rates from diesel LCV in the past two decades. This allows differentiating the performance by emission standards, model years, curb weights, engine loads, manufacturers, vehicle age, and temperature, as well as by measurement devices. We find a general consistency between devices and countries. On-road NO_x emission rates have been much higher than type approval limit values for all manufacturers, but some perform systematically better than others. Emission rates have gone down only with the introduction of Euro 6a-b emission standards since the year 2015. Smoke emission rates are considered a proxy for particulate emissions. Their emissions have decrease substantially from the year 2010 onward for all countries and size classes measured. This is consistent with the substantial tightening of the particulate matter emission limit value that typically forced the introduction of a diesel particulate filter. The average NO_x emission rate increases with engine load and decreasing ambient temperatures, particularly for Euro 4 and 5 emission classes. This explains to a large extent the differences in the absolute level between the measurement sites together with differences in fleet composition. These dependencies have already been observed earlier with diesel passenger cars; they are considered part of an abnormal emission control strategy. Some limited increase of the NO_x emission rate is observed for Euro 3 vehicles older than 10 years. The strong increase for the youngest Euro 6 LCVs might rather reflect technology advances with successively younger models than genuine deterioration. However, the durability of emission controls for Euro 6 vehicles should be better monitored closely. Smoke emission rates continuously increase with vehicle age, suggesting a deterioration of the after-treatment system with use.

Evaluating mobile monitoring of on-road emission factors by comparing concurrent PEMS measurements

Many countries have adopted portable emissions measurement system (PEMS) testing in their latest regulations to measure real-world vehicular emissions. However, its fleetwide implementation is severely limited by the high equipment costs and lengthy setup procedures, posing a need to develop more cost-effective, efficient emission measurement methods, such as mobile chasing tests. We conducted conjoint PEMS-chasing experiments for twelve heavy-duty diesel vehicles (HDDTs) to evaluate the accuracy of mobile measurement results. Two data processing approaches were integrated to automate the calculations of fuel consumption-based emission factors of nitrogen oxides (NO_X). With a total of 245 plume chasing tests conducted, and then averaged by vehicle and road types, we found that the relative errors of vehicle-specific emission factors using an algorithm developed for this project were within approximately ±20% of the PEMS results for all tested vehicles. Stochastic simulations suggested reasonable results could be obtained using fewer chasing tests per vehicle (e.g., 71% for freeways and 94% for local road, equivalent to two chase tests per vehicle). This study improves the understanding of the accuracy of the mobile chasing method, and provides a practical approach for real-time emission measurements for future scaled-up mobile chasing studies.