Highly Fuel-Efficient Transmission and Propulsion Concepts

Transcription

1 Automotive Innovation (2018) 1: Highly Fuel-Efficient Transmission and Propulsion Concepts Ferit Küçükay 1 Andreas Lange 1 Lin Li 1 Received: 20 June 2017 / Accepted: 8 October 2017 / Published online: 8 March 2018 Society of Automotive Engineers of China (SAE-China) 2018 Abstract The integration of one or more electric machines into the drivetrain has resulted in many different powertrain concepts in recent years, ranging from the P2 hybrid to dedicated hybrid transmission (DHT). Two types of DHT with different characteristics are investigated. The first type is the power split hybrid (PS-DHT), which has very low mechanical complexity but needs high electrical effort in the transmission. The second type is multi-mode DHT (MM-DHT), which has a slightly higher mechanical complexity but much less electrical effort when compared with the PS-DHT. A transmission synthesis is used to determine the concept of the MM-DHT. The three different powertrain concepts (i.e., P2, PS-DHT, and MM-DHT) are analyzed and evaluated regarding fuel economy and performance. Legal driving cycles (e.g., Worldwide Harmonized Light vehicles Test Procedure) and the 3D method (driver, driven vehicle, driving environs) are used to investigate the drivetrain in the context of real driving operation. Results show that the two DHT concepts offer better fuel economy than the P2 hybrid drivetrain while still providing the same or even better driving performance. The study also shows that new hybrid concepts created with transmission synthesis can lead to further improvements in hybrid powertrains. Keywords Dedicated hybrid transmission Three-dimensional method Drive train design Fuel economy Driving performance Abbreviations DHT Dedicated hybrid transmission PS Power split MM Multi-mode SOC State of charge ICE Internal combustion engine EM Electric machine HEV Hybrid electric vehicle 1 Introduction Hybrid electric vehicles (HEVs) have lower energy consumption and emissions than conventional vehicles [having internal combustion engines (ICEs)] but do not allow zero-emission driving over long distances. Battery-electric B Lin Li linli@tu-bs.de 1 Institute of Automotive Engineering (IAE), Technische Universität Braunschweig, Braunschweig, Germany vehicles fulfill the requirement of (local) emission-free driving but have a limited range. In this context, the completion of charging infrastructure is important. By contrast, plugin hybrid vehicles (PHEV) allow combined long-range and zero-emission driving and thus meet many of today s customer requirements. The integration of one or more electric machines (EMs) into the drivetrain has resulted in many different powertrain concepts, which differ significantly from one another. A common variant is the P2 topology. As an add-on solution, an EM is integrated into a conventional drivetrain between the ICE and transmission. An additional clutch between the EM and ICE allows electric driving without drag losses of the ICE. This type of hybridization is characterized by a relatively high mechanical complexity, because many gearings, bearings, shafts and shifting elements are present. With the realization of a dedicated hybrid transmission (DHT), this mechanical effort can be reduced. Using only one planetary gear set with two EMs, the powertrain can be realized without any conventional stepped gears. This power split hybrid (PS-DHT) has been in series production for many years. The PS-DHT is characterized by low mechanical effort but high electrical effort.

2 36 F. Küçükay et al. 2 Performance A powertrain has to meet customer requirements as well as possible. The demand is determined by means of a demand map. To create the demand map, the total driving resistance at constant vehicle speed is derived from the air resistance, rolling resistance, and friction resistance. The required maximum speed creates the first limit of the demand map on the right side. The necessary maximum power at the wheels is obtained from the intersection of the v max requirement with the level-road driving resistance curve. The power at the wheels can be supplemented in the form of a constant performance hyperbola toward lower speeds. The maximum traction force then corresponds to the maximum power in the demand map, which is shown in Fig. 1. In addition, the vehicle must be able to drive away from standstill on an incline. The maximum traction force requirement corresponds to the maximum gradeability or acceleration of the vehicle with full use of the traction limit potential. Rather than considering the case of highspeed travel, a fully loaded vehicle is considered because the launching capability has to be ensured even when the vehicle is fully loaded. The upper limit of the demand map results from the traction limit. This is shown schematically in Fig. 1. The present study uses a PHEV with the vehicle parameters given in Table 1. The vehicle has front wheel drive, which is front-loaded when empty. When calculating the demand map, it was considered that the center of gravity moved backward when the drive is fully loaded. Fig. 2 Fig. 3 Demand map: criteria for vehicle launching Demand map coverage for a P2 hybrid at a high SOC Fig. 1 Demand map: traction force and wheel power depending on vehicle speed Table 1 Vehicle parameters of a C-segment PHEV C-segment vehicle parameters (PHEV) m kg 1500 m add kg 400 c D A m f R λ e 1.03 F Fric N 40 P max kw 75 v max km/h 200 The described maximum launching requirement is a demanding criterion, which has to be fulfilled in all driving situations. While hybrid drivetrains can provide additional wheel torque by discharging the battery, this cannot be guaranteed under all circumstances. As a consequence, the launching must be ensured also with a discharged battery. The demand map that has to be met even with a discharged battery is shown in Fig. 2. The coverage of the demand map is mostly not a problem for the PHEV when the battery is charged because the total system power (ICE and EM combined) is usually high. This is due to the fact that the vehicle can be propelled in either hybrid and electric mode alone. For a P2 topology with P ICE = 75 kw, P EM = 45 kw and eight-speed dualclutch transmission (8G-DCT), a clear over-fulfillment of the demand map is seen (see Fig. 3). The peak power of the EM in this example was determined in such a way that the vehicle described in Table 1 can drive through the Worldwide Harmonized Light vehicles Test Procedure (WLTP) electrically. With a high state of charge (SOC), the illustrated P2 power-

3 Highly Fuel-Efficient Transmission and Propulsion Concepts 37 Fig. 4 Demand map coverage for a P2 hybrid at a low SOC Fig. 5 Coverage of the requirement map of a PS-DHT at a low charge level for various power ratings of EM1 train can exceed the required maximum speed of 200 km/h. However, because this maximum speed depends on the SOC, it cannot be guaranteed. Because a constantly sufficient charge level of the battery cannot be guaranteed as already explained above, the coverage of the requirement map is of particular interest at a low SOC when the battery is depleted. A transmission with multiple speeds can be used to meet the requirements (Fig. 4). Owing to high transmission ratios of the lower gears, high traction forces can be generated despite a relatively lowpower ICE in this case. At the same time, small steps between the gears allow small traction gaps. A good approximation to the maximum traction force hyperbola can thus be realized. When using a conventional transmission with stepped gears, a speed gap remains at low speeds when an ICE is used and has to be covered by a launch element (i.e., a clutch or torque converter). It is known in the case of conventional vehicles (having ICEs) that the demand map can be covered by multi-speed transmissions (8G-DCT in this case). No stepped gears are necessary when using a PS-DHT. Instead, a continuously variable transmission is implemented with use of an electric variator the electrical power split. In a PS-DHT, the peak power of the EM in the variator has to be designed in such a way that the power of the ICE can be split depending on the desired gear ratio. Thus, EM1, which is connected to the sun gear, has to provide sufficient torque and at the same time a sufficient speed range. EM2, which is connected to the ring gear, has to receive at least the electrical power of EM1. If this is not the case, the excess power of EM1 has to charge the battery, but this may reduce wheel torque. EM2 is usually designed to be more powerful than EM1. However, this is irrelevant for covering the demand map when the battery is depleted. Figure 5 shows the coverage of the demand map for EM1 peak powers of 43 and 66 kw. The launching capability and maximum speed are ensured with an EM1 output of 43 Fig. 6 Scaling of the DCT torque capacity and PS-DHT variator power as a function of ICE power kw. However, there is a considerable traction force gap. An EM1 output of 66 kw covers the entire demand map. The continuously variable adjustment of the engine speed makes it possible to follow the line of optimal efficiency. In contrast to the case for a P2, a temporary increase in the maximum speed in boosting operation is not possible because the speed limit of EM2 is reached at the designed maximum speed. One disadvantage of the PS-DHT is scalability with regard to increasing the ICE power. For a P2, the torque capacity of the DCT must be increased as the ICE torque increases. Although this involves additional weight, higher costs, and lower transmission efficiency, the necessary modifications to the EM are rather minor in comparison with the necessary modifications to the PS-DHT. Owing to the torque and power coupling of the ICE and EM, an increase in ICE power results in an increase in EM power in the transmission. To realize optimal driving performance with the depleted battery, Fig. 6 shows variator designs for various ICEs ranging from 75 to 225 kw. It is seen that the total power of the two EMs rises

4 38 F. Küçükay et al. from 132 kw (P ICE = 75 kw) to 334 kw (P ICE = 225 kw). Particularly in the context of a high-performance ICE, there are considerable additional costs and additional weight compared with a conventional transmission with stepped gears. Regardless of the conceptual differences, the investigations show that both the P2 and PS-DHT cover the demand map with an appropriate design. This basically depends on the ICE used and the desired coverage of the demand map. However, the driving experiences of the presented concepts are different and therefore attract different customers and market requirements. 3 Fuel and Energy Consumption In addition to driving performance, a powertrain must provide high fuel economy. The presented hybrid concepts P2 and PS-DHT differ regarding their losses. With the P2 hybrid, the efficiency of the 8G-DCT is important to the overall efficiency of the powertrain. Because the DCT has a relatively complex mechanical design and a considerable energy requirement for the actuators, there is relatively large energy loss in the transmission. With recuperation, hybrid vehicles can recover some of the kinetic energy during the thrust phases. The potential energy recuperation in a P2 powertrain is reduced by losses of the transmission. The transmission efficiency in hybrid vehicles is therefore of greater importance than that in conventional vehicles. In case of the PS-DHT, the mechanical losses are lower because the mechanical part of the transmission consists only of a planetary gear set, a reduction gear for EM2 and a differential. Depending on the transmission ratio, part of the ICE power has to be transmitted by an electrical path. In most cases, EM1 generates electrical energy, which EM2 converts into mechanical energy. In this double energy conversion, there are electrical losses of a hybrid transmission. The previous section already explained that the peak power of EM1 and EM2 depends on the ICE power. This affects the operating points of the two EMs and thus also the losses on the electrical path, during electrical driving and recuperation. Figure 7 shows the operating points of EM2 in the WLTP for the different variator designs in Fig. 6. The average EM2 efficiency decreases with increasing ICE power. At the same time, the electric driving and recuperation are less efficient. Moreover, the efficiency of EM1 has to be considered for the efficiency of the electrical path, which also decreases. The share of the electrical path thus determines the overall transmission efficiency. Furthermore, the transmission has to be able to operate the drive units (particularly the ICE) as efficiently as possible. The electrical continuously variable transmission (ecvt) mode of the PS-DHT is advantageous in this respect because the engine speed can be optimally adapted to the driving sit- Fig. 7 Average efficiency for different designs of EM2 as a function of the ICE power of a PS-DHT. The full-load curves for the motor and generator correspond to the power values given above (66, 121, and 167 kw) uation. However, when a DCT is used as in the P2 hybrid, there are always deviations from an efficiency-optimized operation. Nevertheless, this disadvantage can largely be compensated for by a load point shift of the engine. The investigations carried out so far clearly show that both concepts have advantages and disadvantages with regard to driving performance and efficiency. In this respect, it is important to develop new hybrid concepts that combine the advantages of stepped gears with the advantages of ecvt operation. Such concepts can be developed by means of hybrid transmission synthesis. 4 Powertrain Synthesis A tool is used to synthesize various vehicle transmissions. Countershaft transmissions, such as manual transmissions, DCTs, and transmissions in planetary design, such as automatic transmissions and dedicated hybrid transmissions (DHTs), can be synthesized for conventional, hybrid, and electric vehicles [1]. Examples of identified and developed transmissions are shown in Fig. 8. An essential part of the tool is the systematic evaluation of the determined transmissions. The evaluation and selection of the optimal transmissions are carried out on the one hand according to the transmission characteristics, such as number of gears, gear ratio, gear set structure, and loads in the transmission. On the other hand, the transmission is considered and evaluated in the context of the whole vehicle. This is particularly important for hybrid transmissions because traditional evaluation methods can only be partially used for hybrid transmission. Besides the transmission efficiency, the critical operating modes and points and other important properties are determined in the overall vehicle simulation. Many

5 Highly Fuel-Efficient Transmission and Propulsion Concepts 39 Fig. 10 Mix of urban, extra urban, and highway environs of an average driver for a vehicle with a power weight of 12 kg/kw according to the 3D method Fig. 8 Transmission synthesis strategy regarding an optimal selection of operation points. In the following, the three concepts mentioned are compared with respect to their energy and fuel consumption. 5 Concept Comparison Fig. 9 Coverage of the requirement map of a PS-DHT at a low charge level for various power ratings of EM1 transmission concepts are analyzed to identify and design the optimal transmission for hybrid powertrains. The present study synthesized a DHT with two planetary gear sets, four shifting elements (SEs) and two EMs. The DHT can implement two ecvt modes, three stepped gears with parallel EM operation and two electrical modes. For this reason, it is referred to as multi-mode DHT (MM-DHT). In case of a low SOC, the achievable operation ranges of the various modes in the demand map are shown in Fig. 9. As with the PS-DHT, when P EM1 = 21 kw is applied, a high traction force can be achieved, but a clear traction force gap can also be seen. The use of EM1 with power of 49 kw completely covers the demand map. It is shown that the MM-DHT also covers the demand map well, although it has lower mechanical effort than the P2 and less variator power than the PS-DHT. The MM-DHT is therefore categorized between the P2 and PS-DHT. In addition, Fig. 9 shows that the demand map is covered by the ecvt modes as well as with the three speeds from 50 km/h onward. This creates additional degrees of freedom for the operating The concepts P2, PS-DHT, and MM-DHT have previously been compared with respect to their performance. The three different hybrid concepts are compared in the following with regard to their efficiency. The question arises which conceptual interpretations are comparable. A broad variation of basic parameters was thus used. EM maps were scaled in terms of their speed and torque ranges. Furthermore, the basic parameters of the transmissions, such as the stationary ratios of the planetary gear sets, spur gear ratios, and the differential, were varied. A total of 144 different configurations were simulated for the P2 hybrid. The two DHT variants were simulated with 6075 different configurations. A driving cycle needs to be chosen to evaluate fuel consumption. Legal cycles are often selected as these are also relevant to homologation. However, these cycles are not representative for fuel consumption. The representative customer use has been characterized and analyzed for many years using the 3D method (driver, driven vehicle, and driving environs). The 3D method was first introduced in 1990 and has been improved continuously [2 6]. It considers the representative customer use of a vehicle, depending on the load of the driven vehicle, driving style, and driving environs. The variation of all three parameters leads to large diversity that should be considered for fuel and energy consumption, fatigue issues, and vehicle testing [7]. By applying the 3D method in this paper, representative driving cycles are generated for evaluating fuel consumption. Figure 10 shows the generated customer cycle used for the simulated comparison of concepts. The customer cycle can be compared with the WLTP by means of cycle-related characteristic parameters, as shown in Table 2. The customer cycle has more than double the total distance (53 vs 23 km in the WLTP). The average speed is about 14% higher. The average deceleration in thrust phases

6 40 F. Küçükay et al. Table 2 Cycle-related parameters of the 3D customer cycle and WLTP 3D customer cycle WLTP Traction Traction Thrust Traction Thrust Distance (km) Duration (s) Distance share (%) Average acceleration (m/s 2 ) Effective speed (m/s) Standstill (s) Fig. 11 Fuel consumption as a function of hybrid acceleration from 0 to 100 km/h for basic parameter variations of P2, PS-DHT, and MM-DHT concepts Fig. 12 Selection of optimal concepts with comparable acceleration in hybrid mode based on parameterizations of P2, PS-DHT, and MM-DHT concepts is 22% higher than in the WLTP. It is deduced that the driving maneuvers are basically more dynamic than those in the WLTP. The share of the traction phase is higher than in the WLTP owing to the higher highway share. Results of the basic parameter variation are shown in Fig. 11. For each basic parameter configuration, the hybrid fuel consumption is shown as a function of the acceleration from 0 to 100 km/h in hybrid mode. For the best basic parameter configurations, a Pareto front can be identified for each hybrid concept. The front describes the best fuel consumption possible depending on the corresponding performance. It is seen that the MM-DHT achieves the best overall compromise and the absolutely lowest consumption values. The PS-DHT achieves the second-best results. The P2 Pareto front shows that both fuel consumption and driving performance deteriorate below a certain EM power. Although a smaller EM can, in principle, achieve better efficiency, the recuperation is restricted, which prevails in the customer cycle. Because the vehicle has front wheel drive, acceleration from 0 to 100 km/h within 5.8 s is not possible owing to the traction limit at the given center of gravity. A further increase in electrical power thus does not lead to an improvement in terms of the presented criteria. The above parameter study allows the selection and comparison of concepts that have similar performance and can cover the demand map with regard to launching and top speed requirements. An acceleration from 0 to 100 km/h in 6.5 s was specified (Fig. 12). Regarding the selected concepts, it is clear that the P2 has the lowest electrical power and still achieves equivalent driving performance. Through the use of a complex transmission with many gears, it is possible to achieve comparable driving performances with less electrical power. The PS-DHT has the highest total electrical power. Nevertheless, a huge part of the electrical power is required for speed and torque conversion in hybrid mode and does not contribute directly to better driving performance. The MM-DHT has only slightly more electrical power than the P2 hybrid, but the power is divided into two different EMs. This shows that the same driving performance as for the PS-DHT can be achieved despite the lower EM power. Table 3 gives selected basic parameters for the previously considered concepts. The selected concepts are analyzed in

7 Highly Fuel-Efficient Transmission and Propulsion Concepts 41 Table 3 Basic parameters for P2, PS-DHT, and MM-DHT P2 PS-DHT MM-DHT P ICE 75 kw 75 kw 75 kw P EM 89 kw P EM2 = 104 kw P EM2 =78kW P EM1 = 43 kw P EM1 = 21 kw Transmission ϕ TM = 9, 1 i D,1 = 4.14 i D,2 = 3.31 i 0 = 2 i D = 3.48 i 0,1 = 1.87 i 0,2 = 1.66 i D = 3.17 more detail. The performance characteristics of these concepts have already been presented. All concepts meet the requirements of the demand map with regard to the maximum speed and launching ability. For both DHT concepts, there is a traction force gap in the case of a low SOC, as already described above. Finally, the three previously selected hybrid concepts are analyzed in the 3D parameter space in terms of the conceptspecific differences in different driving environments and different driving styles. The following 3D parameters are selected. Fig. 13 Fuel consumption of the P2 hybrid, PS-DHT, and MM-DHT in the 3D parameter space and WLTP 1. Driving style: conservative, average, and sporty 2. Driving environment: urban, extra urban, and highway The vehicle parameters were assumed to be constant in all cases. In addition to customer operation, the WLTP was simulated for each concept. The maximum speed of all hybrid concepts is limited to 200 km/h. Although the vehicle with the P2 powertrain can drive more quickly in boost operation, this would only be possible temporarily and would lead to incomprehensible vehicle behavior when the SOC becomes low. In addition, critical situations arise when overtaking at high speeds if the vehicle suddenly decelerates owing to a discharged battery because the powertrain power is not sufficient. The results of the simulations in the 3D parameter space are shown in Fig. 13. Overall, the MM-DHT has the best fuel consumption. This applies to both the WLTP and urban and extra urban operation. In an urban driving environment, MM-DHT has better fuel economy than the P2 concept by 14%, as shown in Fig. 14. This is explained by relatively high transmission losses in the case of the P2, which reduce the efficiency in both traction (electric driving) and thrust (recuperation) phases. In the urban driving environment, the P2 transmission reaches an average efficiency of around 88%, compared with transmission efficiencies of both DHT concepts exceeding 92%. Furthermore, the ecvt operation of the DHT concepts allows an optimal operation point selection of the ICE, whereas deviations from the optimal efficiency curve occur with a DCT. For the PS-DHT, efficiency potentials arise in an Fig. 14 Fuel consumption of the P2 hybrid, PS-DHT, and MM-DHT in urban areas urban driving environment because the EM2 (driving motor) is connected to the wheel with a relatively low gear ratio due to the v max capability. Because the average speed in an urban environment is low, EM2 is operated with relatively low speeds. As a result, a potential regarding the efficiency of EM2 can be identified. The MM-DHT achieves better operation of the drive units with less mechanical transmission losses than the P2 concept. Overall, the MM-DHT achieves an 8% better consumption in urban areas than the PS-DHT and 14% better consumption than the P2 hybrid. In the WLTP, the MM-DHT achieves 4% better consumption than the PS-DHT and 7% better consumption than the P2 hybrid. In the case of the highway, all concepts have similar efficiency. This is surprising in the case of the PS-DHT regarding the efficiency of the electric path. At higher speeds, and particularly when driving constantly, the share of the electrical path is small, which results in a good transmission efficiency on the highway.

8 42 F. Küçükay et al. In summary, the development of new hybrid concepts using transmission synthesis and optimization is a promising approach of further improving the characteristics of hybrid powertrains with regard to efficiency, driving performance, and scalability. 6 Conclusion Three PHEV powertrains (P2, PS-DHT, and MM-DHT) were compared with respect to driving performance and energy consumption. The concept-specific differences were discussed on the basis of the demand map coverage and the respective dominant loss mechanisms. 1. The P2 topology has the highest mechanical effort and the lowest installed electrical power. 2. The PS-DHT has low mechanical effort but much installed electrical power. 3. Because both the P2 and PS-DHT have concept-specific advantages and disadvantages, the MM-DHT was generated by transmission synthesis, with the purpose of combining the advantages of the two concepts. The generated MM-DHT was also investigated with respect to driving performance and energy consumption. 4. Respective basic parameters were varied for all three concepts. It was shown that the MM-DHT offers the best balance between driving performance and fuel consumption. 5. Ultimately, optimal concepts that provide the same acceleration in hybrid mode from 0 to 100 km/h were selected. 6. The presented concepts were evaluated in the 3D parameter space and the WLTP. It was shown that the two DHT concepts offer better efficiency than the P2 concept, with the MM-DHT achieving the best results. Especially in an urban driving environment, the MM-DHT can operate its drive units more efficiently than the P2 or PS-DHT. 7. The study showed that new hybrid concepts created through transmission synthesis can lead to further improvements in hybrid powertrains. References 1. Li, L., Lange, A.: Bewertung und optimierung neuartiger hybridantriebe. In: WKM Symposium 2016, Graz (2016) 2. Küçükay, F.: Rechnergestützte getriebe-dimensionierung mit repräsentativen lastkollektiven. Automobiltechnische Zeitschrift ATZ (1990) 3. Janßen, A., Küçükay, F.: Entwicklung von Fahrwerks- und Antriebstrangbauteilen durch simulation kundennaher Betriebsbelastungen. Tagungsband 33. Tagung des DVM-Arbeitskreises Betriebsfestigkeit (2006) 4. Fugel, M., Küçükay, F.: Identifikation der Antriebstrangbelastungen im kundenbetrieb 2. In: Internationales Symposium für Entwicklungsmethodik, Optimierung komplexer Antriebsstränge - die Herausforderung der Zukunft (2007) 5. Eghtessad, M., Dietrich, D.: Kundenorientierte dimensionierung von elektrifizierten antrieben. Internationaler VDI-Kongress Getriebe in Fahrzeugen 2010 (2010) 6. Weiler, B.: Kundenrelevante auslegung von nutzfahrzeugen. Schriftenreihe des Instituts für Fahrzeugtechnik TU Braunschweig, Nr, p. 43 (2015) 7. Li, L., Schudeleit, M.: Market-specific dimensioning of drivetrain components. CTI-MAG, the automotive TM, HEV & EV drives magazine by CTI (2013)