A machine learning framework for predicting fuel consumption and CO2 emissions in hybrid and combustion vehicles: comparative analysis and performance evaluation.

Our way of life is on a collision course with geological limitations. Ever since petroleum geologist M. King Hubbard correctly predicted in l956 that U.S. oil production would reach a peak in l973 and then decline (1), scientists and engineers have known that worldwide oil production would follow a similar trend. Today, the only question is when the world peak will occur.The U.S. transportation system depends almost entirely (∼97%) on oil (2), and foreign imports have risen steadily since l973 as the demand increased and domestic supplies decreased. Today, more than 60% of U.S. oil consumption is imported and the dependence on foreign oil is bound to increase. There is no question that once the world peak is reached and oil production begins to drop, either alternative fuels will have to be supplied to make up the difference between demand and supply, or the cost of fuel will increase precipitously and create an unprecedented social and economic crisis for our entire transportation system.Among energy analysts the above scenario is not in dispute. There is, however, uncertainty about the timing. Bartlett (3) has developed a predictive model based on a Gaussian curve similar in shape to the data used by Hubbard as shown in Fig. 1. The predictive peak in world oil production depends only on the assumed total amount of recoverable reserves. According to a recent analysis by the Energy Information Agency (4), world ultimately recoverable oil reserves are between 2.2×1012 barrels (bbl) and 3.9×1012bbl with a mean estimate of the USGS at 3×1012bbl. But changing the total available reserve from 3×1012bbl to 4×1012bbl increases the predicted time of peak production by merely 11yr, from 2019 to 2030. The present trend of yearly increases in oil consumption, especially in China and India, shortens the window of opportunity for a managed transition to alternative fuels even further. Hence, irrespective of the actual amount of oil remaining in the ground, peak production will occur soon and the need for starting to supplement oil as the primary transportation fuel is urgent because an orderly transition to develop petroleum substitutes will take time and careful planning.Some analysts claim that hydrogen can take the place of petroleum in a future transportation system (56). But in previous publications, the authors have shown that hydrogen is inferior as an energy carrier to electricity (7) and that the energy efficiency of hydrogen vehicles, especially if the hydrogen were produced by the electrolysis of water, is considerably less than the efficiency of hybrid electric vehicles or fully electric battery vehicles (7). The results of these analyses have subsequently been confirmed by other studies, particularly those by Hammererschlag and Mazza (8) and Mazza and Hammerschlag (9).Before hydrogen could become a useful automotive fuel, an entirely new system of energy production and distribution on twice the scale of today’s electric power generating stations and distribution grid would have to be built. It has been estimated that a hydrogen transmission and storage system to fuel only 50% of the automotive fleet by the year 2020 would cost at least $600 billion (10) and that to make the hydrogen by electrolysis would require doubling the electric power generation rate (11). There is no question that a paradigm shift in fuel for worldwide transportation is imperative, and before embarking on such a huge investment, it is prudent to compare the hydrogen option with alternative ways to provide the energy and/or fuel needed by the transportation system.This paper presents and analyzes two generic approaches to meet the future demand of the U.S. ground transportation systems that do not require hydrogen, can use existing transmission infrastructure, and can eventually reduce CO2 emission drastically with a renewable energy system. Both these pathways are examined from an energetic and environmental perspective and are shown to be superior to the hydrogen economy on both these criteria. The first approach is a demand-side strategy based on the use of electric hybrid vehicles, an energy-efficient vehicle configuration, combined with a liquid fuel. This approach could use the existing liquid-fuel distribution system, but would need an expanded and robust electric-transmission system, albeit on a smaller and much more economical scale than a hydrogen fuel-cell infrastructure. The second approach is a supply-side strategy, based on synthetic fuel generation that can use initially coal or natural gas as the energy source, but can eventually transition to renewable biomass sources. The two pathways are not mutually exclusive, but can be combined into a secure and efficient future transportation system as will be shown in this paper.Cradle-to-grave energy efficiency is an important criterion for comparing energy-source utilization pathways because if a pathway is less efficient than another pathway that accomplishes the same final goal from the same amount of primary energy, then the less efficient pathway requires more primary energy to accomplish the same end. Hence, if the primary energy source is nonrenewable, then the less efficient pathway leaves less of the energy source for the future. It also means that more pollution is produced and the cost for the final end use is likely higher. However, if the primary energy source is renewable, then the efficiency does not change the amount of primary energy available in the future and energy efficiency does not have the same significance for renewable energy sources as for nonrenewable sources. Efficiency is, of course, important because the cost of delivering the energy is usually strongly influenced by the system efficiency. But a comparison between renewable and nonrenewable pathways should be based on economic and environmental criteria, such as cost and CO2 generation.In order to demonstrate the urgency for initiating a plan to supplement oil as soon as possible, we have made calculations to predict the potential gasoline savings based on the very optimistic scenario that, at an arbitrary starting time, all new light vehicles sold in the U.S. would be either hybrid or electric vehicles. The term “light vehicles” as used here includes all automobiles, family vans, sports utility vehicles, motorcycles, and pickup trucks. This scenario is an extreme case to show that because of the slow turnover of the light-vehicle fleet, it takes a long time for a significant impact on gasoline consumption to occur. The following cases are considered: (i) All new vehicles sold are gasoline-electric hybrid vehicles (HEV); (ii) all new vehicles sold are plug-in, gasoline-electric hybrids with a 20mil electric-only range (PHEV20); (iii) all new vehicles are diesel-electric hybrids (DHEV) with diesel fuel from coal or biomass; (iv) all new vehicles are plug-in, diesel hybrids with a 20mil all-electric range (PDHEV20); or (v) all new vehicles are all-electric vehicles (EV).The calculations use a rate of new vehicle sales of 7% of the fleet per year, a retirement rate of 5%/y, and a resulting net increase in total vehicles of 2%/y. These numbers represent an approximate fit to the light-vehicle sales and total number data for the years 1966 to 2003 reported by the U.S. government (12). All calculated results are presented in percentages and are therefore independent of the time at which all new vehicle sales switch to hybrids or EVs. When new car sales begin to be all hybrids or all EVs, it is assumed that the future rate of retirement of vehicles from the all-gasoline fleet is 5%/y of the remaining gasoline vehicles. The all-gasoline fleet is therefore completely retired 20 years later. The yearly rate of retirement of hybrid or EV vehicles is then 5% of the total number of vehicles at the beginning of that year, less 5% of the number of gasoline vehicles at the beginning of year zero. Thus, in year zero, no hybrid or EVs are retired.The following average vehicle mileage values were used: gasoline fleet, 21mpg (miles per gallon); gasoline HEV, 41mpg; gasoline PHEV 20, 56mpg of gasoline (13). A mileage is not needed for the EVs, or the diesels, since neither use gasoline, and we assume that the diesel fuel will be derived from nonpetroleum sources, as discussed in Secs. 34.The results of these calculations are presented in Figs. 234. Figure 2 shows the ratio of the total number of vehicles in the fleet, the number of all-gasoline vehicles in the fleet, and the number of hybrid or EV vehicles in the fleet to the total number in the fleet as a function of time. The total number of vehicles increases by over 60% in 25 years at the assumed 2%/y net increase while the number of all-gasoline vehicles decreases linearly from 100% initially to 0% after 20y. The number of hybrid or EV vehicles increases from 0% initially to 58% in 10y and 100% in 20y. This graph emphasizes how long it takes for the introduction of a new vehicle type to show a significant impact on the composition of the vehicle fleet, even when only the new vehicle types are sold after a starting point. This slow turnover of the fleet is the fundamental reason that the effects on gasoline consumption show up so slowly.Figure 3 shows the annual reduction in gasoline consumption as a function of time. Note that for HEVs the annual savings in gas consumption is 29% of the gasoline consumption for a conventional fleet in the tenth year and becomes constant at 49% in the twentieth year. Figure 3 also shows that the plug-in gasoline hybrid scenario saves 41% of the usage in the tenth year and increasing to 64% in the twentieth year and thereafter. Clearly, 10y after starting to sell only hybrid or EV vehicles, the impact of the HEV or PHEV20 scenarios on gasoline consumption is still rather small. After 20yr, the impact becomes significant, but gasoline consumption still remains high for gasoline hybrids. The total number of vehicles and the consumption (with the assumption of no efficiency improvement) by an all-gasoline fleet will have increased by more than 60%, but even the PHEV20 savings is only 40% of the zero-time annual-rate of gasoline consumption. The DHEV, DPHEV20, and EV scenarios show 59% annual savings in the tenth year and 100% in the twentieth year and thereafter. As would be expected, the nongasoline vehicles have a much greater impact on gasoline usage than gasoline-using HEVs, and the impact occurs more rapidly.Figure 4 gives the cumulative gasoline savings for the various scenarios compared to an all-gasoline fleet. HEVs save cumulatively 16% after 10yr and 20% after 20 years. Because of the cumulative savings, HEVs would use in 28yr the same amount of gasoline as an all-gasoline fleet would use in 20yr. PHEV20s save 21% after 10yr and 38% after 20yr. These results emphasize the relatively small effect on gasoline consumption that these highly optimistic scenarios have in the first decade after implementation. DHEVs, DPHEV20s, and EVs, the options without any gasoline use, save cumulatively as much as 32% after 10yr and 59% after 20yr.A 2004 report of the Committee on Alternatives and Strategies for Future Hydrogen Production and Use (14), prepared under the auspices of the National Research Council (NRC), concluded that the vision of a hydrogen economy is based on the expectation that hydrogen can be produced from domestic energy sources in a manner that is “both affordable and environmentally benign.” An analysis of currently available technologies for achieving this goal (7) showed that irrespective of whether fossil fuels, nuclear fuels or renewable technologies are used as the primary energy source, hydrogen is inefficient compared to using the electric power or heat from any of these sources directly. Given these facts, it is important to note that the NRC report also stated that “If battery technology improves dramatically, all-electric vehicles might become the preferred alternative (to fuel cell electric vehicles).” The report also noted that “Hybrid vehicle technology is commercially available today and can therefore be realized immediately.” If synthetic fuels made from coal, natural gas, or biomass were used in place of gasoline in hybrid vehicles, the consumption of oil could be reduced immediately and eventually eliminated. In the light of these observations, it is therefore important to examine what the current state of battery technology is, what can be expected in the near future, and how these developments affect the potential of hybrid vehicle performance and economics.To assess the performance of a battery for electric vehicles, the following characteristics have to be considered: Specific energy, a measure of the battery weight in units of watt hours per kilogramEnergy density, a measure of the space the battery occupies in watt hours per cubic meterCapacity, the total quantity of energy a battery can store and later deliver in watt hoursEfficiency, the ratio of energy that can be extracted from the battery to the initial energy input to change the batterySpecific power, the rate at which the battery can deliver the stored energy per unit weight of battery in watts per kilogramBattery lifecycle, the number of charge and discharge cycles that a battery can sustain during its lifeA significant effort to replace oil as a transportation fuel was undertaken ten years ago in California, when the California Air Resources Board [CARB] mandated that a certain percentage of all vehicles sold in California had to have zero tailpipe emissions (15). At that time the only technology available to meet the mandate was the all battery electric vehicle [BEV], which required no gasoline for its operation. The experiment to mandate the use of BEVs in California failed because the technology was not ready for commercialization. The best battery available in 1995 (fluted-tubular lead acid) had an energy storage density of 35Wh∕kg, a specific power of 100W∕kg, and a life cycle of 600-1000cycles. With these battery characteristics, the maximum range of a BEV was only 50mil, and the battery pack required replacement every 25,000mil at a cost of between $7000 and $8000 for an average BEV (16). Since that time, new batteries have been developed by Panasonic, VARTA, and SAFT, that have twice the energy-storage density, three times the specific power, and two or three times the cycle life of the lead acid batteries sold in California, as shown in Table 1 (13).In addition to the advanced batteries, a new concept has been developed that combines the best qualities of hybrid and battery vehicle technologies. This “plug-in hybrid vehicle” can recharge vehicle batteries during off-peak hours, and since most cars are parked 90% of the time, there are plenty of charging opportunities at both home and the workplace. Furthermore, a large portion of the electric generation infrastructure is only needed for peak demands and lays idle much of the time. Hence, if charging automobile batteries occurred during off-peak hours, they would level out the load of the electric production system and reduce the average cost of electricity (17). Moreover, plug-in hybrid vehicles are not range limited because they have an engine that can refuel at existing gas stations to use when the batteries are low.The efficiency of a PHEV depends on the number of miles the vehicle travels on liquid fuel and electricity, respectively, as well as on the efficiency of the prime movers according to1η=energytowheelsenergyfromprimarysource=f1η1η2+f2η3η4where η1 is the efficiency of the primary source of electricity, η2 is the efficiency of transmitting electricity to the wheels, f1 is the fraction of energy supplied by electricity, f2 is the fraction of energy supplied by fuel =(1−f1), η3 is the efficiency of primary source to fuel, and η4 is the efficiency of fuel to wheels.PHEVs can be designed with different all-electric ranges. The distance, in miles, that a PHEV can travel on batteries alone is denoted by a number after PHEV. Thus, a PHEV20 can travel 20mil on fully charged batteries without using the gasoline engine. According to a study by EPRI (13), on average 1/3 of the annual mileage of a PHEV20 is supplied by electricity and 2/3 by gasoline. The percentage depends, of course, on the vehicle design and the capacity of the batteries on the vehicle. A PHEV60 can travel 60mil on batteries alone, and the percentage of electric miles will be greater as will the battery capacity.The tank-to-wheel (more appropriately, battery-to-wheel) efficiency for a battery all-electric vehicle according to EPRI (13) is 0.82. In a previous analysis by the authors (18), the efficiency in 1993 was only 0.49. Comparing these results shows the enormous improvements in the electric component efficiency (controller 87%, battery 90%, charger 90%, drivetrain 90%;). When these numbers are multiplied by a hybrid-weight-times-idle factor of 1.3 (19), the overall efficiency of an electric hybrid is 82%, the same as that used in the EPRI study (13). It is important to note that currently all-electric vehicles can be nearly twice as efficient as when (18) was published.Given the potentials for plug-in hybrid vehicles, the Electric Power Research Institute (13) conducted a large-scale analysis of the cost, the battery requirements, and the economic competitiveness of plug in vehicles today and within the near term future. Table 2 presents the net present value of life-cycle costs over ten years for a midsized combustion vehicle [CV], hybrid vehicle [HEV] and a plug in electric vehicle with a 20mil electric-only range [PHEV20]. The battery module cost in dollars per kilowatt is the cost at which the total life-cycle costs of all three vehicles would be the Figure presents cost for battery as a function of number of units produced per year. According to this a production of about units per year units would the cost reduction to make both hybrid electric vehicles and plug in electric vehicles 3 presents the electric and plug-in hybrid vehicle battery that would be to make electric vehicles cost for vehicles according to EPRI (13). As shown in Table the characteristics of batteries, and batteries are to meet the required cost and performance The battery characteristics shown in Table 1 and Fig. are years and it is likely that more from would show Furthermore, the EPRI study assumed a current gasoline cost of A of the analysis based on a gasoline cost of that the battery at which the net present values of conventional combustion vehicles and battery vehicles are would up from to for an HEV and from to for a PHEV Figure shows the cost for batteries production for Hence, it that the cost of HEVs and with available batteries is with that of engine The EPRI analysis is because it compared the performance of all battery electric and plug in hybrid vehicles only to currently available combustion as shown in the use of diesel in a hybrid would increase the efficiency of compared to a hybrid with engine and the amount of fuel Hence, it be concluded that the EPRI analysis is it includes advanced batteries, it does not the increased efficiency by using diesel of combustion Furthermore, diesel fuel, as will be shown in can be produced from coal or renewable sources as can the electric power required for charging the The introduction of to the energy is the of this it is and can be as renewable technologies become more cost and fossil fuels more natural gas and biomass can be into liquid the most fossil fuel in the is used almost to In order to make coal into a vehicle fuel, it first be to a gas by a of The of this then be to of that can be used as vehicle fuel. biomass and natural gas can be used of coal or combined with coal to make these and are discussed gas can be used as a vehicle fuel, or it can be with to make gas, which can be used to fuels in the same manner as for The technology is well developed as shown by the recent of of which will natural gas, which is currently to liquid fuel. These and a in of which is diesel in With a with an estimated billion and a diesel with the of with an estimated at The of natural gas to make vehicle fuels was discussed in an paper by the authors (18), and of those results are presented later for comparison with coal as the fuel It should also be noted that biomass can be either alone or in with coal and to liquid fuels by the same as coal, or it can also be and then into vehicle fuels as in is a that is a in the production of synthetic liquid fuels from coal for transportation The coal is shown in Fig. It a such as coal or with to and This gas can be to hydrogen or to make or can be used as a transportation fuel in but this study on diesel fuel because are more the first of the coal is with limited to and The in the coal is to hydrogen gas, and are as In the shift is with to and The and hydrogen are from the and to the or into The that is in this is from the in a for Thus, it can be from the and are the costs when liquid fuel is produced from The estimated time of for a is to years. The depends on the production capacity of the the cost of a with a capacity to barrels of liquid fuel per is estimated to be of the order billion of coal claim that there will be gas pollution from the However, in the future vehicle emissions of can be reduced those of vehicles, by the use of plug-in hybrid electric vehicles and by of the from the fuel production is a synthetic diesel fuel that can be made from coal by of The is first to make which can then be to The is to the and the gas is to electricity for the as shown in Fig. is a gas at but can be under and then can be to other liquid of make it an fuel for It is similar to but has a number The number to the of a fuel to With combustion of the fuel occurs after and emissions are as a of combustion The combustion also in by the need for to the shown in Fig. coal into liquid fuel. The was by scientists before and is used today in by to make diesel fuel gas to make a liquid fuel of synthetic diesel fuel, which is similar to and which is used to make synthetic gasoline (7). The is from the liquid diesel and to the The gas resulting from is to electricity for the can be made from coal by by gas After the hydrogen gas and are from the gas, and hydrogen are The hydrogen can be stored and the can be for electricity and/or to the shift as shown in Fig. store and the hydrogen, it is either to it to or to it at a The efficiency of the first option is while the second is efficient (7). Both and hydrogen have been for fuel storage in a of hydrogen fuel-cell vehicles is in the of coal or natural gas into a vehicle fuel. The energy efficiency of these is important in the overall well to efficiency of these alternative Table 4 presents or efficiency for various fuels from coal or natural and have reported the and energy for with of the and values are used (18) presented for natural gas without and estimated that of CO2 the efficiency of by about two percentage Since natural gas only about as much per unit of energy as coal, it has been assumed that will reduce the efficiency of to fuels by percentage point. Thus, percentage has been from values reported by (18) to the values shown in Table In the of data for the of natural gas to the authors assumed that the ratio of the for natural gas is the same as that for coal to estimate this efficiency as shown in Table 4 that the production of liquid fuels from natural gas is more efficient than from But is in and the technology is not a It is however, for the that is currently into the in gasoline The of of these has been But production is the more for the term and does not require hydrogen as a fuel or energy Today, the of fuel from coal, at the only in The of supplies of such fuels as gasoline, and The economic and of coal have been U.S. and for a fuel in using technology and are to the that will have a capacity to of diesel fuel. has in recent NRC study other technologies that could synthetic fuels from biomass and presents a comparison of the energy on energy for production from and These significant in synthetic But the for synthetic fuel production need to be multiplied before synthetic fuels can make up for the between demand and of gasoline after the peak in oil production is on the analysis presented in this we the following and oil production is expected to peak within the and as is liquid fuel are expected to increase This could lead to a crisis in the U.S. transportation system that on 60% of which is options for a transportation crisis by and/or liquid fuels derived from petroleum with synthetic fuels from natural gas, or coal and by demand by increasing the efficiency and mileage of options to have impact they be at least before hybrid vehicles are a option to reduce the liquid fuel consumption of future transportation hybrid vehicles can the existing infrastructure for electric power transmission by charging batteries during peak hours and use liquid fuels only for a fraction of overall power hybrid vehicles can diesel that can be by synthetic fuels derived from coal, natural gas, or use efficiency is increased efficiency alone will not be to the transportation without the production of large of synthetic liquid number of technologies for synthetic diesel that can be used in diesel and reduce emission of that lead to scale of effort required to provide synthetic fuels will require years to and should therefore be as soon as hybrid or all-electric vehicles with available battery technology in an are compared to gasoline of the of the transportation it is that be by government such as for the of synthetic fuels and CO2 high liquid fuel mileage for automobiles, and for efficient plug-in hybrid scenario in this paper for a secure transportation system can be immediately with available technologies and without hydrogen or authors to for as of an independent study for the of at the of

Securing Platinum-Group Metals for Transport Low-Carbon Transition

Author(s): Muehlegger, Erich; Rapson, David | Abstract: This research project explores the plug-in electric vehicle (PEV) market, including both Battery Electric Vehicles (BEVs) and Plug-in Hybrid Electric Vehicles (PHEVs), and the sociodemographic characteristics of purchasing households. The authors use detailed micro-level data on PEV purchase records to answer two primary research questions. Their results confirm that low-income households exhibit a lower share of PEV purchases than they do for conventional, internal combustion engine (ICE) vehicles. Households with annual income less than $50,000 comprise 33 percent of ICE purchases and only 14 percent of PEVS. By comparison, high-income households earning more than $150,000 annually comprise only 12 percent of ICE purchases and 35 percent of PEV purchases over their sample period. Similarly, unsurprising patterns can be seen across ethnicities. For example, non-Hispanic Whites represent 41 percent of ICE purchases but 55 percent of PEV purchases, as compared to Hispanics (38 percent of ICE and 10 percent of PEVs) and African Americans (3 percent of ICEs and 2 percent of PEVS). These differences naturally raise questions about barriers to PEV adoption among low-income and minority ethnic populations. By comparing outcomes in the ICE, hybrid, and PEV markets across income and ethnic groups, the authors are able to test whether price discrimination and barriers to market access are higher in PEV markets for low-income and minority ethnic groups. The authors find that, overall, they are not, although there are mixed results for the used PEV market. In general, non-white, low-income populations face higher prices in the used PEV market, relative to a baseline, than they do in the new PEV market. While some people travel farther to buy used PEVs than they do to buy used ICE vehicles, there is not a pattern that would indicate systematic discrimination (e.g. Hispanics travel farther to buy used PHEVs but less far to buy used BEVs). While the authors admit that their empirical approach cannot control for all potential vehicle composition effects, the authors view their results as being most consistent with a market that provides access to all ethnicities and income groups.View the NCST Project Webpage

Greenhouse gas emissions of conventional and alternative vehicles: Predictions based on energy policy analysis in South Korea

This study examines how driving patterns (distance and conditions) and the electricity generation supply interact to impact well-to-wheel (WTW) energy use and greenhouse gas (GHG) emissions of plug-in hybrid electric vehicles (PHEVs). The WTW performance of a PHEV is compared with that of a similar (nonplug-in) gasoline hybrid electric vehicle and internal combustion engine vehicle (ICEV). Driving PHEVs for short distances between recharging generally results in lower WTW total and fossil energy use and GHG emissions per kilometer compared to driving long distances, but the extent of the reductions depends on the electricity supply. For example, the shortest driving pattern in this study with hydroelectricity uses 81% less fossil energy than the longest driving pattern. However, the shortest driving pattern with coal-based electricity uses only 28% less fossil energy. Similar trends are observed in reductions relative to the nonplug-in vehicles. Irrespective of the electricity supply, PHEVs result in greater reductions in WTW energy use and GHG emissions relative to ICEVs for city than highway driving conditions. PHEVs charging from coal facilities only reduce WTW energy use and GHG emissions relative to ICEVs for certain favorable driving conditions. The study results have implications for environmentally beneficial PHEV adoption and usage patterns.

Problem. The growing demand for sustainable and energy-efficient transportation has intensified interest in plug-in hybrid electric vehicles as an effective alternative to conventional internal combustion engine vehicles. However, the efficiency and flexibility of these systems are significantly constrained by traditional energy management strategies, which lack adaptability to real-world dynamic driving conditions, road scenarios, and individual driving styles. The need for real-time, predictive, and intelligent control algorithms has become critical to optimize energy use, reduce fuel consumption, and ensure the stability of hybrid propulsion systems. Goal. The objective of this study is to analyze and systematize state-of-the-art machine learning methods for optimizing energy management systems in plug-in hybrid electric vehicles, to identify the key technical challenges and to outline the prospects for the development of intelligent, adaptive energy management systems capable of real-time decision-making in uncertain and variable environments. Methodology. This research is based on a comprehensive review of existing literature and technical implementations, focusing on the use of deep learning, reinforcement learning, and predictive control approaches within energy management systems for plug-in hybrid electric vehicles. The study investigates architectural configurations of plug-in hybrid electric vehicles powertrains, their operational modes, functional requirements of energy management systems, and evaluates machine learning algorithms for power distribution optimization, battery health monitoring, predictive energy control, and personalized driving strategies. Results. The analysis confirms that machine learning-enhanced energy management systems can significantly improve fuel economy, adaptability, and operational reliability under varying road and climatic conditions. Reinforcement learning methods enable continuous policy improvement, predictive models allow proactive energy flow planning, and modular energy management systems architectures enhance system scalability and fault tolerance. The integration of machine learning also facilitates fault detection, battery degradation prediction, and utilization of alternative energy sources such as solar or vibrational energy. These solutions collectively contribute to more intelligent and efficient energy management in modern hybrid vehicles. Originality. This study offers a structured classification of machine learning applications in energy management systems for plug-in hybrid electric vehicles, substantiates the advantages of modular control system architectures, and introduces a novel perspective on incorporating behavioral analysis of driver style into energy management systems strategy personalization. It also highlights underexplored areas such as the use of environmental energy inputs (solar, piezoelectric) and outlines methodological considerations for their integration using predictive analytics. Practical value. The findings provide a foundation for the design of next-generation energy management systems in hybrid vehicles, contributing to enhanced fuel efficiency, extended battery life, and improved environmental performance. The proposed insights can be used by researchers and developers to implement intelligent control systems, reduce development time, and align plug-in hybrid electric vehicle design with smart mobility and sustainability goals.

Assessing the life-cycle benefits of vehicle lightweighting requires a quantitative description of mass-induced fuel consumption (MIF) and fuel reduction values (FRVs). We have extended our physics-based model of MIF and FRVs for internal combustion engine vehicles (ICEVs) to electrified vehicles (EVs) including hybrid electric vehicles (HEVs), plug-in hybrid electric vehicles (PHEVs), and battery electric vehicles (BEVs). We illustrate the utility of the model by calculating MIFs and FRVs for 37 EVs and 13 ICEVs. BEVs have much smaller MIF and FRVs, both in the range 0.04-0.07 Le/(100 km 100 kg), than those for ICEVs which are in the ranges 0.19-0.32 and 0.16-0.22 L/(100 km 100 kg), respectively. The MIF and FRVs for HEVs and PHEVs mostly lie between those for ICEVs and BEVs. Powertrain resizing increases the FRVs for ICEVs, HEVs and PHEVs. Lightweighting EVs is less effective in reducing greenhouse gas emissions than lightweighting ICEVs, however the benefits differ substantially for different vehicle models. The physics-based approach outlined here enables model specific assessments for ICEVs, HEVs, PHEVs, and BEVs required to determine the optimal strategy for maximizing the life-cycle benefits of lightweighting the light-duty vehicle fleet.

Well-to-wheel greenhouse gas emissions of electric versus combustion vehicles from 2018 to 2030 in the US

Summary form only given. Since 2000 the price of oil has increased more than two times. Air pollution also has increased. In particular, the increasing emission of CO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> has had a negative influence on the composition of the earth's atmosphere, and it has presumably contributed to the climate changes. Now about 25% of total industrial CO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> emission is produced by vehicles. The world reservoirs of oil are exploited extremely. The oil exploitation peak was about one year ago. Recent prediction assumes oil in known today shape will be exhausted in approximately the next forty years. There are reasons the road transport electrification necessity. The growing market success of hybrid vehicles is achieved as important part of green electric transportation system. In 2020 about 15% of cars total number hybrid and electric vehicles on world roads is expected. This means hybrid engineering and technology permanent improvement is necessary to start now. This presentation concerns two items: Firstly; electric and plug in hybrid vehicles power train energy requirements respectively electricity production and necessity proper infrastructure building. As a temporary solution Plug in Hybrid Electric Vehicles (PHEV)and Battery Powered Electric Vehicles (EV) application adopted to selected urban traffic conditions is depicted. The development of alternative electric energy sources as a element of global PHEV and EV transportation system is emphasized. Secondary; design by modeling and simulations PHEV power trains is considering. The supporting by new solutions automatic transmission traction electric motor operation, permits increasing the entire power train efficiency, causing minimizing energy - fuel and electricity - consumption. Different control strategies are analyzed and compared.

In this study, the impact of improvements and modifications to the chassis framework components, referred to as the Body-in-White (BIW), during the transition from internal combustion engine (ICE) vehicles to electric vehicles (EVs) is examined. Additionally, the material requirements and structural differences compared to ICE vehicles are analyzed. In this context, the literature has been reviewed and presented in a structured flow. The battery requirement of EVs emerges as a factor that increases weight compared to internal combustion engine vehicles. Although advancements in battery technologies have improved the maximum driving range of vehicles, they remain insufficient on their own, necessitating additional efforts towards vehicle lightweighting. The integration of these lightweighting efforts with new technologies has paved the way for the development of new production methods and assembly techniques. Studies examining the compliance of evolving vehicle structures with safety standards, as well as the impact of weight reduction on vehicle emissions, highlight the necessity of addressing this transformation holistically. Therefore, this study investigates the body-in-white structures of vehicles produced by various manufacturers, closely analyzing the changes in materials, weight reduction, and safety considerations during the ICE to EV transition process. Furthermore, the new production methods—such as pressing, welding, and assembly technologies—that companies have integrated into their mass production lines to contribute positively to this transition process and weight reduction have become another focal point of research. These innovations in part manufacturing methods have also played a significant role in the evolution of the body-in-white concept during the ICE to EV transition.

This study employed the Life Cycle Assessment (LCA) method to evaluate the greenhouse gas (GHG) emissions of passenger vehicles across 31 provinces in China. It examined the potential of Battery Electric Vehicles (BEVs), Hybrid Electric Vehicles (HEVs), and Plug-in Hybrid Electric Vehicles (PHEVs) in significantly reducing emissions compared to Internal Combustion Engine Vehicles (ICEVs). The findings revealed that HEVs, PHEVs, and BEVs substantially lower emissions under the national average electricity GHG emissions coefficient, with PHEVs being the most effective, reducing emissions by over 30%. However, disparities exist at the provincial level due to varying electricity structures, with BEVs not fully effective in reducing emissions in about ten provinces, including Liaoning. The study further categorized provinces into six groups using the K-means method to guide the promotion of different types of New Energy Vehicles (NEVs). Sensitivity analysis highlights the critical role of reducing electricity emission coefficients and adjusting PHEV's utility factor in achieving emission reductions. The study also notes that battery replacements in PHEVs and BEVs during their lifecycle could impede emission reduction in most provinces. To maximize the environmental benefits of NEVs, there is a need to move to a greener electricity mix. At the same time, strategies for promoting NEVs should vary for different provinces.

Laboratory characterization of VOC evaporative emissions from light-duty plug-in hybrid electric vehicles: Environmental impact comparison with conventional vehicles.

Cost and energy performance of advanced light duty vehicles: Implications for standards and subsidies

The impact of human behavior on vehicle efficiency has been vastly explored for internal combustion engine (ICE) vehicles. However, human behavioral impacts on vehicle efficiency have not yet transitioned to include battery electric vehicles (BEVs). Understanding the impact of human behavior that achieves BEV efficiency is essential globally, as BEVs begin to retain a significant portion of the automotive market share. BEV sales trends in the US have seen consistent growth since 2010, amounting to over 200,000 units sold by 2015. Globally, the total amount of BEVs and plug-in hybrid electric vehicles (PHEVs) is expected to be 40-70 million by 2025. In light of the growth estimates, defining behavior that induces efficient energy consumption when driving BEVs is essential as these vehicles have a traveling distance constrained to 60-120 miles and can require 1-8 hours to attain a fully charged battery at commercial charging stations. With firm traveling distances and long charging times, defining human behavioral impacts on BEV efficiency will allow drivers to get the most range out of their vehicle. In order to develop categories of BEV drivers in terms of efficiency, an empirical experiment was conducted to determine if clustering drivers on their energy consumption profiles invokes significant categories. The driving attributes that defined the clusters were extracted to compare whether or not efficient BEV driving is similar to eco-driving in ICE vehicles. Furthermore, BEV drivers can suffer from anxiety that stems from limited traveling distance, a phenomenon known as range anxiety. However, there exist other sources of anxiety-related human driving behavior, three of which can be measured using the driving behavior survey (DBS). The three anxiety measures from the DBS were contrasted against the BEV efficiency clusters found from this research, to determine if the anxiety factors defined by the DBS were responsible for efficient BEV driving. The results from this research found two significantly different clusters of BEV driving efficiency, which were defined as efficient and inefficient BEV driving. In comparison to eco-driving in ICE vehicles, both aggressive speed and acceleration were found to be contributing factors to BEV efficiency. The results from the DBS proved that anxiety was not a contributing factor to BEV efficiency, as both clusters had similar answers. The information accumulated through this research can be used to guide new BEV drivers to adopt sustainable driving behaviors, which can help maximize their traveling distance on a single charge. Behavioral contributions to

As many countries transition to electric vehicles (EVs) to reduce tailpipe emissions from internal combustion engine vehicles (ICEVs), both vehicle types continue to generate non-exhaust particulate matter (PM), including tire wear, brake wear, road surface wear, and particularly road dust resuspension. Among these, road dust resuspension is a major contributor to non-exhaust PM. While factors such as vehicle weight and drivetrain configuration have been extensively studied in fleet-level research, direct comparisons between ICEVs and EVs of the same model have not been explored. This study investigates the effects of drivetrain, vehicle weight, and payload on road dust resuspension emissions from ICEV and EV models. Two experimental approaches were employed: (1) acceleration from 0 to 60 km/h, and (2) a simulated real-world driving cycle (RDC). Each test was conducted under both light and heavy payload conditions. The results show that the EV consistently emitted more PM than the ICEV during both acceleration and RDC tests, based on factory-standard vehicle weights. Under identical vehicle weight conditions, the EV demonstrated higher PM resuspension levels, likely due to its higher torque and more immediate power delivery, which increases friction between the tires and the road, particularly during rapid acceleration. Both vehicle types exhibited significant increases in PM emissions under heavy payload conditions. These findings underscore the importance of addressing non-exhaust emissions from EVs, particularly road dust resuspension, and highlight the need for further research into mitigation strategies, such as vehicle lightweighting.