- Email: [email protected]
A novel multimode hybrid energy storage system and its energy management strategy for electric vehicles
Journal of Power Sources 281 (2015) 432e443
Contents lists available at ScienceDirect
Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour
A novel multimode hybrid energy storage system and its energy management strategy for electric vehicles Bin Wang a, Jun Xu a, *, Binggang Cao a, Xuan Zhou b a b
State Key Laboratory for Manufacturing Systems Engineering, School of Mechanical Engineering, Xi'an Jiaotong University, Xi'an 710049, China Department of Electrical and Computer Engineering, Kettering University, Flint, MI 48504, USA
h i g h l i g h t s A novel topology of multimode HESS is proposed for EVs. The rule-based control strategy and the power-balancing strategy are developed for the mode selection and the power distribution. The energy management strategy is proposed to reduce energy losses in the DCeDC converter. The proposed multimode HESS could extend the batteries life and improve the operation efficiency of the HESS.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 January 2015 Accepted 4 February 2015 Available online
This paper proposes a novel topology of multimode hybrid energy storage system (HESS) and its energy management strategy for electric vehicles (EVs). Compared to the conventional HESS, the proposed multimode HESS has more operating modes and thus it could in further enhance the efficiency of the system. The rule-based control strategy and the power-balancing strategy are developed for the energy management strategy to realize mode selection and power distribution. Generally, the DCeDC converter will operate at peak efficiency to convey the energy from the batteries to the UCs. Otherwise, the pure battery mode or the pure ultracapacitors (UCs) mode will be utilized without the DCeDC converter. To extend the battery life, the UCs have the highest priority to recycle the energy and the batteries are isolated from being recharged directly during regenerative braking. Simulations and experiments are established to validate the proposed multimode HESS and its energy management strategy. The results reveal that the energy losses in the DCeDC converter, the total energy consumption and the overall system efficiency of the proposed multimode HESS are improved compared to the conventional HESS. © 2015 Elsevier B.V. All rights reserved.
Keywords: Hybrid energy storage system Energy management strategy Electric vehicles Direct current to direct current converter
1. Introduction To meet the power demands of an electric vehicle (EV), the design of an energy storage system (ESS) with high power and high energy density is of greatest importance [1,2]. There are some power batteries today with high specific power density [3,4], but volume or size problems could not be ignored. Moreover, a massive source of heat will be created when the batteries meet peak power demands, leading to a short battery life [5,6]. Even worse, a singlebattery ESS also causes an increase in cost compared to a multipower sources [1,5]. To solve the problems listed above, ultracapacitors (UCs) can be utilized in the ESS to meet the high power
* Corresponding author. E-mail address: [email protected] (J. Xu). http://dx.doi.org/10.1016/j.jpowsour.2015.02.012 0378-7753/© 2015 Elsevier B.V. All rights reserved.
demand [5,7e9]. Furthermore, the UCs can be recharged or discharged more efficiently and quickly, which can reduce energy losses and recycle braking energy quickly, thus prolong the battery life [2,7]. In this regard, the hybrid energy storage systems (HESSs) of EVs, which include batteries and UCs, have been widely studied in recent years [7,8,10e12]. To achieve low cost of power conversion, the conventional HESS design usually uses a bidirectional DCeDC converter. The HESSs with single DCeDC converter can be divided into UC/battery configuration [1,7], battery/UC configuration [1,13], and the improved configuration based on both UC/battery and battery/UC HESS [5]. However, the UC/battery and the battery/UC HESS fail to achieve the goal that both the batteries and the UCs can provide power directly to the motor inverter without DCeDC converter, which might result in energy losses in DCeDC converter. Although the improved configuration [5] achieves that both the batteries and
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443
the UCs can provide power directly, the UCs voltage must drop below the battery voltage when only the batteries provide power directly. So the improved configuration is a passive configuration and may fail to achieve the highest efficiency of the HESS. To solve problems discussed above, a novel topology of multimode HESS is proposed in this paper. An N-mosfet is connected to the UCs and a P-mosfet is connected to the batteries. The two energy sources are directly connected to the two ends of the bidirectional DCeDC converter respectively. A power diode is used to avoid the batteries being recharged directly. Thus the UCs can provide peak power directly to the motor inverter and the goal to isolate the batteries from providing peak power is achieved. Besides, the batteries are able to provide the power directly to the motor inverter to reduce energy losses of the DCeDC converter. To achieve the high efficiency of the whole system, the configuration and topology of HESS is not enough. Energy management and control is also crucial to the HESS [2,11,14e16]. Many control methods have been used for HESS energy management [17e20]. However, if different operating modes are considered, the HESS may work better. Multimode transmission technology has been successfully used in most hybrid electric vehicles (HEVs) and EVs, [21e23], but few has been reported in HESS. This paper introduce multimode technology to improve the efficiency of the whole system. In the proposed multimode HESS, the pure battery mode would be selected to meet low power demands while the hybrid output mode would be selected to meet peak power demands. Since the HESS has pure battery mode, pure UCs mode, hybrid output mode and recycle mode, etc. Thus, different modes of energy flows need to be analyzed and properly chosen. Otherwise, the batteries may suffer from frequent charge and discharge operations, leading to short battery life [7,24]. In addition, improper mode selection may cause a voltage fluctuation. As a result, the energy losses of the HESS could be increased, and even the efficiency of the motor inverter could be reduced [25]. In this case, the switch rules of mode selection should be reasonable to ensure the high efficiency of energy management strategy. Based on the discussion above, the mode switching strategy will be developed and analyzed. In this paper, a novel topology of multimode HESS is proposed and its energy management strategy is developed. Section 2 presents the configuration of the proposed multimode HESS and its operating modes. Section 3 is the energy management strategy of the proposed HESS. Section 4 focuses on the simulation and its results. Section 5 is the experimental verification. The main conclusion is given in Section 6. 2. Proposed multimode HESS configuration and its operating modes Batteries have a relatively high energy density [26] and UCs have a significantly higher power density [27]. To prolong the battery life, the UCs are suggested to be connected to the dc link directly so that the batteries could be isolated from providing peak power and being recharged directly [8]. The bidirectional DCeDC converter in the HESS has buck mode and boost mode, and the conversion efficiency in different modes is always considered as the most important factor. However, since the energy loss cannot be avoided in both buck mode and boost mode, to improve the efficiency of the whole system, the UCs and the batteries should provide the power directly without DCeDC converter if possible. Based on the aforementioned discussion, this paper proposes a novel multimode HESS configuration. The proposed circuit structure is illustrated in Fig. 1. The main feature of the proposed multimode HESS is that the N-mosfet (SW1) is connected to the UCs and the P-mosfet (SW2) is connected to the batteries. In
433
Fig. 1. Circuit structure of the proposed multimode HESS.
addition, the two energy sources are directly connected to each the two ends of the bidirectional DCeDC converter respectively. A power diode is used to avoid the batteries being recharged directly. What's more, the SW1 and the SW2 are controlled by the same electrical signal. In this configuration, the UCs can provide peak power directly to the motor inverter when the SW1 is ON (the SW2 is OFF). The batteries provide energy through the DCeDC converter only when the SW1 is ON. Thus the goal to isolate the batteries from providing peak power is achieved. When power demands of the motor inverter are not so very high (referred to as “low power demands” in this paper), the SW2 is ON (the SW1 is OFF), and the batteries provide the power directly to the motor inverter to reduce energy losses of the DCeDC converter. The UCs voltage and battery voltage are important factors for the design of HESS. In this paper, the proposed multimode HESS is designed for a SD-EV which is a pure EV developed in our laboratory. The nominal voltage of the batteries is 288 V; the effective operating voltage range of the batteries is from 252 V to 324 V. On the other hand, the effective operating voltage range of the motor inverter is from 190 V to 380 V. Considering the configuration of UCs module, the maximum voltage of the UCs is designed as 375 V. To ensure enough energy from the UCs, the minimum voltage of the UCs is designed as 190 V. The energy stored in UCs is a function of its voltage as shown in (1). It is noted that the actual energy available of the UCs is less than 25% when its voltage is less than 50%. So the lower limit of the UCs voltage is designed as 50% of the maximum voltage.
Ecap ¼
1 2 CV 2 UC
(1)
To improve the efficiency of the HESS, the DCeDC converter should be properly designed. Since the DCeDC converter could boost or buck the battery voltage, the configuration of the DCeDC converter is designed as Fig. 2. In this configuration, the DCeDC converter could also buck the UCs voltage to recharge the batteries, which ensures the safety of the batteries during regenerative braking. Based on the design above, four operation schemes are proposed as follows: (a) the batteries boost scheme, (b) the batteries buck scheme, (c) the UCs buck scheme, (d) the regenerative braking scheme, as shown in Fig. 3. To avoid damages to batteries, the UCs
batteries
UCs
Fig. 2. Configuration of the DCeDC converter.
434
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443
Fig. 3. The operation schemes of the proposed multimode HESS. (a) The batteries boost scheme. (b) The batteries buck scheme. (c) The UCs buck scheme. (d) The regenerative braking scheme.
are designed to provide the power directly to the motor inverter when the proposed peak power mode is demanded. So the schemes (a) and (b) are used to meet the peak power demands. On the other hand, low power demand has little damage to batteries, so the scheme (c) is designed to meet the low power demands. A higher overall system efficiency could be achieved as the batteries work alone. In this scheme, the UCs needs to provide power to the DCeDC converter only when the battery SOC is lower than a certain level, 0.1 for example. The scheme (d) is used to absorb the regenerative braking energy. In this scheme, the UCs have the highest priority to recycle the energy, and the DCeDC converter bucks the UCs voltage to recharge the batteries to extend the battery life. To explain the energy flows of different schemes, the operating modes of the proposed multimode HESS are categorized and analyzed in detail. The batteries boost scheme has three operating modes. In this scheme, the UCs voltage is higher than the battery voltage, and a high level voltage is used to control the SW1 (ON) and the SW2 (OFF). When the DCeDC converter needs not to work, the pure UCs output mode is implemented. When the DCeDC operates at boost mode with discharging the UCs, the hybrid output mode Ⅰ is implemented. On the contrary, when the DCeDC operates at boost mode with recharging the UCs, the UCs recharge mode Ⅰ is implemented. The energy flows of the three modes are shown in Fig. 4 (a)e(c), respectively. The batteries buck scheme has two operating modes. In this scheme, the UCs voltage is lower than the battery voltage. When the SW1 is ON and the SW2 is OFF, the DCeDC converter operates at buck mode. Fig. 5(a) and (b) show the energy flows of the hybrid output mode Ⅱ and the UCs recharge mode Ⅱ in the batteries buck scheme, respectively. The UCs buck scheme also has two operating modes. In this scheme, the UCs voltage is higher than the battery voltage. When the SW1 is OFF and the SW2 is ON, the batteries provide power directly to the motor inverter. Fig. 6(a) and (b) show the energy flows of the pure battery mode and the hybrid output mode Ⅲ. In the hybrid output mode Ⅲ, the DCeDC converter operates at buck mode, this mode is used only when the battery SOC is less than 0.1.
The regenerative braking scheme is designed to give the highest priority to recycle the braking energy. In this scheme, the SW1 is OFF, and it should be noticed that there is no energy flow through the SW2 because the power diode is reverse biased to avoid the batteries being recharged directly as shown in Fig. 1. The proposed multimode HESS commonly operates in the pure UCs recycle mode. The hybrid recycle mode is activated when the UCs voltage is higher than the upper limit. This mode should be used as less as possible to extend the battery life. Fig. 7(a) and (b) show the energy flows of the two modes. 3. Design of the energy management strategy Since the proposed multimode HESS has nine operating modes, the energy management strategy is designed as two modules: mode selection and power distribution. The schematic of the energy management system is shown in Fig. 8. To ensure the optimized control of mode selection and improve the operating efficiency of the proposed multimode HESS, the rule-based control strategy is developed for mode selection and the power-balancing strategy is developed to distribute the power between the batteries and the UCs. 3.1. Modeling From Fig. 8, it can be seen that the input signals are the power demand from the motor inverter Pmotor, the efficiency of the DCeDC converter hdcdc, the battery SOC and the UCs voltage. To design the energy management strategy, the models of the input signals need to be established. When driving the EV, the power demand from the motor Pmotor [8] can be calculated as
Pmotor ¼ Pdrive =hmotor hinverter
(2)
. Pdrive ¼ Fdrive vev hgb
(3)
Fig. 4. The operating modes in the batteries boost scheme. (a) The pure UCs output mode. (b) The hybrid output mode Ⅰ in the batteries boost scheme; (c) The UCs recharge mode Ⅰ in the batteries boost scheme.
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443
435
Fig. 5. The operating modes in the batteries buck scheme. (a) The hybrid output mode Ⅱ in the batteries buck scheme. (b) The UCs recharge mode Ⅱ in the batteries buck scheme.
Fig. 6. The operating modes in the UCs buck scheme. (a) The pure battery mode. (b) The hybrid output mode Ⅲ in the UCs buck scheme.
where Pdrive is the power requirement for driving the EV, hmotor is the efficiency of the motor, hinverter is the efficiency of the motor inverter, Fdrive is the driving force which is the sum of the rolling resistance force, the climbing force, the braking force etc., vev is the speed of the EV, hgb is the gear efficiency. The efficiency of the DCeDC converter hdcdc is related to the battery voltage, the UCs voltage and its operating mode. hidcdc indicates the efficiency of the DCeDC converter in the batteries boost or buck scheme, and hkdcdc indicates the efficiency of the DCeDC converter in the UCs buck scheme or the regenerative
braking scheme.
(
hidcdc ¼ f1 ðVUC ; Vbatteries ; Pdcdc Þ hkdcdc ¼ f2 ðVUC ; Vbatteries ; Pdcdc Þ
(4)
Compared to the UCs voltage, the battery voltage varies little and affects little on the hidcdc . hkdcdc is only used in the hybrid recycle mode or the hybrid output mode in the UCs buck scheme. Since the two modes should be used as less as possible for improving the efficiency of the whole system, we are only
Fig. 7. The operating modes in the regenerative braking scheme. (a) The pure UCs recycle mode. (b) The hybrid recycle mode.
Fig. 8. The schematic of the energy management system.
436
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443
Fig. 9. The equivalent circuit and the internal resistor in different voltage of the battery.
interested in hidcdc , and the functional relationships (5) could be obtained by corresponding experiment.
8 < hidcdc ¼ f ðVUC ; Pdcdc Þ : P peakefficiency ¼ f 1 VUC ; hpeakefficiency dcdc dcdc
Ibat ¼
0:5 2 4R Ebat Ebat bat Pbattery 2Rbat
(5)
Z ktem
The battery SOC and the UCs voltage are very important factors because they decide the operating modes of the DCeDC converter. In this paper, the open-circuit voltage model of the batteries is utilized, which consists of the controlled dc voltage source Ebat in series with the internal resistor Rbat [28,29]. The equivalent circuit of the batteries and the internal resistor in different voltage are shown in Fig. 9. The battery SOC is calculated as
Vbat ¼ Ebat Ibat Rbat
(6)
Pbatteries ¼ Vbat Ibat
(7)
SOCbat ¼ SOC0 þ
(8)
t
0
Ibat dt
Ct
(9)
where Ibat is charging or discharging current of the battery, Pbattery is the power output or input of the battery. SOC0 is the initial SOC. ktem is temperature coefficient. Ct is the capacity of the battery. By using a similar equivalent circuit, we can calculate the UC voltage
IUC ¼
0:5 2 4R P VUC VUC UC UC
Fig. 10. Mode selection and the control rules.
2RUC
(10)
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443
437
Table 1 Power distribution based on power-balancing strategy. Mode selection OperationMode OperationMode OperationMode OperationMode OperationMode OperationMode OperationMode OperationMode OperationMode
¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼
1 2 3 4 5 6 7 8 9
PUCs
Pbatteries
i; k
DCeDC converter
PUCs ¼ Pmotor PUCs ¼ Pmotor Pdcdc
Pbatteries ¼ 0
i; k ¼ 0 i ¼ 1; k ¼ 0
No energy conversion Peak efficiency of the batteries boost mode
Pbatteries ¼ Pdcdc =hidcdc
Peak efficiency of the batteries buck mode
peakefficience Pdcdc ¼ Pdcdc
PUCs PUCs PUCs PUCs
¼0 Pbatteries ¼ Pmotor max1 ¼ Pmotor Pbatteries =hkdcdc , Pbatteries ¼ 0:5Pbatteries ¼ Pmotor Pbatteries ¼ 0 max ¼ PUCs Pbatteries ¼ ðPmotor PUCs Þ=hidcdc
VUC ðn þ 1Þ ¼ VUC ðnÞ IUC Dt=C
UC
(11)
where IUC is the charging or discharging current of the UC, RUC is the internal resistor, PUC is the power output or input of the UCs, VUC(n) is the controlled dc voltage source, CUC is the capacity of the UCs. 3.2. Mode selection The rule-based control strategy is a simple but effective method, which is developed based on theory or project experience and has been widely used to define the operating mode of HEVs [21,23]. The mode selection and the detailed control rules for the proposed multimode HESS are shown in Fig. 10. 1) To ensure that the UCs can absorb more energy and the hybrid recycle mode can be used as less as possible in the next regenerative braking, the Operation Mode1 ¼ 1 should be selected when Pmotor > 0 and VUCs > 0.95. 2) When Vbatteries < VUCs < 0.85, Pmotor > 0, the DCeDC converter operates at peak efficiency with the batteries boost scheme. If Pmotor > Pdcdc, the OperationMode ¼ 2 will be selected, otherwise the OperationMode ¼ 3 will be selected. 3) When VUCs < Vbatteries, Pmotor > 0, the DCeDC converter must operate at peak efficiency with the batteries buck scheme. If Pmotor > Pdcdc, the OperationMode ¼ 4 will be selected, or the OperationMode ¼ 5 will be selected. 4) When 0.85 VUCs 0.95, Pmotor > 0 and Vbatteries > 300 V, the OperationMode ¼ 6 will be selected if the power demand from the motor inverter is lower than the upper limit of the batteries max1 , or the OperationMode ¼ 2 will be selected. The upper Pbatteries
i; k ¼ 0 i ¼ 0; k ¼ 1 i; k ¼ 0 i ¼ 1; k ¼ 0
the
No energy conversion The UCs buck mode No energy conversion The UCs buck mode
limit
of
max1 Pbatteries
peakefficiency Pdcdc =hpeakefficiency . dcdc
batteries
is
designed
as
¼ 5) When the battery SOC is lower than 0.1 and Pmotor > 0, it is needed to limit the power output of the batteries by the limit max1 . The UCs buck scheme will be used and the Pbatteries ¼ 0:5Pbatteries limit OperationMode ¼ 6 will be selected if the Pbatteries > Pmotor , or the OperationMode ¼ 7 will be selected. It should be noted that the UCs voltage is higher than batteries or the OperationMode ¼ 7 is not available. When the battery SOC is lower than 0.1, the power output of the HESS is limited. We define this as a reduced power mode. The reduced power mode should be used as less as possible to extend the battery life, at this mode the driver must drive the vehicle to the nearest EV charging station. 6) It should be noticed that OperationMode ¼ 6 will be activated in two conditions. When the SW1 is switched from ON to OFF, the SW2 has to be switched from OFF to ON. As a result, it may cause a voltage fluctuation in the output side of the HESS. This may seriously degrade the performance of the motor inverter. To reduce the voltage fluctuation, voltage regulation of the batteries is implemented. In this paper, it is expected that the UCs voltage stabilize around 85% (it equals to 319 V). When the battery SOC is higher than 0.1, OperationMode ¼ 6 must meet the requirement of Vbatteries > 300 V to reduce the voltage difference. In addition, a capacitor or a pre-charging circuit is needed in the motor inverter to reduce the voltage fluctuating. 7) When Pmotor < 0, the regenerative braking scheme is implemented. In this scheme, the OperationMode ¼ 8 is always selected if VUCs < 0.95, or the OperationMode ¼ 9 is selected. In the hybrid recycle mode, the DCeDC converter always operates at buck mode to ensure the battery safety. The hybrid recycle mode should be used as less as possible to extend the battery life.
3.3. Power distribution The power-balancing strategy [30] has been successfully used to distribute the power between batteries and internal combustion
Table 2 The important parameters of the SD-EV.
Fig. 11. The SD-EV.
Total mass of the EV Motor Gear efficiency Radius of wheel UCs Batteries DCeDC converter
1250 kg 30 kW (PMSM) [0.93 0.95 0.97 0.97 0.97] 0.29 m 375 V (Kaimai 75 V 250 F in series) 288 V (HuanYu 3.2 V 80 Ah in series) Peak efficiency region: 10e10.5 kW
438
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443
engine (ICE) in HEVs. However, the electric energy of the UCs is different from the internal energy of the ICE, where the UCs provide peak power, but the ICE provides main power to the drivetrain. In addition, the efficiency of the DCeDC converter should be considered to design the power-balancing strategy. To the proposed multimode HESS, the power demand from motor inverter can be calculated as
Pmotor ¼ Pbatteries hidcdc þ PUCs hkdcdc
(12)
From equation (12), the power-balancing of the proposed multimode HESS is determined by the relationships between the Pmotor, PUCs, Pbatteries, hidcdc and hkdcdc . The power distribution based on power-balancing strategy and the explanations are shown in Table 1. 1) OperationMode ¼ 1, the pure UCs output mode. In this mode, the UCs provide all the energy to the motor inverter. i,k ¼ 0 means that the HESS works without use of the DCeDC converter. 2) OperationMode ¼ 2e3, the hybrid output mode and the UCs recharge mode with the batteries boost scheme. In these two modes, the DCeDC converter operates at boost mode. Since Vbatteries < VUCs < 0.85, the UCs can be recharged or provide peak power. The DCeDC converter operates at peak efficiency to reduce energy losses. For the power-balancing strategy, the power output of the batteries is peakefficience Pdcdc =hpeakefficience . dcdc
Pbatteries ¼ 3) OperationMode ¼ 4e5, the hybrid output mode and the UCs recharge mode with the batteries buck scheme. In these two modes, the DCeDC converter must operate at buck mode since the UCs voltage is lower than the battery voltage. The DCeDC converter also operates at peak efficiency. It should be noted max ¼ that the power demand of the motor must be limited by Pmotor max1 when the UCs voltage is less than 200 V. Pbatteries
4) OperationMode ¼ 6, the pure battery mode. In this mode, the UCs voltage is in the region [0.85, 0.95] and the power demand max1 . The UCs needs not to from motor inverter is lower than Pbatteries be charged, and the batteries can provide all the energy to the motor inverter in this condition. So Pmotor ¼ Pbatteries, PUCs ¼ 0. 5) OperationMode ¼ 7, the hybrid output mode with the UCs buck scheme. It is a reduced power mode because the battery SOC is less than 0.1. The power output of the DCeDC converter is limit Pdcdc ¼ Pmotor Pbatteries . This mode should be used as less as possible, and the driver must drive the vehicle to the nearest EV charging station. 6) OperationMode ¼ 8, the pure UCs recycle mode. The UCs have the highest priority to recycle the energy, so it is designed to recycle all the energy from the motor inverter when VUCs < 0.95. In this condition, PUCs ¼ Pmotor. 7) OperationMode ¼ 9, the hybrid recycle mode. In this mode, VUCs > 0.95. To ensure the safety of the batteries, the regenerative braking power of the motor should be limited, so limit ¼ P max þ P limt hi and the DCeDC converter operPmotor UCs batteries dcdc ates at buck mode to convey the energy from the UCs to the batteries. i ¼ 1 means that the batteries are recharged.
4. Simulation analysis To validate the configuration of the proposed multimode HESS and its energy management strategy, the simulation is established according to the power demand of the SD-EV. The SD-EV and its important parameters are shown in Fig. 11 and Table 2. The simulation model of the drivetrain configuration for the SD-EV with the proposed multimode HESS based on ADVISOR is shown in Fig. 12(a). The Urban Dynamometer Driving Schedule (UDDS) and the New European Driving Cycle (NEDC) are chosen to demonstrate the proposed multimode HESS and its energy management strategy. The power demands from the motor inverter for the SD-EV in
Fig. 12. Simulation setup. (a)The simulation model of the drivetrain configuration for SD-EV with the proposed multimode HESS. (b) The UDDS. (c) The NEDC.
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443
the UDDS and the NEDC are shown in Fig. 12(b) and (c). In the UDDS, the maximum driving power demand and the maximum braking power are 42.9 kW and 11.3 kW respectively. In the NEDC, the maximum driving power demand and the maximum braking power are 57.5 kW and 24.3 kW, respectively. The average power can be calculated as
1 t
Zt Pmotor ðtÞdt ¼ P motor
(13)
0
where P motor is the average power from the start of the trip to the current time instant “t”,Pmotor(t) is the instantaneous power. The P motor in the UDDS and the NEDC can be calculated by (13), which is 7.08 kW and 8.29 kW, respectively. On the other hand, numerous tests of the SD-EV in suburban road have demonstrated that the average power is less than 10 kW. So the peak efficiency power output of the DCeDC converter in the simulation model is designed as 10 kW. The simulation results are shown in Fig. 13. In the Fig. 13(a) and (b), the initial value of the UCs voltage is set to 375 V. Since the UCs voltage is higher than 95% in the inception phase, the pure UCs output mode is selected in both the UDDS and the NEDC. Then the UCs voltage is in the region [85%, 95%], so the
439
hybrid output mode is selected to meet the peak power demands in the UDDS. On the contrary, the pure battery mode is selected when the power demand is lower than the upper limit of the batteries in the NEDC. In addition, the pure UCs recycle mode is selected when the power demand is lower than zero because the UCs can absorb all the energy from the motor inverter. When the UCs voltage is lower than 85% (319 V), the UCs recharge mode in the batteries boost scheme should be selected to meet low power demands. In the NEDC, it can be seen that the power output of the batteries can be maintained at 10 kW after 1000 s because the UCs voltage is lower than the battery voltage. In this condition, the hybrid output mode in the batteries buck scheme is selected. At the end of the NEDC, the UCs voltage increases rapidly because the UCs recharge mode is selected. In the Fig. 13(c) and (d), the initial value of the UCs voltage is set to 340 V. Since the UCs voltage is in the region [85%, 95%], the hybrid output mode or the pure battery mode is selected in the inception phase in the UDDS or the NEDC, respectively. In the UDDS, the UCs voltage drops to 296 V to meet peak power demands. In this condition, the power output of the batteries should be maintained at 10 kW and the hybrid output mode or the UCs recharge mode should be maintained for a long time. The UCs voltage is maintained at around 319 V at last. In the NEDC, we can
Fig. 13. Simulation results. (a) Simulation results in the UDDS when the initial voltage of the UCs is set to 375 V. (b) Simulation results in the NEDC when the initial voltage of the UCs is set to 375 V. (c) Simulation results in the UDDS when the initial voltage of the UCs is set to 340 V. (d) Simulation results in the NEDC when the initial voltage of the UCs is set to 340 V.
440
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443
Fig. 14. Experimental platform.
see the UCs voltage reducing constantly. The power output of the batteries is maintained at 10 kW constantly after 820 s. It should be noticed that the UCs voltage is lower than 200 V in 1120 s. Although the UCs voltage is higher than 50%, it may fail to provide enough energy to meets a constant peak power demands. So if the SD-EV needs a better power performance, it needs a larger size UCs or larger size batteries with high power output, and it also needs a larger size DCeDC converter. The different modes and the power distribution between the batteries and the UCs in different drive conditions seen in Fig. 13(a)e(d), preliminarily prove the availability of the mode selection and the power distribution. 5. Experimental verification To further validate the configuration of the proposed multimode HESS and its energy management strategy, an experimental platform is constructed. The whole system is scaled down for safer experiments. The experimental platform is illustrated in Fig. 14 and the major parameters of its components are listed in Table 3. It should be noticed that Table 3 presents the normal voltage of the batteries. The actual voltage of the batteries is 11.2e14.4 V. In the experiment, the measured efficiency of the DCeDC converter is shown in Fig. 15. It can be seen that the points in the blue line (in the web version) are the highest efficiency operating points of the DCeDC converter at any corresponding voltage. These points are stored in the controller for controlling the DCeDC converter. For this reason, the DCeDC converter will operate at peak efficiency to convey the energy from the batteries to the UCs. So the power
peakefficience
peakefficience
output of the batteries is Pbatteries ¼ Pdcdc =hdcdc . In the experiment, to ensure the safety of the batteries, we define that the up limit of the power output of the batteries is 55 W, which is close to the highest efficiency operating points of the DCeDC converter. Fig. 16(a) illustrates the experimental results with only the batteries. It shows that when the power load is changed from 40 W to 60 W, the battery voltage drops quickly. It is no doubt that batteries will be damaged without any other energy source to meet a higher power load. Fig. 16(b) illustrates the experimental results of the proposed multimode HESS when the UCs voltage is higher than that of the batteries. It can be seen that the power output of the batteries is less than 55 W at any time, which ensures the safety of the batteries at any time and prolongs the life of the batteries. On the other hand, when the UCs voltage is 14.51 Ve14.45 V and the power demand is 40 W, the proposed multimode HESS operates at the UCs recharge mode. When the UCs voltage is 22.05 Ve21.95 V and the power demand is 40 W, the pure battery mode is selected. In the other conditions, the hybrid output mode is selected. Furthermore, it is
Table 3 Experimental parameters. Electric load Power supply Transmission time interval Data type UCs Batteries DCeDC converter
150 V (KL6101 100 A, 1200 W Max) 32 V (IPD3305-SLU 5.2A, 0.03%) 50 Hz Hexadecimal 25 V (Kaimai 25 V 94 F) 12.8 V (HuanYu 3.2 V 80 Ah in series) Peak efficiency region: 49e53 W
Fig. 15. The measured efficiency of the DCeDC converter.
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443 peakefficience
peakefficience
known that Pbatteries ¼ Pdcdc =hdcdc at the UCs recharge mode and the hybrid output mode, so the DCeDC converter operates at peak efficiency. Fig. 16(c) illustrates the experimental results of the proposed multimode HESS when the UCs voltage is lower than batteries voltage. In this condition, the batteries buck scheme is always used. When the power demand is 60 W, the hybrid output mode is selected. When the power demand is changed to 40 W, the UCs recharge mode is selected. In the batteries buck scheme, the DCeDC converter also operates at peak efficiency, so the equation peakefficience Pbatteries ¼ Pdcdc =hpeakefficience also works. dcdc To further validate the rule-based control strategy for the mode selection, the power supply is used to carry out the charging test (input voltage or current). First, the proposed multimode HESS is charged by a 10 W constant power load until the UCs voltage is equal to 25 V. It can be seen that the pure UCs recycle mode is selected, and there are no power flew into batteries in Fig. 16(d). Thus, the batteries are isolated from frequent charge and discharge operations. Second, the 40 W constant power load is applied to the proposed multimode HESS. It can be seen that at this condition the pure UCs output mode is selected, and even the power demand is 60 W, the pure UCs output mode is also selected because the UCs
441
voltage is higher than 23.75 V (95% of the maximum voltage). At last, the hybrid output mode is activated when the UCs voltage is lower than 23.75 V. The mode selection and the power distribution are shown in Fig. 16(d). Fig. 16(b)e(d) show the mode selection and the power distribution in different conditions. The pure battery mode is used to avoid the energy loss of the DCeDC converter when the proposed multimode HESS meets low power demands. The DCeDC converter is always operating at peak efficiency at the hybrid output mode or the recharge mode. The UCs have the highest priority to recycle the energy during regenerative braking for protecting the batteries. So the proposed multimode HESS can reduce the energy loss of the DCeDC converter and extend the battery life. It should be noticed that the fluctuating current (in Fig. 16(b)e(c)) cannot be avoided during the 20 s of mode switching in. Because the input voltage of the load either equals to the battery voltage or the UC voltage, the voltage fluctuating is unavoidable in the output side of the HESS. It is a difficult technical problem when the HESS is applied to the SD-EV, and it might reduce the efficiency of the inverter. Therefore, a capacitor or a pre-charging circuit is needed in motor inverter to reduce the voltage fluctuating. Indeed, the efficiency of the motor inverter is decreased when the HESS switches to the pure battery mode. However, the HESS has no
Fig. 16. Experimental results. (a) Experimental results only with the batteries. (b) Experimental results in the batteries boost scheme. (c) Experimental results in the batteries buck scheme. (d) Experimental results when the UCs voltage is higher than 85%.
442
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443
Table 4 Comparative results with 1/500 of the power demands in the UDDS and the NEDC. Driving cycle
UDDS NEDC
Energy losses in the DCeDC
Total energy consumption
Overall system efficiency
Multimode HESS
Improved HESS [8]
Multimode HESS
Improved HESS [8]
Multimode HESS
Improved HESS [8]
4.7% 4.9%
8.6% 9.1%
21,398 J 21,451 J
21,464 J 21,893 J
90.66% 91.59%
90.38% 89.74%
energy loss in the DCeDC converter with the pure battery mode. So when the battery voltage is close to the UCs voltage, the pure battery mode can significantly improve the overall system efficiency. To have more specific analysis, the proposed multimode HESS and the improved HESS [5] are tested with 1/500 of the power demands in the UDDS and the NEDC. The energy loss in the DCeDC converter, the total energy consumption and the overall system efficiency of the two HESSs are listed in Table 4. It can be seen that the energy loss in the DCeDC converter of the proposed multimode HESS is less than the improved HESS' in both the UDDS and NEDC. However, the total energy consumption and the overall system efficiency of the two HESSs in the UDDS are nearly equal due to the energy loss increased in switching ON/OFF states of the SW1 and the SW2. The less time of switching ON/OFF states of the SW1 and the SW2, the higher efficiency of the overall system to the proposed multimode HESS, are successfully demonstrated by the test results in the NEDC. The comparative results show that the proposed multimode HESS has a better performance than the improved HESS. 6. Conclusion A novel multimode HESS and its energy management strategy for electric vehicles (EVs) have been proposed in this paper. Compared to the conventional HESS, the proposed multimode HESS has more operating modes by controlling the P-mosfet, the Nmosfet and the DCeDC converter. Both the batteries and the UCs can provide power directly to the motor inverter. The DCeDC converter will operate at peak efficiency to convey the energy from the batteries to the UCs or the motor inverter. In addition, the UCs have the highest priority to recycle the energy, and the batteries are isolated from being recharged directly during regenerative braking. Based on the proposed multimode HESS, the operating modes and energy flows have been categorized and analyzed to design the energy management strategy, including mode selection and power distribution. The rule-based control strategy has been developed for mode selection and the power-balancing strategy has been proposed to distribute the power between the batteries and the UCs. The simulation model of the proposed multimode HESS in the SD-EV based on ADVISOR was established. Simulation results validated the rule-based control strategy for mode selection and the power-balancing strategy for power distribution. Furthermore, a scaled-down experimental platform was established to validate the proposed multimode HESS and its energy management strategy. The experimental results successfully demonstrated that the proposed multimode HESS could reduce the energy losses of the system and protect the batteries. The overall system efficiency has been improved much in the UDDS and the NEDC, which were 0.28% and 1.85% respectively compared to the improved HESS. The recommended future developments are: (i) adaptive control strategy will be applied to the mode selection to obtain better performances in different driving cycles, and (ii) a suitable circuit and switching algorithms will be developed to reduce the fluctuating current/voltage in the design of HESS with its motor/inverter
drive integration. Acknowledgment This work was supported by the National Natural Science Foundation of China (Grant No. 51405374) and the Postdoctoral Science Foundation of China (Grant No. 2014M560763). References [1] S.F. Tie, C.W. Tan, A review of energy sources and energy management system in electric vehicles, Renew. Sustain. Energy Rev. 20 (2013) 82e102. [2] L. Lu, X. Han, J. Li, J. Hua, M. Ouyang, A review on the key issues for lithium-ion battery management in electric vehicles, J. Power Sources 226 (2013) 272e288. [3] H. Nagata, Y. Chikusa, A lithium sulfur battery with high power density, J. Power Sources 264 (2014) 206e210. [4] J.G. Zhu, Z.C. Sun, X.Z. Wei, H.F. Dai, A new electrochemical impedance spectroscopy model of a high-power lithium-ion battery, RSC Adv. 4 (2014) 29988e29998. [5] J. Cao, A. Emadi, A new battery/ultracapacitor hybrid energy storage system for electric, hybrid, and plug-in hybrid electric vehicles, IEEE Trans. Power Electron. 27 (2012) 122e132. [6] Y. Abdul-Quadir, T. Laurila, J. Karppinen, K. Jalkanen, K. Vuorilehto, L. Skogstrom, et al., Heat generation in high power prismatic Li-ion battery cell with LiMnNiCoO2 cathode material, Int. J. Energy Res. 38 (2014) 1424e1437. [7] J.M. Blanes, R. Gutierrez, A. Garrigos, J.L. Lizan, J.M. Cuadrado, Electric vehicle battery life extension using ultracapacitors and an FPGA controlled interleaved buck-boost converter, IEEE Trans. Power Electron. 28 (2013) 5940e5948. [8] J.P.F. Trovao, V.D.N. Santos, P.G. Pereirinha, H.M. Jorge, C.H. Antunes, A simulated annealing approach for optimal power source management in a small EV, IEEE Trans. Sustain. Energy 4 (2013) 867e876. [9] A. Kuperman, I. Aharon, S. Malki, A. Kara, Design of a semiactive batteryultracapacitor hybrid energy source, IEEE Trans. Power Electron. 28 (2013) 806e815. [10] F. Ongaro, S. Saggini, P. Mattavelli, Li-ion battery-supercapacitor hybrid storage system for a long lifetime, photovoltaic-based wireless sensor network, IEEE Trans. Power Electron. 27 (2012) 3944e3952. [11] O. Laldin, M. Moshirvaziri, O. Trescases, Predictive algorithm for optimizing power flow in hybrid ultracapacitor/battery storage systems for light electric vehicles, IEEE Trans. Power Electron. 28 (2013) 3882e3895. [12] A. Khaligh, Z. Li, Battery, ultracapacitor, fuel cell, and hybrid energy storage systems for electric, hybrid electric, fuel cell, and plug-in hybrid electric vehicles: state of the art, IEEE Trans. Veh. Technol. 59 (2010) 2806e2814. [13] M. Michalczuk, L.M. Grzesiak, B. Ufnalski, A lithium battery and ultracapacitor hybrid energy source for an urban electric vehicle, Electrotech. Rev. 88 (2012) 158e162. [14] H. Jung, H.F. Wang, T.S. Hu, Control design for robust tracking and smooth transition in power systems with battery/supercapacitor hybrid energy storage devices, J. Power Sources 267 (2014) 566e575. [15] J. Curti, X. Huang, R. Minaki, Y. Hori, A simplified power management strategy for a supercapacitor/battery hybrid energy storage system using the halfcontrolled converter, in: IECON 2012-38th Annual Conference on IEEE Industrial Electronics Society, 2012, pp. 4006e4011. [16] L. Gao, R.A. Dougal, S. Liu, Power enhancement of an actively controlled battery/ultracapacitor hybrid, IEEE Trans. Power Electron. 20 (2005) 236e243. [17] C. Romaus, J. Bocker, K. Witting, A. Seifried, O. Znamenshchykov, Optimal energy management for a hybrid energy storage system combining batteries and double layer capacitors, Energy Convers. Congr. Expo. (2009) 1640e1647. ECCE 2009. IEEE, 2009. [18] X. Hu, N. Murgovski, L. Johannesson, B. Egardt, Comparison of three electrochemical energy buffers applied to a hybrid bus powertrain with simultaneous optimal sizing and energy management, IEEE Trans. Intell. Transp. Syst. 15 (2014) 1193e1205. [19] M. Ansarey, M. Shariat Panahi, H. Ziarati, M. Mahjoob, Optimal energy management in a dual-storage fuel-cell hybrid vehicle using multi-dimensional dynamic programming, J. Power Sources 250 (2014) 359e371. [20] X. Hu, L. Johannesson, N. Murgovski, B. Egardt, Longevity-conscious dimensioning and power management of the hybrid energy storage system in a fuel cell hybrid electric bus, Appl. Energy (2014).
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443 [21] F. Zhu, L. Chen, C. Yin, Design and analysis of a novel multimode transmission for a HEV using a single electric machine, IEEE Trans. Veh. Technol. 62 (2013) 1097e1110. [22] P. Taesik, K. Taehyung, Novel energy conversion system based on a multimode single-leg power converter, IEEE Trans. Power Electron. 28 (2013) 213e220. [23] L. Chen, F. Zhu, M. Zhang, Y. Huo, C. Yin, H. Peng, Design and analysis of an electrical variable transmission for a serieseparallel hybrid electric vehicle, IEEE Trans. Veh. Technol. 60 (2011) 2354e2363. [24] R. Adany, D. Aurbach, S. Kraus, Switching algorithms for extending battery life in electric vehicles, J. Power Sources 231 (2013) 50e59. [25] S. Lu, K.A. Corzine, M. Ferdowsi, A New battery/ultracapacitor energy storage system design and its motor drive integration for hybrid electric vehicles, IEEE Trans. Veh. Technol. 56 (2007) 1516e1523. [26] J. Moreno, M.E. Ortúzar, J.W. Dixon, Energy-management system for a hybrid
[27]
[28]
[29]
[30]
443
electric vehicle, using ultracapacitors and neural networks, IEEE Trans. Ind. Electron. 53 (2006) 614e623. M. Ortúzar, J. Moreno, J. Dixon, Ultracapacitor-based auxiliary energy system for an electric vehicle: implementation and evaluation, IEEE Trans. Ind. Electron. 54 (2007) 2147e2156. J. Xu, C.C. Mi, B. Cao, J. Cao, A new method to estimate the state of charge of lithium-ion batteries based on the battery impedance model, J. Power Sources 233 (2013) 277e284. J. Xu, C.C. Mi, B.G. Cao, J.J. Deng, Z. Chen, S.Q. Li, The state of charge estimation of lithium-ion batteries based on a proportional-integral observer, IEEE Trans. Veh. Technol. 63 (May 2014) 1614e1621. S. Adhikari, S. Halgamuge, H.C. Watson, “An online power-balancing strategy for a parallel hybrid electric vehicle assisted by an integrated starter generator, IEEE Trans. Veh. Technol. 59 (2010) 2689e2699.
Contents lists available at ScienceDirect
Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour
A novel multimode hybrid energy storage system and its energy management strategy for electric vehicles Bin Wang a, Jun Xu a, *, Binggang Cao a, Xuan Zhou b a b
State Key Laboratory for Manufacturing Systems Engineering, School of Mechanical Engineering, Xi'an Jiaotong University, Xi'an 710049, China Department of Electrical and Computer Engineering, Kettering University, Flint, MI 48504, USA
h i g h l i g h t s A novel topology of multimode HESS is proposed for EVs. The rule-based control strategy and the power-balancing strategy are developed for the mode selection and the power distribution. The energy management strategy is proposed to reduce energy losses in the DCeDC converter. The proposed multimode HESS could extend the batteries life and improve the operation efficiency of the HESS.
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 January 2015 Accepted 4 February 2015 Available online
This paper proposes a novel topology of multimode hybrid energy storage system (HESS) and its energy management strategy for electric vehicles (EVs). Compared to the conventional HESS, the proposed multimode HESS has more operating modes and thus it could in further enhance the efficiency of the system. The rule-based control strategy and the power-balancing strategy are developed for the energy management strategy to realize mode selection and power distribution. Generally, the DCeDC converter will operate at peak efficiency to convey the energy from the batteries to the UCs. Otherwise, the pure battery mode or the pure ultracapacitors (UCs) mode will be utilized without the DCeDC converter. To extend the battery life, the UCs have the highest priority to recycle the energy and the batteries are isolated from being recharged directly during regenerative braking. Simulations and experiments are established to validate the proposed multimode HESS and its energy management strategy. The results reveal that the energy losses in the DCeDC converter, the total energy consumption and the overall system efficiency of the proposed multimode HESS are improved compared to the conventional HESS. © 2015 Elsevier B.V. All rights reserved.
Keywords: Hybrid energy storage system Energy management strategy Electric vehicles Direct current to direct current converter
1. Introduction To meet the power demands of an electric vehicle (EV), the design of an energy storage system (ESS) with high power and high energy density is of greatest importance [1,2]. There are some power batteries today with high specific power density [3,4], but volume or size problems could not be ignored. Moreover, a massive source of heat will be created when the batteries meet peak power demands, leading to a short battery life [5,6]. Even worse, a singlebattery ESS also causes an increase in cost compared to a multipower sources [1,5]. To solve the problems listed above, ultracapacitors (UCs) can be utilized in the ESS to meet the high power
* Corresponding author. E-mail address: [email protected] (J. Xu). http://dx.doi.org/10.1016/j.jpowsour.2015.02.012 0378-7753/© 2015 Elsevier B.V. All rights reserved.
demand [5,7e9]. Furthermore, the UCs can be recharged or discharged more efficiently and quickly, which can reduce energy losses and recycle braking energy quickly, thus prolong the battery life [2,7]. In this regard, the hybrid energy storage systems (HESSs) of EVs, which include batteries and UCs, have been widely studied in recent years [7,8,10e12]. To achieve low cost of power conversion, the conventional HESS design usually uses a bidirectional DCeDC converter. The HESSs with single DCeDC converter can be divided into UC/battery configuration [1,7], battery/UC configuration [1,13], and the improved configuration based on both UC/battery and battery/UC HESS [5]. However, the UC/battery and the battery/UC HESS fail to achieve the goal that both the batteries and the UCs can provide power directly to the motor inverter without DCeDC converter, which might result in energy losses in DCeDC converter. Although the improved configuration [5] achieves that both the batteries and
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443
the UCs can provide power directly, the UCs voltage must drop below the battery voltage when only the batteries provide power directly. So the improved configuration is a passive configuration and may fail to achieve the highest efficiency of the HESS. To solve problems discussed above, a novel topology of multimode HESS is proposed in this paper. An N-mosfet is connected to the UCs and a P-mosfet is connected to the batteries. The two energy sources are directly connected to the two ends of the bidirectional DCeDC converter respectively. A power diode is used to avoid the batteries being recharged directly. Thus the UCs can provide peak power directly to the motor inverter and the goal to isolate the batteries from providing peak power is achieved. Besides, the batteries are able to provide the power directly to the motor inverter to reduce energy losses of the DCeDC converter. To achieve the high efficiency of the whole system, the configuration and topology of HESS is not enough. Energy management and control is also crucial to the HESS [2,11,14e16]. Many control methods have been used for HESS energy management [17e20]. However, if different operating modes are considered, the HESS may work better. Multimode transmission technology has been successfully used in most hybrid electric vehicles (HEVs) and EVs, [21e23], but few has been reported in HESS. This paper introduce multimode technology to improve the efficiency of the whole system. In the proposed multimode HESS, the pure battery mode would be selected to meet low power demands while the hybrid output mode would be selected to meet peak power demands. Since the HESS has pure battery mode, pure UCs mode, hybrid output mode and recycle mode, etc. Thus, different modes of energy flows need to be analyzed and properly chosen. Otherwise, the batteries may suffer from frequent charge and discharge operations, leading to short battery life [7,24]. In addition, improper mode selection may cause a voltage fluctuation. As a result, the energy losses of the HESS could be increased, and even the efficiency of the motor inverter could be reduced [25]. In this case, the switch rules of mode selection should be reasonable to ensure the high efficiency of energy management strategy. Based on the discussion above, the mode switching strategy will be developed and analyzed. In this paper, a novel topology of multimode HESS is proposed and its energy management strategy is developed. Section 2 presents the configuration of the proposed multimode HESS and its operating modes. Section 3 is the energy management strategy of the proposed HESS. Section 4 focuses on the simulation and its results. Section 5 is the experimental verification. The main conclusion is given in Section 6. 2. Proposed multimode HESS configuration and its operating modes Batteries have a relatively high energy density [26] and UCs have a significantly higher power density [27]. To prolong the battery life, the UCs are suggested to be connected to the dc link directly so that the batteries could be isolated from providing peak power and being recharged directly [8]. The bidirectional DCeDC converter in the HESS has buck mode and boost mode, and the conversion efficiency in different modes is always considered as the most important factor. However, since the energy loss cannot be avoided in both buck mode and boost mode, to improve the efficiency of the whole system, the UCs and the batteries should provide the power directly without DCeDC converter if possible. Based on the aforementioned discussion, this paper proposes a novel multimode HESS configuration. The proposed circuit structure is illustrated in Fig. 1. The main feature of the proposed multimode HESS is that the N-mosfet (SW1) is connected to the UCs and the P-mosfet (SW2) is connected to the batteries. In
433
Fig. 1. Circuit structure of the proposed multimode HESS.
addition, the two energy sources are directly connected to each the two ends of the bidirectional DCeDC converter respectively. A power diode is used to avoid the batteries being recharged directly. What's more, the SW1 and the SW2 are controlled by the same electrical signal. In this configuration, the UCs can provide peak power directly to the motor inverter when the SW1 is ON (the SW2 is OFF). The batteries provide energy through the DCeDC converter only when the SW1 is ON. Thus the goal to isolate the batteries from providing peak power is achieved. When power demands of the motor inverter are not so very high (referred to as “low power demands” in this paper), the SW2 is ON (the SW1 is OFF), and the batteries provide the power directly to the motor inverter to reduce energy losses of the DCeDC converter. The UCs voltage and battery voltage are important factors for the design of HESS. In this paper, the proposed multimode HESS is designed for a SD-EV which is a pure EV developed in our laboratory. The nominal voltage of the batteries is 288 V; the effective operating voltage range of the batteries is from 252 V to 324 V. On the other hand, the effective operating voltage range of the motor inverter is from 190 V to 380 V. Considering the configuration of UCs module, the maximum voltage of the UCs is designed as 375 V. To ensure enough energy from the UCs, the minimum voltage of the UCs is designed as 190 V. The energy stored in UCs is a function of its voltage as shown in (1). It is noted that the actual energy available of the UCs is less than 25% when its voltage is less than 50%. So the lower limit of the UCs voltage is designed as 50% of the maximum voltage.
Ecap ¼
1 2 CV 2 UC
(1)
To improve the efficiency of the HESS, the DCeDC converter should be properly designed. Since the DCeDC converter could boost or buck the battery voltage, the configuration of the DCeDC converter is designed as Fig. 2. In this configuration, the DCeDC converter could also buck the UCs voltage to recharge the batteries, which ensures the safety of the batteries during regenerative braking. Based on the design above, four operation schemes are proposed as follows: (a) the batteries boost scheme, (b) the batteries buck scheme, (c) the UCs buck scheme, (d) the regenerative braking scheme, as shown in Fig. 3. To avoid damages to batteries, the UCs
batteries
UCs
Fig. 2. Configuration of the DCeDC converter.
434
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443
Fig. 3. The operation schemes of the proposed multimode HESS. (a) The batteries boost scheme. (b) The batteries buck scheme. (c) The UCs buck scheme. (d) The regenerative braking scheme.
are designed to provide the power directly to the motor inverter when the proposed peak power mode is demanded. So the schemes (a) and (b) are used to meet the peak power demands. On the other hand, low power demand has little damage to batteries, so the scheme (c) is designed to meet the low power demands. A higher overall system efficiency could be achieved as the batteries work alone. In this scheme, the UCs needs to provide power to the DCeDC converter only when the battery SOC is lower than a certain level, 0.1 for example. The scheme (d) is used to absorb the regenerative braking energy. In this scheme, the UCs have the highest priority to recycle the energy, and the DCeDC converter bucks the UCs voltage to recharge the batteries to extend the battery life. To explain the energy flows of different schemes, the operating modes of the proposed multimode HESS are categorized and analyzed in detail. The batteries boost scheme has three operating modes. In this scheme, the UCs voltage is higher than the battery voltage, and a high level voltage is used to control the SW1 (ON) and the SW2 (OFF). When the DCeDC converter needs not to work, the pure UCs output mode is implemented. When the DCeDC operates at boost mode with discharging the UCs, the hybrid output mode Ⅰ is implemented. On the contrary, when the DCeDC operates at boost mode with recharging the UCs, the UCs recharge mode Ⅰ is implemented. The energy flows of the three modes are shown in Fig. 4 (a)e(c), respectively. The batteries buck scheme has two operating modes. In this scheme, the UCs voltage is lower than the battery voltage. When the SW1 is ON and the SW2 is OFF, the DCeDC converter operates at buck mode. Fig. 5(a) and (b) show the energy flows of the hybrid output mode Ⅱ and the UCs recharge mode Ⅱ in the batteries buck scheme, respectively. The UCs buck scheme also has two operating modes. In this scheme, the UCs voltage is higher than the battery voltage. When the SW1 is OFF and the SW2 is ON, the batteries provide power directly to the motor inverter. Fig. 6(a) and (b) show the energy flows of the pure battery mode and the hybrid output mode Ⅲ. In the hybrid output mode Ⅲ, the DCeDC converter operates at buck mode, this mode is used only when the battery SOC is less than 0.1.
The regenerative braking scheme is designed to give the highest priority to recycle the braking energy. In this scheme, the SW1 is OFF, and it should be noticed that there is no energy flow through the SW2 because the power diode is reverse biased to avoid the batteries being recharged directly as shown in Fig. 1. The proposed multimode HESS commonly operates in the pure UCs recycle mode. The hybrid recycle mode is activated when the UCs voltage is higher than the upper limit. This mode should be used as less as possible to extend the battery life. Fig. 7(a) and (b) show the energy flows of the two modes. 3. Design of the energy management strategy Since the proposed multimode HESS has nine operating modes, the energy management strategy is designed as two modules: mode selection and power distribution. The schematic of the energy management system is shown in Fig. 8. To ensure the optimized control of mode selection and improve the operating efficiency of the proposed multimode HESS, the rule-based control strategy is developed for mode selection and the power-balancing strategy is developed to distribute the power between the batteries and the UCs. 3.1. Modeling From Fig. 8, it can be seen that the input signals are the power demand from the motor inverter Pmotor, the efficiency of the DCeDC converter hdcdc, the battery SOC and the UCs voltage. To design the energy management strategy, the models of the input signals need to be established. When driving the EV, the power demand from the motor Pmotor [8] can be calculated as
Pmotor ¼ Pdrive =hmotor hinverter
(2)
. Pdrive ¼ Fdrive vev hgb
(3)
Fig. 4. The operating modes in the batteries boost scheme. (a) The pure UCs output mode. (b) The hybrid output mode Ⅰ in the batteries boost scheme; (c) The UCs recharge mode Ⅰ in the batteries boost scheme.
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443
435
Fig. 5. The operating modes in the batteries buck scheme. (a) The hybrid output mode Ⅱ in the batteries buck scheme. (b) The UCs recharge mode Ⅱ in the batteries buck scheme.
Fig. 6. The operating modes in the UCs buck scheme. (a) The pure battery mode. (b) The hybrid output mode Ⅲ in the UCs buck scheme.
where Pdrive is the power requirement for driving the EV, hmotor is the efficiency of the motor, hinverter is the efficiency of the motor inverter, Fdrive is the driving force which is the sum of the rolling resistance force, the climbing force, the braking force etc., vev is the speed of the EV, hgb is the gear efficiency. The efficiency of the DCeDC converter hdcdc is related to the battery voltage, the UCs voltage and its operating mode. hidcdc indicates the efficiency of the DCeDC converter in the batteries boost or buck scheme, and hkdcdc indicates the efficiency of the DCeDC converter in the UCs buck scheme or the regenerative
braking scheme.
(
hidcdc ¼ f1 ðVUC ; Vbatteries ; Pdcdc Þ hkdcdc ¼ f2 ðVUC ; Vbatteries ; Pdcdc Þ
(4)
Compared to the UCs voltage, the battery voltage varies little and affects little on the hidcdc . hkdcdc is only used in the hybrid recycle mode or the hybrid output mode in the UCs buck scheme. Since the two modes should be used as less as possible for improving the efficiency of the whole system, we are only
Fig. 7. The operating modes in the regenerative braking scheme. (a) The pure UCs recycle mode. (b) The hybrid recycle mode.
Fig. 8. The schematic of the energy management system.
436
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443
Fig. 9. The equivalent circuit and the internal resistor in different voltage of the battery.
interested in hidcdc , and the functional relationships (5) could be obtained by corresponding experiment.
8 < hidcdc ¼ f ðVUC ; Pdcdc Þ : P peakefficiency ¼ f 1 VUC ; hpeakefficiency dcdc dcdc
Ibat ¼
0:5 2 4R Ebat Ebat bat Pbattery 2Rbat
(5)
Z ktem
The battery SOC and the UCs voltage are very important factors because they decide the operating modes of the DCeDC converter. In this paper, the open-circuit voltage model of the batteries is utilized, which consists of the controlled dc voltage source Ebat in series with the internal resistor Rbat [28,29]. The equivalent circuit of the batteries and the internal resistor in different voltage are shown in Fig. 9. The battery SOC is calculated as
Vbat ¼ Ebat Ibat Rbat
(6)
Pbatteries ¼ Vbat Ibat
(7)
SOCbat ¼ SOC0 þ
(8)
t
0
Ibat dt
Ct
(9)
where Ibat is charging or discharging current of the battery, Pbattery is the power output or input of the battery. SOC0 is the initial SOC. ktem is temperature coefficient. Ct is the capacity of the battery. By using a similar equivalent circuit, we can calculate the UC voltage
IUC ¼
0:5 2 4R P VUC VUC UC UC
Fig. 10. Mode selection and the control rules.
2RUC
(10)
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443
437
Table 1 Power distribution based on power-balancing strategy. Mode selection OperationMode OperationMode OperationMode OperationMode OperationMode OperationMode OperationMode OperationMode OperationMode
¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼ ¼
1 2 3 4 5 6 7 8 9
PUCs
Pbatteries
i; k
DCeDC converter
PUCs ¼ Pmotor PUCs ¼ Pmotor Pdcdc
Pbatteries ¼ 0
i; k ¼ 0 i ¼ 1; k ¼ 0
No energy conversion Peak efficiency of the batteries boost mode
Pbatteries ¼ Pdcdc =hidcdc
Peak efficiency of the batteries buck mode
peakefficience Pdcdc ¼ Pdcdc
PUCs PUCs PUCs PUCs
¼0 Pbatteries ¼ Pmotor max1 ¼ Pmotor Pbatteries =hkdcdc , Pbatteries ¼ 0:5Pbatteries ¼ Pmotor Pbatteries ¼ 0 max ¼ PUCs Pbatteries ¼ ðPmotor PUCs Þ=hidcdc
VUC ðn þ 1Þ ¼ VUC ðnÞ IUC Dt=C
UC
(11)
where IUC is the charging or discharging current of the UC, RUC is the internal resistor, PUC is the power output or input of the UCs, VUC(n) is the controlled dc voltage source, CUC is the capacity of the UCs. 3.2. Mode selection The rule-based control strategy is a simple but effective method, which is developed based on theory or project experience and has been widely used to define the operating mode of HEVs [21,23]. The mode selection and the detailed control rules for the proposed multimode HESS are shown in Fig. 10. 1) To ensure that the UCs can absorb more energy and the hybrid recycle mode can be used as less as possible in the next regenerative braking, the Operation Mode1 ¼ 1 should be selected when Pmotor > 0 and VUCs > 0.95. 2) When Vbatteries < VUCs < 0.85, Pmotor > 0, the DCeDC converter operates at peak efficiency with the batteries boost scheme. If Pmotor > Pdcdc, the OperationMode ¼ 2 will be selected, otherwise the OperationMode ¼ 3 will be selected. 3) When VUCs < Vbatteries, Pmotor > 0, the DCeDC converter must operate at peak efficiency with the batteries buck scheme. If Pmotor > Pdcdc, the OperationMode ¼ 4 will be selected, or the OperationMode ¼ 5 will be selected. 4) When 0.85 VUCs 0.95, Pmotor > 0 and Vbatteries > 300 V, the OperationMode ¼ 6 will be selected if the power demand from the motor inverter is lower than the upper limit of the batteries max1 , or the OperationMode ¼ 2 will be selected. The upper Pbatteries
i; k ¼ 0 i ¼ 0; k ¼ 1 i; k ¼ 0 i ¼ 1; k ¼ 0
the
No energy conversion The UCs buck mode No energy conversion The UCs buck mode
limit
of
max1 Pbatteries
peakefficiency Pdcdc =hpeakefficiency . dcdc
batteries
is
designed
as
¼ 5) When the battery SOC is lower than 0.1 and Pmotor > 0, it is needed to limit the power output of the batteries by the limit max1 . The UCs buck scheme will be used and the Pbatteries ¼ 0:5Pbatteries limit OperationMode ¼ 6 will be selected if the Pbatteries > Pmotor , or the OperationMode ¼ 7 will be selected. It should be noted that the UCs voltage is higher than batteries or the OperationMode ¼ 7 is not available. When the battery SOC is lower than 0.1, the power output of the HESS is limited. We define this as a reduced power mode. The reduced power mode should be used as less as possible to extend the battery life, at this mode the driver must drive the vehicle to the nearest EV charging station. 6) It should be noticed that OperationMode ¼ 6 will be activated in two conditions. When the SW1 is switched from ON to OFF, the SW2 has to be switched from OFF to ON. As a result, it may cause a voltage fluctuation in the output side of the HESS. This may seriously degrade the performance of the motor inverter. To reduce the voltage fluctuation, voltage regulation of the batteries is implemented. In this paper, it is expected that the UCs voltage stabilize around 85% (it equals to 319 V). When the battery SOC is higher than 0.1, OperationMode ¼ 6 must meet the requirement of Vbatteries > 300 V to reduce the voltage difference. In addition, a capacitor or a pre-charging circuit is needed in the motor inverter to reduce the voltage fluctuating. 7) When Pmotor < 0, the regenerative braking scheme is implemented. In this scheme, the OperationMode ¼ 8 is always selected if VUCs < 0.95, or the OperationMode ¼ 9 is selected. In the hybrid recycle mode, the DCeDC converter always operates at buck mode to ensure the battery safety. The hybrid recycle mode should be used as less as possible to extend the battery life.
3.3. Power distribution The power-balancing strategy [30] has been successfully used to distribute the power between batteries and internal combustion
Table 2 The important parameters of the SD-EV.
Fig. 11. The SD-EV.
Total mass of the EV Motor Gear efficiency Radius of wheel UCs Batteries DCeDC converter
1250 kg 30 kW (PMSM) [0.93 0.95 0.97 0.97 0.97] 0.29 m 375 V (Kaimai 75 V 250 F in series) 288 V (HuanYu 3.2 V 80 Ah in series) Peak efficiency region: 10e10.5 kW
438
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443
engine (ICE) in HEVs. However, the electric energy of the UCs is different from the internal energy of the ICE, where the UCs provide peak power, but the ICE provides main power to the drivetrain. In addition, the efficiency of the DCeDC converter should be considered to design the power-balancing strategy. To the proposed multimode HESS, the power demand from motor inverter can be calculated as
Pmotor ¼ Pbatteries hidcdc þ PUCs hkdcdc
(12)
From equation (12), the power-balancing of the proposed multimode HESS is determined by the relationships between the Pmotor, PUCs, Pbatteries, hidcdc and hkdcdc . The power distribution based on power-balancing strategy and the explanations are shown in Table 1. 1) OperationMode ¼ 1, the pure UCs output mode. In this mode, the UCs provide all the energy to the motor inverter. i,k ¼ 0 means that the HESS works without use of the DCeDC converter. 2) OperationMode ¼ 2e3, the hybrid output mode and the UCs recharge mode with the batteries boost scheme. In these two modes, the DCeDC converter operates at boost mode. Since Vbatteries < VUCs < 0.85, the UCs can be recharged or provide peak power. The DCeDC converter operates at peak efficiency to reduce energy losses. For the power-balancing strategy, the power output of the batteries is peakefficience Pdcdc =hpeakefficience . dcdc
Pbatteries ¼ 3) OperationMode ¼ 4e5, the hybrid output mode and the UCs recharge mode with the batteries buck scheme. In these two modes, the DCeDC converter must operate at buck mode since the UCs voltage is lower than the battery voltage. The DCeDC converter also operates at peak efficiency. It should be noted max ¼ that the power demand of the motor must be limited by Pmotor max1 when the UCs voltage is less than 200 V. Pbatteries
4) OperationMode ¼ 6, the pure battery mode. In this mode, the UCs voltage is in the region [0.85, 0.95] and the power demand max1 . The UCs needs not to from motor inverter is lower than Pbatteries be charged, and the batteries can provide all the energy to the motor inverter in this condition. So Pmotor ¼ Pbatteries, PUCs ¼ 0. 5) OperationMode ¼ 7, the hybrid output mode with the UCs buck scheme. It is a reduced power mode because the battery SOC is less than 0.1. The power output of the DCeDC converter is limit Pdcdc ¼ Pmotor Pbatteries . This mode should be used as less as possible, and the driver must drive the vehicle to the nearest EV charging station. 6) OperationMode ¼ 8, the pure UCs recycle mode. The UCs have the highest priority to recycle the energy, so it is designed to recycle all the energy from the motor inverter when VUCs < 0.95. In this condition, PUCs ¼ Pmotor. 7) OperationMode ¼ 9, the hybrid recycle mode. In this mode, VUCs > 0.95. To ensure the safety of the batteries, the regenerative braking power of the motor should be limited, so limit ¼ P max þ P limt hi and the DCeDC converter operPmotor UCs batteries dcdc ates at buck mode to convey the energy from the UCs to the batteries. i ¼ 1 means that the batteries are recharged.
4. Simulation analysis To validate the configuration of the proposed multimode HESS and its energy management strategy, the simulation is established according to the power demand of the SD-EV. The SD-EV and its important parameters are shown in Fig. 11 and Table 2. The simulation model of the drivetrain configuration for the SD-EV with the proposed multimode HESS based on ADVISOR is shown in Fig. 12(a). The Urban Dynamometer Driving Schedule (UDDS) and the New European Driving Cycle (NEDC) are chosen to demonstrate the proposed multimode HESS and its energy management strategy. The power demands from the motor inverter for the SD-EV in
Fig. 12. Simulation setup. (a)The simulation model of the drivetrain configuration for SD-EV with the proposed multimode HESS. (b) The UDDS. (c) The NEDC.
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443
the UDDS and the NEDC are shown in Fig. 12(b) and (c). In the UDDS, the maximum driving power demand and the maximum braking power are 42.9 kW and 11.3 kW respectively. In the NEDC, the maximum driving power demand and the maximum braking power are 57.5 kW and 24.3 kW, respectively. The average power can be calculated as
1 t
Zt Pmotor ðtÞdt ¼ P motor
(13)
0
where P motor is the average power from the start of the trip to the current time instant “t”,Pmotor(t) is the instantaneous power. The P motor in the UDDS and the NEDC can be calculated by (13), which is 7.08 kW and 8.29 kW, respectively. On the other hand, numerous tests of the SD-EV in suburban road have demonstrated that the average power is less than 10 kW. So the peak efficiency power output of the DCeDC converter in the simulation model is designed as 10 kW. The simulation results are shown in Fig. 13. In the Fig. 13(a) and (b), the initial value of the UCs voltage is set to 375 V. Since the UCs voltage is higher than 95% in the inception phase, the pure UCs output mode is selected in both the UDDS and the NEDC. Then the UCs voltage is in the region [85%, 95%], so the
439
hybrid output mode is selected to meet the peak power demands in the UDDS. On the contrary, the pure battery mode is selected when the power demand is lower than the upper limit of the batteries in the NEDC. In addition, the pure UCs recycle mode is selected when the power demand is lower than zero because the UCs can absorb all the energy from the motor inverter. When the UCs voltage is lower than 85% (319 V), the UCs recharge mode in the batteries boost scheme should be selected to meet low power demands. In the NEDC, it can be seen that the power output of the batteries can be maintained at 10 kW after 1000 s because the UCs voltage is lower than the battery voltage. In this condition, the hybrid output mode in the batteries buck scheme is selected. At the end of the NEDC, the UCs voltage increases rapidly because the UCs recharge mode is selected. In the Fig. 13(c) and (d), the initial value of the UCs voltage is set to 340 V. Since the UCs voltage is in the region [85%, 95%], the hybrid output mode or the pure battery mode is selected in the inception phase in the UDDS or the NEDC, respectively. In the UDDS, the UCs voltage drops to 296 V to meet peak power demands. In this condition, the power output of the batteries should be maintained at 10 kW and the hybrid output mode or the UCs recharge mode should be maintained for a long time. The UCs voltage is maintained at around 319 V at last. In the NEDC, we can
Fig. 13. Simulation results. (a) Simulation results in the UDDS when the initial voltage of the UCs is set to 375 V. (b) Simulation results in the NEDC when the initial voltage of the UCs is set to 375 V. (c) Simulation results in the UDDS when the initial voltage of the UCs is set to 340 V. (d) Simulation results in the NEDC when the initial voltage of the UCs is set to 340 V.
440
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443
Fig. 14. Experimental platform.
see the UCs voltage reducing constantly. The power output of the batteries is maintained at 10 kW constantly after 820 s. It should be noticed that the UCs voltage is lower than 200 V in 1120 s. Although the UCs voltage is higher than 50%, it may fail to provide enough energy to meets a constant peak power demands. So if the SD-EV needs a better power performance, it needs a larger size UCs or larger size batteries with high power output, and it also needs a larger size DCeDC converter. The different modes and the power distribution between the batteries and the UCs in different drive conditions seen in Fig. 13(a)e(d), preliminarily prove the availability of the mode selection and the power distribution. 5. Experimental verification To further validate the configuration of the proposed multimode HESS and its energy management strategy, an experimental platform is constructed. The whole system is scaled down for safer experiments. The experimental platform is illustrated in Fig. 14 and the major parameters of its components are listed in Table 3. It should be noticed that Table 3 presents the normal voltage of the batteries. The actual voltage of the batteries is 11.2e14.4 V. In the experiment, the measured efficiency of the DCeDC converter is shown in Fig. 15. It can be seen that the points in the blue line (in the web version) are the highest efficiency operating points of the DCeDC converter at any corresponding voltage. These points are stored in the controller for controlling the DCeDC converter. For this reason, the DCeDC converter will operate at peak efficiency to convey the energy from the batteries to the UCs. So the power
peakefficience
peakefficience
output of the batteries is Pbatteries ¼ Pdcdc =hdcdc . In the experiment, to ensure the safety of the batteries, we define that the up limit of the power output of the batteries is 55 W, which is close to the highest efficiency operating points of the DCeDC converter. Fig. 16(a) illustrates the experimental results with only the batteries. It shows that when the power load is changed from 40 W to 60 W, the battery voltage drops quickly. It is no doubt that batteries will be damaged without any other energy source to meet a higher power load. Fig. 16(b) illustrates the experimental results of the proposed multimode HESS when the UCs voltage is higher than that of the batteries. It can be seen that the power output of the batteries is less than 55 W at any time, which ensures the safety of the batteries at any time and prolongs the life of the batteries. On the other hand, when the UCs voltage is 14.51 Ve14.45 V and the power demand is 40 W, the proposed multimode HESS operates at the UCs recharge mode. When the UCs voltage is 22.05 Ve21.95 V and the power demand is 40 W, the pure battery mode is selected. In the other conditions, the hybrid output mode is selected. Furthermore, it is
Table 3 Experimental parameters. Electric load Power supply Transmission time interval Data type UCs Batteries DCeDC converter
150 V (KL6101 100 A, 1200 W Max) 32 V (IPD3305-SLU 5.2A, 0.03%) 50 Hz Hexadecimal 25 V (Kaimai 25 V 94 F) 12.8 V (HuanYu 3.2 V 80 Ah in series) Peak efficiency region: 49e53 W
Fig. 15. The measured efficiency of the DCeDC converter.
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443 peakefficience
peakefficience
known that Pbatteries ¼ Pdcdc =hdcdc at the UCs recharge mode and the hybrid output mode, so the DCeDC converter operates at peak efficiency. Fig. 16(c) illustrates the experimental results of the proposed multimode HESS when the UCs voltage is lower than batteries voltage. In this condition, the batteries buck scheme is always used. When the power demand is 60 W, the hybrid output mode is selected. When the power demand is changed to 40 W, the UCs recharge mode is selected. In the batteries buck scheme, the DCeDC converter also operates at peak efficiency, so the equation peakefficience Pbatteries ¼ Pdcdc =hpeakefficience also works. dcdc To further validate the rule-based control strategy for the mode selection, the power supply is used to carry out the charging test (input voltage or current). First, the proposed multimode HESS is charged by a 10 W constant power load until the UCs voltage is equal to 25 V. It can be seen that the pure UCs recycle mode is selected, and there are no power flew into batteries in Fig. 16(d). Thus, the batteries are isolated from frequent charge and discharge operations. Second, the 40 W constant power load is applied to the proposed multimode HESS. It can be seen that at this condition the pure UCs output mode is selected, and even the power demand is 60 W, the pure UCs output mode is also selected because the UCs
441
voltage is higher than 23.75 V (95% of the maximum voltage). At last, the hybrid output mode is activated when the UCs voltage is lower than 23.75 V. The mode selection and the power distribution are shown in Fig. 16(d). Fig. 16(b)e(d) show the mode selection and the power distribution in different conditions. The pure battery mode is used to avoid the energy loss of the DCeDC converter when the proposed multimode HESS meets low power demands. The DCeDC converter is always operating at peak efficiency at the hybrid output mode or the recharge mode. The UCs have the highest priority to recycle the energy during regenerative braking for protecting the batteries. So the proposed multimode HESS can reduce the energy loss of the DCeDC converter and extend the battery life. It should be noticed that the fluctuating current (in Fig. 16(b)e(c)) cannot be avoided during the 20 s of mode switching in. Because the input voltage of the load either equals to the battery voltage or the UC voltage, the voltage fluctuating is unavoidable in the output side of the HESS. It is a difficult technical problem when the HESS is applied to the SD-EV, and it might reduce the efficiency of the inverter. Therefore, a capacitor or a pre-charging circuit is needed in motor inverter to reduce the voltage fluctuating. Indeed, the efficiency of the motor inverter is decreased when the HESS switches to the pure battery mode. However, the HESS has no
Fig. 16. Experimental results. (a) Experimental results only with the batteries. (b) Experimental results in the batteries boost scheme. (c) Experimental results in the batteries buck scheme. (d) Experimental results when the UCs voltage is higher than 85%.
442
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443
Table 4 Comparative results with 1/500 of the power demands in the UDDS and the NEDC. Driving cycle
UDDS NEDC
Energy losses in the DCeDC
Total energy consumption
Overall system efficiency
Multimode HESS
Improved HESS [8]
Multimode HESS
Improved HESS [8]
Multimode HESS
Improved HESS [8]
4.7% 4.9%
8.6% 9.1%
21,398 J 21,451 J
21,464 J 21,893 J
90.66% 91.59%
90.38% 89.74%
energy loss in the DCeDC converter with the pure battery mode. So when the battery voltage is close to the UCs voltage, the pure battery mode can significantly improve the overall system efficiency. To have more specific analysis, the proposed multimode HESS and the improved HESS [5] are tested with 1/500 of the power demands in the UDDS and the NEDC. The energy loss in the DCeDC converter, the total energy consumption and the overall system efficiency of the two HESSs are listed in Table 4. It can be seen that the energy loss in the DCeDC converter of the proposed multimode HESS is less than the improved HESS' in both the UDDS and NEDC. However, the total energy consumption and the overall system efficiency of the two HESSs in the UDDS are nearly equal due to the energy loss increased in switching ON/OFF states of the SW1 and the SW2. The less time of switching ON/OFF states of the SW1 and the SW2, the higher efficiency of the overall system to the proposed multimode HESS, are successfully demonstrated by the test results in the NEDC. The comparative results show that the proposed multimode HESS has a better performance than the improved HESS. 6. Conclusion A novel multimode HESS and its energy management strategy for electric vehicles (EVs) have been proposed in this paper. Compared to the conventional HESS, the proposed multimode HESS has more operating modes by controlling the P-mosfet, the Nmosfet and the DCeDC converter. Both the batteries and the UCs can provide power directly to the motor inverter. The DCeDC converter will operate at peak efficiency to convey the energy from the batteries to the UCs or the motor inverter. In addition, the UCs have the highest priority to recycle the energy, and the batteries are isolated from being recharged directly during regenerative braking. Based on the proposed multimode HESS, the operating modes and energy flows have been categorized and analyzed to design the energy management strategy, including mode selection and power distribution. The rule-based control strategy has been developed for mode selection and the power-balancing strategy has been proposed to distribute the power between the batteries and the UCs. The simulation model of the proposed multimode HESS in the SD-EV based on ADVISOR was established. Simulation results validated the rule-based control strategy for mode selection and the power-balancing strategy for power distribution. Furthermore, a scaled-down experimental platform was established to validate the proposed multimode HESS and its energy management strategy. The experimental results successfully demonstrated that the proposed multimode HESS could reduce the energy losses of the system and protect the batteries. The overall system efficiency has been improved much in the UDDS and the NEDC, which were 0.28% and 1.85% respectively compared to the improved HESS. The recommended future developments are: (i) adaptive control strategy will be applied to the mode selection to obtain better performances in different driving cycles, and (ii) a suitable circuit and switching algorithms will be developed to reduce the fluctuating current/voltage in the design of HESS with its motor/inverter
drive integration. Acknowledgment This work was supported by the National Natural Science Foundation of China (Grant No. 51405374) and the Postdoctoral Science Foundation of China (Grant No. 2014M560763). References [1] S.F. Tie, C.W. Tan, A review of energy sources and energy management system in electric vehicles, Renew. Sustain. Energy Rev. 20 (2013) 82e102. [2] L. Lu, X. Han, J. Li, J. Hua, M. Ouyang, A review on the key issues for lithium-ion battery management in electric vehicles, J. Power Sources 226 (2013) 272e288. [3] H. Nagata, Y. Chikusa, A lithium sulfur battery with high power density, J. Power Sources 264 (2014) 206e210. [4] J.G. Zhu, Z.C. Sun, X.Z. Wei, H.F. Dai, A new electrochemical impedance spectroscopy model of a high-power lithium-ion battery, RSC Adv. 4 (2014) 29988e29998. [5] J. Cao, A. Emadi, A new battery/ultracapacitor hybrid energy storage system for electric, hybrid, and plug-in hybrid electric vehicles, IEEE Trans. Power Electron. 27 (2012) 122e132. [6] Y. Abdul-Quadir, T. Laurila, J. Karppinen, K. Jalkanen, K. Vuorilehto, L. Skogstrom, et al., Heat generation in high power prismatic Li-ion battery cell with LiMnNiCoO2 cathode material, Int. J. Energy Res. 38 (2014) 1424e1437. [7] J.M. Blanes, R. Gutierrez, A. Garrigos, J.L. Lizan, J.M. Cuadrado, Electric vehicle battery life extension using ultracapacitors and an FPGA controlled interleaved buck-boost converter, IEEE Trans. Power Electron. 28 (2013) 5940e5948. [8] J.P.F. Trovao, V.D.N. Santos, P.G. Pereirinha, H.M. Jorge, C.H. Antunes, A simulated annealing approach for optimal power source management in a small EV, IEEE Trans. Sustain. Energy 4 (2013) 867e876. [9] A. Kuperman, I. Aharon, S. Malki, A. Kara, Design of a semiactive batteryultracapacitor hybrid energy source, IEEE Trans. Power Electron. 28 (2013) 806e815. [10] F. Ongaro, S. Saggini, P. Mattavelli, Li-ion battery-supercapacitor hybrid storage system for a long lifetime, photovoltaic-based wireless sensor network, IEEE Trans. Power Electron. 27 (2012) 3944e3952. [11] O. Laldin, M. Moshirvaziri, O. Trescases, Predictive algorithm for optimizing power flow in hybrid ultracapacitor/battery storage systems for light electric vehicles, IEEE Trans. Power Electron. 28 (2013) 3882e3895. [12] A. Khaligh, Z. Li, Battery, ultracapacitor, fuel cell, and hybrid energy storage systems for electric, hybrid electric, fuel cell, and plug-in hybrid electric vehicles: state of the art, IEEE Trans. Veh. Technol. 59 (2010) 2806e2814. [13] M. Michalczuk, L.M. Grzesiak, B. Ufnalski, A lithium battery and ultracapacitor hybrid energy source for an urban electric vehicle, Electrotech. Rev. 88 (2012) 158e162. [14] H. Jung, H.F. Wang, T.S. Hu, Control design for robust tracking and smooth transition in power systems with battery/supercapacitor hybrid energy storage devices, J. Power Sources 267 (2014) 566e575. [15] J. Curti, X. Huang, R. Minaki, Y. Hori, A simplified power management strategy for a supercapacitor/battery hybrid energy storage system using the halfcontrolled converter, in: IECON 2012-38th Annual Conference on IEEE Industrial Electronics Society, 2012, pp. 4006e4011. [16] L. Gao, R.A. Dougal, S. Liu, Power enhancement of an actively controlled battery/ultracapacitor hybrid, IEEE Trans. Power Electron. 20 (2005) 236e243. [17] C. Romaus, J. Bocker, K. Witting, A. Seifried, O. Znamenshchykov, Optimal energy management for a hybrid energy storage system combining batteries and double layer capacitors, Energy Convers. Congr. Expo. (2009) 1640e1647. ECCE 2009. IEEE, 2009. [18] X. Hu, N. Murgovski, L. Johannesson, B. Egardt, Comparison of three electrochemical energy buffers applied to a hybrid bus powertrain with simultaneous optimal sizing and energy management, IEEE Trans. Intell. Transp. Syst. 15 (2014) 1193e1205. [19] M. Ansarey, M. Shariat Panahi, H. Ziarati, M. Mahjoob, Optimal energy management in a dual-storage fuel-cell hybrid vehicle using multi-dimensional dynamic programming, J. Power Sources 250 (2014) 359e371. [20] X. Hu, L. Johannesson, N. Murgovski, B. Egardt, Longevity-conscious dimensioning and power management of the hybrid energy storage system in a fuel cell hybrid electric bus, Appl. Energy (2014).
B. Wang et al. / Journal of Power Sources 281 (2015) 432e443 [21] F. Zhu, L. Chen, C. Yin, Design and analysis of a novel multimode transmission for a HEV using a single electric machine, IEEE Trans. Veh. Technol. 62 (2013) 1097e1110. [22] P. Taesik, K. Taehyung, Novel energy conversion system based on a multimode single-leg power converter, IEEE Trans. Power Electron. 28 (2013) 213e220. [23] L. Chen, F. Zhu, M. Zhang, Y. Huo, C. Yin, H. Peng, Design and analysis of an electrical variable transmission for a serieseparallel hybrid electric vehicle, IEEE Trans. Veh. Technol. 60 (2011) 2354e2363. [24] R. Adany, D. Aurbach, S. Kraus, Switching algorithms for extending battery life in electric vehicles, J. Power Sources 231 (2013) 50e59. [25] S. Lu, K.A. Corzine, M. Ferdowsi, A New battery/ultracapacitor energy storage system design and its motor drive integration for hybrid electric vehicles, IEEE Trans. Veh. Technol. 56 (2007) 1516e1523. [26] J. Moreno, M.E. Ortúzar, J.W. Dixon, Energy-management system for a hybrid
[27]
[28]
[29]
[30]
443
electric vehicle, using ultracapacitors and neural networks, IEEE Trans. Ind. Electron. 53 (2006) 614e623. M. Ortúzar, J. Moreno, J. Dixon, Ultracapacitor-based auxiliary energy system for an electric vehicle: implementation and evaluation, IEEE Trans. Ind. Electron. 54 (2007) 2147e2156. J. Xu, C.C. Mi, B. Cao, J. Cao, A new method to estimate the state of charge of lithium-ion batteries based on the battery impedance model, J. Power Sources 233 (2013) 277e284. J. Xu, C.C. Mi, B.G. Cao, J.J. Deng, Z. Chen, S.Q. Li, The state of charge estimation of lithium-ion batteries based on a proportional-integral observer, IEEE Trans. Veh. Technol. 63 (May 2014) 1614e1621. S. Adhikari, S. Halgamuge, H.C. Watson, “An online power-balancing strategy for a parallel hybrid electric vehicle assisted by an integrated starter generator, IEEE Trans. Veh. Technol. 59 (2010) 2689e2699.