Technology Comparison for Spark Ignition Engines of New Generation.
Figure 1 shows the macro-trends that are shaping the future of the automotive industry. Some of them have a considerable impact on the vehicle propulsion system. The most significant ones are:
* sustainability, namely the simultaneous reduction of C[O.sub.2]/ fuel consumption and emissions;
* autonomous driving;
* powertrain electrification.
In this scenario, as shown in Figure 2, internal combustion engines will still have a key role, as confirmed by many studies [3,2].
In long-term scenarios, it is still assumed the presence of a thermal unit in most powertrain configurations, albeit with varying degrees of electrification. The pure electric propulsion can mitigate the problem of the cost of fuel, as shown in [5,6,7], and the local emission of vehicles, even though it does not seem to solve the problem of fine dust, mainly due to asphalt wear, tires and brakes, and as such directly linked to the weight of the vehicle, as shown in .
In purely electric vehicles, it is necessary to replace functions that are not closely related to propulsion, which are carried out by the thermal unit, as described in Figure 3, namely:
* production of electricity for devices on board the vehicle;
* mechanical energy production for the air conditioning system [8,9,10];
* production of thermal energy to heat the passenger compartment.
In the absence of an internal combustion engine that converts chemical energy stored in the fuel, the additional functions [9,10] require an electrical energy storage on board possibly much larger than what is necessary for traction only, with negative impacts on cost, weight, range and charging times.
The effectiveness of a purely electric powertrain for the global reduction of C[O.sub.2] is another item that poses some doubts. Indeed, the most correct way to evaluate the environmental impact is to consider the so-called Well-to-Wheel (WTW) [13,14,15] efficiency, which is the product between the production yield from the primary source and distribution of the fuel (or energy carrier), said Well-to-Tank (WTT) and the conversion efficiency of the fuel/ energy carrier into energy for propulsion, said Tank-To-Wheel (TTW):
WTW = WTT x TTW (1)
Considering this approach, the effectiveness of the pure electric propulsion becomes significant when the energy production comes mainly from renewable sources . In addition, as already mentioned above, to consider the real use of the vehicle, TTW should also include useful energy not used for traction, leading to a more significant variable, that is the Tank-To-Use (TTU) efficiency:
TTU = [Energy Used in Vehicle/Energy Taken from Tank] (2)
Moreover, a more detailed evaluation should include both the production efforts and the user scenario, such as the proportion of short/long distances, the intermediate charging as well as the supply of raw materials and recycling .
The analysis of the impacts on C[O.sub.2] emissions of different types of powertrain using the approach just described, will be presented in a future work.
After highlighting the importance of powertrains that include an internal combustion engine, it is necessary to evaluate how they can cope with the more and more stringent regulatory constraints, relating to C[O.sub.2] emissions and toxic gases, as detailed in . It should be considered that there are significant differences in the various world markets, and after year 2020 the scenario is still uncertain, as summarized in Figure 4.
Figure 4. Trend of CO2 emission legislation (dashed lines are only forecast) Years NEDC eqiv. CO2 emissio limit g/km 2015 130 2020 95 2025 75 2016 154 2025 97 2015 167 2020 117 2025 96 2030 72 Note: Table made from bar graph.
Lower C[O.sub.2] emissions may be achieved by reducing fuel consumption, or by using fuels with smaller carbon content, such as bio-fuels or natural gas . Furthermore, the efficiency increase should proceed in parallel with the reduction of toxic emissions, both during regulated driving cycles and real driving conditions . These constraints are the main challenges confronting the design and development of new internal combustion engines, together with safety, comfort, drivability/performance, and costs.
Downsizing concepts, including turbocharging in combination with direct injection, have significantly contributed to the recent improvements of internal combustion engines [15, 16, 17, 39]. Among these, gasoline engines, operated with stoichiometric mixture, appear particularly promising for the simultaneous reduction of fuel consumption and toxic emissions. The Three-Way-Catalytic converter is a well-proven, affordable and very efficient after-treatment system. Diesel engines are more efficient, but N[O.sub.x] and PM reduction is much more complex, expensive and potentially highly impacting on end user use since the SCR system will be used more frequently to respect the new N[O.sub.x] limits under real driving conditions, thus requiring urea refills more often [18,19].
Nonetheless, SI engine efficiency is still to be significantly increased to respect the next and future C[O.sub.2] regulations. To better identify and understand the most effective technologies that could be introduced or adopted, one should firstly focus on the main limiting factors affecting efficiency and transient response. Referring to Figure 5 and [34, 35]:
* pumping losses at partial loads in zone n.1;
* limited thermodynamic efficiency due to reduced compression ratio, to avoid knocking damages at high loads, mainly in zones 2 and 3;
* combustion chamber and exhaust system thermo-mechanical stresses at high loads load in zone n.4 where an enrichment of the mixture is necessary to protect the engine components such as the turbine;
* turbo-lag for turbo-downsized engines, in zone n.2 where scavenging strategies should be adopted with negative effects on the Three-Way Catalyst (TWC).
This paper presents a conceptual comparison among the most promising technologies for SI engines, considering their effectiveness, their technological maturity and their impact on powertrain production costs. The considered timeframe horizon is 2025, and for brevity, this work will focus on powertrains designed for C-segment vehicles, which constitute the largest portion of the market, both in European and emerging markets.
The main investigated technologies are related to combustion management and air charging systems, the key elements for improving the previously mentioned limiting factors. To develop an objective comparison of the possible benefits, the evaluation criteria have been defined based on the requirements described in the previous paragraphs, and considering the most promising powertrain architectures. Different powertrain electrification levels have also been considered, thus implying different operational requirements for the internal combustion engine.
Below, a general overview of this work is provided. The main objective is twofold: technological assessment for SI ICEs, and definition of a methodology to objectify the comparison between technological solutions for complex powertrain architectures.
The first section presents a detailed analysis of the powertrain requirements, using a Quality Function Deployment (QFD) with a modified and personalized approach [20, 21]. Two vehicle concepts have also been considered, namely human-driven and computer-driven, by integrating and providing more details than what has already been published, for example in . Indeed, semi-autonomous and autonomous driving systems will be progressively introduced in the market, and therefore the analysis of their impact on future powertrains cannot be ignored. The following section introduces the key powertrain architectures to achieve the desired targets, and focuses on their impacts on the internal combustion engines. Finally, evaluation criteria for a quantitative comparison are defined.
Then the different ICE technologies are described and critically analyzed, highlighting their effectiveness versus the main requirements. Such considerations are mainly based on the most significant technical and scientific literature.
The subsequent section introduces a simplified ICE-Vehicle model, which has been used to validate and integrate bibliographic data, and to further extend the analysis, to consider technologies and vehicle-powertrain architectures that could not be found in the literature. The attention is then mainly focused on technology costs, with a 2025 timeframe constraint.
A comparison among the most promising technologies for SI engines is finally presented, by making use of a selection matrix that defines the overall ranking based on benefits/costs ratio.
NEW POWERTRAIN AND ENGINE REQUIREMENTS
To define the requirements of the new powertrain architectures, a modified QFD approach has been used. QFD is an effective way to link the requests and needs perceived by the end-user point of view to the technical features of a product. Specifically, needs are evaluated by the users at the vehicle high level, therefore a multi-level QFD approach, as in , is used, until the powertrain level is reached, where engine and other powertrain components are evaluated. A brief illustration of the process is given in Figure 6.
User's needs and their priority (1-low, 7-high), can be obtained from market research and they represent the inputs for QFD. Specifically, in HoQ1 (house of quality, level 1), driver's needs are evaluated in an objective way by means of engineering variables. For instance, the demand for high performances can be objectified by the maximum vehicle speed, its acceleration or the time from 0 to 100 km/h. Table1 shows the measurable requirements, which, in turn, represent the inputs for the second level, coming from driver's point of view.
Furthermore, requirements can be divided in two sub-groups, both for standard and autonomous vehicles:
* explicit requirements by end-user;
* requirements from legislation related to C[O.sub.2] and toxic emissions, which are considered implicit by the end-user.
Legislation requirements are clearly of great importance because vehicle needs to fulfill them to be saleable. They are not explicitly considered by the user although their impact on customer's choice is becoming greater due to the driving bans imposed in many cities to the vehicles complying with old legislations.
Furthermore, the requirements for autonomous vehicles change, as already shown in ; specifically, the requirements related to performances (such as vehicle speed or acceleration time) become less important. The opposite is true for those related to comfort, such as NVH, torque regularity, vibration.
However, the constraints on C[O.sub.2] and toxic emissions remain at high priority and are the main driver of the technological change requested to new vehicles and powertrains.
The priorities and requirements targets in Table1 depend on the vehicle segment, legislation standards and the manufacturer position among the competitors. Moreover, it is important to underline that costs are not considered here. The reason is that cost is one of the main requirements for the end-user and its weight would not be enough important if considered at this level. It instead will be considered by means of the Technology Value Ratio, as defined in the section Results of Engine Technology Comparison.
As mentioned before, this work refers to a C-segment vehicle and regulation standards on emissions and fuel consumption by 2025 as shown in Figure 4. The priorities were introduced based on the study , adjusted with Magneti Marelli internal analysis.
On QFD level 2, in HoQ2.1, the requirements and priorities for the powertrain, described in Table1, are mapped with their technical characteristics in Table 2, with appropriate specifications to effectively satisfy the requirements. In Appendix 3, the HoQ2.1 correlation matrix is illustrated. In Table 2, the priority specifications in case of driver or driverless vehicle as well as the most significant differences between the two cases are reported. The relevance of the different characteristics is highlighted by the green histogram on the left. Among all, the powertrain weight, including the contribution of the energy storage system, represents the most important factor in addition to the energy generation efficiency for the propulsion. Among the technical characteristics with higher ranking, it must be noted that the energy refilling time is critical for pure electric powertrains but not for hybrid ones. Clearly, emissions still have a crucial role. In case of vehicle with autonomous driving, the powertrain characteristics related to the performances are less important, whereas the comfort features are more relevant. This is true in case of the vehicles being used exclusively in driverless mode.
The reported characteristics and their targets are the inputs for the next phase of definition and selection of the powertrain architecture.
POWERTRAIN AND ENGINE ARCHITECTURE
ICE technical features and powertrain requirements depend on the architecture chosen for the analysis. As shown in the previous chapter, efficiency (tank to wheel) is one of the key features of the powertrain and electrification is one of the most effective ways to improve it. Several architectures exist , however the main ones are summarized in Figure 7. From B to F, the degree of electrification of the powertrain increases, which implies a greater availability of electrical power with respect to the total power used for traction and an higher pure electric mileage range. In hybrid vehicles, the electric machine can be mounted in different positions and can be used for different functions. In Figure 7, the electric machine positioning and the power flows for propulsion and kinetic energy recovery during braking, which represent a key feature to improve the efficiency of the vehicle, are highlighted. Electric machine positioning is denoted by P0/P1/P2 or P3. P0 means that the Electric Motor (EM) is coupled to the crankshaft via a belt in a configuration known as Belt Starter Generator (BSG). When the EM is in P1 position, it is directly coupled to the crankshaft by a toothed wheel or a chain. P1 configuration allows reaching a higher maximum torque than the belt driven electric machine, due to the higher power from recuperation and the more robust connection. In P2 configuration, the EM is located upstream the transmission between two clutches, in parallel to the ICE. P2 configuration allows the functionality of pure electric driving, with higher fuel consumption saving than P0/P1 architectures. The torque delivered by the EM is limited mainly by the maximum torque of transmission, hence in case of contemporary use of ICE and EM, the gearbox must transmit the sum of the torques delivered by the two machines. In the P3 architecture, the EM is connected in parallel to the ICE downstream the gearbox, without any adaptation, and typically a BSG (> 4 kW) is also adopted in P0/P1 position to charge the battery during engine stops. In all previous hybrid configurations, the worse efficiency regions of ICE at part load can be avoided, leading to a lower fuel consumption. Furthermore, braking energy recovery can be performed.
In the so called Mild Hybrid powertrain (typically P0 and P1 configurations), the EM cannot drive the vehicle in pure electric mode, however it can support the internal combustion engine. P0 configuration main advantage is low cost integration, however P1 allows higher torque density.
In Full Hybrid powertrain, the most effective solutions are P2 and Power Split. In Power Split (configuration E), ICE is connected at the same time to the generator and the electric motor by a planetary gearbox, and the motor shaft is connected to the wheels with a final gear reduction. Power split concept combines features of both series and parallel hybrid. Full hybrid can be obtained also in P3 configuration (scheme D), with the highest energy recuperation potential.
Scheme F in Figure 7, describes a serial hybrid, where ICE operates as a range extender of the autonomy. The battery capacity allows to cover, without recharge, a longer distance than for a parallel hybrid (>250 km). ICE and generator have typically very high efficiency in a limited operating area  and rate power lower than 50% of the electric traction motor, depending on vehicle class and battery capacity.
The architectures from C (only P2) to F can be plug-in, with the possibility to recharge the battery like a pure electric vehicle. The functions of each described architecture are summarized in Table 3.
Using the selection matrix, described in Table 2, a comparison among the most interesting variants of the proposed architectures is performed using the technical requirements described before as selection criteria. In the selection, the values of the technical specifications have been disregarded, considering the capacity of each architecture of satisfying the requirements. Focusing on C-segment, the reference powertrain consists of a 4-cyilinder GDI turbocharged engine with a 6-speed DCT gearbox.
In , a comparison between P2 hybrid and Power Split architectures is shown and it is considered to define the selection matrix. In Table 4, architectures from 2 to 6 have a growing score due to the electrification, however they are characterized by rising costs due to the increased complexity and the larger size of the battery, which is necessary to ensure a long range in purely electric mode.
Moreover, as underlined by the high score, electrification affects not only the overall efficiency of the powertrain but also NOx as well as Soot/PN emissions. Indeed, the required torque is shared between electrical motor and internal combustion engine, and EM may relieve ICE during transient phases, where EGR management (internal or external) is critical, allowing the reduction of NOx peaks. Clearly, the impact of the electrification is higher when the proportion of the torque assigned to the EM is higher, which happens for HV configurations. As a matter of fact, the presence of TWC when ICE operates in stoichiometric conditions reduces the global ranking in the selection matrix.
The plug-in penetration in C-segment is expected only after 2025 to fulfill the C[O.sub.2] requirements. Indeed, as it will be shown in the next sections, C[O.sub.2] targets to 2020 and 2025 can be reached for C-segment vehicle by improving internal combustion engine efficiency and by a 48V electrification.
ENGINE TECHNOLOGIES DESCRIPTION AND ANALYSIS
This chapter summarizes the most relevant technologies for SI-ICE highlighting their advantages and the main drawbacks. The considered technologies can be classified according to the engine subsystem, as described in the following table:
The aforementioned technologies have been chosen for their degree of maturity with respect to the horizon of year 2025; moreover, they are considered the most effective for overcoming the limits of SI-ICEs, exposed previously.
Engine Subsystem Technologies Combustion System GDI Lean Combustion Miller/Atkinson Cycle Variable Compression Ratio Water Injection Cylinder Deactivation Air System External EGR Multistage Air Charging
Furthermore, the technologies to reduce ICE mechanical friction and the power from auxiliaries, especially in the regulated cycles when the system starts from cold conditions, can be considered already state-of-art in many applications. They can be summarized as follows:
* electrification of coolant and lubricating systems, allowing to optimize pumping requirements for the coolant and oil, depending on the real needs of the engine, in contrast with non-variable mechanical pumps; furthermore, the combination of the electrification with split cooling, allows the engine to achieve warm conditions sooner, while the charge air cooler is kept colder in order to keep the filling efficiency at an appropriate level.
* crankshaft offset adjustments, bearings design or steel pistons (instead of aluminum).
After-treatment systems are not considered in this work because the options are quite defined; in case of stoichiometric combustion, the after-treatment system is composed of a three-way catalytic converter (TWC) in addition to a gasoline particulate filter (GPF); in case of lean combustion another catalyst to reduce N[O.sub.x], such as SCR or LNT must be added. Furthermore, the benefits achievable with biofuels are not considered in this work and they will be investigated in future works.
GDI Lean Combustion
Lean combustion in gasoline engine has proven to significantly increase the overall efficiency . Unfortunately, expensive exhaust after-treatment is necessary because the three-way catalyst cannot reduce N[O.sub.x] in case of excess of oxygen in the exhaust gas. To mitigate the catalyst cost increase, engine has to operate in extreme lean conditions (e.g. lambda > 1.8), thus obtaining a drastic reduction of engine out N[O.sub.x] emissions and further engine efficiency improvement, thanks to the higher ratio of specific heats and lower heat losses to the cylinder walls. Figure 8, from , is an example of the operating modes of a lean concept in the engine working points, highlighting the operating range in the WLTC cycle.
In addition to the increased demands on the charging system, further challenges arise with the extreme lean operation. The ignition of the diluted mixture is strongly hindered by the poor thermal boundary conditions that might lead to misfiring, therefore ignition system technology development is often required. An additional limitation under lean conditions is the increased cycle-to-cycle variations (CoV) due to the slower combustion. CoV can be minimized with higher turbulent combustion velocity by enhancing the in-cylinder charge motion. Hence, port and combustion chamber design as well as the spray pattern layout play a key role in the development of extreme lean combustion concepts. Furthermore, the transient operations including the switching between operating modes, e.g. from stoichiometric to lean air/fuel ratios or vice-versa, have to be properly managed. As for all lean gasoline engines, the control of N[O.sub.x] emissions remains a challenge, therefore LNT or SCR solutions are needed. LNT was already employed at end of the '90s in the first vehicles (Toyota, Mitsubishi, VW) equipped with GDI lean engines, and in the recent years it has been widely adopted in diesel engine applications.
In , Ricardo shows a lean burn combustion concept that uses spray guided lean stratified operation. N[O.sub.x] emissions control is carried out by means of a Lean N[O.sub.x] Trap (LNT) catalyst. The study highlights the BSFC improvement achievable in stratified lean burn conditions (2.0 L turbo, 200 kW rated power) for a D-segment vehicle. Over NEDC, the solution has been simulated producing a 14% reduction of C[O.sub.2] emissions compared with the equivalent stoichiometric turbocharged GDI engine, considering LNT regeneration. Over WLTC, the fuel consumption reduction is 12%.
Miller-Cycle describes a combustion system with early or late intake valve closing [30, 41, 47, 49, 51, 63]. Figure 9 illustrates the two possibilities used to perform the Miller cycle.
Miller cycle is based on Otto cycle, having an expansion stroke which is longer than the compression stroke. In late intake valve closing Miller cycle (LIVC), the intake valve is kept open during a significant part of the piston movement during compression, so that part of the mixture contained in the cylinder is sent back to the inlet manifold, allowing the reduction of mixture mass that is "trapped" in the cylinder. Moreover, pumping losses are reduced, and the volume of the 'trapped' mixture corresponds to the cylinder volume at the closure of the intake valve. In early intake valve closing Miller cycle (EIVC), the descent of the piston creates a vacuum inside the cylinder, achieving the manifold pressure somewhere during compression, near the position where the intake valve was closed on the downward movement of the piston.
The major benefits of Miller cycle can be obtained in combination with boosting and charge air cooling, with an higher compression ratio (which increases part load operation efficiency). Moreover, in medium load conditions, lower peak pressure and temperature are reached, leading to a lower knock tendency.
Atkinson-Cycle is instead a combustion system with increased expansion stroke realized by a different crank train, which involves a system redesign. Atkinson, as Miller, has a lower effective compression ratio compared to the expansion ratio.
On the other hand, both for Miller and Atkinson cycles, depending on the gasoline engine configuration, a turbocharger upgrading or installation might be necessary to ensure the specific power and the low-end torque. In naturally aspirated engines, the lower engine power and torque might also be compensated by an increased displacement which in turns increases the friction losses.
In , FEV estimates that the Miller cycle reduces fuel consumption by 3.9%-5.7% over a baseline, 1.0 L downsized turbocharged engine with variable valve lift and timing. Part of the efficiency increase is due to the increase of geometric compression ratio from 10:1 to 12:1.
VW recently implemented the Miller cycle concept in EA211 1.5L TSI EVO engine family  and the claimed average efficiency improvement is 10% compared with the previous generation engine (1.4L TSI), including also a VGT (variable geometry turbocharger), cylinder deactivation and other in-cylinder improvements to increase engine efficiency. However, the reduction of BSFC, due to the Miller cycle implementation, was estimated to be at least half of the overall benefit. The new engine shows a reduction in fuel consumption between 5-10%, compared to the previous generation over most of its engine map. Furthermore, for a significant portion of the low loads region, Miller cycle enables 10-30% reduction in fuel consumption.
In , Audi shows a significant fuel economy improvement for the 2.0L TFSI with Miller cycle, implemented by means of the EIVC, in which, some additional gas expansion occurs, which helps reducing in-cylinder temperatures. In this configuration, engine compression ratio was increased from 9.6:1 to 11.7:1. Compared to the previous generation, fuel consumption is reduced up to 21% (on the NEDC), while power output is increased up to 25%.
Variable Compression Ratio
Increasing the compression ratio is an effective way to improve the combustion efficiency, as clearly stated by Equation (3) in the Powertrain System Model section of the paper. However, this solution is limited by high peak cylinder pressures and temperatures, which affect the powertrain design (friction and materials) as well as increasing knocking tendency in gasoline engines. Alternatively, the reduction of compression ratio helps solving the problems mentioned before, reducing the overall friction, but also getting some improved powertrain efficiency at full load operation. However, this possibly leads to cold start problems, combustion stability issues and worse efficiency at part load.
Currently, in the market, both tendencies of increasing and decreasing the compression ratio can be found, requiring different side measures, as shown in Figure 10.
To get the best from both trends, a variable compression ratio (VCR) system should be employed, using a higher CR at engine part load operations and a lower CR at higher loads. Two possible VCR technical solutions have been recently developed:
* a 2-step compression ratio [30,52] that improves full load performance, reducing emissions and friction with lower compression ratios, while it increases the combustion efficiency at lower loads with higher compression ratio;
* a fully variable VCR [30,44,53] that allows adjusting the optimal CR for each operating point, using the full potential of this technology.
In addition, there are no significant disadvantageous interdependencies with other technologies. Compared to the fully variable VCR, the cost of the 2-step VCR is much lower , and at the same time more than 80% of the fuel consumption reduction potential of the continuous system can be achieved in gasoline engines. Moreover, in 2-step VCR, the compression ratio can be maximized at lower loads without redesigning the different systems which are subjected to high stresses, e.g. piston or crankshaft, with important advantages in terms of packaging, modification of production and friction. Clearly, theoretical potential will not be fully available; indeed, due to the piston design for maximum compression ratio, disadvantages in the combustion efficiency and knock limit arise. Moreover, the modified conrod is usually heavier than the original one.
In , AVL presents its 2-stage VCR system based on the connecting rod length variation. The concept allows modular integration in the existing architecture without significant modifications, with only minor adaption of the production lines and thus it represents a highly attractive solution with respect to costs. Depending on the vehicle class, the load profile and the power rating, a C[O.sub.2] reduction between 5% and 9% in the WLTC is achievable.
In , Hyundai presents its fully variable VCR system and evaluates its benefits in 1.6L inline 4-cylinder gasoline engines both NA PFI and TC DI. VCR, by the combination of four valve timings (IVO, IVC, EVO, and EVC), has the potential to improve the fuel efficiency by 4 to 5 %. Low speed torque, in the NA PFI engine, is increased by more than 5 to 10%, while in the TC DI engine, it is increased by more than 10%.
Water injection (WI) is a cost-effective solution to minimize knocking in (downsized) gasoline engines, especially with high compression ratio. This technology is already well-known because it has been used in motorsports and aviation industry. The current trends in legislation related to the automotive sector make it particularly relevant for serial production in passenger vehicles. Water injection can be exploited as a solution to reduce fuel consumption and C[O.sub.2] emissions, especially at medium and high load, as well as to increase the engine power output [36, 37, 38, 39, 59]. The main concept is to inject water and to use its high latent heat of vaporization to reduce gas temperatures before their combustion. In fact, the addition of vaporized water would result in an efficiency decrease due to the reduction of [c.sub.p]/[c.sub.v] ratio. However, the evaporation of water results in a significant temperature reduction of the mixture of air or air/fuel and water, permitting the adoption of an higher spark advance, thus possibly reaching the optimum MFB50 position. This effect can overcompensate the efficiency decrease expected when adding gaseous water. At the same time, the dilution of the mixture with water and the reduced temperature level in the combustion chamber have minor influences on the losses due to unburned fuel and the losses due to heat transfer.
Furthermore, the cooling effect may be utilized in different ways according to the desired benefits:
* knock tendency reduction[right arrow] higher compression ratio enabled/advanced spark ignition permitted[right arrow] fuel efficiency increase at the same desired brake torque;
* reduction of exhaust temperatures under full load operation[right arrow] mixture enrichment can be avoided[right arrow] fuel efficiency increases at the same desired brake torque;
* knock tendency reduction[right arrow] advanced spark ignition[right arrow] torque increase at the same fuel consumption.
The ability of WI to lower the exhaust gas temperature is also of interest since it may be used as an enabler for employing turbochargers with variable geometry turbines (VGT) even in gasoline engines, thus allowing further downsizing potential. Alternatively, it may be used to reduce material costs on the turbochargers, due to reduced thermal stresses on the component.
As shown in Figure 11, three different implementations of water injection systems are possible:
1. Port Water Injection (PWI). In this approach, water is injected on the inlet side with a low-pressure system (5/20 bar). The advantage of this solution is its simplicity because it includes a simple pressure supply with an electrically driven water pump. Furthermore, corrosion and freezing issues caused by water are relatively easy to handle.
2. Direct Water Injection Mixture (DWI-M). Water is mixed with gasoline and it is directly injected via a modified high-pressure injector. Water quantity is metered based on the intake air mass measurement; the water-fuel mixture is compressed to an emulsion by the high-pressure pump after which it is injected via the high-pressure water-fuel injectors directly into the combustion chamber. While the amount of water required is slightly lower than PWI, the system suffers from several technical challenges and risks. Firstly, the presence of high-strength steels in such high-pressure systems is not trivial and thus magnifies the challenges associated with corrosion damage. This implies a major and complete overhaul of the existing high-pressure fuel system. Secondly, to enable a quick availability of water in the system, the connecting pipes and components need to have small cross-sectional areas and volumes, respectively. This creates high pressure pulsations and thus high-pressure peaks in the injectors and in the high-pressure pump. Therefore, a completely new and complex high-pressure system is required, which has a negative effect on costs and risks. Furthermore, when the engine is stopped, water needs to be removed from the fuel system by running the engine and flushing it with pure gasoline. This makes the solution less effective when used together with hybrid powertrain solutions and start-stop technology. Finally, the challenges associated with freezing and corrosion are magnified in such a high-pressure system.
3. Direct Water Injection Separate (DWI-S). A high-pressure water injection system is installed in parallel to the existing high-pressure fuel injection system requiring additional high pressure direct water injectors to be integrated in increasingly smaller and complex cylinder heads. Component costs are then higher than the equivalent low-pressure variant. As in the previous case, the presence of high-strength steels magnifies the problem of corrosion damage, while keeping the robustness against freezing, i.e. due to anomalous water expansion. In addition, reversing the flow through the high pressure mechanical pump is achieved only with additional valves and a high system complexity which adversely affects weight and costs.
As a result of the previous considerations, the Port Water Injection concept is the best candidate for series production. Its main drawback with respect to the other possible solution is the higher water consumption.
The general limit of WI is the availability of good quality of water (deionized) on-board the vehicle. Three possible solutions are under investigation; refilling by the user, A/C condensation and rainwater harvesting, exhaust gas condensation. The first one is the most promising because it is cheap and accepted by the end-customer, as reported in the survey . The other solutions are considered to be developed to minimize the end-user impact and refilling costs.
In , the benefits of PWI with a CR increased by 2 points were studied, resulting in 4% of fuel saving on WLTC, 13% in real word conditions and up to 20% in full load. Water rate related to fuel was up to 60%, with a water consumption under real world driving conditions between 1.3-2.8 L/100 km, considering a large family car equipped with a turbo GDI engine with 115-130 kW rated power.
In , the DWI-S technology combined with Miller cycle, cooled exhaust gas recirculation and variable compression ratio was investigated. After a single cylinder experimental investigation, driving cycle simulations were carried out in order to understand both the fuel and water reduction potentials. The combination of water injection with the Miller cycle and cooled EGR has allowed to improve the indicated specific fuel consumption up to 197 g/kWh together with a significant enhancement of the region of very attractive efficiency values, especially at low engine speeds and high loads. Increasing the VCR up to 10.7/14.7 from 9.5/13, the resulting fuel consumption benefit in WLTP was up to 6.7 %. At the same time, water consumption was up to 3 l/100km. Therefore, onboard water generation or, at least, the utilization of tap water for a refill tank are required, in order to ensure an affordable cost also from the ownership's perception.
Cylinder Deactivation (CD) allows a significant reduction of the pumping as well as heat transfer losses at lower engine loads, by reducing the number of active operating cylinders. The active displacement is reduced, increasing manifold pressure and reducing pumping losses. The requested load on the cylinder (BMEP) is also increased, which reduces the relative heat transfer to the cylinder walls with respect to the available heat. Since other technologies (e.g. downsizing or VVT) can reduce pumping and friction losses, the coupling of CD with other technologies could be not so effective.
There are two possible main solutions to implement cylinder deactivation:
1. A variable valvetrain is used to shutoff the intake and exhaust valves of the deactivated cylinders.
2. Electronic Cylinder Deactivation; in which, the exhaust system is duplicated (for instance, each cylinder pair 1-4 and 2-3 has a dedicated exhaust line) and the injection is switched off for one cylinder group.
Mode 2 is an alternative method without the need for additional flexibility in the valve train and it additionally offers the chance to utilize the deactivated cylinders with the characteristic sound of the engine. The main drawback is the packaging impact of the exhaust line.
The conventional cylinder deactivation, especially the first solution, is normally applied only to large engines with an even number of cylinders. In this way, the cylinders deactivate symmetrically in order to avoid intense torque fluctuations and vibrations. Furthermore, higher improvements can be attained with the dynamic deactivation of individual cylinders. Regarding this particular technology, many systems are being currently developed. These systems continually change the active cylinders and have many potential advantages over conventional cylinder deactivation, such as:
* Maintaining uniform engine operation temperatures;
* Allowing the throttle to remain nearly fully open by controlling the engine power by varying the firing cylinders;
* Handling noise, vibration and harshness by dynamically controlling the active cylinders, allowing the adoption of cylinder deactivation at lower engine rpm;
* Expanding the range of applicability to smaller engines, even to 3-cylinder engines or with an odd number of cylinders.
An example of this new CD system is the DSF (Dynamic Skip Fire) system, described in , that claims to achieve a fuel benefit between 10/20%, depending on engine type.
In , the benefits achievable with the Electronic Cylinder Deactivation in a 4 Cyl 1.4L TC GDI engine are shown. In the NEDC cycle, fuel consumption improvements are up to 7%, with small penalty in comparison with a 2 Cyl 0.9L TC GDI engine.
The external exhaust gas recirculation (EGR)  is an alternative to the internal one, made using VVA systems. It can be carried out according to the three layouts described in Figure 12:
* High Pressure (HP-EGR)
* Low Pressure (LP-EGR)
* Dedicated EGR (D-EGR)
High pressure and low-pressure split is also possible. In the HP-EGR solution, exhaust gases are recycled from upstream the turbine to the intake manifold; in the LP-EGR version, exhaust gases are recycled from downstream the catalyst to the compressor intake. In both cases, gas is cooled with an air/water cooler before entering the control valve, typically driven by a DC-motor. For the LP-EGR case, a lower pressure ratio for the gas flow is available, however, the main advantage is the lower temperature in the manifold . Indeed, exhaust gases are cooled both in LP-Cooler and turbocharger intercooler, which is designed for the maximum engine power, a faraway condition from the operating area of LP-EGR. Furthermore, LP-EGR leads the compressor to work in an area with greater efficiency, with benefits on fuel consumption for the lower exhaust pressure at turbine upstream due to the reduced compressor requested power. In SI engines, EGR can be used also for the reduction of N[O.sub.x] emissions, the de-throttling at partial loads and knock mitigation. An additional benefit is at high engine power, thanks to the avoiding of the enrichment of the mixture with advantages on BSFC and PN.
The main EGR drawbacks are the reduction of maximum power with a given turbocharger layout and the negative impact on transient response at low engine speeds because of the increased turbocharger size to compensate for maximum power loss with EGR. Furthermore, cooling capacity of the engine cooling system as well as for the charge air intercooler (for LP-EGR) has to be adapted and integrated to the front end of the vehicle. The third variant for the external EGR is called "dedicated EGR", as shown in Figure 12. Dilution of the intake charge, with the traditional EGR, provides benefits in terms of cycle efficiency and knock resistance. However, it also poses challenges in terms of combustion stability, condensation and power density. With D-EGR, one cylinder, which operates in rich condition, produces EGR for all four cylinders of the engine, introducing reformates such as CO and H2 into the intake charge by means of a mixer installed upstream the intake manifold, bringing back some of the stability lost for EGR dilution, leading to a higher ratio of specific heats and a benefit on effective RON of the fuel. To enable the technology, in addition to the high-pressure EGR loop and EGR cooler, a supercharger with a bypass valve and an after-cooler is used. A cold start valve is installed as alternative path for exhaust gases when the EGR valve is closed. A PFI injector can be added to the intake manifold which allows flexibility in the way the extra fuel to the 4th cylinder is delivered.
In , the following advantages with the adoption of LP-EGR are shown:
* 1% BSFC improvement at partial loads;
* more than 4% BSFC improvement, with the same PN-emission, by shifting the 50% MFB50 to the optimum in knock limited engine operating points.
In , an extreme downsizing from a 3.7L V6 NA engine to a I4, 2.5L turbocharged engine is allowed by a cooled HP-EGR in combination with scavenging, achieving an increase in the low-end torque and maximum efficiency by more than 30%, as well as a significant weight and size reduction.
In , D-EGR improves the knock resistance of a 2.L GDI engine allowing a compression ratio equal to 11.7. In this configuration, the improvement in engine efficiency was at least 10% across the whole performance map, with substantially higher improvements for certain engine operating points. For instance, BSFC @ 2000 rpm/ 2 bar BMEP improved from 385 g/kWh to 330 g/kWh and the lowest BSFC in the engine map was 212 g/kWh compared to 236 g/kWh of the original engine. The addition of 2-stage boosting also allowed the engine to meet its BMEP target of 17 bar from 1500 to 5500 rpm while maintaining good transient response and low engine-out emissions. The main drawbacks of D-EGR are the need to control combustion stability and the complexity due to a second stage of air boosting to recover the power loss at high loads.
Multistage Air Charging
The most effective way to reach a higher specific power output is increasing the boost pressure. This allows more air (oxygen) into the cylinders. Furthermore, increased boost pressure is necessary in combination with increased EGR and the application of Miller Cycle to avoid reduced volumetric efficiency and reduced full load torque.
The extension of boosting operative range can increase the low-end torque, which is a critical issue especially in extreme downsized engines , with fun to drive benefits and/or possibility to implement down-speeding strategies. Furthermore, the use of complex charging systems is necessary to fulfill the new European RDE standards where scavenging strategy will not be implementable, because it can cause the decrease of TWC efficiency.
The extension of the boosting range is possible employing VGT turbochargers, as long as exhaust gas temperature is low enough, however better results can be attained employing multistage charging layouts, according to two main solutions:
* 2-stage turbochargers;
* mechanical supercharger or electric turbocharger in addition to the first turbocharger stage.
In the first solution, two differently sized turbochargers are used and chosen to be operative in different areas of the engine map. This solution has huge impact on cost and layout. In the second solution, the tendency is to use an electric turbocharger (or an e-booster only), that can deliver air at maximum boosting pressure within 300-1000 ms [42,57]. This technology is particularly suitable when used in conjunction with a 48 Volt BSG, allowing to achieve a fuel economy improvement over 10% and a significant gain in the fun to drive, as shown in [16,57,58].
Summary of Technology Benefits and Drawbacks
The evaluated technology benefits, drawbacks and the influences in the main zones of engine map referring to Figure 5 are summarized in Table 5. Moreover, in Appendix 1, a detailed summary of the outcomes from literature analysis is reported.
POWERTRAIN SYSTEM MODEL
Much information to evaluate new technologies relating to the powertrain can be found in literature; however, the effects due to the technology mixing for various powertrain configurations and in different vehicle segments can be predicted only thanks to the simulation approach.The adopted models must be at the same time adequately accurate and fast, to carry out evaluations on long driving profiles and to be used for system optimization. In [29, 30], several examples of simplified models used for evaluations are considered. In this study, 0-D models, whose scheme is depicted in Figure 13, are used.
The outputs are fuel consumption, emissions and performance, evaluated on different driving profiles (e.g. NEDC, WLTC, etc.). The details of the model have already been the subject of other works [25, 26, 27, 28], to which some of the authors have contributed; therefore, in this section, only specific parts related to the impacts given by the engine technologies under consideration are shown. Specifically, the brake effective torque and the brake specific fuel consumption (BSFC) have been evaluated using the Willan's line method, as described in .
The definitions and the mathematical steps leading up to the torque model can be found in Appendix 2.
Specifically, brake effective torque ([T.sub.e]) can be expressed as follows (3):
[T.sub.e] = e * [[eta].sub.idc] * [H.sub.LHV] * [[??].sub.f]/[omega] - [p.sub.loss] * [V.sub.d]/4[pi] - ([p.sub.exh] - [p.sub.intake])[V.sub.d]/4[pi] (3)
where: e is the product of thermodynamic and combustion efficiencies, [H.sub.LHV] is the fuel lower heating value, [[??].sub.f] is the fuel rate, [omega] is the engine speed in rad/s, [p.sub.loss] takes into account the mechanical losses for friction and auxiliaries, [V.sub.d] is the engine displacement, [p.sub.exh] and [p.sub.intake] are the exhaust and intake pressures, respectively.
Moreover, [[eta].sub.idc] is the ideal efficiency which is theoretically dependent on compression ratio only. However, in this work, a modification to ta[k.sub.e] into account the Miller cycle is considered, as proposed in :
[[eta].sub.idc] = 1 - 1/[r.sup.[gamma]-1.sub.g] * ([sigma]) (4)
where [sigma]=[r.sub.g]/[r.sub.tr] is the expansion - compression ratio, with [r.sub.g] geometrical and [r.sub.tr] trapped compression ratio, respectively. Moreover, f([sigma]) takes into the account the effect of the Miller cycle on the theoretical efficiency; its formula can be found in Appendix 2.
Introducing [lambda], the volumetric efficiency [[eta].sub.v] and the air density p, (3) can be re-written as:
[T.sub.e] = e * [[eta].sub.v] * [[eta].sub.idc] [rho] * [V.sub.d]/[lambda] * [alpha] * [H.sub.LHV] - [p.sub.loss] * [V.sub.d]/4[pi] - * ([p.sub.exh] - [p.sub.inta[k.sub.e]])[V.sub.d]/4[pi] (5)
From (5), the bra[k.sub.e] specific fuel consumption BSFC, which is correlated to the C[O.sub.2] emission, is calculated:
BSFC = [[??].sub.f]/[T.sub.e] * [[omega].sub.eng] (6)
Notice that, the first term of the equation (5) is the indicated gross torque, the second is the torque term for friction and auxiliaries and the third is the component due to the pumping losses. To consider the effects lin[k.sub.e]d to the applied technologies, the generating torque equation can be modified through several coefficients. In particular, the effects of the technologies investigated can affect all the terms of Equation (5) that can be corrected as follows:
[T.sub.e] = [k.sub.e] * e * [[eta].sub.v] * [[eta].sub.idc] [rho] * [V.sub.d]/[lambda] * [alpha] * [H.sub.LHV] - [k.sub.f] * [p.sub.loss] * [V.sub.d]/4[pi] - [k.sub.p] * ([p.sub.exh] - [p.sub.inta[k.sub.e]])[V.sub.d]/4[pi] (7)
Where [k.sub.e] takes into account the correction on the thermodynamic efficiency (e.g. for lower/higher heat losses, MFB50 position, knock phenomena), [k.sub.f] is the correction for the friction losses (e.g. for heavier/lighter cranktrain, bigger/smaller bearings, etc..), [k.sub.p] is the correction for the pumping losses (e.g. for part load operations throttle adoption or not, turbocharger improvements, VGT or WG adoption, etc..).
The corrective parameters [k.sub.f], [k.sub.e] and [k.sub.p] have been evaluated starting from literature analysis and by means of a 1D engine model, calibrated and validated against experimental data, as described in . A 1.4L - 4 cylinder - Turbo- GDI engine with VVT and CR =10 has been chosen as baseline engine. Specifically, the 1D model contains, in addition to the air path with turbocharger and supercharger, a predictive combustion modelling, which can also predict MAPO statistical distribution. In case of water injection, as an example, the corrective parameters can be found by firstly establishing a trade-off between MAPO percentile and indicated efficiency for knock limited engine operating points .
Subsequently, water injection is simulated and the new trade-off can be defined; [k.sub.e] is therefore calculated and extrapolated for the nearest points, obtaining the static map to be used in the 0D model. In Table 6, the effects of the ICE investigated technologies on the parameters are illustrated:
In Figure 14 and Figure 15, the BSFC map and the correlations between measured and estimated values of BMEP and BSFC for the baseline engine are shown, demonstrating the feasibility of the approach.
The advantage of the approach is its modularity. Indeed, from the baseline case, the BSFC curve derived for any combination of technologies can be estimated. As an example, Figure 16 shows the BSFC map for the baseline case, adding the Miller cycle with an increased CR (12.5), cylinder displacement and friction reduction of 20%. The results are quite similar than those shown in Figure 17, describing the measured BSFC map of the evolution of the baseline engine with the technologies investigated in .
The cumulative trends in fuel consumption on NEDC cycle for a C-segment vehicle equipped with the baseline engine and with some of the technologies studied are shown in Figure 18; moreover, in Table 7, all the most meaningful technologies and their combinations are analyzed; their savings in fuel/C[O.sub.2] both in NEDC and WLTC are shown. The C[O.sub.2] estimation by means the model is 125 g/km in NEDC cycle against 123 g/km of the real emission of the reference vehicle. The mass of the vehicle in the baseline configuration (only ICE) was considered equal to 1350 kg, whereas it was increased by 60 kg for a 48 Volt 22kW hybrid in P2 configuration and 1 kWh of battery capacity (source: MM data).
From Figure 18 and Table 7, it is clearly shown that the benefits given by each technology are not additive, and that the contribution of the electrification decreases with increasing the efficiency of ICE, especially in the WLTC cycle which is characterized by a higher average load.
As it can be observed, the estimation of the improvements related to the analyzed technologies is aligned with the values highlighted in the bibliography.
ENGINE TECHNOLOGY COST ANALYSIS
Cost estimations for powertrain manufactures during the early phase of technologies investigation or product development are subject to several uncertainties. These uncertainties are related to information concerning the product and production including the production process and its resources. Nevertheless, especially when developing a new technology, the assumed product costs must be tracked in order to lead the product to a successful market introduction.
The main sources of technology costs are [16, 30, 46, 52]. Information coming from these works was reviewed and integrated with Magneti Marelli data. The estimated costs at 2025 for the technologies considered for C-segment vehicle are summarized in the following Table 8.
RESULTS OF ENGINE TECHNOLOGY COMPARISON
The engine technologies studied in the previous chapters by analyzing the literature and using the models, can be compared both in terms of C[O.sub.2]/energy consumption on the reference cycles and, in a more effective and complete way, in terms of fulfillment of the design criteria and priorities coming from HoQ2.1, taking into account not only the GHG emission or fuel consumption but also other technical characteristics linked to the purchase reasons of the vehicle (e.g. performance, fun to drive, etc.). In addition to the individual technologies, some interesting combinations among them and some powertrain architectures listed in the previous chapter are considered in this work. The combinations and architectures have been chosen considering the time horizon of 2025 and certain assumptions on the emission legislation. The comparison was made with the selection matrix shown in Table 9. The values shown in the matrix depend on the technology effectiveness in satisfying the different design criteria. In the lower part of the matrix, the technology ranking can be found. The ranking expresses the effectiveness of the technology and it is graphically summarized by the green and light blue histograms for human and autonomous driving, respectively.
As it is clearly shown, the powertrain characteristics that impact on C[O.sub.2] and emissions are still the most important both for human and autonomous driven vehicles. Conversely, the characteristics linked to the performances are more relevant for the human driven vehicles. The most effective technology combinations include a mix able to improve the limits of gasoline engines shown in the previous chapters. In particular, the downsizing from 4 to 3 cylinders, with a second turbocharger stage (e.g. electric type), in conjunction with Miller and port water injection is a particularly interesting solution. This best engine configuration has been assessed in terms of C[O.sub.2] reduction by means of the simulation, obtaining the results shown in Figure 19. Starting from the engine and vehicle baseline, the European limits at 2020 are achievable, with a slight electrification (BSG) that can ensure a margin. Instead, to face the limits of 2025, assumed to be set to 75 g/km, the introduction of a full hybrid P2 48V 22kW architecture is not enough. Other improvements should be introduced, for example a vehicle mass reduction ( -20%), even though other technology alternatives will be explored in future works.
Figure 19. Technological path to face CO2 requirements, without penalty on engine performance and emissions. Technologies are additive, excluding the path highlighted by the arrows where the adoption of BSG or P2 48 V layout are considered as alternative. CO2 [g/km] in NDEC 1.4L 4Cyl TC 125 GDI S&S Miller 117 (CR+2) VGT 1.0L 3cyl 98 MT6+ WI DCT 7 91 BSG 85 e-sc P2 48 V 82 -20% 74 Vehicle Mass Reduction Note: Table made from bar graph.
However, another essential aspect is the value analysis (cost/benefit), where the benefit is weighted by means of the costs, summarized in the previous chapter. The Technology Value Ratio describes the relationship between the satisfaction of a requirement (benefits) and the use of resources (costs):
Technology Value Ratio = Function Effectiveness/Cost (8)
For this aim, the Function Effectiveness is measured by means of the ranking in the selection matrix, Table 9, whereas the costs are evaluated according to Table 8. By this procedure, the function/cost value, for each technology and their combinations, can be obtained. In Figure 20, the comparison of the technologies is shown both in terms of ranking from the selection matrix and of Value Ratio. According to the last index, water injection, Miller cycle and cylinder deactivation are very promising technologies. However, to achieve the C[O.sub.2] emission target a mix of engine technologies are needed even if these combinations (right side of figure) have a lower value, showing a growing powertrain cost in order to meet the legislation limits.
Alternatively, the technologies value can be defined as ratio between the percentage of C[O.sub.2] savings and cost. This evaluation is depicted in Figure 21. Assuming that 1% of C[O.sub.2] saving is slightly above 1 g/km, considering the baseline engine (125 g/km), it can be observed that all technologies and their mix are below the cost of the penalty (95[euro] /g) foreseen by the legislation. It is also important to remark that the affordable cost per gram of C[O.sub.2] saving depends on the distance of the powertrain emissions from the legislation limit. This means that the application of the technologies at higher cost could be mandatory to achieve the legislation limits, as long as the cost itself remains lower than the legislation penalty.
Future C[O.sub.2] scenarios represent a great challenge to the automotive industry. However, other end-user requirements (e.g. fun to drive, performance) have to be ta[k.sub.e]n into account. Many technological opportunities are available to improve the current powertrain, including not only electrification but also several technologies for ICE, especially for SI engines. A structured approach to link end-user requirements to powertrain technical features was proposed. In this way, an objective comparison of architectures and technologies can be carried out. The effectiveness of the technical solutions must be weighted taking into account the cost, which is one of the main purchasing drivers. The case study was a C-segment vehicle, for which a possible technology path to fulfill customer requirements and C[O.sub.2] limits in the timeframe of 2025 was presented. From the analysis, some technological improvements, like downsizing/down-speeding increase, Miller cycle, and water injection can help to achieve the 2020 standards, without an high grade of electrification. To face the further steps of C[O.sub.2] reduction at 2025, an higher electrification grade is needed; however it does not seem to be enough without other vehicle improvements (i.e. mass reduction). Moreover, other powertrain improvements, considering for instance bio-fuels or plug-in hybrid architectures, can be introduced, however these topics will be analyzed in future works.
To complete the analysis of future powertrain scenarios, the following topics will be investigated:
* improvement of the models to ta[k.sub.e] into account all significant vehicle loads (e.g. HVAC) for real driving conditions, with the aim to evaluate also the Tank To Use;
* analysis of the natural gas and bio-fuels impact , in term of performance and C[O.sub.2]/consumption;
* analysis of the benefits coming from new Low Temperature Combustion concepts (e.g. Gasoline Compression Ignition ), that could improve emissions and overall fuel efficiency up to 25%, greater than state-of-art of gasoline engine, but at lower cost than a diesel with similar efficiency;
* analysis of benefits of engine Heat Recovery systems [26,27,66], especially in hybrid electric architectures.
[3.] Sugiyama M., "The new generation of Toyota Powertrain", 25rd Aachen Colloquium Automobile and Engine Technology 2016
[4.] Bonnel, P., Giechaskiel, B., Vlachos, G., Weiss M., et al. "The Euro 6 Real-Driving Emissions (RDE) procedure for light-duty vehicles: Effectiveness and practical aspects", International Vienna Motor Symposium 2016
[5.] "Roadmap ICT for the Fully Electric Vehicle", ICT4FEV. FP7 Grant Agreement Number 260116
[6.] Contestabile, M., Offer, G., Slade, R., Jaegar, F., "Battery electric vehicles, hydrogen fuel cells and biofuels. Which will be the winner?" ICEPT Working Paper, June 2011, Ref: ICEPT/WP/2011/008
[7.] Uwe Wagner et al., "48 V P2 Hybrid Vehicle with an Optimized Engine Concept - Optimum Drivability with Excellent Fuel Economy and Cost-Efficiency", International Vienna Motor Symposium 2016
[8.] Shete, K., "Influence of Automotive Air Conditioning load on Fuel Economy of IC Engine Vehicles", International Journal of Scientific & Engineering Research, Volume 6, Issue 8, August-2015
[9.] Johnson, V., "Fuel Used for Vehicle Air Conditioning: A State-by-State Thermal Comfort-Based Approach," SAE Technical Paper 2002-01-1957, 2002, doi:10.4271/2002-01-1957.
[10.] Farrington, R., Rugh, J., "Impact of Vehicle Air-Conditioning on Fuel Economy,
[11.] Tailpipe Emissions, and Electric Vehicle Range", , Earth Technologies Forum Washington, D.C., October 31, 2000
[12.] Victor, R., Timmers, J.H., Peter A.J. Achten "Non-exhaust PM emissions from electric vehicles", , Atmospheric Environment, Elsevier, 11 March 2016
[13.] Department Of Energy,"Well-to-Wheels Greenhouse Gas Emissions and Petroleum Use for Mid-Size Light-Duty Vehicles", May 10, 2013
[14.] Edwards, R., Hass, H., Larive, J.F., Lonza, L., Maas, H., Ric[k.sub.e]ard, D. "Well-To-Wheels Report" (Version 4.a), JRC Technical Report 2014, Report EUR 26236 EN
[15.] Brunner, H., Hirz, M., Fischer, P., "C[O.sub.2] emissions of different technologies in passenger cars at real user scenarios in the product life cycle", International Vienna Motor Symposium 2016
[16.] Kirwan, J., Shost, M., Roth, G., and Zizelman, J., "3-Cylinder Turbocharged Gasoline Direct Injection: A High Value Solution for Low CO2 and NOx Emissions," SAE Int. J. Engines 3(1):355-371, 2010, doi:10.4271/2010-01-0590.
[17.] Boggs, D., Dorobantu, M., German, J., Isenstadt A., Watson, T., "Downsized, boosted gasoline engines", ICCT Working Paper 2016-21
[18.] Frost & Sullivan, "Role of Exhaust After-treatment Solution in Reducing Tailpipe Emissions 2016 and 2020 Global Outlook", August 2015
[19.] Frost & Sullivan, "Analysis of European Mar[k.sub.e]t Potential for Selective Catalytic Reduction (SCR) Technology for Passenger Vehicles", December 2014
[20.] Habs Moy, M., "Commercial Gas Turbine Engine Platform Strategy and Design", Master of Science In Engineering & Management, MIT, 2000
[21.] Reichert, E., Combustion Engine Development Utilizing Design for Six Sigma
[22.] Deloitte, "Global Automotive Consumer Study Exploring consumers' mobility choices and transportation decisions", 2014
[23.] Hirsch, R., Nase, A., Straschill, R., Will, P., "Connected and Automated Vehicles: New Degrees of Freedom to Improve the Powertrain", 25th Aachen Colloquium Automobile and Engine Technology 2016
[24.] Gohlich, D., Grabener, S., "Identification of User-Oriented Electric Commercial Vehicle Concepts with a Particular Focus on Auxiliaries", 25th Aachen Colloquium Automobile and Engine Technology 2016
[25.] Fiorani, P., Gambarotta, A., Lucchetti, G., Ausiello, F. et al., "A detailed Mean Value Model of the exhaust system of an automotive Diesel engine," SAE Technical Paper 2008-28-0027, 2008, doi:10.4271/2008-28-0027.
[26.] Gambarotta, A., Lucchetti, G., Fiorani, P., De Cesare, M. et al., "A thermodynamic Mean Value Model of the inta[k.sub.e] and exhaust system of a turbocharged engine for HiL/SiL applications.," SAE Technical Paper 2009-24-0121, 2009, doi:10.4271/2009-24-0121.
[27.] Arsie, I., Cricchio, A., Pianese, C., De Cesare, M. et al., "A Comprehensive Powertrain Model to Evaluate the Benefits of Electric Turbo Compound (ETC) in Reducing C[O.sub.2] Emissions from Small Diesel Passenger Cars," SAE Technical Paper 2014-01-1650, 2014, doi:10.4271/2014-01-1650.
[28.] Arsie, I., Cricchio, A., Marano, V., Pianese, C. et al., "Modeling Analysis of Waste Heat Recovery via Thermo Electric Generators for Fuel Economy Improvement and C[O.sub.2] Reduction in Small Diesel Engines," SAE Int. J. Passeng. Cars - Electron. Electr. Syst. 7(1):246-255, 2014, doi:10.4271/2014-01-0663.
[29.] Sorrentino, M., Mauramati, F., Arsie, I., Cricchio, A. et al., "Application of Willans Line Method for Internal Combustion Engines Scalability towards the Design and Optimization of Eco-Innovation Solutions," SAE Technical Paper 2015-24-2397, 2015, doi:10.4271/2015-24-2397.
[30.] Brooker, A., Gonder, J., Wang, L., Wood, E. et al., "FASTSim: A Model to Estimate Vehicle Efficiency, Cost and Performance," SAE Technical Paper 2015-01-0973, 2015, doi:10.4271/2015-01-0973.
[31.] "FEV-ICCT, "2025 Passenger Car and Light Commercial Vehicle Powertrain Technology Analysis, Final Report / September 2015
[32.] Esposito, S., Hoppe, P., Kumagai, T., Mally, M., et al., "Model Based Development for Extreme Lean Burning Gasoline Powertrains-Combining Best Efficiency with Lowest Emissions", 2016 JSAE Congress, Sapporo, Japan, 19 Oct 2016 21 Oct 2016, 907-912.
[33.] Andersson, J., Keenan, M., Osborne, R., Rouaud, C., "Development of lean stratified turbocharged gasoline engines for 2018 SOP", SIA Technical paper, R-2013-06-22.
[34.] Lenga, H., Hartmann, R., Hyouk Min, B., Grimm, J., Winkler, M., "Low Pressure EGR for Downsized Gasoline Engines, 23rd Aachen Colloquium Automobile and Engine Technology 2014.
[35.] Heywood, J.B., Internal combustion engine fundamentals, New York, 1988
[36.] Frohnmaier, M., Hettinger, A., Pauer, T., Schenk, P., Kampmann, S., Walther, J. "Optimization of Gasoline Engines by Water Injection", International Vienna Motor Symposium 2016.
[37.] Thewes, M., Baumgarten, H., Scharf, J., Birmes, G., et al., "Water Injection - High Power and High Efficiency Combined", 25th Aachen Colloquium Automobile and Engine Technology 2016.
[38.] Hoppe, F., Thewes, M., Baumgarten, H., Dohmen, J., "Water injection for gasoline engines: Potentials, challenges, and solutions", International Journal of Engine Research, 2015, doi: 10.1177/1468087415599867
[39.] Cavina N., Businaro A., Rojo N., De Cesare M, "Investigation of Water Injection Effects on Combustion Characteristics of a GDI TC Engine," SAE Int. J. Engines 10(4):2209-2218, 2017.
[40.] Niizato, T., Yasui, Y., Urata, Y., Wada, Y., Jono, M., Nakano, K., Taguchi, M. "Honda's New Turbo-GDI Engine Series for global application", International Vienna Motor Symposium 2016.
[41.] Hwang, I., et al., "Hyundai-Kia's Highly Innovative 1.6L GDI Engine for Hybrid Vehicle", International Vienna Motor Symposium 2016
[42.] Heiduk, T., Wei[beta], U., Frohlich, A., Helbig, J., Ziilch, S., Lorenz, S., "The new Audi V8 TDI engine", International Vienna Motor Symposium 2016.
[43.] Ohta, Y., Fushiki, S., Matsuo, S., "The New PRIUS Powertrain: The New 1.8L ESTEC 2ZR-FXE Engine with the New Generation Hybrid System", International Vienna Motor Symposium 2016.
[44.] Kyoung-Pyo Ha, Woo Tae Kim, In Sang Ryu, You Sang Son, "Development of Continuously Variable Valve Duration (CVVD) Engine", 25th Aachen Colloquium Automobile and Engine Technology 2016.
[45.] Kapus P.E., Spanner C., Graf B., Fraidl G.K. "Cylinder Deactivation with 4 Cylinder Engines - An Alternative to 2 Cylinders?", International Vienna Motor Symposium 2011
[46.] Isenstadt A., German J., Dorobantu M.," Naturally aspirated gasoline engines and cylinder deactivation", ICCT WORKING PAPER 2016-12
[47.] Martins, J., Uzuneanu, K., Ribeiro, B., and Jasasky, O., "Thermodynamic Analysis of an Over-Expanded Engine," SAE Technical Paper 2004-01-0617, 2004, doi:10.4271/2004-01-0617.
[48.] Eichler, F., et al, "The New EA211 TSI[R] evo from Volkswagen", 37th International Vienna Motor Symposium, 2016
[49.] Song, S., and Zhang, H., "Performance Study for Miller Cycle Natural Gas Engine
[50.] Based on GT-Power", Journal of Clean Energy Technologies, Vol. 3, No. 5, September 2015
[51.] "2017 Audi A4 ultra with Millerized 2.0 TFSI offers 31 mpg combined;highest EPA-estimated fuel economy in competitive segment," Green Car Congress, accessed 17 August 2016, http://www.greencarcongress.com/2016/08/20160817-audi.html
[52.] Helfried Sorger, Wolfgang Schoffmann, Siegfried Losch, Andreas Krobath, Wolfgang Unzeitig, Gunter Fraidl, Paul Kapus, Alois Furhapter, "AVL's VCR System Modular and Cost Efficient C[O.sub.2] Reduction" AVL List GmbH, Graz, Austria, 2017-01-0634
[53.] Berger, J., Nowak, L., Pogam, M., Groupe, B. De Gooijer "PSA Gasoline Engine Next Generation: Gomecsys VCR Concept as a Solution?", 25th Aachen Colloquium Automobile and Engine Technology 2016.
[54.] Ferrey, P., Miehe, Y., Constensou, C., Collee V., "Potential of a Variable Compression Ratio gasoline SI Engine with very high Expansion Ratio and Variable Valve Actuation", Technical paper, SIA, 12/04/2013
[55.] Hirose, I., "Mazda 2.5L SKYACTIV-G Engine with New Boosting Technology", 37th International Vienna Motor Symposium, 2016
[56.] Thewes, M., Baumgarten, H., Dohmen, J., Uhlmann, T., et al., "Gasoline Combustion Systems Beyond 2020", 23rd Aachen Colloquium Automobile and Engine Technology 2014
[57.] Durrieu D., Criddle M., Menegazzi P., Wu Y., "Electric supercharger: new electric and boosting architecture for downsizing and downspeeding", SIA Technical paper, R-2013-05-14
[58.] Stapelbroek, M., Birmes, G., Thewes, M., Espig, M. et al., "Fuel Consumption Reduction and Performance Improvement by Electric Driven Supercharger", 25th Aachen Colloquium Automobile and Engine Technology 2016
[59.] Bevilacqua, V., Jacobs, E., Grauli, G., Wust J. Future of Downsizing: Fuel Consumption Improvement in NEDC as well as in Customer Operating Conditions, SIA Technical paper 2013, R-2013-06-30
[60.] Chadwell, C., Alger, T., Zuehl, J., and Gu[k.sub.e]lberger, R., "A Demonstration of Dedicated EGR on a 2.0 L GDI Engine," SAE Int. J. Engines 7(1):434-447, 2014, doi:10.4271/2014-01-1190.
[61.] Kapadia, J., Kok, D., Jennings, M., Kuang, M. et al., "Powersplit or Parallel - Selecting the Right Hybrid Architecture," SAE Int. J. Alt. Power. 6(1):68-76, 2017, doi:10.4271/2017-01-1154.
[63.] Osborne, R., Downes, T., O'Brien, S., Pendlebury, K. et al., "A Miller Cycle Engine without Compromise - The Magma Concept," SAE Int. J. Engines 10(3):2017, doi:10.4271/2017-01-0642.
[64.] Darlington, T., Herwick, G., Kahlbaum, D., and Drake, D., "Modeling the Impact of Reducing Vehicle Greenhouse Gas Emissions with High Compression Engines and High Octane Low Carbon Fuels," SAE Technical Paper 2017-01-0906, 2017, doi:10.4271/2017-01-0906.
[65.] Sellnau, M., Foster, M., Moore, W., Sinnamon, J. et al., "Second Generation GDCI Multi-Cylinder Engine for High Fuel Efficiency and US Tier 3 Emissions," SAE Int. J. Engines 9(2):1002-1020, 2016, doi:10.4271/2016-01-0760.
[66.] Arsie, I., Cricchio, A., Pianese, C., Ricciardi, V. et al., "Modeling and Optimization of Organic Rankine Cycle for Waste Heat Recovery in Automotive Engines," SAE Technical Paper 2016-01-0207, 2016, doi:10.4271/2016-01-0207.
Matteo De Cesare
AD - Autonomous Driving
AFR - Air-Fuel Ratio
AMEP - Available Mean Effective Pressure
AMT - Automatic Manual Transmission
BMEP - Brake Mean Effective Pressure
BSFC - Brake Specific Fuel Consumption
BSG - Belt Starter Generator
CO - Carbon monoxide
CoV - Coefficient of Variation
C[O.sub.2] - Carbon dioxide
CR - Compression Ratio
DCT - Dual Clutch Transmission
DSF - Dynamic Skip Fire
DWI-M - Direct Water Injection - Mixture
DWI-S - Direct Water Injection - Separate
EGR - Exhaust Gas Recirculation
EM - Electric Motor
FR - Friction Reduction
GDI - Gasoline Direct Injection
GPF - Gasoline Particulate Filter
HC - Hydrocarbon
HD - Human Driving
HLHV - Fuel Lower Heating Value
HoQ - House of Quality
HP - High Pressure
ICE - Internal Combustion Engine
LP - Low Pressure
LNT - Lean N[O.sub.x] Trap
MR - Market Research
NEDC - New European driving cycle
N[O.sub.x] - Nitrogen Oxides
NVH - Noise Vibration Harshness
OEMs - Original Equipment Manufacturers
PM - Particulate mass
PN - Particulate Number
[p.sub.exh] - Exhaust Manifold Pressure
[p.sub.intake] - Intake manifold pressure
[p.sub.loss] - Mean effective pressure lost
PWI - Port Water Injection
PWT - Powertrain
QFD - Quality Function Deployment
r - Compression ratio
RDE - Real Driving Emissions
SCR - Selective Catalytic Reduction
S - Piston stroke
SI - Spark Ignition
Te - Effective torque
TTU - Tank to Use
TTW - Tank to Wheel
TWC - Three-way catalyst
VCR - Variable Compression Ratio
Vd - Displacement
[V.sub.pist] - Average piston speed
VVA - Variable valve actuator
WLTP - Worldwide harmonized light vehicles test procedure
WHR - Waste heat recovery
WI - Water Injection
WTT - Wheel to Tank
WTW - Well to Wheel
[[epsilon].sub.tr] - Trapped compression ratio
[[eta].sub.idc] - Thermodynamic efficiency
[gamma] - Specific heat ratio
[sigma] - Expansion - Compression ratio
[omega] - Engine angular velocity
BMEP can be expressed as:
BMEP = [[eta].sub.cv] -AMEP-[p.sub.loss] - [p.sub.pumping] (9)
BMEP = 4[pi]/[V.sub.d] * [T.sub.e] (10)
AMEP = 4[pi][H.sub.LHV]/[V.sub.d] * [[??].sub.f]/[omega] (11)
the available mean effective pressure 'AMEP' represents the available chemical energy granted by the fuel flow rate [[??].sub.f]
[p.sub.loss] = [p.sub.loss]([v.sub.pist])
[P.sub.pumping] [p.sub.exh] - [p.sub.intake] (12,13)
the [[eta].sub.cv] is the global engine conversion efficiency as function of the 'e' energy conversion efficiency parameter, which is the product between the thermodynamic and combustion efficiency and [[eta].sub.idc], the corrected theoretical efficiency:
[[eta].sub.cv] = e([v.sub.pist],amep) * [[eta].sub.idc]
(14)[v.sub.pist] is the average piston speed defined as:
[v.sub.pist] = S/[pi] * [omega] (15)
where S is the engine stro[k.sub.e]. The term [[eta].sub.idc] is corrected to ta[k.sub.e] into account the Miller/Atkinson cycle:
[[eta].sub.idc] = 1- 1/[r.sup.[gamma]-1.sub.g] - f([sigma]) (16)
where, as described in :
f([sigma]) = [[sigma].sup.[gamma]]([gamma]-1)-[gamma]*[[sigma].sup.[gamma]-1]+1/([gamma]-1)*[[sigma].sup.[gamma]-1]*B (17)with
(B = [H.sub.LHV]/1+AFR/[RT.sub.air])
(18)and [sigma]=[r.sub.g]/[r.sub.tr], where [r.sub.g] and [r.sub.tr] are the geometric and trapped compression ratios, respectively.
By replacing (10), (11), (12), (13) and (14) in (9) it is possible to obtain the effective torque [T.sub.e]:
[T.sub.e] = e[[eta].sub.idc][H.sub.LHV][[??].sub.f]/[omega] - [p.sub.loss] [V.sub.d]/4[pi] - ([p.sub.exh]-[p.sub.intake])[V.sub.d]/4[pi]
Matteo De Cesare
MAGNETI MARELLI SpA - Div. Powertrain
University of Bologna
MAGNETI MARELLI SpA - Div. Powertrain
Table 1. Measurable Powertrain Requirements and Priority from HoQ1 Priority from MR Needs/ Measurable Needs Vehicle Requests and Request Controlled (from HoQ1, Vehicle Level) by Driver End-User Low Noise 1 Noise Emission Level 4 High 2 Accel.Time 0 -->100 Km/h 3 Performance 3 Accel.Time 80 -->120 Km/h 3 (elasticity) 4 Max Vehicle Speed 3 High 5 Accel. Time 0 -->50 Km/h 4 Comfort (direct transmission) 6 Time w/o acceleration 4 during load Req, 7 Low Vibration 4 8 Cabin Comfort 4 High 9 Mileage w/o failure 4 Reliability Easy to use 10 Energy Refilling autonomy 5 11 Maintenence intervall/cost 4 12 Energy Refilling Time 4 Low Fuel 13 Fuel Economy 4 Consumption Standards Low Emission 14 Toxic Emission according 7 to Standard (WLTP and RDE) 15 EOBD/OBD2 compliance 7 16 CO2 according to Standard 7 Safety 17 Mileage w/o critical event 7 Priority from MR Needs/ Autonomous Unit Requests Driving End-User Low Noise 5 [dB] High 2 [sec] Performance 2 [sec] 2 [km/h] High 2 [sec] Comfort 5 [ms] 5 [dB] 5 [[degrees]C] & [hum] High 5 [km] Reliability Easy to use 5 [km] 5 [km] 4 [min] Low Fuel 5 [km/l] Consumption Standards Low Emission 7 [mg/km] 7 [mg/km] 7 [g/km] Safety 7 [km] Table 3. Functions in hybrid electric powertrains Functions ICE P0 Belt P0 Belt Hybrid Hybrid Range 12V 48V P2/P3, Plug-in Extender Power EV Split Cold Start Y P Y Y Y Stop & Start Y Y Y Y Y Y Coasting Y (*) Y Y Y Y Y High Efficiency Y Y Y Y Y Electric Power Generation Regenerative P Y Y Y Y Braking Torque Assist P Y Y Y - Electric Take-off Y Y Y Y Electric Driving Y Y Y Plug-in Electric Y Y Charging (*) with e-clutch or AMT/DCT Y=Yes P=Partially Table 4. Selection Matrix for Powertrain Architectures POWERTRAIN Relevance 1 2 ARCHITECTURES (0/100) ICE GDI Hybrid Turbo + DCT P0/P2 48V Powertrain Charcteristics Human (from HOQ2.1) Driving 1 Powertrain Noise Level 10 1 2 2 Powertrain Time to Full Power 15 3 5 3 Low-end Torque (at wheel) 20 3 7 4 Speed Range of High Torque 16 5 7 5 Maximum Power 21 9 9 7 Time w/o Torque delivery 18 5 7 during load Req. 8 Powertrain Vibration 13 1 2 9 Tank to Wheel Maximum 25 5 5 Efficiency 10 Tank to Wheel Efficiency at 29 3 7 part/low load 11 NOx Emission according to 14 1 3 RDE 12 HC/CO Emission according to 16 5 7 RDE 13 Soot/PN according to RDE 15 1 3 14 CO2 on WLTC 23 5 7 15 Heating Power 5 9 9 17 Efficiency of Electric Power 21 5 5 Gen. 18 Energy Storage Capacity 15 7 7 19 Energy Refilling Time 14 9 9 25 % Braking energy recovered 23 0 5 26 PWT Size 5 7 5 27 PWT Weight 38 7 5 - PWT Architecture Ranking [0/100] 55 72 POWERTRAIN 3 4 ARCHITECTURES Hybrid Hybrid Power P2 HV Split HV Powertrain Charcteristics (from HOQ2.1) 1 Powertrain Noise Level 3 7 2 Powertrain Time to Full Power 7 9 3 Low-end Torque (at wheel) 7 9 4 Speed Range of High Torque 7 9 5 Maximum Power 9 9 7 Time w/o Torque delivery 7 9 during load Req. 8 Powertrain Vibration 3 7 9 Tank to Wheel Maximum 7 9 Efficiency 10 Tank to Wheel Efficiency at 9 9 part/low load 11 NOx Emission according to 5 5 RDE 12 HC/CO Emission according to 9 9 RDE 13 Soot/PN according to RDE 5 5 14 CO2 on WLTC 7 7 15 Heating Power 9 9 17 Efficiency of Electric Power 5 7 Gen. 18 Energy Storage Capacity 7 7 19 Energy Refilling Time 9 9 25 % Braking energy recovered 7 7 26 PWT Size 5 3 27 PWT Weight 5 5 PWT Architecture Ranking [0/100] 82 93 POWERTRAIN 5 6 ARCHITECTURES Hybrid P2/Power Range Extender Split Plugin EV Plugin Powertrain Charcteristics (from HOQ2.1) 1 Powertrain Noise Level 7 7 2 Powertrain Time to Full Power 9 9 3 Low-end Torque (at wheel) 9 9 4 Speed Range of High Torque 9 9 5 Maximum Power 9 9 7 Time w/o Torque delivery 9 9 during load Req. 8 Powertrain Vibration 7 7 9 Tank to Wheel Maximum 9 9 Efficiency 10 Tank to Wheel Efficiency at 9 9 part/low load 11 NOx Emission according to 5 5 RDE 12 HC/CO Emission according to 9 9 RDE 13 Soot/PN according to RDE 5 5 14 CO2 on WLTC 9 9 15 Heating Power 9 7 17 Efficiency of Electric Power 7 9 Gen. 18 Energy Storage Capacity 5 7 19 Energy Refilling Time 5 3 25 % Braking energy recovered 7 7 26 PWT Size 3 3 27 PWT Weight 3 3 PWT Architecture Ranking [0/100] 89 90 Table 5. Summary of Technology benefits and drawbacks ICE Advantages Influenced Technology Engine Zones (Fig. 5) GDI lean - Pumping losses reduction at All engine Combustion low loads operating area - Higher ratio of specific heats with - Knock mitigation higher impact - Lower heat losses towards on 1, 2 and 3 cylinder walls Miller - Pumping losses reduction at All engine Cycle low loads operating area - Higher geometric CR enabled with - Knock mitigation higher impact - Lower peak pressure and on 1 and 2 temperature - VGT enabling with higher cylinder displacement VCR - Higher efficiency at low loads Zones 1 and 3 due to the high CR - Very effective coupling with Miller Water - Very effective way to Zones 2,3 and 4. Injection mitigate knock phenomena If higher CR is - CR can be increased adopted,the - Lower exhaust temperature impact is on the in full load avoiding whole engine enrichment and enabling VGT operative area Cylinder - Pumping and heat losses Zone 1 Deactivation reduction - Higher efficiency at low loads External EGR - Knock mitigation Zone 2,3 and 4 - Lower heat losses - De-throttling at partial loads Multistage - Low end torque increase All the engine Air Charging - Downsizing and down-speeding operative area enabler - Scavenging reduction or elimination - Drivability improvement ICE Drawbacks Technology GDI lean - Expensive exhaust after-treatment Combustion - Higher cycle-to-cylce variation - Higher boosting demand - Combustion chamber and piston re-design - More advanced ignition system requirements Miller - Lower volumetric efficiency, Cycle more boosting required - Displacement needs to be increased to recover the engine power leading to higher friction losses VCR - High cost and complexity - Complex handling during switching from high CR to low CR Water - Water consumption leads to Injection the need of producing it on board - Corrosion needs to be handled Cylinder - Torque fluctuation, vibrations Deactivation and noise to be handled - Low effectiveness in coupling with other technologies - Cost and packaging impact External EGR - Reduction of maximum power with a given turbocharger layout - Negative impact on transient response Multistage Cost and packaging impact Air Charging Added complexity for the engine control Table 6. ICE technologies and main parameters ICE Technology Linked Parameter Cylinder deactivation (VVA) [V.sub.d] [k.sub.f],[k.sub.p] in part load operations Cylinder deactivation [V.sub.d] [k.sub.f],[k.sub.p] in part load (2CATs) operations Miller/Atkinson Cycle [sigma] and therefore [[eta].sub.idc], [k.sub.p] in the part load operations, [[eta].sub.v], [k.sub.e] depending on the strategies used External EGR [k.sub.e] for the lower heat losses and knock mitigation, [k.sub.p],[gamma] Lean Combustion [k.sub.e], [k.sub.p] especially at part low operations and knock resistance Water Injection [k.sub.e],y and knock resistance [right arrow] higher [r.sub.g] allowed [right arrow] higher [[eta].sub.idc] Variable Compression R. [[eta].sub.idc], [k.sub.f] 2stage-Air Charging Extreme downsizing [right arrow] [V.sub.d] and [pme.sub.req] Table 7. Fuel benefit of ICE technologies vs. reference engine estimated and from literature data Technology CO2/Fuel Saving in NEDC [%] Model Literature BaselineTC, GDI, S&S 1.4L, 4Cyl 1.01/2.01, TC, 3/4 Cyl Miller Cycle (CR+2) 6 5/12 VCR- 2stage (CR+2) 4 4/9 Port Water Inj. (CR +2) 4 4.4 Electronic Cyl Deactivation 5 7 Cylinder Deactivation 7 - LP-EGR/ HP-EGR 3.5 - Miller Cycle+Wl 10 - Miller Cycle+VCR 10 - 2st.-Turbo (3cyl + downsp) 12 12/18 eBooster (3cyl + downsp) 12 12/18 BSG 8 5/10 BSG+e Booster+ 14 12/18 3 cyl+ downspeeding BSG+eBooster+WI (CR +2) 15 - 3 cyl+ downspeeding P2 48Volt 22kw 20 15/20 P2 48V22kw+Miller+WI 28 - Technology CO2/Fuel Saving in WLTC [%] Model Literature BaselineTC, GDI, S&S 1.4L 4Cyl 1.0L/2.0L, TC, 3/4 Cyl Miller Cycle (CR+2) 5 5/12 VCR- 2stage (CR+2) 5.5 4/9 Port Water Inj. (CR +2) 6.0 4/6.5 Electronic Cyl Deactivation 5 Cylinder Deactivation 6 10 LP-EGR/ HP-EGR 3 3/4 Miller Cycle+Wl 9 - Miller Cycle+VCR 9 5/9 2st.-Turbo (3cyl + downsp) 12 - eBooster (3cyl + downsp) 12 - BSG 6 5/6 BSG+e Booster+ 10 6 3 cyl+ downspeeding BSG+eBooster+WI (CR +2) 12 - 3 cyl+ downspeeding P2 48Volt 22kw 18 - P2 48V22kw+Miller+WI 24 - Table 8. Cost of analyzed technologies Technology Baseline [DELTA] Remark Average Cost [[euro]] GDI Lean 1-stageTC, GDI, 4 Cyl 385 including LNT catalyst Miller Cycle 1-stage TC, GDI, 4 Cyl, 200 considering a VVA 2st of TC or desplacement increasing VCR- 2stages 1-stage TC, GDI, 4 Cyl 125 VCR- Continous 1-stageTC, GDI, 4Cyl 350 Port Water Injection 1-stage TC, GDI, 4 Cyl 95 DWI - Separate 1-stage TC, GDI, 4 Cyl 180 DWI - Mixture 1-stage TC, GDI, 4 Cyl 130 Electronic Cyl Deact. 1-stage TC, GDI, 4 Cyl 100 Cylinder Deactivation 1-stage TC, GDI, 4 Cyl, 200 WA LP-EGR 1-stage TC, GDI, 4 Cyl 115 HP-EGR 1-stage TC, GDI, 4 Cyl 115 2stages-Turbo 1-stage TC, GDI, 4 Cyl 200 eBooster 1-stage TC, GDI, 4 Cyl 400 including Li battery BSG 1-stage TC, GDI, 4 Cyl 700 including Li battery BSG+eBooster 1-stage TC, GDI, 4 Cyl 1000 including Li battery Table 9. Selection Matrix of Technologies to ICE improvement (2025 timeframe) SELECTION MATRIX COMBUSTION TECHNOLOGIES 1 ENGINE Relevance Port Water ARCHITECTURES (0+100) Injection Powertrain Charcterlstics Human Auton. (from HoQ2.1) Driving Driving 1 Powertrain Noise Level 10 13 2 Powertrain Time to Full Power 15 9 3 3 Low-end Torque (at wheel) 20 14 7 4 Speed Range of High Torque 16 9 3 5 Maximum Power 21 14 7 7 Time wlo Torque delivery 18 15 during toad Req. 8 Powertrain Vibration 13 16 9 Tank to Wheel Maximum 23 22 5 Efficiency 10 Tank to Wheel Efficiency at 29 24 3 part/low load 11 NOx Emission according to 14 16 3 RDE 12 HC/CO Emission according to 16 18 RDE 13 Soot/PN according to RDE 15 17 14 CO2 on WLTC 23 19 5 15 Heating Power 5 5 26 PWT Size 5 8 -1 27 PWT Weight 38 35 - - Tecnology Ranking [0/100] 25 - Human D. Tecnology Ranking [0/100] 21 - Auton. D. SELECTION MATRIX COMBUSTION TECHNOLOGIES 2 3 4 5 ENGINE DWI- DWI- GDI Miller ARCHITECTURES Saparate Mixture Lean /Atkinson Cycle Powertrain Charcterlstics (from HoQ2.1) 1 Powertrain Noise Level 2 Powertrain Time to Full Power 3 3 5 3 Low-end Torque (at wheel) 7 7 -1 4 Speed Range of High Torque 3 3 5 Maximum Power 7 7 7 7 Time wlo Torque delivery during toad Req. 8 Powertrain Vibration 9 Tank to Wheel Maximum 5 5 3 9 Efficiency 10 Tank to Wheel Efficiency at 3 3 7 7 part/low load 11 NOx Emission according to 3 3 -1 1 RDE 12 HC/CO Emission according to 3 RDE 13 Soot/PN according to RDE 1 -1 14 CO2 on WLTC 5 5 7 9 15 Heating Power 26 PWT Size -1 -1 -3 -5 27 PWT Weight -1 Tecnology Ranking [0/100] 25 25 17 25 - Human D. Tecnology Ranking [0/100] 21 21 15 22 - Auton. D. SELECTION MATRIX AIR DELIVERY TECHS. 9 10 11 ENGINE 3cyl + 3cyl + Cylnder ARCHITECTURES 2-stage 2-stage Deact. TC TC (VVA) (BMEP (BMEP 27 bar) 35 bar) Powertrain Charcterlstics (from HoQ2.1) 1 Powertrain Noise Level -1 -1 -1 2 Powertrain Time to Full Power 3 3 -1 3 Low-end Torque (at wheel) 5 7 4 Speed Range of High Torque 3 7 5 Maximum Power 7 9 7 Time wlo Torque delivery 1 3 during toad Req. 8 Powertrain Vibration -1 9 Tank to Wheel Maximum 5 Efficiency 10 Tank to Wheel Efficiency at 7 9 9 part/low load 11 NOx Emission according to RDE 12 HC/CO Emission according to 1 RDE 13 Soot/PN according to RDE 1 1 14 CO2 on WLTC 7 9 5 15 Heating Power 26 PWT Size 27 PWT Weight -3 -1 Tecnology Ranking [0/100] 25 31 15 - Human D. Tecnology Ranking [0/100] 20 24 14 - Auton. D. SELECTION MATRIX AIR DELIVERY TECHS. MIX TECHS. 12 13 14 ENGINE Electronic 1+9+11 1+5+9 ARCHITECTURES CD Powertrain Charcterlstics (from HoQ2.1) 1 Powertrain Noise Level -1 -1 2 Powertrain Time to Full Power 5 7 3 Low-end Torque (at wheel) 7 7 4 Speed Range of High Torque 5 5 5 Maximum Power 7 5 7 Time wlo Torque delivery 1 during toad Req. 8 Powertrain Vibration -1 -1 9 Tank to Wheel Maximum 5 5 9 Efficiency 10 Tank to Wheel Efficiency at 9 7 7 part/low load 11 NOx Emission according to 3 3 RDE 12 HC/CO Emission according to 1 1 1 RDE 13 Soot/PN according to RDE 1 1 14 CO2 on WLTC 5 7 7 15 Heating Power 26 PWT Size -1 -1 -1 27 PWT Weight -1 -1 Tecnology Ranking [0/100] 16 32 37 - Human D. Tecnology Ranking [0/100] 15 26 31 - Auton. D. SELECTION MATRIX TECHS. MIX 15 16 ENGINE 5+6+8+9 4+6+9 ARCHITECTURES Powertrain Charcterlstics (from HoQ2.1) 1 Powertrain Noise Level -1 -1 2 Powertrain Time to Full Power 5 5 3 Low-end Torque (at wheel) 5 5 4 Speed Range of High Torque 5 5 5 Maximum Power 5 5 7 Time wlo Torque delivery during toad Req. 8 Powertrain Vibration -1 -1 9 Tank to Wheel Maximum 9 5 Efficiency 10 Tank to Wheel Efficiency at 7 7 part/low load 11 NOx Emission according to 3 RDE 12 HC/CO Emission according to RDE 13 Soot/PN according to RDE 1 1 14 CO2 on WLTC 5 5 15 Heating Power 26 PWT Size -1 -3 27 PWT Weight -1 -1 Tecnology Ranking [0/100] 31 25 - Human D. Tecnology Ranking [0/100] 25 20 - Auton. D. Table 10. Summary of technology impact from literature analysis Analyzed Reference Vehicle Engine Other Engine Technology Work Segment Refernce Modification Port Water 30 TC,GDI, CR=+2 130 Kw Injection 50 1.0LTC,GDI, SA optimizatin CR=11 6 Direct 32 E 1.6L,TCGDI, VCR 14.7/10.7 Water VCR 13/9,5 Injection AT 8-speed GDI Lean 29 D 2.0 L, TC, GDI, 200KW 27 c 1.0L.TC,GDI, CR=12 WL,CR=10 44 1.4L,TC,GDI 1.5L,CR-12.5, VGT,CD Miller 46 D 2.0 L, TC, CR- 9.6 CR-11.7 Cycle 58 D 2.0L,TC,GDI CR=13.1 4 Cyl 4cyl, CR=10.5 Supercharger CR=13.1. 3Cyl, Supercharger 37 c 1.4L.TC,GDI 1.6L, CR=13. RC=10,S/B=l.17 S/B=1.35, Friction Reduction LP-EGR 30 C 1.2L,TC,GDI RC = 10.5 Dedicated 54 2.0 L,TC Supercharger EGR added, CR=11.7 HP-EGR 49 SUV 3.7L,V6, NA,CR=13 2.SL,TC,Scav. GDI,CR=10 Variable 47 2.0L.TC, GDI, Miller. 2step Compres. CR=9.5 VCR=9.5/14 Ratio 40 1.6L,TC.GDI,CR=10 Continous VCR=11 27 CR=13.1 2 step VCR=12.1 Cylinder 41 c 1.4L,TC.GDI Electronic Cylinder Deactivation Deactiv 57 Dynamic Skip Fire 48V BSG 14 51 B NA,PFI 14 14 c 2.0 L, 4 cyl, PFI 1.01, 3 Cyl, TC, GDI 48V BSG + 14 c 1.01, 3Cyl,TC, GDI Combustion & eBooster Thermal opt. 14 1.7EL,TC,ODI 14 TC, GDI 51 Sport TC, GDI, AT Engine Characteristics Analyzed Low-end [DELTA] [DELTA] Tank [DELTA]Tank Technology Torque Maximum to Wheel to Wheel Improvement Power Maximum Efficiency at Efficiency part/low load [%] [%] Port Water 13 Injection 1.8 Direct Water Injection GDI Lean 5 10/30 Miller 25 Cycle 1.8/9.4 5.3/21.8 2 LP-EGR 1 Dedicated 10 EGR HP-EGR 30 30 Variable Compres. Ratio 10 4/5 Cylinder Deactiv 48V BSG 48V BSG + eBooster 200 g/kwh 50 Engine Characteristics Analyzed CO2 Benefit CO2 Benefit Technology in WLTC in NEDC [%] [%] Port Water 4 Injection Direct 6.5 4.4 Water Injection GDI Lean 12 14 3.8 4.4 10 Miller 21 Cycle 6.3 12.5 LP-EGR 4 Dedicated EGR HP-EGR Variable 5/9 Compres. Ratio 4.2/6.2 Cylinder 7 Deactiv 10/20 48V BSG 5.1/6.2 4 10 50 48V BSG + 17 eBooster 6 15 Table 11. House of Quality 2.1 Priority from MR Needs/ Measurable Needs Vehicle Requests and Request Controlled (from HoQ1, by Driver Vehicle Level) Low Noise 1 Noise Emission Level 2 Accel.Time 0-->100Km/h High 3 Accel Time 80->120 Km/h Performance elasticity) 4 Max Vehicle Speed 5 Accel. Time 0-->50Km/h direct transmission) End-User High 6 Time w/o acceleration Comfort during load Req, Low Vibration Cabin Comfort High 9 Mileage w/o failure Reliability 10 Energy Refilling autonomy Easy to use 11 Maintenence intervalt/cost 12 Energy Refilling Time Low Fuel 13 Fuel Economy Consumption Standards 14 Toxic Emission according to Standard (WLTP and RDE) Low Emission 15 EOBD/OBD2 compliance 1G C02 according to Standard Safety 17 Mileage w/o critical event Powertrain characteristic ranking - Human driving Powertrain Characteristic ranking - Autonomous driving Needs/ Autonomous Unit Requests Driving Low Noise 4 5 [dB] 3 2 [sec] High 3 2 [sec] Performance 3 2 [km/h] 4 2 [sec] End-User High 4 5 [mo] Comfort 4 8 [dB] 4 5 [[degrees]C] & [hum] High 4 5 [km] Reliability 5 5 [km] Easy to use 4 5 [km] 4 4 [min] Low Fuel 4 5 [km/l] Consumption Standards 7 7 [mg/km] Low Emission 7 7 [mg.'km] 7 7 [g/km] Safety 7 7 [km] Powertrain characteristic ranking - Human driving Powertrain Characteristic ranking - Autonomous driving HoQ2.1 Powertrain Level Traction Power Generation/Control 1 2 3 Powertrain Powertrain Time Low-end Torque Noise Level to Full Power (at wheel) [down arrow] [up arrow] [up arrow] Needs/Requests Low Noise 9 9 9 High 3 Performance 9 9 End-User High 3 1 Comfort 9 High Reliability Easy to use Low Fuel 1 5 Consumption Standards 1 1 Low Emission 1 5 Safety Powertrain 10 15 20 characteristic ranking - Human driving Powertrain 13 9 14 Characteristic ranking - Autonomous driving HoQ2.1 Powertrain Level Traction Power Generation/Control 4 5 6 Speed Range Maximum Max Amplitude of High Power of jerking Torque Oscillation [up arrow] [up arrow] [down arrow] Needs/Requests Low Noise 9 9 High 5 9 Performance 5 9 9 1 End-User High 1 1 9 Comfort 9 High Reliability 5 Easy to use 5 Low Fuel 1 -5 Consumption Standards 1 Low Emission 1 6 Safety Powertrain 16 21 10 characteristic ranking - Human driving Powertrain 9 14 11 Characteristic ranking - Autonomous driving HoQ2.1 Powertrain Level Traction Power Generation/Control 7 8 9 Time w/o Powertrain Tank to Torque Vibration Wheel Maximum delivery Efficiency during load Req. [up arrow] [down arrow] [up arrow] Needs/Requests Low Noise 5 5 1 High 5 1 Performance 5 5 1 End-User High 9 Comfort 9 9 9 -1 High Reliability 9 Easy to use 1 Low Fuel 9 Consumption Standards 1 1 Low Emission 9 Safety Powertrain 18 13 25 characteristic ranking - Human driving Powertrain 15 16 22 Characteristic ranking - Autonomous driving HoQ2.1 Powertrain Level Traction Power Emissions Control Generation/Control 10 11 Tank to NOx Emission Wheel according Effieteney to RDE at part/low load [up arrow] [down arrow] Needs/Requests Low Noise 5 -1 High 5 -1 Performance 1 -1 5 -1 End-User High Comfort -1 High Reliability 9 1 Easy to use 1 Low Fuel 9 1 Consumption Standards 1 9 Low Emission 9 9 -1 Safety Powertrain 29 14 characteristic ranking - Human driving Powertrain 24 16 Characteristic ranking - Autonomous driving HoQ2.1 Powertrain Level Emissions Control 12 13 14 HC/CO Emission Soot/PN CO2 on according according to WLTC to RDE RDE [down arrow] [down arrow] [down arrow] Needs/Requests Low Noise -1 -1 High -1 -1 Performance 1 1 5 -1 -1 End-User High Comfort High Reliability -1 -1 9 Easy to use Low Fuel 1 -1 9 Consumption Standards 9 9 Low Emission 9 9 1 -1 9 Safety Powertrain 18 15 23 characteristic ranking - Human driving Powertrain 18 17 19 Characteristic ranking - Autonomous driving HoQ2.1 Powertrain Level HVAC eGen. 15 16 17 Heating Mech Power Efficiency Power for Cooling of Electric Power Gen. [down arrow] [down arrow] [up arrow] Needs/Requests Low Noise 3 1 High 3 1 Performance 3 1 8 1 End-User High 1 Comfort 9 5 5 High Reliability 5 5 Easy to use 5 Low Fuel -1 5 5 Consumption Standards 1 8 8 Low Emission -1 5 5 Safety Powertrain 5 29 21 characteristic ranking - Human driving Powertrain 5 24 18 Characteristic ranking - Autonomous driving HoQ2.1 Powertrain Level Energy Storage Self-diagnosis 18 19 20 Energy Energy PWT hours Storage Refilling w/o failure Capacity Time [up arrow] [down arrow] [up arrow] Needs/Requests Low Noise 3 High 3 Performance 1 3 3 End-User High Comfort High 9 Reliability 9 -3 Easy to use -1 9 Low Fuel 3 Consumption Standards 1 Low Emission 5 9 3 Safety 5 Powertrain 15 14 15 characteristic ranking - Human driving Powertrain 13 12 15 Characteristic ranking - Autonomous driving HoQ2.1 Powertrain Level Self-diagnosis 21 22 23 Firing Extra PWT on board Risk Torque Failure Recovery Recovery Detection time Time Time [down arrow] [down arrow] [down arrow] Needs/Requests Low Noise 1 High Performance End-User High Comfort 1 High 5 Reliability Easy to use Low Fuel 1 Consumption Standards 1 Low Emission 7 7 9 Safety 9 9 5 Powertrain 16 16 20 characteristic ranking - Human driving Powertrain 17 17 23 Characteristic ranking - Autonomous driving HoQ2.1 Powertrain Level Self-diagnosis eBrake 24 25 26 Fuctions % Braking PWT Size /Components energy failure recovered prevention [up arrow] [up arrow] [down arrow] Needs/Requests Low Noise 3 1 High 1 Performance 1 1 End-User High Comfort 3 High 5 3 Reliability 9 1 Easy to use Low Fuel 1 9 1 Consumption Standards 1 1 1 Low Emission 9 9 1 Safety 5 Powertrain 22 23 5 characteristic ranking - Human driving Powertrain 25 21 8 Characteristic ranking - Autonomous driving HoQ2.1 Powertrain Level eBrake 27 PWT Weight [down arrow] Needs/Requests Low Noise 5 High 5 Performance 3 5 End-User High Comfort High Reliability 9 Easy to use 1 Low Fuel 9 Consumption Standards 9 Low Emission 9 Safety Powertrain 38 characteristic ranking - Human driving Powertrain 35 Characteristic ranking - Autonomous driving
|Printer friendly Cite/link Email Feedback|
|Author:||De Cesare, Matteo; Cavina, Nicolo; Paiano, Luigi|
|Publication:||SAE International Journal of Engines|
|Date:||Dec 1, 2017|
|Previous Article:||Evaluation of the Potential of Water Injection for Gasoline Engines.|
|Next Article:||Development of a High Performance Natural Gas Engine with Direct Gas Injection and Variable Valve Actuation.|