Next generation electric drives for HEV/EV propulsion systems: Technology, trends and challenges

Next generation electric drives for HEV/EV propulsion systems: Technology, trends and challenges I. L´opeza,∗, E. Ibarrab, A. Matallanab, J. Andreub, I. Kortabarriab aDepartment of Electrical Engineering, UPV/EHU, C. Rafael Moreno Pitxitxi, 48013 Bilbao, Spain bDepartment of Electronic Technology, UPV/EHU, C. Rafael Moreno Pitxitxi, 48013 Bilbao, Spain Abstract In recent decades, several factors such as environmental protection, fossil fuel scarcity, climate change and pollution have driven the research and development of a more clean and sustainable transport. In this context, several agencies and associations, such as the European Union H2020, the United States Council for Automotive Research (USCAR) and the United Nations Economic and Social Commission for Asia (UN ESCAP) have defined a set of quantitative and qualitative goals in terms of efficiency, reliability, power losses, power density and economical costs to be met by next generation hybrid and full electric vehicle (HEV/EV) drive systems. As a consequence, the automotive electric drives (which consists of the electric machine, power converter and their cooling systems) of future vehicles have to overcome a number of technological challenges in order to comply with the aforementioned technical objectives. In this context, this paper presents, for each component of the electric drive, a comprehensive review of the state of the art, current technologies, future trends and enabling technologies that will make possible next generation HEV/EVs. Keywords: EV; HEV; HSEM; SiC; power electronics; power converter; advanced cooling. ; SBD, Solar Electric Vehicle (SEV), Schottky Barrier Diode; SiC, Silicon Carbide; SiFe, Silicon-iron; SRM, Switched Reluctance Machine; SVM, Space Vector Modulation; ∗Corresponding author Email addresses: [email protected], Tel. +0034 94 601 39 15 (I. L´opez), [email protected] (E. Ibarra), [email protected] (A. Matallana), [email protected] (J. Andreu), [email protected] (I. Kortabarria) Abbreviations:Al-Cap, Aluminium Electrolytic Capacitor; BJT, Bipolar Union Transistor; CoFe, Cobalt-Ion; CuBe, Copper-Berylium; CuZr, Copper-Zirconium; DC, Direct Current; DOE, U.S. Department of Energy; EMI, Electromagnetic Emissions; ESR, Cpacitor parasitic resistance; ESL, Stray Inductance; EV, Electric Vehicle; FC, Fuel Cell; FOC, Field Oriented Control; GaN, Galium Nitride; GHG, Greenhouse Gas emissions; HEV, Hybrid Electric Vehicle; HSEM, High Speed Electric Machine; ICE, Internal Combustion Engine; IGBT, Insulated Gate Bipolar Transistor; IM, Induction Machine; JBD, Junction Barrier Diode; JFET, Junction Field-Effect Transistor; MLC-Cap, Multi-layer Ceramic capacitor; MPPF-Cap, Metallized Polypropylene Film Capacitor; MOSFET, Metal-oxide-semiconductor Field-effect transistor; NREL, National Renewable Energy Laboratory; PMSM, Permanent Magnet Synchronous Machine; PWM, Pulse Width Modulation; RBS, Regenerative Braking System Preprint submitted to Elsevier May 15, 2019 This is the accepted manuscript of the article that appeared in final form in Renewable and Sustainable Energy Reviews 114 : (2019) // Article ID 109336, which has been published in final form at https:// doi.org/10.1016/j.rser.2019.109336. © 2019 Elsevier under CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/) List of symbols feElectric frequency pPole number PDTotal Power Dissipation Pmech Mechanical Power Rc−sCase to heat sink thermal resistance Rj−cJunction to case thermal resistance Rs−aHeat sink to ambient thermal resistance Rth Thermal resistance TaAmbient Temperature TcCase Temperature TjJunction Temperature TsSink Temperature Tmech Mechanical Torque td,on Turn on delay td,off Turn off delay VDSsat Collector source saturation voltage VGSsth Gate threshold voltage ωmech Mechanical speed SynRM, Synchronous Reluctance Machine; TE, Thermoelectric Cooling; THD, Total Harmonic Distortion; TIM, Thermal Interface Material; USCAR, United States Council for Automotive Research; UN ESCAP, United Nations Economic and Social Commission for Asia; WGB, Wide Bandgap. 1. Introduction In this decade, environmental protection and alternative green energies have become one of the main concerns for social and political agents and for the scientific community due to a number of factors, such as greenhouse gas (GHG) emissions, fossil fuels scarcity and their price volatility, or high pollution in modern cities. These factors are accelerating the development of more efficient, sustainable and renewable energy systems [1–6]. Transportation is one of the sectors that most contribute to GHG emissions, producing approximately the 27 % of the total [7]. Within the transportation sector (air, rail and road transport), road transport accounts for the 75 % of transport GHG emissions [2]. Nowadays, population is growing at a remarkable rate. According to [8], the world population will rise up to 9.8 billion in 2050, which supposes an increase of 30 % with regard to 2017 population (7.6 billion). As a consequence, the number of road vehicles is expected to be of around 2 billion in 2050 [9]. In this context, road vehicle electrification becomes crucial in order to overcome the aforementioned societal and environmental issues. This last is forcing the research and development of novel concepts and innovations to make electric vehicles (EV) and hybrid electric vehicles (HEV) more efficient, reliable and safe at an affordable cost [1]. Nowadays, there is a great number of original equipment manufacturers (OEMs) producing EVs, HEVs and fuel cell (FC) vehicles (table 1). Battery powered and Plugin hybrid EV stock surpassed 2 million vehicles in 2016 [10]. The EV market beat a new record in 2016, with over 750 thousand sales around the world [10], leading to an increase of 33% compared to 2015 [11]. As it can be seen in figure 1, China is the 2 China EE.UU Japan Norway Netherlands United Kingdom France Germany Norway Netherlands France Island Sweden Switzerland Belgium Austria 29.0% 6.3% 6.0% 3.6% 1.9% 1.7% 1.5% 1.5% 646 570 147 116 110 88 83 74 Ranking of countries by VE stock (thousand of units, 2016) Ranking of countries by market share by total fleet (%, 2016) Figure 1: Electric vehicle market in 2016 [10]. Electric cars in the vehicle stock (millions) Consistent with the ambition of the EV30@30 campaign IEA B2DS Paris Declaration IEA RTS Historical Cumulative country targets (as of 2016) Cumulative OEMs announcements (estimate) 200 220 180 140 160 120 80 100 60 20 40 0 2010 2015 2020 2025 2030 IEA 2DS Figure 2: Deployment scenarios for the stock of electric cars to 2030 forecast by relevant agents [10]. country with the largest electric car stock, with the United States in a second place. According to a number of agents, the future market evolution in the following years is promising, and some forecasts expect a global EV stock of 20 million of units in 2020 (figure 2). However, the current global electric car stock corresponds to just 0.2% of the total number of passenger light-duty vehicles in circulation [10]. Considering that the previous forecasts can be fulfilled as long as they meet all the requirements from which they have been based, it is more realistic to define a scenario with an electrified vehicle increase between 9 million and 15 million units by 2020 [11]. In order to achieve all these market objectives, it becomes clear that it is necessary to provoke a change in the society and make the EV an attractive road transport alternative. From the end user point of view, the main technical aspects that are considered to decide whether or not purchase an EV or HEV are vehicle power, efficiency, maximum speed, dynamic response, reliability, autonomy and, last but not least, economical costs. All these items are related with the vehicle electric drives, which include electric machines, power electronics and their respective cooling systems. From a technical point of view, significant advances will be required in order to achieve the required cost reductions for the popularization of HEV/EV technologies. Historically, advances in transport drives have been conducted following progressive steps. From 1995 to 2005, electric drive systems research targets focused on the development of basic automotive components (integrated electric motor drives and power 3 Table 1: Example of electrified vehicles on the market, including their configuration, machine technology and power ratings. Model Year Vehicle Config. Motor Power (kW) Total/Electric Audi 2009 Q5 Hybrid Hybrid PMSM 182 / 40 BMW 2014 i8 Hybrid PMSM 265 / 96 BYD 2008 F3DM Plug-in Hyb. PMSM 125 / 75 Honda 2009 Insight Hybrid PMSM 83 / 10 Honda 2001 Civic Hybrid PMSM 69 / 10 Audi 2009 Q5 FCEV Fuel cell IM 80 Ford 2000 Ford P2000 Fuel cell IM 67 Honda 2008 FCX Clarity Fuel cell PMSM 100 Hyundai 2013 ix35 FCEV Fuel cell IM 100 Mercedes 2010 Clase B F-Cell Fuel cell PMSM 100 Toyota 2015 Mirai Fuel cell PM 113 BMW 2013 i3 BEV Electric PMSM 125 BYD 2014 E6 Electric PMSM 90 Citr¨oen 2011 C-Zero Electric PMSM 47 Citr¨oen 2016 E-Mhari Electric PMSM 50 Ford 2011 Focus Electric Electric PMSM 107 Kia 2014 Soul EV Electric PMSM 81.4 Land Rover 2013 Defender Electric SRM 70 Mercedes 2014 SLS AMG ED Electric PMSM 550 Mercedes 2014 Clase B ED Electric PMSM 132 Mitsubishi 2009 i-MIEV Electric PMSM 47 Nissan 2010 Leaf Electric PMSM 80 Peugeot 2010 iOn Electric PMSM 49 Peugeot 2014 Partner Electric Electric PMSM 49 Porsche 2020 Mission E Electric PMSM 440 Renault 2001 Kangoo I Electric PMSM from 22 up to 29 Renault 2011 Kangoo ZE Electric PMSM 44 Renault 2011 Fluence ZE Electric PMSM 70 Renault 2012 Twizy Electric PMSM 8 Renault 2012 Zoe Electric PMSM 65 Smart 2011 Fortwo ED Electric PMSM 55 Tazzari 2009 EM1 Electric IM 20 Tazzari 2009 Zero Classic Electric IM 20 Tesla 2011 Roadster Electric IM 185 Tesla 2012 Model S Electric IM from 235 up to 568 Tesla 2015 Model X Electric IM from 193 up to 375 Think 2008 Think City Electric IM 34 Toyota 2012 RAV4 EV Electric IM 115 Volkswagen 2014 e-Golf Electric PMSM 85 Volkswagen 2013 e-UP Electric PMSM 60 Table notes: PMSM: Permanent Magnet Synchronous Machine. IM: Induction Motor. SRM: Switched Reluctance Machines. modules with maximum junction temperatures of 125 ◦C and switching frequencies between 2 kHz and 10 kHz) [12]. From 2005 to 2015 great efforts were conducted in order to achieve higher junction temperatures (150 ◦C), high switching frequency operation (up to 12 kHz) and integrated motor and power electronics [12]. Regenerative braking systems (RBS), which allow recharging the energy source by feeding back energy from the electric machine during braking, have also been developed and extensively investigated over the last years, as their implementation can improve fuel efficiencies between 30 % and 40 % for HEVs [13], and driving range can be extended between 8 % to 25 % for EVs [14]. Relevant examples of recent investigations aiming to improve regenerative braking include the dynamic detection of the lowest speed threshold at which electric braking is effective [15], control algorithms for independent regenerative braking torquevectoring on vehicles with in-wheel machines [16, 17], or investigations where motor and 4 Table 2: Current status and future targets for EV drive main components (electric machine and power electronics) in terms of power density, efficiency and costs (from [23, 24]). Electric motor Power electronics Characteristic 2010 2015 2020 2010 2015 2020 Specific power (kW/kg) 1.2 1.3 1.6 10.8 12 14.1 Power density (kW/l) 3.7 5.0 5.7 8.7 12 13.4 Efficiency (%) 90 92 93 91 94 97 Cost ( $ /kW) 11.1 7 4.7 7.9 5 3.3 hydraulic braking forces are coordinated [18]. Currently, the U.S Department of Energy (DOE) identifies a number of technical aspects in the report called EV Everywhere [19]. Based on the aforementioned features required by the consumers, EV Everywhere also defines a set of goals and technical targets. In this context, table 2 shows quantitatively power density, efficiency and cost targets for future HEV/EV electric drives. Other national and international programmes, adopted by each community or country, have also similar qualitative and quantitative targets to be reached by the future electric propulsion technologies. Among them, Horizon 2020 [20], the United States Council for Automotive Research (USCAR) [21] and the United Nations Economic and Social Commission for Asia (UN ESCAP) [22] represent global HEV/EV technology trends, as they cover Europe, North America and Asia. According to these programmes, the expected improvements in future EV propulsion systems can be summarized as [20–22]: 1. Increase of torque production capability of the electric machine by 30% and speed by 50%. 2. Reduction of 50% on motor losses. 3. Power density increase of 50% in power converters and, at the same time, reduction of power converter losses by 50%. 4. Overall efficiency optimization of 20%. 5. Powertrain system (motor, power converter and energy source) weight and volume reduction of 40%. 6. Costs reductions (four times reduction) on electric machine and power electronics. 7. Simplification of the thermal management systems using on-board coolants with minimal additional components as possible. As a consequence of these specific targets, the main elements to be integrated in future HEV/EV electric drives will face a number of challenges that should be overcome. In this paper, the aforementioned challenges are identified, and technological solutions required to constitute these future drives are also thoroughly reviewed. In this sense, section 2 overviews vehicle electrification architectures and lists the enabling technologies for the required future innovations, section 3 focuses on the electric machine and torque control challenges and solutions, while sections 4 and 5 determine power converter and cooling system challenges and solutions, respectively. 5 ICE Mechanical transmission ICE vehicle Power Electronic converter M/G Transmission Battery electric vehicle Power Electronic converter ICE GENERATOR Transmission Series hybrid vehicle Power Electronic converter Transmission Parallel hybrid vehicle ICE ICE M/G Power Electronic converter GENERATOR Transmission Series-parallel hybrid vehicle ICE M/G M/G Power Electronic converter Transmission Complex hybrid vehicle ICE M/G Power Electronic converter Transmission Plug in hybrid vehicle M/G Hydrogen cylinder Transmission Fuel cell vehicle M/G M/G M/G Hydrogen cylinder Transmission Fuel cell hybrid vehicle Power Electronic converter Transmission Solar hybrid vehicle ICE Solar panel DC/DC converter M/G Power Electronic converter B/UC DC/DC converter Power Electronic converter Figure 3: Architecture and configuration of ICE and EV, FC and HEV (adapted from [1]). 2. EV and HEV propulsion architectures and enabling technologies for next generation drives In general, three electric vehicular architectures that compete with internal combustion engine (ICE) vehicles can be distinguished: battery powered electric vehicles (EV), hybrid electric vehicles (HEV), fuel cell vehicles (FC), colorredfuel cell hybrid electric vehicles (FCHEV) and hybrid solar electric vehicles (HSEV)1. Figure 3 shows the aforementioned powertrain configurations. One of the main differences between those architectures relies on the energy source. For example, EVs and HEVs rely on batteries, while FCs rely on hydrogen fuel cell stacks (zero emission vehicle, only emits water and heat). FC supplies a constant power, but it does not adapt properly to a rapid change of power demand, being its mayor application slow speed vehicles (buses, trams...) [29, 30]. However, using a FC as the main energy source and battery as an additional storage system, a FCHEV configuration 1Considering current technology, solar vehicles without hybridization do not have enough autonomy for current mobility requirements. 6 Table 3: Characteristic of ICE vehicle, EV, FC and HEV [25–28]. Feature ICE vehicle EV HEV Plug-in HEV FC Propulsion ICE based ED based(1) ICE & ED based ICE & ED based ED based System Energy Fuel tank Battery Fuel tank Fuel tank Fuel cell tank storage Ultra capacitor Battery Battery Battery Flywheel Ultra capacitor Ultra capacitor Ultra capacitor Flywheel Flywheel Flywheel Energy Petrol Electric Petrol & electric Petrol & electric Hydrogen source Energy source Refueling Charging Refueling Charging station Hydrogen refiner infrastructure station station station & refueling station & refueling station Well-to-tank(2) 88.0% 37.0% 88.0% -(3) 58.4% Tank-to-wheel(2) 12.1% 83% 22.3% -(3) 46.6% Well-to-wheel(2) 10.6% 31.3% 19.6% -(3) 27.2% Commercialized Yes Yes Yes Partially No Smooth operation(4) No Yes Yes Yes Yes Emissions Very high No Very low Low Ultra low System complexity Very low low Medium High Very high Bulky Yes No Yes Yes No (1) ED based: Electric Drive system based. (2) Approximate values. Gasoline has been considered as fuel for hybrid configurations. (3) These numbers will highly depend on the specific vehicle and will be between HEV and EV values. (4) Regarding torque ripple. is obtained, where the operation range and speed are increased, but the power train needs to be modified. For example, some manufacturers such as Honda, Toyota and Hyundai manufacture high performance FCHEVs [30]. With the aim of extend the range autonomy and reduce the environmental impact, there are also architectures based on solar panels installed on the top of the vehicle (HSEVs), which combine the benefits of solar energy (available, sustainable, renewable and clean) with the HEV characteristics, obtaining a fuel-efficient vehicle [29, 31, 32]. Although a number of vehicular topologies can be found in the literature, the majority of manufacturers rely on EVs and/or HEVs. EVs have a simple structure where energy flows from or to the battery through the bidirectional power converter (figure 3). On the other hand and depending of its internal configuration, hybrid vehicles can be classified as series HEV, parallel HEV, series-parallel HEV, complex HEV and plug in HEV [1], which differences rely on the way energy flows from the storage sources of energy [32–34]. With independence of the vehicular architecture, the electric drive (power converter and electric machine) is the core of all these electrified powertrains (figure 3). The development of next generation green vehicles based on advanced electric drives requires to focus on the following aspects: cost reduction, efficiency improvement and achievement of high power densities [35]. The key enabling technologies that will improve the aforementioned features on electric machines can be summarized as [12, 35–39]:  Cost reduction: (a) usage of new magnet materials without rare earths and heavy rare earths and (b) simplification of the cooling system (air cooling is considered among other solutions).  High efficiency: (a) development of high performance alloys, such as magnetic steels 7 and copper alloys and (b) the design of machines with low copper and iron losses.  High power density: (a) increase of the electric machine operation speed and (b) high performance cooling for increased power capabilities. On the other hand, the following enabling technologies must be considered for future power converters [12, 35–38]:  Cost reduction: (a) revolutionary and/or evolutionary changes to designs or manufacturing techniques and (b) simplification of the cooling system (as in the case of the electric motor, air cooling is considered as a possibility).  High efficiency targets, with the implementation of wide bandgap (WBG) semiconductor technology with low switching losses.  High power density targets, (a) by introducing WBG technology including advanced packaging and (b) by implementing optimum thermal management. Note that some of the requirements lead to opposite design concepts. For example, regarding the cooling system, a trade off between simplification (cost reduction) and thermal management optimization (high power density targets) must be followed, being this particular aspect challenging for researchers. The same applies to non rare-earth based machine technologies and power density targets. In the following sections, a deep and extensive study of the core components and technologies of electric drive is performed, which include the electric motor, the bidirectional converter and the thermal management, pointing out both the advantages and disadvantages of each technology and the new challenges that these entail to get the future technical goals. 3. Next generation electric machines for EV and HEV propulsion systems 3.1. Current electric machine technologies: overview and comparison of most relevant features The most established EV and HEV electric machine technologies are, by far, Permanent Magnet Synchronous Machines (PMSM) and Induction Machines (IM) [40]. As it can be derived from table 1, PMSM [41–46] are the preferred option in current EV and HEV vehicles. The main reason is their high efficiency and superior power density, which is particularly a critical factor in HEVs (due to their tight space constraints). Specially, Interior PMSMs are employed, as the additional reluctant torque allows to achieve higher power density than their Surface Mounted PMSM counterparts [41]. However, technologies that do not rely on rare-earth based magnets are gaining popularity [2, 41, 47], due to the scarcity, price fluctuations and high costs associated with these materials (neodimium and, in lower quantities, dysprosium and terbium) [47–51]. Among them, the mature squirrel cage IM [52–56] has a real market penetration (table 1) in the automotive industry, while other electric machine technologies, such as Ferrite based Permanent Magnet Assisted Synchronous Reluctance Machines (PM-assisted SynRM) [57–61] and Switched Reluctance Machines (SRM) [62–66] are gaining attraction from the scientific community. 8 In PM-assisted SynRMs the rotor is manufactured in a way that the asymmetry between the dand q-axes is maximized. Thanks to this, the reluctant torque of the electric machine is maximized, allowing to achieve power densities of around 75% of an interior PMSM for the same size and liquid cooling technology [67]. As a drawback, the ferrite-PM-assisted SynRM machine may suffer from demagnetization at low temperatures, which can be avoided or minimized by a proper design or by preheating the magnets before machine start [58, 68]. This type of machine is being extensively researched [58–60, 69] and, although there are virtually no commercial solutions equipped with this technology [47], it represents a promising alternative for the future. SRMs have a double salient structure (saliency in both stator and rotor) and no magnets nor windings in the rotor [70]. This saliency is used for electromagnetic torque production, requiring an specific control approach that takes into account the non-linearities of the machine [71]. As opposite to synchronous and induction machines, SRMs require an H-bridge converter topology (generally in asymmetric configuration) instead of a common Voltage Source Inverter (VSI) [72]. SRMs exhibit a number of advantages such as simple structure, flexibility of control, high efficiency, lower cost and intrinsic fault tolerance [62, 72]. Although SRMs can be considered as a low cost solution for EV and HEV applications, they have significant disadvantages that must be considered, such as high torque ripple, medium power density, high DC bus current ripple, high acoustic noise and high electromagnetic emissions (EMI) [72, 73]. The 6/4 pole SRM can be considered as the standard configuration for EV applications [74]. However, new design approaches that partially overcome the aforementioned drawbacks are being currently researched [63–65, 75]. Generally speaking, the main drawbacks of these advanced structures are their added manufacturing complexity and additional costs. According to [47], SRMs are now being introduced into vehicle prototypes. Land Rover and Toyota (in partnership with Renault-Nissan) are currently working on SRM drives (table 1). Quantitative and qualitative comparisons of the most relevant machines (PMSM, PMassisted SynRM, IM and SRM) can be found in the scientific literature [76–85]. Table 4 collects both general features and quantitative values of each technology. It becomes clear that the PMSM technology is superior regarding power density. Although PM technology shows the best efficiencies bellow base speed, the common operation range of the target electric vehicle must be considered to determine the solution that gives a better overall efficiency during the life-cycle of the vehicle. Generally speaking, in efficiency terms PMSM could be preferred for urban driving, as extra current is required at field weakening operation (high speeds), while IMs and SRMs would be more efficient for high speed sport cars (figure 4). Regarding cost and simplicity, SRM technology could be considered the best option, being this a critical aspect in the automotive industry. However, the disadvantages associated with its technology (high torque ripple, noise, vibrations) seem to favour the IM technology, as it can be confirmed from the market penetration of each technology (table 1). From the reviewed literature [76–85], it can be concluded that the suitability of each technology depends on the specific context of a given propulsion system. Thus, it is not possible to determine the absolute superiority of one technology over others and a technological decision must be conducted after analysing a number of factors for a given application. In this context and once the most significant requirements of a given EV 9 POWER CONVERTER DC Source Cdc Cy Cy Gate Drive Controller A B C D MOTOR Cold plate Power module (A) Driver board (C) DC link capacitors (D) Bus bar (B) Figure 6: Typical layout and elements of a three-phase automotive power converter. capacitor current ripple should be increased. As a consequence, a capacitor with higher RMS current withstanding capability (which is usually bulkier) could be required. Taking into account the aforementioned points, it can be concluded that the most effective approaches that can be followed in order to cope with high speed control issues is the combination of the following solutions:  Include the delay compensation control approach proposed in [127] in the control structure, ensuring by programming that possible measurement synchronization loss is minimized.  Increase of the switching frequency of the power converter [128] in order to reduce the fundamental-to-samplingfrequency ratio and improve both torque and field weakening regulation. This increase on the switching frequency has significant technological consequences, specially in the elements that constitute the power converter, i.e., power semiconductors, driver board, control board, reactive elements and cooling system. In the following, the consequences of this challenges and the technological approaches that overcome them will be thoroughly reviewed. 4. Wide bandgap based power control units for automotive drive systems The power converter can be considered as the core element of the electric drive, as it is responsible for controlling the bidirectional power flow between the electric machine and the battery pack. Figure 6 shows the general diagram of an HEV/EV power converter, which, in this particular case, incorporates a three-phase voltage source inverter. This diagram also shows the common layout of an automotive power converter, including its main functional blocks, i.e., power semiconductor devices, driver board, dc-link capacitor(s) and bus bar. In the following paragraphs, the primary requirements and the major trends of these functional blocks for next generation HEV/EV automotive power converters will be described. The thermal management technology (cold plate in figure 6) is also a key element of a power converter. This block will be separately reviewed in section 5. 16 Table 6: SiC and GaN material properties compared with Si [134]. Material property Si GaN SiC-4H(1) Band Gap (eV) 1.1 3.4 3.2 Critical field (106V/cm) 0.25 0.3 2.2 Electron mobility (cm2/V-sec) 1350 1000 950 Thermal conductivity (Watts/cm2K) 1.5 1.3 3-4 (1) The 4H in SiC-4H refers to the crystal structure of the SiC materials. 4.1. Power semiconductors Power semiconductor devices are the key components in any power electronics system. Their optimization entails improvements in relevant aspects of the electric drive, such as efficiency, reliability, specific mass and volume of power converter, power losses, power density, or quality of synthezised voltage and current waveforms [132–135]. In the particular case of the automotive industry, automotive grade power semiconductors are required for the industrialization of power converters. The automotive grade certificate ensures the quality, performance and safety of the product under the stringent automotive operation conditions. The Automotive Electronics Council has developed the AEC Q101 (Stress Test Qualification for Discrete elements) standard, which establishes common part-qualification and quality-system standards for automotive power semiconductors in terms of life-cycle, operation temperature, humidity and vibrations. In general, current automotive power converters rely on the well established and mature silicon (Si) technology [136–138]. However, with regard to the technological and cost targets for next generation HEV/EV drives (section 1), Si exhibits a number of limitations in terms of blocking voltage capability, operation temperature and switching frequency [133, 139]. As control of future HSEMs will require to operate at high switching frequencies (section 3.3), the relatively high switching losses of Si devices will reduce drastically the efficiency of power converters, leading to the need of complex and expensive cooling systems [133, 140], or even making this technology infeasible for the application. Consequently, the introduction of a new generation of power devices based on wide band gap (WBG) semiconductor materials will be required in future automotive power converters. The larger band gap of WBG materials leads to a lower intrinsic carrier concentration and an increase of the maximum operation junction temperature. Their high thermal conductivity also reduces the thermal resistance of the device [134]. Thus, Si drawbacks could be overcome to a great extend with this technology [135]. Silicon carbide (SiC) and gallium nitride (GaN) are currently considered as the most matured WBG technologies [132–135, 141–144]. The high critical field of SiC and GaN allows to operate at higher voltages when compared to Si devices [145, 146]. Likewise, GaN has higher electron mobility than SiC, meaning that, theoretically, GaN devices are the best suited for high switching frequency operation. However, SiC exhibits a higher thermal conductivity. Consequently, SiC devices could be preferred for high power density applications [145]. Although GaN technology shows superior features than SiC in most parameters, its lower thermal conductivity becomes a big challenge for system designers [147]. As a summary, table 6 shows a comparison of the most relevant physical parameters of Si, SiC and GaN materials. In the following, SiC and GaN diode and active switch semiconductors are reviewed. 17 The requirements to be met for them in automotive electric drives will be discussed. A. SiC devices As in silicon technology, there are different SiC devices, among which stant out SiC based Schottky barrier diodes (SBD) and junction barrier diodes (JBD), PiN diodes, JFETs, BJTs and MOSFETs, which have been available in the market since some years ago. Regarding diodes, the junction knee voltage of SiC PiN diodes makes them ineffective for blocking voltages below 3.3 kV [140, 148]. Likewise, the reverse recovery current of SiC PiN devices results in large reverse recovery losses [148]. Consequently, SiC Schottky diodes could be preferred for automotive applications. Both SBD and JBD diodes are well suited for applications in which high switching frequency operation is required, and they also match perfectly as freewheeling diodes to be paired with Si IGBTs [133, 140]. However, JBDs have two significant advantages over SBDs. On the one hand, they are able to handle higher voltages and, on the other hand, they exhibit a lower leakage current [148, 149]. Typically, blocking voltages of automotive SiC Schottky diodes lay between 600 V and 1700 V, featuring current ranges between 2 A and 47 A (table 7). Regarding SiC power transistors, SiC MOSFETs are preferred as active power switches, mainly because they are able to swicht at high frequencies, because their gate charge is similar to Si MOSFETs and IGBTs, being normally off, and also because they require relatively simple driver circuits [148, 149]. MOSFETs have an intrinsic antiparallel diode (body diode) between the drain terminal and the substrate. For that reason, if necessary, the use of the external freewheeling diode could be discarded, being the body diode which performs this function. There are some modules that follow that criteria. Nevertheless, the use of the external Schottky freewheling diode is recommended, because the overall efficiency and the reliability is improved, and also because the freewheeling dissipation occurs in a different device from the MOSFETs themselves. Current automotive SiC MOSFETs have voltage ranges between 400 V and 1700 V, with current capabilities between 2.6 A and 100 A (table 7). It is expected that commercial products up to 3.3 kV will be available in the near future [148]. Finally, it is important to note that SiC IGBTs are still under development due to reliability problems (mainly due to forward voltage drift). However, some authors consider this technology as the one with the highest potential for future high-voltage applications [139]. Moreover, some manufacturers such as CREE, Infineon, Microsemi and others can make full SiC modules of different topologies (table 8). Their commercialization will depend on overcoming reliability problems through the improvement of the quality of the initial semiconductor material [133]. The scientific community is performing significant advances in that field, being relevant examples the converters based on 15 kV/40 A SiC N-IGBT presented in [150, 151], or the series connected 15 kV SiC IGBTs and MOSFETs presented in [152] for MV power conversion systems. B. GaN devices GaN technology is in its first development stages concerning to power applications. Vertical GaN devices are being considered for research due to their potential. GaN switches are expected to have a 100 times performance improvement over Si-based de18 Table 7: Some of automotive grade and automotive oriented SiC discrete devices available on the market. Company Device Packaging Product name Voltage (V) I (A) @ Tc (ºC) C2M0040120D 1200 40 @ 100 C2M0045170D 1700 48 @ 100 C2M0025120D 1200 60 @ 100 CPM2-1200-0025B 1200 71 @ 100 CPM2-1200-0040B 1200 40 @ 100 CPM2-1700-0045B 1700 48 @ 100 C3D20060D 600 26 @ 135 C3D30065D 650 36 @ 135 C5D50065D 650 50 @ 130 C4D40120D 1200 54 @ 135 TO-247-2 C4D30120D 1200 43 @ 135 TO-220-2 C4D20120A 1200 25 @ 135 CPW5-0650-Z030B 650 30 @ 150 CPW5-0650-Z050B 650 50 @ 175 CPW5-1200-Z050B 1200 50 @ 175 CPW5-1700-Z050B 1700 51 @ 150 GP1T036A060B 1200 50 @ 100 GP1T025A120B 1200 50 @100 GP1T040A120B 1200 31 @ 100 GP1T072A060B 600 25 @ 100 GP2D020A120B 1200 24 @ 150 GP2D020A120B 600 29 @ 150 GP2D050A120B 1200 58 @ 150 IXFN70N120SK 1200 48 @ 100 IXFN50N120SK 1200 33 @ 100 IXFN50N120SiC 1200 35 @ 80 TO-268AA MCB60I1200TZ 1200 60 @ 100 DCG45X1200NA 1200 40 @ 100 DCG85X1200NA 1200 76 @ 100 DCG100X1200NA 1200 94 @ 100 DCG130X1200NA 1200 114 @100 APT40SM120B 1200 29 @ 100 APT70SM70B 700 41 @ 100 APT80SM120B 1200 55 @ 100 APT70SM70J 700 34 @100 APT40SM120J 1200 23 @ 100 APT80SM120J 1200 36 @ 100 TO-247 MSC030SDA120B 1200 30 @ 140 S4001 650 70 @ 25 S4002 650 93 @ 25 S4003 650 118 @ 25 S2301 1200 40 @ 25 S4101 1200 55 @ 25 SCT3017AL 650 83 @ 100 SCT3022AL 650 65 @ 100 SCT3022KL 1200 67 @ 100 SCT3030AL 650 49 @ 100 SCT3030KL 1200 51 @ 100 SCT3040KL 1200 39 @ 100 TO-247 SCH2080KE 1200 28 @ 100 SCS215AG 650 15 @ 175 SCS230AE2 650 30 @ 130 SCS215AE 650 15 @ 130 SCS215AM 650 15 @55 Embossed tape SCS215AJ 650 15 @ 120 SCT50N120 1200 50 @ 100 SCT30N120 1200 34 @ 100 SCTW100N65G2AG 650 85 @ 100 TO-247 STPSC40065C-Y 650 40 @ 130 TO-220AC / D2PAK STPSC20H12-Y 1200 20 @ 150 TO-247 LL STPSC40H12C 1200 40 @ 150 TO-247 / TO-220 AB STPSC20H065C-Y 650 20 @ 150 TO-247 / TO-220 AC / D2PAK STPSC20065-Y 650 20 @ 140 PG-TO220-2-1 IDH20G120C5 1200 20 @ 150 IDW30G120C5B 1200 30 @ 150 IDW40G120C5B 1200 40 @ 148 IDW40G65C5 Scho�ky diode 650 40 @ 110 TO-220AC TO-247 TO-220FM ST Microelectronics MOSFET HIP247 INFINEON PG-TO247-3 Microsemi MOSFET TO-247 SOT-227 ROHM MOSFET Bare Die TO-247N IXYS MOSFET SOT-227B SOT-227B Global Power MOSFET TO-247 TO-247-2L CREE MOSFET TO-247-3 Bare Die Scho�ky TO-247-3 Bare Die * * * * * * * * * * * * * * * * * * * * * * * * * * diode Scho�ky diode Scho�ky diode Scho�ky diode Scho�ky diode Scho�ky diode (∗) Product with automotive grade certification. 19 Table 8: Some of automotive oriented SiC modules devices available on the market. Device Company Product name Voltage (V) I (A) @ Tc (ºC) Half Bridge CAS120M12BM2 1200 138 @ 90 CAS300M12BM2 1200 293 @ 90 CAS300M17BM2 1700 225 @ 90 APTSM120AM14CD3AG 1200 180 @ 80 APTMC120AM08CD3AG 1200 190 @ 80 APTSM120AM09CD3AG 1200 268 @ 80 APTSM120AM08CT6AG 1200 293 @ 80 APTMC120AM12CT3AG 1200 165 @ 80 APTMC120AM09CT3AG 1200 220 @ 80 BSM180D12P2C101 1200 180 @ 60 BSM120D12P2C005 1200 120 @ 60 BSM180D12P3C007 1200 180 @ 60 BSM300D12P2E001 1200 300 @ 60 Semikron SKM500MB120SC 1200 431 @ 80 Six Pack Triple phase leg APTSM120TAM33CTPAG 1200 89 @ 80 APTMC120TAM17CTPAG 1200 110 @ 80 APTMC120TAM12CTPAG 1200 165 @ 80 Rec�fier GHXS020A060S-D1 600 25 @ 125 GHXS030A060S-D1 600 36 @ 150 GHXS030A120S-D1 1200 30 @ 125 GHXS045A120S-D1 1200 45 @ 150 APT40DC60HJ 600 40 @ 100 APT40DC120HJ 1200 40 @ 100 MOSFET module Scho�ky diode module CCS020M12CM2 SK45MAHT12SCp MOSFET module MOSFET module 1200 20 @ 90 Semikron 1200 38 @ 70 Microsemi Global Power Microsemi CREE Microsemi ROHM CREE u v w vices, and a 10 times improvement over SiC, due to the excellent material characteristics such as high electron mobility, high breakdown field and high electron velocity [153]. GaN shows the highest breakdown voltage level with the lowest conduction resistance for the same material area [148]. However, one of the major drawbacks of GaN devices resides in one of their main theoretical advantages, i.e., the ultra-fast switching speeds of these devices (of around 1-2 ns), which makes parasitic inductances become a serious problem [149]. Currently, most of the commercialized GaN Schottky diodes (600 V/10-15 A) are either lateral or quasi-vertical, due to the lack of electrically conducting GaN substrates [133, 140]. Moreover, GaN material is very expensive, so that manufacturers prefer to design diodes with heterogeneous structures such as silicon, SiC, and even zafire [147]. On the other hand, the great majority of the existing GaN active switches are High Electron Mobility Transistors (HEMTs) and their derivatives [133, 139, 140]. An HEMT is an heterojunction device composed by an AlGaN layer formed on top of a GaN substrate. These devices represent a remarkable tradeoff between specific on-resistance and 20 Table 9: Some of automotive grade and automotive oriented GaN discrete devices available on the market. Company Device Packaging Product name Voltage (V) I (A) @ Tc (ºC) GS66516T 650 V 47 @ 100 GS66516B 650 V 47 @ 100 Transphorm GaN FET TO-247 TPH3205WSBQA 650 V 22 @ 100 GaN Systems E-HEMT GaNPX packaging * (∗) Product with automotive grade certification. breakdown voltage [133]. GaN HEMTs are inherently normally-on devices, but high power applications require normally-off (safe) devices with high currents and voltage capability [133, 153]. One solution for the normally-on challenge is to make a cascode connection of a GaN HEMT with a low voltage Si MOSFET. In that way, the gate drive is the same as for a Si MOSFET (normally-off) [146, 149]. However, the main disadvantages or current automotive GaN devices are the low voltage and current ratings (table 9) and, on the other hand, the need to package two switching devices in a leaded multichip package [146, 149]. Additionally, there are only a few GaN power devices available in the market (table 9), as most of them are prototypes. C. Electrical requirements for automotive WBG power stage dimensioning Once the most suitable power semiconductor devices available in the market have been identified, it is important to define the electrical requirements (in terms of voltage blocking and current handling capabilities) to be met by next generation HEV/EV power converters. Voltage requirements mainly depend on the nature of the vehicle and electric machine. In this context, nominal voltages ranging from 100 V to 150 V can be tipically found in mild HEV battery packs [154]. Regarding heavily electrifiec vehicles, higher voltages are the norm. For example, nominal voltages between 200 V and 360 V are common in light to medium weight vehicles mounting medium voltage electric machines [44, 103, 154–156], while voltages close to 800 V to 900 V are the norm in heavy duty vehicles including high power voltage machines [156, 157]. It is important to note that this voltages can become higher depending on the battery State of Charge (SoC) or during regenerative braking, as well as overvoltages that appear in the switching path due to the parasitic inductances. Depending on the battery pack maximum voltage ratings, SiC or GaN devices belonging to the 600 V, 1200 V or 1700 V families should be selected. The determination of the current requirements is not straightforward, as it will depend on a number of factors, such as the machine maximum torque and power ratings, and also on its torque per ampere production capabilities [44, 158, 159]. In synchronous machines based on PMs, field weakening current injection requirements must also be determined considering the worst operation scenario (i.e., considering the minimum dclink operation voltage) [42, 44]. Thus, an analysis of the specific drive would be required for the determination of the exact current requirements. From the available literature, it can be concluded that maximum currents ranging from 200 A to 255 A are common for drive systems including 200 V -400 V battery packs and synchronous machines with power ratings between 50 kW and 70 kW [44, 103, 160]. Similar requirements can be found for IM technologies, as current ratings of around 235 A are common for 50 kW IMs [47]. 21 Significant current handling capabilities will be required for higher power vehicles or overload operation conditions (80 kW-125 kW). For example, maximum currents between 400 A and 480 A are required for the synchronous machines presented in [160, 161]. For these power ratings, less current would be required in electric drives including high voltage electric machines and battery packs [156]. However, this type of machines/battery packs are, in most cases, intended for higher power applications, requiring significantly higher current ratings. Thus, considering the actual HEV/EV requirements and automotive SiC and GaN devices in the market, and taking into account the current ratings of discrete devices, converters based on this technology could cover the aforementioned requirements by means of discrete/bare die device parallelization. The main objective of the parallelization is to increase current capability and, therefore, the converter power capability is also increased. The vast majority of manufacturers offer power modules in which each switch is composed of several semiconductors (bare dies) arranged in parallel, to be able to reach the currents required by the application. To achieve this objective, current distribution for each parallelized device must be as equal as possible. However, there are several factors that cause current imbalances, among others [162]:  Differences in the internal parameters which characterize each device to be parallelized. On the one hand, these parameters affect the static behaviour. The variation of collector-source saturation voltage (VDSsat) and the gate threshold voltage (VGSth) with the temperature involves a current change. In order to prevent high current imbalances, power transistors must have similar characteristic curves. On the other hand, dynamic behaviour is also affected by the variation of internal parameters with the junction temperature, since devices can present different turn on (td,on) and turn off (td,off ) delays which can produce current imbalances [163, 164]  Driver circuit design. An optimal drive circuit gate design is essential, since it has a direct influence over current balance. In this sense, the design must be symmetrical. In order to achieve this objective, driver output impedance must be controlled. There are also in the technical literature various control strategies of common or separate gate [165, 166]  Power circuit layout design. It is fundamental to minimize, as much as possible, parasitic inductances to reduce the voltage peaks caused by di/dt through the switching path [167]. 4.2. Reactive elements and bus bar As shown in figure 6, the power conversion stage may include a number of passive elements between the vehicle battery and the power semiconductors, such as Y-type capacitors (CY) and dc-link capacitors (Cdc). The dc-link capacitor (or dc-link capacitor bank) can be considered as the most relevant reactive element in an automotive power converter. In order to prevent significant battery degradation, battery current ripple should be kept bellow 10% of the rated battery current [168, 169]. The dc-link capacitor must filter the high frequency current 22 MPPF-Caps or MLC-Caps Case I - high ripple current applications Al-Caps Case II - low ripple current applications IC2 C1C2 IC1 Ripple current rating Capacitance Figure 7: Capacitance requirements of various capacitor technologies under high current ripple operation (adapted from [172]). ripple produced by semiconductor switching at the converter input side. On the other hand, as power electronics are one of the major EMI sources in a vehicle, Y capacitors can be mounted from the dc-bus terminals to the vehicle chassis (figure 6) for effective EMI reduction [170]. A number of design aspects must be considered when selecting or designing this critical element. Firstly, the capacitor must be rated according to the battery voltage and must be able to withstand the maximum rms current that will circulate through it. The capacitance value must be selected to keep the current ripple bellow the aforementioned levels. At the same time, the high frequency voltage ripple across the dc-link capacitor should be limited to ±10% of the rated voltage for all expected load conditions [168, 171]. This second aspect is relevant to guarantee a proper torque control [171]. Another critical aspect is the stray inductance (ESL). The capacitor ESL must be low enough to avoid overvoltage failures during semiconductor’s turn-off switching and also to reduce switching losses on the semiconductor [169]. This is of particular importance when using WBG semiconductors with high speed switching capabilities (section 4.1). The dc-link is a bulky element, and one of the main fault sources in modern power converters [172–174], being, in general, its life time shorter than that of the other components of the converter [175]. Thus, proper capacitor dimensioning and thermal managing must be carried out in order to extend the capacitor life cycle [172]. The capacitor parasitic resistance ESR is responsible for its heating; thus, the lower the ESR the better the dc-link reliability. Aluminum Electrolytic Capacitors (Al-Caps), Metallized Polypropylene Film Capacitors (MPPF-Caps) and high capacitance Multi-Layer Ceramic Capacitors (MLC-Caps) are commonly used in high voltage dc-link applications. In practice, Al-Caps and MLCCaps allow high energy densities up to 2 J/cm3[172]. However, MPPF-Caps are preferred for automotive applications, as they provide a well-balanced performance in terms of cost, ESR value, required capacitance vs current ripple (figure 7) and reliability [172, 174, 176– 23 Table 10: Automotive grade metallized polypropylene film capacitors capacitors available on the market for dc-link applications. Manufacturer Family Vnom (V) Capacity (µF) Imax (Arms) ESR (mΩ) ESL (nH) Description Epcos PCC M651/M652 450 300-500 110-140 0.6 25 For M651/M652 modules Epcos PCC HP1 450 300-500 80 0.5-1.0 25-30 For HybridPack 1 packaging Epcos PCC HP2 450 500-1000 120-135 0.4-1.0 15 For HybridPack 2 packagingKement C4E series 450 500 120 1.0 15 nH Vishay Dale MKP1849 500 500-1000 120-160 0.25-0.3 15-20 Vishay Dale MKP1848 450-1200 1-400 2.5-54.0 1.3-54 >1 For PCB mounting Superior Intermediate Inferior Relative performance Capacitance Voltage Ripple current ERS and DF Frequency Cap. Stability Vol. derating Temperature Reliability Energy density Cost Al-Caps MPPF-Caps MLC-Caps Figure 8: Performance comparison of the main types of capacitors for dec-link applications (adapted from [172].) 178]. Figure 8 shows a performance comparisons between Al-Caps, MPPF-Caps and MLC-Caps. MPPF-Caps have a superior performance, they provide a well-balanced performance for high voltage applications (above 500 V) in terms of ESR and cost. The cost of MPPF-Caps is about 1/3 of Al-Caps, and it implies the possibility to achieve a lower cost, higher power density dc-link design with MPPF-Caps in high ripple current applications, such as in electric vehicles [172]. Table 10 summarizes various automotive grade MPPF dc-link capacitors available on the market, including their most significant features. In general, two packaging technologies can be distinguished: (a) Capacitors with integrated bus bars (figure 9(a)), which are specifically designed for various particular automotive power module layouts. These solutions provide easy integration and high capacitances with a very low ESL (table 10). (b) Capacitors for PCB or bus bar mounting (table 10 and figure 9(b)). They allow flexible capacitor bank and bus bar desings, which can effectivelly reduce the value of ESL and improve semiconductors switching dynamics. For example, in [169], the DC link inductance has been reduced from 15,4 nH to 2,8 nH using an optimized switching cell. Thus, proper bus bar desing with low stray inductance is required when following the PCB or bus bar approach and fast switching WBG devices are used in the power 24 (a) Epcos B25655J4507K005 automotive grade capacitor with integrated bus bar. (b) Vishay MKP1848 family through-hole automotive grade capacitor. Figure 9: Automotive grade high power MPPF-Caps. stage [179]. A number of considerations must be taken into acount regarding this critical element. On the one hand, power semiconductors and DC-link capacitor geometry must be properly selected in order to optimize the power density, as well as to minimize the bus bar complexity. The bus bar shape will be directly influenced by the power conversion stage required by the specific vehicle drive topology (three-phase, dual three-phase, asymmetric H-bridge, etc.). If capacitor cooling is required, this will also influence the design geometry [180]. On the other, terminal connections have a significant impact on stray inductance and their influence must be considered, and special care must be taken regarding sharp corners and bends, as they can cause eddy currents and, consequently, voltage drops which result in additional losses and heat generation [180]. Finally, it must be borne in mind that the ESR of an MPPF capacitor depends on both capacitor temperature and current ripple frequency. In this case, the temperature dependency is low. However, the ESR significantly increases while increasing the switching frequency of the converter [168]. Thus, this aspect should be carefully analysed in high WBG power converters with high switching frequency operation. 5. Advanced power converter thermal management An effective vehicle cooling system architecture and performance enables a more compact power converter packaging, reducing the propulsion system size and weight, while increasing system reliability and power density [181–184]. Nevertheless, current approaches for power converter cooling, in general, do not simultaneously meet the future cost, performance and size targets (table 2). For this reason, thermal management represents one of the major technical challenges for HEV/EV drive designers and manufacturers [181, 183]. Thermal resistance (Rth) is one of the most important parameters to be considered in the design of an automotive power converter. It indicates its heat transfer capability and determines the temperature gradient between the heat source and heatsink. For a layer of a given material, Rth is determined by the material thickness (t), thermal 25 Liquid Vapor Heat Source 123 4 Working fluid evaporates to vapour absorbing thermal energy. Vapour goes along cavity to lower temperature end. Vapour condenses back to fluid and is absorbed by the wick, releasing thermal energy. Working fluid flows back to the higer temperature end. 1 2 3 4 Heat sink High temperature (Evaporator) Low temperature (Evaporator) Environment temperature Wick Casing Figure 14: Heat pipe based cooling operation principle. by phase change, i.e., the absorption of thermal energy in the process of evaporation (liquid to vapour) and release of heat in the condensation process (vapour to liquid). A number of two-phase cooling alternatives, such as heat pipes, thermosyphon cooling and thermoelectric cooling can be highlighted [216, 244–246]. Heat pipe technology relies on natural forces to transfer heat. Heat pipes are composed by three main elements: container, fluid and wicked structure (figure 14). The cooling liquid is evaporated by the heat source at one end of the heat pipe. The vapour is transferred to the opposite end by convection, where it is condensed and the heat is transferred to the environment through a heat exchanger. The vapour cools and condenses into liquid, and is carried back to the heat source through the capillary wick structure along the perimeter of the heat pipe. To form the capillary structure of the heat pipe, a porous material is applied on the inner wall of the pipe. This can be done using either metal foams (such as steel, aluminium, copper or nickel) or using carbon fibres [182, 247]. Due to the importance of the wick structure, many works are focused in its development and manufacturing in order to improve the heat transfer [247, 248]. The main advantages of heat pipes are their flexibility to be produced in all forms and sizes and the obtained high heat transfer capability (heat fluxes in the range of tens and hundreds of W/cm2). Operation temperature limitations and vertical height are their main drawbacks [216, 244]. The applicability of this cooling technology in EVs is demonstrated in [244, 249]. However, there are more advanced methods able to increase operation range while improving heat dissipation, as is the case of pulsating heat pipes (PHP) [250–253]. On the other hand, thermosyphon cooling (figure 15) is a special type of heat pipe in which the fluid is driven only by gravity forces [216]. The heated liquid is less dense than the cooler water. Therefore, it rises to the top of the cooling system, forcing natural circulation of the cooling liquid. Then, the heat is dissipated to the ambient air through a condenser. Finally, the cold liquid returns to the source and the cooling process starts 32 Heat source Evaporator Condenser Fan Heat Column with hot fluid Column with cold fluid Figure 15: Thermosyphon cooling operation principle. again. Unlike the heat pipe, it does not incorporate a wick structure through which the condensed working fluid returns. The main disadvantage of this technology is that the system must always be positioned in a vertical direction requiring a large amount of space. Thermosyphon technology achieves heat flux transfer values of around 230-240 W/cm2, with moderate temperature fluctuations [245]. Many efforts have been done to improve the heat dissipation capability. In [245], a pump has been inserted in the loop to vary the flow rate of the coolant, allowing a cooling capacity up to 500 W/cm2with a thermal resistance of 0.125 (Kcm2)/W. Likewise, NREL (National Renewable Energy Laboratory) has developed a thermosyphon based system to cool the power electronics asociated to an EV motor. With this approach, 3.5 kW of semiconductor losses have been dissipated [254]. In [255], a power module cooling system consisting of a thermosyphon circuit composed by double-sided evaporator has been presented allowing a total dissipation power of 1500 W. Thermoelectric cooling (TE) must also be considered. TEs are based on the Peltier effect [256]. When a dc current flows through the module, a heat transference is produced so that each side of the module is cooled or heated respectively (figure 16). This allows to cool each semiconductor device individually and uniformly, leading to semiconductor isothermalization [256]. From the electric design point of view, this improves the prediction of device failures, thermal stress, reliability and lifetime of the power module [216, 246, 249]. Precise temperature control, fast dynamic response, reduced weight and small size are the main advantages of this technology, although the maximum power dissipation is limited to several tens of watts, the efficiency is also low and its cost is high [216, 246]. The application of TE cooling to cool power electronics circuits represents a serious handicap [216, 246]. However, some recent approaches increase both the efficiency and operation temperature [257]. Hybrid solutions have been also proposed [258], where cold 33 Power device TE element NNPP Heat sink DC source Electrical insulation Heat dissipation Figure 16: Diagram of a thermoelectric cooler. plate liquid cooling is combined with TE cooling. The cold plate is used to cool the power module globally, while the embedded TE is used to ensure the temperature uniformity in the semiconductor. Finally, the solution presented in [259] is also worth to mention, as it proposes an innovative two-phase cooling system that uses conventional air conditioning components already available in vehicles, together with a conventional water cooled cold plate, not requiring the development of new technology, while achieving the benefits of two-phase cooling. 5.3. Other cooling technologies for high power applications Spray cooling [260] and jet impingement [261] technologies are somehow popular in the industry. Up to date, they have not been used in marketed HEV/EVs, but they can be considered as promising options, since they are a cooling technologies capable of dissipating the large heat fluxes required in a high-power electronics devices [262]. With spray cooling, very high heat transfer coefficients can be achieved with relatively low coolant flow, leading to light and compact cooling systems [260]. Following this approach, the refrigerant is sprayed into fine droplets that individually impinge on the surface area to be cooled (figure 17). Spraying reduces flow rate requirements, but requires a high pressure at the spray nozzle. NREL has developed a cooling system based on spray cooling [203]. This approach can dissipate heat fluxes between 150 and 200 W/cm2. Likewise, studies carried out in [263–266] demonstrate the effectiveness of this cooling approach. Jet impingement cooling is similar to spray cooling, but is performed with a lower pressure drop at the spray nozzle and a higher coolant flow, reducing noise and the possible spray nozzle clogging [261, 267]. As an example, [268] and [269] present a power module cooling system based on this technology. Figure 18 shows a comparison of heat transfer coefficient shapes of both jet and spray cooling technologies. Danfoss Silicon Power has developed a system called Shower Power wich is based on this approach [201]. This eliminates the temperature gradients, allowing to homogeneously cool large power modules, and the parallel placement of many power chips and improving their operation life-cycle [201]. Another advantage of this technology is 34 Solder Cu IGBT Diode Substrate: Al2O3 or AlN Cooler Spray nozzle Figure 17: A power module cooled by spray cooling technology. Heat Source h Jet Spray Figure 18: Heat tranfer coefficient shapes of jet impingement and spray cooling. the possibility of using low-cost materials, such as plastics, to manufacture the cooling device [234]. In [181, 261, 267, 270], the integration of jet impingement technology in transport applications has been studied with promising results. 6. Conclusions From the reviewed literature, it becomes clear that the development of a sustainable road transport system is of capital importance due to environmental, societal and economical factors. Optimistic forecasts related with future HEV/EV stock evolution should be achieved in order to significantly reduce in road transport GHG emissions. In order to make HEV/EV an attractive option for consumers, a number of technical aspects, such as efficiency, reliability, autonomy and cost must be improved. Being the electric drives a key element of HEV/EV technologies, improvements on the following aspects will be required: 35 1. A relevant evolution in electric machine technologies will be required for next generation drives. In this context, a significant cost reduction is pursued and, at the same time, power densities must be improved while reducing power losses and improving efficiency. In some aspects, these targets lead to opposite design concepts. Cost reduction requires the investigation on non rare earth based machine technologies (IMs, SRMs and PM-assisted SynRMs are being mainly considered). However, this technologies have a significantly lower power density than rare earth based machines, making it difficult to achieve the volume reductions targeted by the international agents. Thus, it can be concluded that an extensive research in HSEM technology (with mechanical speeds beyond 10000 rpm) would be conducted in the following years in order to simultaneously achieve the aforementioned targets. Due to the specific requirements of automotive machines, a number of electromechanical aspects and the usage of advanced materials should be investigated for the success of this high speed technology. 2. Torque control of automotive HSEMs can be challenging, mainly due to the high fundamental-to-sampling frequency ratios for proper current regulation, and due to the difficulties to implement reliable field weakening algorithms with such ratios. As it could difficult to implement reliable advanced regulation structures under high non-linearities (common in automotive machines), it can be concluded that a significant reduction of the execution time of the controller is required for next generation HEV/EV drive controllers. 3. As this controller execution time reduction implies the increase of the switching frequency of the power converter, and taking into account that international agents require a significant improvement in the converter efficiency, it becomes mandatory to introduce WBG technologies with very low switching losses in automotive drive systems power conversion stages. From the available WBG technology, SiC Schottky diodes and SiC MOSFETs could be preferred due to their features. Current handling requirements could be achieved by means of the parallelization of available automotive grade discrete devices, bare dies, or using high power modules from the major WBG power semiconductor manufacturers. The particularities derived from the high switching speed of such WBG devices will require a proper power converter layout design, specially regarding bus bar and power semiconductor placement. In this context, it is concluded that MPPF-Caps could be preferred due to their well-balance performance for high voltage applications. It will be also required to use capacitors with the lowest ESL possible in order to reduce the stray inductances and, consequently, minimize the overvoltages produced during fast switching. 4. Finally, a number of relevant considerations regarding future HEV/EV drive cooling systems must be taken into account. Being the high power density a mandatory requirement, current state of the art shows that most commercial HEV/EVs use separate liquid-cooling systems for the power electronics and the electric machine. Generally, direct cooling structures are used in the power converters in order to reduce the thermal resistance and improve the reliability of the drive. However, the costs associated with this cooling architecture are high. 36 One clear tendency for cost reduction leads to the elimination of one of the liquid cooling loops. However, this implies that the nominal coolant temperature of the power semiconductors is increased from 65 ◦C to 105 ◦C. This requires an optimization of the cooling system for thermal resistance minimization, and a number of cooling concepts such as double sided cooling or microchanneled cold plates are being implemented and investigated. On the other hand, high performance aircooling is also being considered to achieve significant cost reductions. However, the cooling efficiency is very low when compared to liquid-cooling, and optimized designs are required in order to make use of the circulating air flow produced when the vehicle is in motion. In both scenarios, reduced power losses of WBG devices would make them mandatory for the power conversion stage. Last but not least, it is important to point out that a great number of investigations regarding alternative cooling architectures, such as spray cooling, jet impingement cooling, heat pipes, thermosyphon and thermoelectric cooling are being considered, among others. This technologies exhibit interesting features, but they must be further investigated for their implementation in real HEV/EVs. 7. Acknowledgements This work has been partially supported by the Department of Education, Linguistic Policy and Culture of the Basque Government within the fund for research groups of the Basque university system IT978-16, by the Ministerio de Econom´ıa y Competitividad of Spain within the project DPI2014-53685-C2-2-R and FEDER funds and by the Government of the Basque Country within the research program ELKARTEK as the project KT4TRANS (KK-2015/00047 and KK-2016/00061), as well as by the program to support the specialization of Ph.D researchers at UPV/EHU ESPDOC16/25. Bibliography References [1] L. Kumar, S. Jain, Electric propulsion system for electric vehicular technology: A review, Renewable & Sustainable Energy Reviews 29 (2014) 924–940. [2] J. Riba, C. Lopez-Torres, L. Romeral, A. Garcia, Rare-earth-free propulsion motors for electric vehicles: A technology review, Renewable and Sustainable Energy Reviews 57 (2016) 367–379. [3] M. Kumar, S. Revankar, Development scheme and key technology of an electric vehicle: An overview, Renewable and Sustainable Energy Reviews 70 (2017) 1266–1285, cited By 1. [4] Paris declaration on electro-mobility and climate change and call to action, Tech. rep. (2015). [5] International Energy Agency, Energy technology perspectives 2017, Tech. rep., International Energy Agency (June 2017). URL https://www.iea.org/etp2017/summary [6] International Energy Agency, Energy climate change & environment, Tech. rep., International Energy Agency (2016). [7] United States Environmental Protection Agency (EPA), Inventory of u.s. greenhouse gas emissions and sinks: 1990-2015, Tech. rep., U.S. Government (2017). [8] United Nations, Department of Economic and Social Affairs, Population Division , World Population Prospects: The 2017 Revision, Tech. rep., United Nations (2018). [9] U.S. Energy Infomation Administration, Annual Energy Outlook of 2018, Tech. rep., U.S. Energy Infomation Administration (2018). [10] International Energy Agency, Global EV outlook, beyond on million electric cars, Tech. rep., International Energy Agency (2017). 37 [11] Basque energy cluster, Estudio de mercado de veh´ıculo el´ectrico y sus infraestructuras de recarga, Tech. rep., Basque energy cluster. [12] S. Rogers, S. Boyd, Overview of the DOE VTO Electric Drive Technologies R&D Program, Tech. rep., Vehicle Technologies office (U.S Department of Energy) (2016). [13] Ko, J. and Ko, S. and Son, H. and Yoo, B. and Cheon, J. and Kim, H., Development of Brake System and Regenerative Braking Cooperative Control Algorithm for Automatic-TransmissionBased Hybrid Electric Vehicles, IEEE Transactions on Vehicular Technologies 64 (2) (2015) 431– 440. [14] Xu, G. and Li, W. and Xu, K. and Song, Z., An Intelligent Regenerative Braking Strategy for Electric Vehicles, Energies 4 (2011) 1461–1477. [15] Heydari, S. and Fajiri, P. and Rasheduzzaman, M. and Sabzehgar, R., Maximizing Regenerative Braking Energy Recovery of Electric Vehicles Through Dynamic Low-Speed Cutoff Point Detection, IEEE Transactions on Transportation Electrification 5 (1) (2019) 262–270. [16] Chen, Y. and Chen, S. and Zhao, Y. and Gao, Z. and Li, C., Optimized Handling Stability Control Strategy fora Four In-Wheel Motor Independent-Drive Electric Vehicle, IEEE Access 7 (2019) 17017–17032. [17] Sun, H. and Wang, H. and Zhao, X., Line Braking Torque Allocation Scheme for Minimal Braking Loss of Four-Wheel-Drive Electric Vehicles, IEEE Transactions on Vehicular Technology 68 (1) (2019) 180–192. [18] Yuan, Y. and Zhang, J. and Li, Y. and Li, C., A Novel Regenerative Electrohydraulic Brake System: Development and Hardware-in-Loop Tests, IEEE Transactions on Vehicular Technology 67 (12) (2018) 11440–11452. [19] J. F. Miller, D. Howell, The ev everywhere grand challenge, EVS27 International Battery, Hybrid and Fuel Cell Electric Vehicle Symposium (2013) 1–6. [20] Horizon 2020, The EU framework programme for research and innovation. [link]. URL https://ec.europa.eu/programmes/horizon2020 [21] United States Council for Automotive Rresearch LLC. [link]. URL http://www.uscar.org/guest/index.php [22] United nations economical and social commission for Asia and the Pacific. [link]. URL http://www.unescap.org/ [23] Energy Efficiency & Renewable Energy, Multi-year program plan 2011-2015: Vehicle technologies program, Tech. rep., Office of Energy Efficiency & Renewable Energy (2010). [24] L. Kumar, S. Jain, Electric propulsion system for electric vehicular technology: A review, Renewable and Sustainable Energy Reviews 29 (2014) 924–940. [25] C. Chan, The state of the art of electric, hybrid and fuel cell vehicles., in: IEEE, Vol. 95, 2007, pp. 704–718. [26] S. Williamson, A. Emadi, Comparative assessment of hybrid electric and fuel cell vehicles based on comprehensive well-to-wheels efficiency analysis, IEEE Transactions on Vehicular Technology 54 (3) (2005) 856–862. [27] H. Righlot, F. Rieck, Energy chain and efficiency in urban traffic for ice and ev, in: Proc. of the EVS Conference, 2013. [28] A. Rousseau, R. Ahluwahlia, B. Deville, Q. Zhang, Well-to-wheels analysis of advanced suv fuel cell vehicles, Tech. rep., Society of Automotive Engineers, Inc. (2003). [29] C, Mahmoudi. and A, Flah. and L, Sbita, An overview of electric Vehicle concept and power management strategies, in: International Conference on Electrical Sciences and Technologies in Maghreb (CISTEM) , no. 1-6, 2015. [30] Das, H.S. and Tan, C.W. and Yatim, A.H.M., Fuel cell hybrid electric vehicles: A review on power conditioning units and topologies, Renewable and Sustainable Energy Reviews 76 (2017) 268–291. [31] Conti, S. and Di Mauro, S. and Raciti, A. and Rizzo, S. and Susinni, G. and Musumeci, S. and Tenconi, A., Solar electric vehicles: state-of-the-art and perspectives, in: International Annual Conference, AEIT, 2018, pp. 1–6. [32] Krishna, S. and Hari, B. and Dheerendra, S., A comprehensive review on hybrid electric vehicles: architectures and components, Journal of Modern Transportation 27 (2) (2019) 77–107. [33] Enang, W. and Bannister, C., Modelling and control of hybrid electric vehicles (A comprehensive review), Renewable and Sustainable Energy Reviews 74 (2017) 1210–1239. [34] T. Alagarsamy and B. Moulik, A Review on Optimal Design of Hybrid Electric Vehicles and Electric Vehicles, in: International Conference for Convergence in Technology (I2CT), 2018. [35] Driving Research and Innovation for Vehicle efficiency and Energy sustainability (USDRIVE), Electrical and electronics technical team roadmap, Tech. rep., Driving Research and Innovation 38 for Vehicle efficiency and Energy sustainability (2013). [36] S. Waye, High temperature air-cooled power electronics thermal design, Tech. rep., National Renewable Energy Laboratory (2014). [37] B. Ozpineci, Annual progress report for the electric drive technologies program, Tech. rep., Oak Ridge National Laboratory (2016). [38] C. Whaling, Ev power electronics cost analysis, IEEE transportation, electrification comunity newsletter (2014). [39] D. Gerada, A. Mebarki, N. Brown, C. Gereda, A. Cavagnino, A. Boglietti, High-speed electrical machines: Technologies, trends, and developments, IEEE Transactions on Industrial Electronics 61 (6) (2014) 2946–2959. [40] A. Bazi, Electric machines and energy storage technologies in EVs and HEVs for over a century, in: Proc. of the International Electric Machines and Drives Conference, 2013, pp. 212–219. [41] T. Finken, M. Hombitzer, K. Hameyer, Study and comparison of several permanent-magnet excited rotor types regarding their applicability in electric vehicles, in: Proc. of the Electrical Power Train Conference (Emobility), 2010. [42] S. Morimoto, Y. Takeda, T. Hirasa, K. Taniguchi, Expansion of operating limits for permanent magnet morot by current vector control considering inverter capacity, IEEE Transactions on Industry Applications 26 (5) (1990) 866–871. [43] E. Trancho, E. Ibarra, A. Arias, C. Salazar, I. L´opez, A. D´ıaz de Guere˜nu, A. Pe˜na, IPMSM Torque Control Strategies based on LUTs and VCT feedback for Robust Control under Machine Parameter Variations, in: Proc. of the IEEE Industrial Electronics Conference (IECON), 2016, pp. 2833–2838. [44] S. Jung, J. Hong, K. Nam, Current minimizing torque control of the IPMSM using ferrari’s method, IEEE Transactions on Power Electronics 28 (12) (2013) 5603–5617. [45] Y. Inoue, S. Morimoto, M. Sanada, Control scheme for wide-speed-range operation of synchronous reluctance motor in m-t frame synchronized with stator flux linkage, IEEJ Journal of Industry Applications 2 (2) (2013) 98–105. [46] Iyer, K.L.V. and Lai, C. and Mukundan, S. and Dhulipati, H. and Mukherjee, K. and Kar, N.C., Investigation of Interior Permanent Magnet Motor With Dampers for Electric Vehicle Propulsion and Mitigation of Saliency Effect During Integrated Charging Operation, IEEE Transactions on Vehicular Technology 68 (2) (2019) 1254–1265. [47] I. Boldea, L. Tutelea, L. Parsa, D. Dorrell, Automotive electric propulsion systems with reduced or no permanent magnet: An overview, IEEE Transactions on Industrial Electronics 61 (10) (2014) 5696–5711. [48] M. Burwell, Performance cost comparison of induction-motor & permanent-magnet-motor in a hybrid electric car, Tech. rep. (2013). [49] M. Soderznik, K. Rozman, S. Kobe, P. McGuiness, The grain-boundary diffusion process in NdFe-B sintered magnets based on the electrophoretic deposition of DyF3, Intermetallics 23 (2012) 158–162. [50] M. Thomson, E. Chang, A. Foto, J. Citron-Rivera, D. Haddad, R. Waldo, F. Pinkerton, Grainboundary-diffused magnets. the challenges in obtaining reliable and representative bh curves for electromagnetic motor design, IEEE Electrification Magazine (2017) 19–27. [51] T. Jahns, Getting rare-earth magnets out of ev traction machines, IEEE Electrification Magazine (2017) 6–18. [52] Y. Guan, Z. Zhu, I. Afinowi, J. Mipo, P. Farah, Difference in maximum torque-speed characteristics of induction machine between motor and generator operation modes for electric vehicle application, Electric Power Systems Research 136 (2016) 406–414. [53] H. Sira-Ramirez, F. Gonzalez-Monta˜nez, J. Cortes-Romero, A. Luviano-Juarez, A robust linear field-oriented voltage control for the induction motor: Experimental results, IEEE Transactions on Industrial Electronics 60 (8) (2013) 3025–3033. [54] S. Bozhko, S. Dymko, S. Kovbasa, S. Peresada, Maximum torque-per-amp control for traction im drives: Theory and experimental results, IEEE Transactions on Industry Applications 53 (1) (2017) 181–193. [55] K. Chau, W. Li, Overview of electric machines for electric and hybrid vehicles, International Journal of Vehicle Design 64 (1) (2014) 46–71. [56] Chinmaya, K.A. and Singh, G.K., Integrated onboard single-stage batterycharger for PEVs incorporating asymmetricalsix-phase induction machine, IET Electrical Systems on Transportation 9 (1) (2019) 8–15. [57] S. Morimoto, S. Ooi, Y. Inoue, M. Sanada, Experimental evaluation of a rare-earth-free 39 PMASynRM with ferrite magnets for automotive applications, IEEE Transactions on Industrial Electronics 61 (10) (2014) 5749–5756. [58] N. Bianchi, Z. Mahmoud, An analytical approach to design the PM in PMAREL motors robust toward the demagnetization, IEEE Transactions on Energy Conversion 31 (2) (2016) 800–809. [59] N. Bianchi, S. Bolognani, E. Carraro, M. Castiello, E. Fornasiero, Electric vehicle traction based on synchronous reluctance motors, IEEE Transactions on Industry Applications 52 (6) (2016) 4762–4769. [60] P. Reddy, K. Grace, A. El-Refaie, Conceptual design of sleeve rotor synchronous reluctance motor for traction applications, IET Electric Power Applications 10 (5) (2016) 368–347. [61] Trancho, E. and Ibarra, E. and Arias, A. and Kortabarria, I. and Prieto, P. and Martinez de Alegria, I. and Andreu, J. and Lopez, I., Sensorless control strategy for light-duty EVs and efficiency loss evaluation of high frequency injection under standardized urban driving cycles, Applied Energy 224 (2018) 647–658. [62] W. Ding, S. Yang, Y. Hu, S. Li, T. Wang, Z. Yin, Design consideration and evaluation of a 12/8 high-torque modular-stator hybrid excitation switched reluctance machine for ev applications, IEEE Transactions on Industrial Electronics (2017) 1–6. [63] E. Afjei, H. Toliyat, A novel multilayer switched reluctance motor, IEEE Transactions on Energy Conversion 17 (2) (2002) 217–221. [64] P. Desai, M. Krishnamurthy, N. Schofield, A. Emadi, Novel switched reluctance machine configuration with higher number of rotor poles than stator poles: concept to implementation, IEEE Transactions on Industrial Electronics 57 (2) (2010) 649–659. [65] C. Lee, R. Krishnan, New designs of a two-phase e-core switched reluctance machine by optimizing the magnetic structure for a specific application: concept, design, and analysis, IEEE Transactions on Industry Applications 45 (5) (2009) 1804–1814. [66] Nguyen, D.M. and Bahri, I. and Krebs, G. and Berthelot, E. and Marchand, C. and Ralev, I. and Burkhart, B. and De Donker, R.W., Efficiency Improvement by the Intermittent Control for Switched Reluctance Machine in Automotive Application, IEEE Transactions on Industry Applications DOI 10.1109/TIA.2019.2906860 (2019). [67] S. Ooi and S. Morimoto and M. Sanada and Y. Inoue, Performance evaluation of a high power density pmasynrm with ferrite magnets, IEEE Transactions on Industry Applications 49 (3) (2013) 1308–1315. [68] A. Vagati and B. Boazzo and P. Guglielmi and G. Pellegrino, Ferrite assited synchronous reluctance machines: a general approach, proceedings of the international conference of electric machines (2012) 1315–1321. [69] Y. Wang, D. Ionel, M. Jiang, S. Stretz, Establishing the relative merits of synchronous reluctance and pm-assisted technology through systematic design optimization, IEEE Transactions on Industry Applications 52 (4) (2016) 2971–2978. [70] B. Bilgin, A. Emadi, M. Kishanamurthy, Comprehensive evaluation of the dynamic performance of a 6/10 srm for traction application in phevs, IEEE Transactions on Industrial Electronics 60 (7) (2013) 2564–2575. [71] J. Ahn, Torque control strategy for high performance sr drive, Journal of Electrical Engineering and Technology 3 (4) (2008) 538–545. [72] D. Cabezuelo, J. Andreu, I. Kortabarria, E. Ibarra, J. Garate, Srm converter topologies for ev application: State of the technology, in: Proc. of the International Symposium on Industrial Electronics (ISIE), 2017, pp. 1–6. [73] R. Suryadevara, B. G. Fernandes, Control techniques for torque ripple minimization in switched reluctance motor: An overview, 8th IEEE International Conference on In Industrial and Information Systems (ICIIS) (24-29) (2013). [74] V. Petrus, A. Pop, C. Martis, J. Gyselinck, V. Iancu, Design and comparison of different switched reluctance machine topologies for electric vehicle propulsion, in: Proc. of the International Conference on Electrical Machines (ICEM), 2010, pp. 1–6. [75] M. Abbasian, B. Fahimi, M. Moallem, High torque double-stator switched reluctance machine for electric vehicle propulsion, in: Vehicle Power and Propulsion Conference (VPPC), 2010 IEEE, 2010, pp. 1–5. [76] T. Finken, M. Felden, K. Hameyer, Comparison and design of different electrical machine types regarding their applicability in hybrid electrical vehicles, Proceedings of the Electrical Machines Conference (2008) 1–5. [77] W. Cao, B. Mecrow, G. Atkinson, J. Bennett, D. Atkinson, Overview of electric motor technologies used for more electric aircraft (mea), IEEE Transactions on Industrial Electronics 59 (9) (2012) 40 3523–3531. [78] M. Zeraoulia, M.and Benbouzid, D. Diallo, Electric motor drive selection issues for hev propulsion systems: A comparative study, IEEE Transactions on Vehicular Technology 55 (6) (2006) 1756 – 1764. [79] D. Dorrel, A. Knight, M. Popescu, L. Evans, D. Staton, Comparison of different motor design drives for hybrid electric vehicles, in: Proc. of the IEEE ECCE conference, 2010, pp. 3352–2259. [80] Z. Yang, F. Shang, I. Brown, M. Krishnamurthy, Comparative study of interior permanent magnet, induction, and switched reluctance motor drives for ev and hev applications, IEEE Transactions on Transportation Electrification 1 (3) (2015) 245–254. [81] P. Zheng, F. Wu, Y. Lei, Y. Sui, B. Yu, Investigation of a novel 24-slot/14-pole six-phase faulttolerant modular permanent-magnet in-wheel motor for electric vehicles, Energies 6 (2013) 4980– 5002. [82] J. Wang, X. Yuan, K. Atallah, Design optimization of a surface-mounted permanent-magnet motor with concentrated windings for electric vehicle applications, IEEE Transactions on Vehicular Technology 62 (3) (2013) 1053–1064. [83] D. Fodorean, L. Idoumghar, M. Brevilliers, P. Minciunescu, C. Irima, Hybrid differential evolution algorithm employed for the optimum design of a high-speed PMSM used for ev propulsion, IEEE Transactions on Industrial Electronics 64 (12) (2017) 9824–9833. [84] D. Fodorean, Study of a high-speed motorization with improved performances dedicated for an electric vehicle, IEEE Transactions on Magnetics 50 (2) (2014) 921–924. [85] E. Oksuztepe, In-wheel switched reluctance motor design for electric vehicles by using a paretobased multiobjective differential evolution algorithm, IEEE Transactions on Vehicular Technology 66 (6) (2017) 4706–4715. [86] K. Kiyota, A. Chiba, Design of switched reluctance motor competitive to 60-kw ipmsm in thirdgeneration hybrid electric vehicle, IEEE Transactions on Industry Applications 48 (6) (2012) 2303– 2309. [87] J. Juergens, A. Fricasse, L. Marengo, J. Gragger, M. De Gennaro, B. Ponick, Innovative design of an air cooled ferrite permanent magnet assisted synchronous reluctance machine for automotive traction application, in: Proc. of the International Conference on Electrical Machines (ICEM), 2016, pp. 803–810. [88] M. Obata, S. Morimoto, M. Sanada, Y. Inoue, Performance of pmasynrm with ferrite magnets for ev/hev applications considering productivity, IEEE Transactions on Industry Applications 50 (4) (2014) 2472–2434. [89] W. Zhao, D. Chen, T. Lipo, B. Kwon, Performance improvement of ferrite-assisted synchronous reluctance machines using asymmetrical rotor configurations, IEEE Transactions on Magnetics 51 (11) (2015) 1–4. [90] M. Obata, S. Morimoto, M. Sanada, Y. Inoue, High-performance pmasynrm with ferrite magnet for ev/hev applications, in: Proc. of the European Power Electronics and Applications Conference (EPE), 2013, pp. 1–9. [91] M. Olszewski, Oak ridge national laboratory annual progress report for the power electronics and electric machinery program, Tech. rep., Oak Ridge National Laboratory (2011). [92] H. Cai, B. Guan, L. Xu, Low-cost ferrite pm-assisted synchronous reluctance machine for electric vehicles, IEEE Transactions on Industrial Electronics 61 (10) (2014) 5741–5748. [93] D. Guerada, X. Zu, H. Zhang, M. Galea, C. Guerada, S. Pickering, High torque-density in-wheel electrical machine for an electric bus, in: Proc. of the IEEE Vehicle Power and Propulsion Systems, 2016, pp. 1–6. [94] F. Barrero, J. Duran, Recent Advances in the Design, Modeling, and Control of Multiphase Machines - Part I, IEEE Transactions on Industrial Electronics 63 (1) (2016) 449–458. [95] F. Barrero, J. Duran, Recent Advances in the Design, Modeling, and Control of Multiphase Machines - Part II, IEEE Transactions on Industrial Electronics 63 (1) (2016) 459–468. [96] P. Zheng, Y. Sui, Z. Fu, P. Tang, F. Wu, P. Wang, Investigation of a five-phase 20-slot/18-pole pmsm for electric vehicles, in: Proc. of the International Conference on Electrical Machines and Systems (ICEMS), 2014, pp. 1168–1172. [97] J. Karttunen, S. Kallio, P. Peltoniemi, P. Silventoinen, O. Pyrhonen, Decoupled vector control scheme for dual three-phase permanent magnet synchronous machines, IEEE Transactions on Industrial Electronics 61 (5) (2013) 2185–2196. [98] S. Kallio, M. Andriollo, A. Tortella, J. Karttunen, Decoupled dq model of double-star interiorpermanent-magnet synchronous machines, IEEE Transactions on Industrial Electronics 60 (6) (2013) 2486–2494. 41 [231] N. Sahiti, A. Lemouedda, D. Stojkovic, F. Durst, E. Franz, Performance comparison of pin fin in-duct flow arrays with various pin cross-sections, Applied Thermal Engineering 26 (11) (2006) 1176 – 1192. [232] M. I. Hasan and H. L. Tbena, Enhancing the cooling performance of micro pin fin heat sink by using the phase change materials with different configurations, in: International Conference on Advance of Sustainable Engineering and its Application (ICASEA), 2018, pp. 205–209. [233] K. A. Moores and Y. K. Joshi and G. H. Schiroky, Thermal characterization of a liquid cooled alsic base plate with integral pin fins, IEEE Transactions on Components and Packaging Technologies 24 (1) (2001) 213–219. [234] J. Schulz-Harder, Efficient cooling of power electronics, in: International Conference on Power Electronics Systems and Applications, 2009, pp. 1–4. [235] Infineon, HybridpackT M 2 power module for hybridand electric vehicles, Tech. rep., Infineon (2012). [236] [237] R. Staunton, T. Burress, L. Marlino, Evaluation of 2005 honda accord hybrid electric drive system, Tech. rep., Oak Ridge National Laboratory (2006). [238] AC Propulsion partners with BMW to build 500 electric vehicles, Tech. rep., AC Propulsion (2008). [239] J. Hsu, R. Staunton, M. Starke, Barriers to the application of high-temperature coolants in hybrid electric vehicles, Tech. rep., Oak Ridge National Laboratory Technical Report ORNL/TM2006/514 (2006). [240] A. Pe˜na, Final report summary - eunice (eco-design and validation of in-wheel concept for electric vehicles), Tech. rep. (2015). [241] M. De Gennaro, Syrnemo report summary, Tech. rep., Austrian Institute of Technology (2016). URL http://cordis.europa.eu/result/rcn/192394 en.html [242] M. Chinthavali, J. Cristopher, R. Arimilli, Feasibility study of a 55-kw air-cooled automotive inverter, in: Proc. of the IEEE Aplied Power Electronics Conference (APEC), 2012, pp. 2246 – 2253. [243] B. Wrzecionko, D. Bortis, J. Kolar, A 120c ambient temperature forced air-cooled normally-off sic jfet automotive inverter system, IEEE Transactions on Power Electronics 29 (5) (2014) 2345–2358. [244] T. Hirasawa, M. Ikeda, C. Sasaki, Thermal management system with heatpipe for in vehicle electric devices 44 (2013) 20–25. [245] F. Battaglia, F. Singer, S. V. Dessiatoun, M. M. Ohadi, Comparison of near source two-phase flow cooling of power electronics in thermosiphon and forced convection modes, in: IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm), 2017, pp. 752–758. [246] C. Li, D. Jiao, J. Jia, F. Guo, J. Wang, Thermoelectric cooling for power electronics circuits: Modeling and active temperature control, IEEE Transactions on Industry Applications 50 (6) (2014) 3995–4005. [247] M. Nishikawara, H. Nagano, Optimization of wick shape in a loop heat pipe for high heat transfer, International Journal of Heat and Mass Transfer 104 (Supplement C) (2017) 1083–1089. [248] L. Vasiliev, D. Lossouarn, C. Romestant, A. Alexandre, Y. Bertin, Y. Piatsiushyk, V. Romanenkov, Loop heat pipe for cooling of high-power electronic components, International Journal of Heat and Mass Transfer 52 (1) (2009) 301 – 308. [249] F. P. Brito, J. Martins, L. M. Goncalves, R. Sousa, Modelling of thermoelectric generator with heat pipe assist for range extender application, in: Annual Conference of the IEEE Industrial Electronics Society (IECON), 2011, pp. 4589–4595. [250] M. Mameli, M. Marengo, S. Zinna, Thermal simulation of a pulsating heat pipe: Effects of different liquid properties on a simple geometry, Heat Transfer Engineering 33 (14) (2012) 1177–1187. [251] G. Burban, V. Ayel, A. Alexandre, P. Lagonotte, Y. Bertin, C. Romestant, Experimental investigation of a pulsating heat pipe for hybrid vehicle applications, Applied Thermal Engineering 50 (1) (2013) 94 – 103. [252] S. de Vries, D. Florea, F. Homburg, A. Frijns, Design and operation of a tesla-type valve for pulsating heat pipes, International Journal of Heat and Mass Transfer 105 (Supplement C) (2017) 1 – 11. [253] H, Xu. and P, Zhang. and L, Yan. and D, Xu. and W, Ma. and L, Wang., Thermal Characteristic and Analysis of Microchannel Structure Flat Plate Pulsating Heat Pipe With Silver Nanofluid, IEEE Access 7 (2019) 51724–51734. [254] S. Narumanchi, G. Moreno, J. Jeffers, A passive two-phase cooling system for automotive power electronics, in: Annual IEEE Semiconductor Thermal Measurement and Management Symposium, 48 2014, pp. 1–6. [255] F. Agostini, T. Gradinger, C. de Falco, Simulation aided design of a two-phase thermosyphon for power electronics cooling, in: Annual Conference of the IEEE Industrial Electronics Society (IECON), 2011, pp. 1560–1565. [256] C. J. Mole, D. V. Foster, R. A. Feranchak, Thermoelectric cooling technology, IEEE Transactions on Industry Applications IA-8 (2) (1972) 108–125. [257] E. Strickland, A new kind of cool, IEEE Spectrum 48 (6) (2011) 16–16. [258] P. Wang, P. McCluskey, A. Bar-Cohen, Hybrid solidand liquid-cooling solution for isothermalization of insulated gate bipolar transistor power electronic devices, IEEE Transactions on Components, Packaging and Manufacturing Technology 3 (4) (2013) 601–611. [259] Aranzabal, I. and Mart´ınez de Alegr´ıa, I. and Delmonte, N. and Cova, P. and Kortabarria, I., Comparison of the Heat Transfer Capabilities of Conventional Singleand Two-Phase Cooling Systems for an Electric Vehicle IGBT Power Module, IEEE Transactions on Power Electronics 34 (5) (2019) 4185–4194. [260] J. Kim, Spray cooling heat transfer: The state of the art, International Journal of Heat and Fluid Flow 28 (4) (2007) 753 – 767. [261] K. Gould, S. Q. Cai, C. Neft, A. Bhunia, Liquid jet impingement cooling of a silicon carbide power conversion module for vehicle applications, IEEE Transactions on Power Electronics 30 (6) (2015) 2975–2984. [262] M. D. Clark and J. A. Weibel and S. V. Garimella, Identification of the Dominant Heat Transfer Mechanisms during Confined Two-phase Jet Impingement, in: IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm), 2018, pp. 424–428. [263] H. Bostanci, D. V. Ee, B. A. Saarloos, D. P. Rini, L. C. Chow, Thermal management of power inverter modules at high fluxes via two-phase spray cooling, IEEE Transactions on Components, Packaging and Manufacturing Technology 2 (9) (2012) 1480–1485. [264] D. Bharathan, V. Hassani, Spray cooling: An assessment for use with automotive power electronics applications, Tech. rep., National Renewable Energy Laboratory (2005). [265] L. J. Turek, D. P. Rini, B. A. Saarloos, L. C. Chow, Evaporative spray cooling of power electronics using high temperature coolant, in: Conference on Thermal and Thermomechanical Phenomena in Electronic Systems, 2008, pp. 346–351. [266] R. Mertens, C. L., K. e. a. Sundaram, Spray cooling of igbt devices, Journal of Electronic Packaging (ASME) 129 (3) (2007) 316–323. [267] P. R. Parida, S. V. Ekkad, K. Ngo, Impingement-based high performance cooling configurations for automotive power converters, International Journal of Heat and Mass Transfer 55 (4) (2012) 834 – 847. [268] B. Mouawad and R. Skuriat and J. Li and C. M. Johnson and C. DiMarino, Development of a highly integrated 10 kV SiC MOSFET power module with a direct jet impingement cooling system, in: International Symposium on Power Semiconductor Devices and ICs (ISPSD), 2018, pp. 256–259. [269] F. Zhou and K. W. Jung and Y. Fukuoka and E. M. Dede, Chip-scale cooling of power semiconductor devices: Fabrication of Jet impingement design, in: International Symposium on Power Semiconductor Devices and ICs (ISPSD), 2018, pp. 516–519. [270] A. Bhunia, C.-L. Chen, Jet impingement cooling of an inverter module in the harsh environment of a hybrid vehicle, in: Technical Conference and Exhibition on Integration and Packaging of MEMS, Vol. 4, 2005, pp. 561–567. 49