Magazines/Newspapers

A review of vehicular emission models

Description
A review of vehicular emission models
Published
of 12
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
   13  –  15 June 2013 Celje, Slovenia The International Conference on Logistics & Sustainable Transport 2013, website: http://iclst.fl.uni-mb.si/ A review of vehicular emission models Katja BAŠKOVIČ 1   and dr. Matjaž KNEZ 1 *   1  Faculty of logistics, University of Maribor, Celje, Slovenia  Abstract   —  This paper reviews the latest literature about vehicular emission models. Vehicular emission models are methods for calculating the level of pollutant emissions, regarding emission factors, average speed, fuel consumption and the amount of traffic on the defined type of road. In the paper we are turning towards urban roads and looking for all types of vehicular emission models and good practice examples. Key words  —  emissions, practice examples, urban roads, vehicular emission models. I.   I NTRODUCTION  All over the world we can see how the climate is changing. After few years when winters were almost snowless, we have snow from the end of October till the middle of the April, regardless we live in continental climate part of Slovenia. Temperatures can decrease from 12°C to –9°C over the night. Waters are flooding in larger areas than usual. Even if we look around the world, nowhere is better. Hurricanes and typhoons change the land, oceans level is rising and the Arctic sea ice is reaching its record melt. All presented facts are the consequence of the enhanced greenhouse effect, because human activities are releasing additional amounts of greenhouse gases (GHGs) into the atmosphere. The most commonly produced GHG is carbon dioxide (CO 2 ). Since the Industrial Revolution the concentration of CO 2  in the atmosphere has increased by around 41% and it rises. One of the main sources of CO 2  is the combustion of fossil fuels to power machines, generate electricity, heat buildings and above all to transport people and goods. Other GHGs are emitted in smaller quantities than CO 2 , but they trap heat far more effectively [1]. Transport is responsible for around a quarter of EU GHG emissions. As can be seen from Fig. 1, this makes it the second biggest GHG emitting sector after energy. While emissions from other sectors are generally falling, those from transport have increased 36% between 1990 and 2007. The EU has policies in place to reduce emissions from a range of modes of transport, such as including aviation in the EU Emissions Trading System (EU ETS) and CO 2  emissions targets for cars [2]. Figure 1: EU27 greenhouse gas emissions by sector and mode of transport, 2007 [3].   13  –  15 June 2013 Celje, Slovenia The International Conference on Logistics & Sustainable Transport 2013, website: http://iclst.fl.uni-mb.si/ Road transport contributes about one-fifth of the EU's total emissions of CO 2 . Its emissions from road transport increased by nearly 23% between 1990 and 2010, and without the economic downturn growth could have been even bigger [4]. Other pollutant emissions emitting in road transport are carbon monoxide (CO), sulfur dioxide (SO 2 ), lead, hydrocarbons (HC), oxides of nitrogen (NO  x ), non-methane volatile compounds (NMVOC) and particulate matter (PM). European Commission (EC) implements different strategy following regulations on reducing CO 2  and other pollutant emissions from vehicles. Their reduction target is under the Kyoto Protocol and beyond [4]. There are many possibilities how to measure, calculate and control the pollutant emission. From many possibilities we peak emission models which are very useful to calculate and show the amount of emissions in selected area. In the following chapter can be seen the descriptions of all covered emission models classified in different groups and the descriptions of selected, after 2002 presented, vehicular emission models. II.   V EHICULAR E MISSION MODELS  Vehicular emission models are methods for calculating the level of pollutant emissions, regarding emission factors, average speed, fuel consumption and the amount of traffic on the defined type of road. There are two different arrangements of vehicular emission models. In reference [5] emission models are primary classified on modelling approaches used when calculating hot, cold start and evaporative emissions. Secondary hot emission models are classified into three main groups of increasing level of complexity: (a)   emission factor models, (b)   average speed models, (c)   modal emission models. In reference [6] emission models are classified according to the input data, the scale of the study and the type of pollutants being considered: (a)   model relying on fuel quantities, (b)   model relying on average traffic volumes per detailed categories of vehicles, (c)   model relying on average speed of the traffic, (d)   model implying detailed description of traffic situation, (e)   model providing the emissions from traffic related variable, (f)   model representing a detailed description of speeds experienced, (g)   model relying on chronological speed (instantaneous model).  A.   Emission model groups descriptions Groups from both classifications can be combined in one list of groups and subgroups. How are they connected can be seen from their following description. 1)   Emission factor model Emission factor models function with a simple calculation method and do not require large amounts of input data. The estimation of the emissions is expressed by the use of an emissions factor related to one type of vehicle and a specific driving mode (i.e. urban, rural or motorway). Emission factors are derived from the mean values of repeated measurements over a particular driving cycle and are usually expressed in mass of pollutant per unit distance. Emission factor models are commonly used in the development of national and regional emission inventories. This approach is not accurate on microscale, regarding the emission factors are based on average driving characteristics [5]. 2)    Average speed model Average speed models are based on speed-related emission functions, generated by the measurements of the emission rates over a variety of trips at different speed levels. These models are often used in emission inventories on a road network scale. Their use on a microscale level is inappropriate, because they usually do not include changes in operational modes [5].   13  –  15 June 2013 Celje, Slovenia The International Conference on Logistics & Sustainable Transport 2013, website: http://iclst.fl.uni-mb.si/ 3)    Modal emission model Modal emission models operate at a higher level of complexity. Modal models based on speed and acceleration present the emission rates as a function of different levels of speed as well as of the various operational modes (i.e. acceleration, deceleration, steadly-speed cruise and idle). They provide more accurate emissions estimation at a microscale level than emission factor and average speed models. Modal models based on speed and acceleration can be specific enough to provide emission levels or fuel consumption, second-by-second, for a particular type of vehicle from a given driving cycle. Modal models based on engine power present emissions as a function of engine demand and other physical parameters related to vehicle operation. This type of model is usually highly complex due to the large amount of data required [5]. 4)    Model relying on fuel quantities Models use fuel consumption data (i.e., fuel sale data) and categories of vehicles. Models can only be used for large-scale inventories [6]. 5)    Model relying on average traffic volumes per detailed categories of vehicles Models use a single emission factor to represent a particular type of vehicle and driving cycle. The emission factors are calculated as mean values of measurements on a number of vehicles over given driving cycles, and are usually stated in terms of the mass of pollutant emitted per vehicle distance of fuel. These factors are mostly used in national and regional emission inventories [6]. 6)    Model relying on average speed of the traffic These models predict average emission factors for a vehicle class that is driven over a number of different driving patterns, which are a function of the mean travelling speed. The total emissions can be calculated by the sum of the exhaust emissions (hot and cold), evaporative emissions and for some models emissions due to road vehicle tires and brakes as well as road wear resulting from the vehicles’ motion. The results of these models may not be very precise, but they cover the major emission processes and most pollutants of interest [6]. 7)    Model implying detailed description of traffic situation These models use discrete emission factors for predefined traffic situations (e.g. stop-and-go, saturated, heavy, free flow). They allow the estimation of pollutants emitted by the hot and cold exhaust emissions as well as fuel evaporation. The methodologies were also developed for small scales (i.e. single street). Traffic situation models require vehicle-kilometre-travelled (VKT) data per driving situation as input. These models provide emissions for a large number of different regulated and non-regulated pollutants [6]. 8)    Model providing the emissions from traffic related variable Emission factors obtained from traffic-variable models are defined by traffic flow variables such as average speed, traffic density, queen length and signal settings. They use a correction of the average speed to assume the effects of the traffic onto pollutant emissions [6]. 9)    Model representing a detailed description of speeds experienced These models are based on tests on a large number of vehicles according to various driving cycles. Within the model, each driving cycle is characterized by a large number of descriptive parameters (e.g. average speed, number of stops per km, kinematics of vehicles). For each pollutant and vehicle category, a regression model is fitted to the average emission values over the different driving cycles. These models require detailed information on the movement of the vehicles (instantaneous speed, acceleration) [6]. 10)    Model relying on chronological speed (instantaneous model) These models represent explicitly the vehicle emission behaviour by relating emission rates to vehicle operation during series of short time steps. In some models, vehicle operation is defined in terms of a relatively small number of modes (i.e. idle, acceleration, deceleration and cruise). For   13  –  15 June 2013 Celje, Slovenia The International Conference on Logistics & Sustainable Transport 2013, website: http://iclst.fl.uni-mb.si/ each of the modes, the emission rate for a given vehicle category and pollutant is fixed and the total emission rate is calculated by weighting each model emission rate by the time spent in each mode. Some instantaneous models relate vehicle engine power, speed and acceleration during a driving cycle. The model estimates only hot running emissions [6]. B.   List of groups and subgroups Regarding presented descriptions we combined all groups in the following list: Emission factor model o   Model relying on fuel quantities, o   Model relying on average traffic volumes per detailed categories of vehicles. Average speed model o   Model relying on average speed of the traffic. Modal emission model o   Model implying detailed description of traffic situation, o   Model providing the emissions from traffic related variable, o   Model representing a detailed description of speeds experienced, o   Model relying on chronological speed (instantaneous model). C.   Description of vehicular emission models In the following paragraphs examples of new, after 2002 presented, vehicular emission models are described. 1)    Microscale Emission Model POLY Microscale emission models [7] were developed due to the difficulty of presenting acceleration or deceleration in a macroscale emission model. These models can estimate second-by-second emissions. Microscale emission models  e i,j,k,m (t) where developed for each emission type  m  (e.g. CO, HC or NO  x ) and light-duty vehicle categories (vehicle size ( i ), model year (  j ), emitter type ( k )) by adopting least-square regression. The proportion  P i,j,k  for each category ( i, j, k ) was derived based on the national level of vehicle distributions. The emission of type m,  that is produced by a vehicle with size i  at time t , e i,m (t)  can be estimated as (1) Vehicles were classified into categories by size: light-duty gasoline vehicle (LDGV) (i.e., passenger cars), light-duty gasoline trucks under 6,000 lbs. gross vehicle weight (LDGT1) and light-duty gasoline trucks 6,000 lbs. to 8,500 lbs. gross vehicle weight (LDGT2); by model year (e.g. before 1975, 1975-1980, 1981-1986, 1987-1990, 1991-1993, 1994-1997 for LDGV) and into five different emitter type groups: one normal emitter and four high-emitter types. Result of classification is 41 groups of vehicles made before 1997. The type m  emission rate (g/s) for vehicle group ( i, j, k ) at time t , e i,j,k,m (t) in modelling is represented as (2) where three factors were taken into account: tractive power, grade and time dependence. Tractive power is represented by using variables W(t), V(t), V   2 (t)  and V  3 (t) . W(t)  is related to kinetic power, which is a part of tractive power: (3) where  M is the vehicle mass with appropriate inertial correction for rotating and reciprocating parts (kg); g 0 = gravitational constant (9.81 m/s); φ  is the grade of roadway; and a(t) is the acceleration or deceleration rate at time t . The effect of grade of a roadway,  A(t)  (i.e. combined acceleration or deceleration rate) is formulated as (4)   13  –  15 June 2013 Celje, Slovenia The International Conference on Logistics & Sustainable Transport 2013, website: http://iclst.fl.uni-mb.si/ where a(t)  is a vehicle’s actual acceleration or deceleration rate; and g(t)  represents the grade (percentage) of the roadway segment where a vehicle is travelling at time t.  Time dependence in emissions response to vehicle operation (e.g. the use of timer to delay command enrichment or oxygen storage in the catalytic converter) was taken into account in this project by employing time-series variables. Researcher figured out that the acceleration or deceleration in the preceding time periods, not in current time, has the most obvious impact on the emissions at time t.  Patterns of impact are not the same for different emission types. To consider the impact in model variables of combined accelerations or decelerations rates in the current and past periods (i.e.  A(t),…,A(t -9) ) were used. There are also two variables used in (2), T’(t)  to represent the duration of acceleration and T’’(t)  to represent the duration of deceleration. At specific point in time, only one value can be greater than zero. The emission model (2) was validated through the root mean squared error (RMSE), the aggregated total prediction error (ATPE) and the correlation coefficient (R) between the predicted and actual emissions. Model POLY (1) was compared with the model CMEM and the emissions model Integration that was adopted in the microscopic traffic simulation model. Because the emission models developed in this project were based on the data set of the Federal Test Procedure (FTP) cycle, researchers used second-by-second emissions data of the modal emission cycle (MEC) and the US06 cycle. From results of comparisons through the RMSEs, ATPEs and correlation coefficient it can be observed that the POLY models has the same trend of change as the measured emissions and predicts more accurately than the other two models most of the time. 2)    Microscale Emission Factor Model for Particulate Matter (MicroFacPM) A microscale emission factor model for PM (MicroFacPM) [8] for predicting real-world real-time motor vehicle emissions for total suspended PM (TSP), PM less than 10μm aerodynamic diameter (PM 10 ) and PM less than 2.5μm aerodynamic diameter (PM 2.5 ) has been developed to support studies about human exposure near roadways and inside vehicles travelling along the roadways. The research was funded by Environmental Protection Agency (EPA). The model is written in FORTRAN 90 for calculating emission factors from vehicular traffic in United States. It uses modelling concepts and structure similar that is used in the development of a microscale emission factor model for CO (MicroFacCO). MicroFacPM uses available information concerning the vehicle fleet composition. General models structure of the MicroFacPM can be seen from Fig. 2. Figure 2: The MicroFacPM general model structure
Search
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks