https://doi.org/10.6113/JPE.2018.18.2.343

ISSN(Print): 1598-2092 / ISSN(Online): 2093-4718

A Wide Voltage-Gain Range Asymmetric H-Bridge Bidirectional DC-DC Converter with a Common Ground for Energy Storage Systems

Yun Zhang^†, Yongping Gao^*, Jing Li^**, and Mark Sumner^***

^†,*School of Electrical and Information Engineering, Tianjin University, Tianjin, China

^**Department of Electrical and Electronic Engineering, University of Nottingham Ningbo China, Ningbo, China

^***Department of Electrical and Electronic Engineering, University of Nottingham, Nottingham, ENG, UK

Abstract

A wide-voltage-conversion range bidirectional DC-DC converter is proposed in this paper. The topology is comprised of one typical LC energy storage component and a special common grounded asymmetric H-bridge with four active power switches/anti-parallel diodes. The narrow output PWM voltage is generated from the voltage difference between two normal (wider) output PWM voltages from the asymmetric H-bridge with duty cycles close to 0.5. The equivalent switching frequency of the output PWM voltage is double the actual switching frequency, and a wide step-down/step-up ratio range is achieved. A 300W prototype has been constructed to validate the feasibility and effectiveness of the proposed bidirectional converter between the variable low voltage side (24V~48V) and the constant high voltage side (200V). The slave active power switches allow ZVS turn-on and turn-off without requiring any extra hardware. The maximum conversion efficiency is 94.7% in the step-down mode and 93.5% in the step-up mode. Therefore, the proposed bidirectional topology with a common ground is suitable for energy storage systems such as renewable power generation systems and electric vehicles with a hybrid energy source.

Key words: Asymmetric H-bridge, Bidirectional DC-DC converter, Common ground, Energy storage systems, Wide voltage-gain range

Manuscript received May 29, 2017; accepted Oct. 15, 2017

Recommended for publication by Associate Editor Yan Xing.

^†Corresponding Author: zhangy@tju.edu.cn Tel: +86-0130-3221-0767, Tianjin University

^*School of Electrical and Information Eng., Tianjin University, China

^**Dept. Electrical Electron. Eng., Univ. of Nottingham Ningbo China, China

^***Dept. Electrical Electron. Eng., Univ. of Nottingham, ENG, UK

Ⅰ. INTRODUCTION

The twin challenges associated with fossil fuels, the diminishing resources available worldwide and the increasingly serious environment pollution associated with these fuels, mean that renewable power generation and the electrification of transportation can be seen as efficient ways to address these challenges [1]-[4]. In renewable power generation systems, high energy/power density storage devices, such as batteries and super-capacitor banks, are very important for the smoothing of energy fluctuations through peak load shifting [5], [6]. In hybrid energy source electric vehicles, high power density super-capacitor banks are used to provide the high instantaneous power required for acceleration and braking processes, and high energy density battery banks deal with the stable or low frequency components of energy to and from the DC-link side [7], [8] or “steady state” power. As a result, the battery bank can be operated in a healthy manner to maintain a long-life. In addition, an excellent dynamic response and energy conversion efficiency can also be achieved. However, bidirectional DC-DC converters are required to buffer the energy between the high voltage DC-link side (200V to 400V) and the low voltage energy storage side (24V to 48V) in renewable power generation systems and hybrid energy source electric vehicles. Therefore, the voltage-conversion range of the selected bidirectional DC-DC converter is approximately between 5 and 10.

In applications that require galvanic isolation, the dual active full-bridge DC-DC converter can obtain the required high voltage-gain by using an appropriate turns ratio between the primary and secondary sides of the high frequency transformer [9]. However, the large turns ratio of this transformer at high power levels creates the following problems [10]: reduced coupling, core losses from non-sinusoidal (i.e. high frequency AC square wave) excitation and dielectric losses in the insulation. In addition, the distributed capacitance of the winding turns reduces the efficiency. When the input and output voltages cannot match the turns ratio of the transformer, the switching loss increases dramatically [11]. In applications that do not require isolation, DC-DC converters with a coupled-inductor can have a higher voltage-gain, and reduce the reverse recovery losses of the diodes [12], [13]. However, it is difficult to design coupled-inductors (and high frequency transformers) at these high power levels. Although the converter in [14] achieved ZVS for all of the active switches, it still needs a coupled-inductor, and fails to achieve a bidirectional power flow. In addition, the input current ripple is considerable due to the operation of the coupled-inductor, which may result in a shorter cycle life of the energy storage devices, as well as a reduction in efficiency. For the conventional bidirectional DC-DC converter, there are extreme voltage stresses across the active power switches and diodes when extreme duty cycles are used [15]. This results in serious EMI issues and reverse recovery losses, as well as a lower conversion efficiency. The converters in [16] and [17] can achieve a higher voltage-gain. However, the maximum voltage stress of the power switch in [16] is higher than that of the high voltage side, and the converter in [17] needs an auxiliary circuit that includes a capacitor and an inductor to achieve high efficient power conversion. The Cuk and Sepic/Zeta conversion efficiencies are lower, due to the cascaded configurations of the two power stages [18], [19]. In [20], a high efficiency and high voltage-gain DC-DC converter with soft-switching was proposed. However, it has a complex coupled-inductor and associated circuits. The soft-switching DC-DC converter proposed [21] can operate over a wide load range. However, its conversion ratio is not high enough for the proposed applications. In [22], a converter is proposed that combines soft-switching and hard-switching techniques and can improve the conversion efficiency. However, it needs extra semiconductors and a coupled-inductor. An interleaved high conversion-ratio bidirectional DC-DC converter was proposed for distributed energy-storage systems in [23]. However, its high conversion-ratio was obtained by adding switched- capacitors and coupled-inductors in series. It also requires more active power switches, capacitors and coupled-inductors, and is not cost effective. In order to reduce the input current ripple, interleaved switched-capacitor converters have been proposed in [24], [25]. However, the converter in [24] fails to achieve soft-switching, and the converter in [25] suffers from a huge current ripple in the low-voltage side. Through a switched- capacitor cell, the converter in [26] achieved a high voltage conversion ratio. Unfortunately, a potential difference between the output and the input side grounds exists in the high frequency PWM voltage, which makes it unsuitable for energy storage systems.

In addition to the DC-DC converters with transformers, the coupled-inductors and the switched-capacitors mentioned above, a group of DC-DC converters exists which uses the Z-source structure, which can improve the step-up/step-down ratio [27]. High voltage-gain DC-DC converters with a Z-source structure do not have a common ground between the input side and the output side, because the boost inductor is directly replaced by the Z-source. However, common grounded DC-DC converters are required for renewable power generation systems and hybrid energy source electric vehicles, due to safety concerns (especially during maintenance). In addition, they can reduce EMI issues. Therefore, a common grounded Z-source DC-DC converter with a high voltage-gain was proposed in [28] based on the above mentioned Z-source DC-DC converter. However, this common grounded Boost DC-DC converter has not been applied to energy storage systems or hybrid energy source electric vehicles. Moreover, its input current is discontinuous due to the input diode for the Z-source.

In this paper, a bidirectional DC-DC converter for the energy storage systems in renewable energy and electric vehicle systems is proposed. It is comprised of one LC energy storage component and four active power switches with the anti- parallel diodes. This avoids the need for a transformer, a coupled-inductor, and a switched-capacitor or a Z-source impedance net. A high step-up/step-down ratio can be obtained by using non-extreme duty cycles for the four active power switches. In addition, the low and high voltage sides have a common ground, and the input/output currents on the low voltage side (i.e. battery or super-capacitor banks) are continuous. This paper is structured as follows. In Section II, the proposed topology is demonstrated. The operating principle is explained in Section III, and experimental results and analysis are presented in Section IV. Finally, some conclustions are given in Section V.

Ⅱ. TOPOLOGY

The evolution process of the proposed topology in this paper is presented in Fig. 1. As with the conventional common grounded bidirectional DC-DC converter (i.e. the buck/boost DC-DC converter), it is comprised of one bidirectional power cell (i.e. a half bridge) as shown in Fig. 1(a). The bidirectional H-bridge DC-DC converter can be synthesized by connecting two bidirectional power cells in parallel [29]. The output PWM voltage U_ab can be given as:

     (1)

where the points "a" and "b" are the output ports of the left and right half bridges, respectively, and point "g" is the ground of the high voltage side U_h. In addition, U_ag is the voltage stress of the active power switch Q₂, and U_bg is the voltage stress of the active power switch Q₄. In terms of (1), the non-extreme output PWM voltages U_ag and U_bg can be obtained when the duty cycles of the active power switches are around 0.5. The PWM voltage difference U_ab between U_ag and U_bg becomes extremely small, i.e. a high step-up or step-down ratio is achieved. However, there is a problem in the topology of the bidirectional H-bridge DC-DC converter shown in Fig. 1(a). The high and the low voltage sides do not have a common ground. Therefore, EMI issues can occur and maintenance can be unsafe in real applications. Furthermore, the bidirectional H-bridge DC-DC converter without a common ground in Fig. 1(a) cannot be extended to the interleaved bidirectional DC-DC converters for higher power levels.

(b)

Fig. 1. The evolution process of the proposed topology. (a) Bidirectional H-bridge DC-DC converter without a common ground [29]. (b) Evolution process of the proposed topology. (c) Proposed bidirectional DC-DC converter with a common grounded asymmetric H-bridge.

(a)

(b)

(c)

Therefore, a common grounded topology is derived as shown in Fig. 1(b). First, the ground "d" of the low voltage side U_l should be disconnected from point "b" (i.e. Cut I), and point "d" connects the ground "g" of the high voltage side U_h directly (Connection I). Then, the left point "e" of the inductor L should be also disconnected from point "a" (i.e. Cut II), and point "e" connects the output port "b" of the right half bridge directly (Connection II). Finally, in order to obtain an extremely narrow PWM voltage U_bg between the points "b" and "g", U_bg can be achieved by (1) as:

    (2)

Therefore, point "c" should still be disconnected from the positive port of the high voltage side (i.e. Cut III), and point "c" connects the output port "a" of the left half bridge directly (Connection III). As a result, a bidirectional DC-DC converter with a common grounded asymmetric H-bridge is proposed as shown in Fig. 1(c). D₁~D₄ are the corresponding anti-parallel diodes of the MOSFETs Q₁~Q₄, and S₁~S₄ are the gate signals for Q₁~Q₄. L is the energy storage inductor, and C_h and C_l are the filter capacitors of U_h and U_l, respectively.

Ⅲ. OPERATING PRINCIPLE

A. Bidirectional Operation States

In the proposed converter, shown in Fig. 1(c), the switching state "S_x=1" stands for the active power switch Q_X "ON", where x=1, 2, 3 and 4. Otherwise, "S_x=0" represents Q_X "OFF". It is worth noting that Q₁ and Q₃ act as master active power switches when the power flow is from U_h to U_l, i.e. Q₂ and Q₄ are slave active power switches (S₂S₄=00) in the step-down mode. In this mode, the states of the converter's components are shown in Table I. Therefore, Q₁ and Q₃ are ON and Q₂, Q₄ and D₁~D₄ are all OFF when S₁S₂S₃S₄=1010. The inductor current i_L flows in Q₁ and Q₃, L is charged by U_h, and the output PWM voltage is U_bg=U_h. When S₁S₂S₃S₄=1000, Q₁ and D₄ are ON, and Q₂~Q₄ and D₁~D₃ are all OFF. However, i_L flows in D₄ rather than Q₁, due to the freewheeling requirement of the inductor. Therefore, L is discharged into the load, and U_bg=0. When S₁S₂S₃S₄=0010, Q₃, D₂ and D₄ are ON, and Q₁, Q₂, Q₄, D₁ and D₃ are OFF. One part of i_L flows in D₂ and Q₃ in series, and the other part of i_L flows in D₄. As a result, L continues to discharge into the load and U_bg=0.

When the power flow is from U_l to U_h, Q₁ and Q₃ are slave active power switches (S₁S₃=00) in the step-up mode, and Q₂ and Q₄ act as master active power switches. In this step-up mode, the states of the converter's components are shown in Table II. When S₁S₂S₃S₄=0000, D₁ and D₃ are ON, Q₁~Q₄, D₂ and D₄ are OFF. i_L flows in D₃ and D₁ in series, L discharges into the load, and U_bg=U_h. When S₁S₂S₃S₄=0100, Q₂ and D₃ are ON, Q₁, Q₃, Q₄, D₁, D₂ and D₄ are OFF. i_L flows in D₃ and Q₂ in series, L is charged by U_l, and U_bg=0. When S₁S₂S₃S₄=0001, only Q₄ is ON, L is charged by U_l through Q₄, and U_bg=0.

TABLE I COMPONENT STATES WHEN THE POWER FLOW IS FROM UH TO UL (STEP-DOWN)

Master power switches	S1S2S3S4	L	D1	D2	D3	D4	Ubg
Q1 and Q3	1010	ch.	OFF	OFF	OFF	OFF	Uh
	1000	dis.	OFF	OFF	OFF	ON	0
	0010	dis.	OFF	ON	OFF	ON	0
*Annotate: “ch.” and “dis.” mean “charged” and “discharged” energy, respectively.

TABLE II COMPONENT STATES WHEN THE POWER FLOW IS FROM U_L TO U_H (STEP-UP)

Master power switches	S1S2 S3S4	L	D1	D2	D3	D4	Ubg
Q2 and Q4	0000	dis.	ON	OFF	ON	OFF	Uh
	0100	ch.	OFF	OFF	ON	OFF	0
	0001	ch.	OFF	OFF	OFF	OFF	0

B. Operating for a Wide-Voltage-Conversion Range

The PWM modulation strategy for the proposed converter in the step-down mode is shown in Fig. 2. The gate signals S₁ and S₃ for the master active power switches Q₁ and Q₃ are defined as:

      (3)

Fig. 2. PWM modulation strategy in the step-down mode (S₂S₄=00).

Where m_a and m_b are the modulation indices for the left and right half bridges of the asymmetric H-bridge, is the limited condition, and is the range of values for the carrier. As a result, the duty cycles of Q₁ and Q₃ move towards 0.5 when m_a and m_b tend to 0.5 according to Fig. 2(a)-(c) and (3). In addition, the frequency of the output PWM voltage U_bg is twice the actual switching frequency. The narrower the pulse is, the shorter the inductor charging time exists. In this case, the step-down ratio can be very high. It is noted that the frequency of the inductor current is twice the actual switching frequency, and that the inductor current ripple is half that of the Buck/Boost converter. As a result, a smaller inductance can be adopted to reduce cost and to increase the power density. In the continuous current mode (CCM), the energy W_ch1 charged in the inductor is equal to the energy W_dis1 discharged from it in each carrier period T. Equation (4) can be described as:

      (4)

where t_off1 and t_off3 are the "OFF" time intervals for Q₁ and Q₃ in each carrier period. In terms of (4), the relationship between the low voltage side U_l and the high voltage side U_h can be written as follows in the step-down mode:

                   (5)

where M_buck is the step-down ratio, and d₁=m_a=(T－t_off1)/T and d₃=1－m_b=(T－t_off3)/T are the duty cycles for the master active power switches Q₁ and Q₃, respectively. In addition, the relationship between m_a and m_b in the step-down mode can be refrained as follows, taking into account the fact that all of the duty cycles are close to 0.5.

   (6)

When the power flow is from U_l to U_h, the PWM modulation strategy in the step-up mode is shown in Fig. 3. The gate signals S₂ and S₄ for the master active power switches Q₂ and Q₄ are defined by (7).

     (7)

where . The duty cycles of Q₂ and Q₄ move towards 0.5 when m_a and m_b approach 0.5 according to Fig. 3(a)-(c) and (7). Furthermore, the output PWM voltage U_bg with double the switching frequency can be achieved by means of (2) and Fig. 3(d)-(g). It is noted that the narrower the pulses of U_bg are, the shorter the inductor discharge time exists. A high step-up ratio can be obtained. In addition, the frequency of the inductor current is twice the actual switching frequency, and the inductor current ripple is also half that of the Buck/Boost converter, as well as that in the step-down mode. In the CCM, the energy W_ch2 charged in the inductor is equal to the energy W_dis2 discharged in each carrier period, and equation (8) can be written as:

      (8)

where t_on2 and t_on4 are the "ON" time intervals of Q₂ and Q₄. According to (8), the relationship between U_l and U_h can be obtained as follows in the step-up mode:

                 (9)

where M_boost is the step-up ratio, and d₂=1－m_a=t_on2/T and d₄=m_b=t_on3/T are the duty cycles for Q₂ and Q₄, respectively. Furthermore, the relationship between m_a and m_b in the step-up mode can be obtained as follows, considering that all of the duty cycles are as close to 0.5 as possible.

      (10)

Combining (5) and (9), the relationship between U_l and U_h in both directions can be unified as follows:

           (11)

where . According to Fig. 2, and Fig. 3, the equivalent carrier frequency of the output PWM voltage is double the actual switching frequency. The wide voltage-gain range of the proposed bidirectional converter always exists without the need for extreme duty cycles as shown in Fig. 4, by means of (6) and (10).

Fig. 3. PWM modulation strategy in step-up mode (S₁S₃=00).

Fig. 4. Comparison of voltage-conversion ranges for the proposed converter and the Buck/Boost converter. (a) In the step-down mode. (b) In the step-up mode.

(a)

(b)

According to Fig. 4, the conventional Buck/Boost converter has a narrower voltage conversion range and suffers from an extreme duty cycle (i.e. over 0.8~0.9) when it operates at a high voltage gain. As for the proposed converter, in the step-down mode, high step-down ratios (e.g. 0.1~0.3) change linearly with the duty cycle. As a result, the required duty cycles d₁ and d₃ are around 0.5 (approximately in the range of 0.55~0.65), as shown in Fig. 4(a). In the step-up mode, although the high step-down ratios (e.g. 3~20) vary in a non-linear manner with the duty cycles d₂ and d₄, the required duty cycles are also around 0.5 (approximately in the range of 0.3~0.475), as shown in Fig. 4(b). Based on the analysis above, the proposed converter has a wider voltage gain range. The duty cycles for all of the semiconductors are kept close to 0.5.

Even if it operates with a large conversion ratio (i.e. U_h/U_l =20), more appropriate duty cycles (0.475, 0.525) appear, rather than the extremely low duty cycles (i.e. d=0.05 in the step-down mode, and d=0.95 in the step-up mode) in the Buck/Boost converter. It is beneficial to reduce the losses and improve the conversion efficiency.

C. Synchronous Rectification of Slave Active Power Switches

According to Fig. 2 and Fig. 3, master active power switches change according to the power flow direction. There are three switching states for each power flow direction when the slave active power switches operate as a diode rectifier (DR).

   (12)

In addition, all four of the active power switches (MOSFETs) of the common grounded asymmetric H-bridge are controlled in a complementary manner as:

               (13)

Therefore, the three DR switching states in each power flow direction can be described by (14), namely the slave active power switches operate as synchronous rectifiers (SR) in each power flow direction. As a result, the slave active power switches can use zero-voltage-switching (ZVS) turn-on and turn-off, in terms of the dead time t_d for Q₁ and Q₂ in the left half bridge, and Q₃ and Q₄ in the right half bridge.

     (14)

The synchronous rectification operation principle of the proposed bidirectional converter is shown in Fig. 5. When it operates in the step-down mode, Q₁ and Q₃ are master active power switches, and Q₂ and Q₄ are slave active power switches, as shown in Fig. 5(a). The anti-parallel diodes D₂ and D₄ follow currents during the dead time t_d for Q₂ and Q₄, leading to zero voltages across Q₂ and Q₄. As a result, Q₂ and Q₄ can obtain the ZVS turn-on and turn-off with the "step-down-SR" switching states in (14).

Fig. 5. Synchronous rectification operation principle of the proposed bidirectional converter. (a) Current-flow path in the step-down mode. (b) Current-flow path in the step-up mode.

(a)

(b)

Similarly, Q₁ and Q₃ change to be slave active power switches, and Q₂ and Q₄ become master active power switches when the converter operates in the step-up mode, as shown in Fig. 5(b). The anti-parallel diodes D₁ and D₃ also follow currents during the dead time t_d for Q₁ and Q₃, resulting in zero voltages across Q₁ and Q₃. Therefore Q₁ and Q₃ can also achieve the ZVS turn-on and turn-off with the "step-up-SR" switching states in (14).

D. Parameters Design of the Capacitor and Inductor

According to Fig. 1(c), Fig. 2 and the operation principle of the bidirectional DC-DC converter in the step-down mode, when S₁S₃=11, Q₁ and Q₃ are turned on, Q₂ and Q₄ are turned off, and the inductor is charging. The charging time in each carrier period is (d₁+d₃-1)×T, and the high frequency component current of the inductor current flows through the low voltage side capacitor. Then, (15) can be obtained as:

              (15)

where ∆U_l is the voltage ripple of the low voltage side capacitor, and ∆I_L-Buck is the inductor current ripple in the step-down mode. In addition, U_l=(d₁+d₃-1)×U_h. Then, the capacitance of the low voltage side capacitor and the inductance L_Buck in the step-down mode can be obtained as:

      &bsp;               (16)

Similarly, in the step-up mode, the inductor discharges when S₂S₄=00. The discharging time in each carrier period is (1-d₂-d₄)×T. Then, (17) can be achieved as:

             (17)

where ∆U_h is the voltage ripple of the high voltage side capacitor, and ∆I_L-Boost is the inductor current ripple in the step-up mode. In addition, I_o=(1-d₂-d₄)×I_L-Boost is the load current in the high voltage side, U_l=(1-d₂-d₄)×U_h. Then, the capacitance of the high voltage side capacitor and the inductance L_Boost in the step-up mode can be achieved as:

   (18)

In terms of (16) and (18), the capacitance of the high and low voltage-side capacitors and the inductance of the inductor can be designed as above.

E. Comparisons with Other Bidirectional Solutions

According to the analysis above, comparisons can be drawn among the proposed and other bidirectional solutions in the step-up mode, as shown in Table III. The traditional Buck/ Boost converter only needs two semiconductors, and its maximum conversion efficiency is about 94.4%. However, its ideal voltage-gain 1/(1-d) is limited due to the effects of parasitic resistance and extreme duty cycles. As a result, it cannot meet the requirements of energy storage system applications. The bidirectional DC-DC converters in [16] and [26] achieved a higher voltage-gain. However, these converters need two inductors. In addition, the maximum voltage stress across the semiconductors of the converter in [16] is U_h+U_h(1-d), which increases the switching losses and reduces the conversion efficiency. The potential difference between the output and the input side grounds of the converters in [26] and [29] exists the high frequency PWM voltage (i.e. without a common ground), which may result in more EMI issues. In terms of the proposed asymmetric H-bridge bidirectional DC-DC converter, the number of main components is equal to those of the converters in [16] and [29], and its voltage gain is higher than those in [16] and [26]. In Table III, it is shown that the efficiency of the proposed converter is almost the same as that of the Buck/Boost converter in [30], and that the lowest efficiency increases from 71% with the converter in [29] to 83%, which is a great improvement. The voltage/current stresses on the semiconductors of the proposed converter are the same as those of the Buck/Boost converter under the same input and output voltage/power. In addition, the low and high voltage sides have a common ground, and a high step-up/step-down ratio can be achieved when all of the active power switches operate with duty cycles close to 0.5, which improves the converter’s performance and reliability. When compared to the Buck/Boost converter, although two additional semiconductors are required, the proposed converter can achieve a higher voltage gain and a wider voltage conversion range with proper duty cycles. In addition, an effective modulation method reduces the inductor current ripple to half that of the Buck/Boost converter. As a result, a smaller inductance can be adopted to reduce the volume and enhance the dynamic response.

TABLE III COMPARISONS OF THE PROPOSED AND OTHER BIDIRECTIONAL SOLUTIONS

Bidirectional Solutions	Voltage Gain	Amount of Semiconductors	Amount of Inductors	Maximum Voltage Stress across Semiconductors	Maximum Current Stress across Semiconductors	Common Ground	Efficiency
Buck/Boost Converter in [30]		2	1	Uh	Ih/(1-d)	YES	83.5% - 94.4%
Converter in [16]		4	2	Uh+Uh(1-d)		YES	95% - 97.1%
Converter in [26]		3	2			NO	86% - 98%
Converter in [29]		4	1	Uh		NO	71% - 93.6%
Proposed Converter		4	1	Uh		YES	83% - 93.5%

F. Control Strategy of Bidirectional Power Flows

Based on the operation principles in Section III (A~C), a bidirectional power flow control strategy can be achieved as shown in Fig. 6. The feedback voltages U_h and U_l, and the feedback current i_L are sampled by sensors. According to the mode selection signal U_c, the operation modes of the bidirectional DC-DC converter switch between the step-up and the step-down modes. It operates in the step-up mode when U_c=1, and the inductor current i_L is controlled by the Boost current controller with the reference current I_ref-Boost in the current-loop. Meantime, the voltage U_h is controlled by the Boost voltage controller with a reference voltage U_ref-Boost in the voltage-loop. The corresponding PWM scheme, as shown in Fig. 3 and Fig. 5(b), is selected to generate the gate signals S₁~S₄ in the step-up mode. If U_c is changed from "1" to "0", the inductor current i_L is controlled by the Buck current controller with an opposite direction reference current I_ref-Buck, and the voltage U_l is controlled by the Buck voltage controller with a reference voltage U_ref-Buck. The corresponding PWM scheme, as shown in Fig. 2 and Fig. 5(a), is selected to generate the gate signals S₁~S₄ in the step-down mode. As a result, the inductor current rises reversely after falling to zero.

Fig. 6. Control strategy of bidirectional power flows.

Ⅳ. EXPERIMENT RESULTS AND ANALYSIS

In order to validate the feasibility and effectiveness of the proposed converter, a 300W prototype was developed, as shown in Fig. 7. The low voltage side is variable (U_l =24~48V), and the high voltage side is constant (U_h=200V). The bidirectional voltage and current loops are controlled by a TMS320F28335 DSP, and MOSFETs (IXYS-IXFK64N50P) are selected as the active power switches. The switching frequency is f_s=10kHz, the dead time is t_d=1㎲, and the initial value of the inductor is L=306μΗ. The experimental parameters are shown in Table IV.

TABLE IV EXPERIMENTAL PARAMETERS

Parameters	Values
Rated power P	300W
Storage/filter capacitor Cl	200uF
Storage/filter capacitor Ch	330uF
Storage/filter inductor L	306uH
High voltage side Uh	200V
Low voltage side Ul	24~48V
Switching frequency fs	10kHz
Power semiconductors Q1~Q4	IXYS-IXFK64N50P

Fig. 7. Experimental prototype of the asymmetric H-bridge bidirectional DC-DC converter.

The voltage stress and the gate signals of the slave active power switches in the SR mode of operation are shown in Fig. 8. In the step-down mode, the slave active power switches Q₂ and Q₄ achieve ZVS turn-on and turn-off. The gate signal S₂ and the voltage stress U_Q2 for Q₂ are shown in Fig. 8(a). In the step-up mode, the slave active power switches Q₁ and Q₃ obtain ZVS turn-on and turn-off. The gate signal S₃ and the voltage stress U_Q3 for Q₃ are shown in Fig. 8(b). Therefore, the slave active power switches of the proposed bidirectional converter can use ZVS turn-on and turn-off without any extra hardware. This is beneficial to improve conversion efficiency.

Fig. 8. Voltage stress and gate signals of the slave active power switches in SR operation. (a) Gate signal and voltage stress of Q₂ in the step-down mode. (b) Gate signal and voltage stress of Q₃ in the step-up mode.

(a)

(b)

The voltages on the high voltage side (constant U_h=200V) and on the continuous variable low voltage side (U_l is between 24V and 48V) are shown in Fig. 9. In the step-down mode, the input voltage is constant at 200V, and the output voltage changes from 24V to 48V continuously over 8 seconds due to the action of the voltage control loop, as shown in Fig. 9(a). The proposed converter can operate over a wide step-down voltage-conversion range (from 0.12 to 0.24). When it operates in the step-up mode, the input voltage changes from 48V to 24V continuously over 8 seconds, and the output voltage remains at a constant 200V due to the action of the voltage control loop, as shown In Fig. 9(b). Therefore, the proposed converter can also operate in a wide step-up voltage-conversion range (from 4.2 to 8.4).

Fig. 9. Voltages on the high voltage side (constant 200V) and the continuous variable low voltage side (between 24V and 48V). (a) In the step-down mode. (b) In the step-up mode.

(a) In the step-down mode.

(b) In the step-up mode

When the proposed converter operates in the step-down mode (converting 200V to 24V), a narrow pulse output PWM voltage U_bg=U_ag - U_ab is needed. The output PWM voltages, the inductor current i_L and the corresponding gate signal- voltage stress are shown in Fig. 10. The PWM voltages U_ag and U_ab from the half bridges are shown in Fig. 10(a). Then, the narrow pulse output PWM voltage U_bg is obtained from U_ag - U_ab as shown in Fig. 10(b). Therefore, in each switching period T_s=100, the narrow pulse output PWM voltage U_bg is double the actual switching frequency. The inductor is charged when U_bg is at 200V (the narrow pulse), and discharged when U_bg is at zero, as shown in Fig. 10(b). Although a high step-down ratio is achieved, all of the active power switches operate with proper duty cycles, where d₁≈0.58 is close to 0.5, taking the active power switch Q₁ as an example as shown in Fig. 10(c). It is noted that the inductor of the proposed converter is charged and discharged twice during each switching period. When compared with the Buck/Boost converter, there are two additional power semiconductors in the proposed converter. However, the equivalent switching frequency of the proposed converter is double that of the Buck/Boost converter. The volumes of all the capacitors and inductors in the proposed converter can be reduced by almost half when compared with those of the Buck/Boost converter.

Fig. 10. Output PWM voltages, inductor current i_L and the corresponding gate signal-voltage stress in the step-down mode. (a) U_ag and U_ab. (b) U_bg and the inductor current i_L. (c) S₁ and U_Q1.

(a)

(b)

(c)

When running in the step-up mode (converting 24V to 200V), a narrow pulse output PWM voltage is also needed. The output PWM voltages, the inductor current i_L and the corresponding gate signal-voltage stress are shown in Fig. 11. The PWM voltages U_ag and U_ab from the half bridges are shown in Fig. 11(a), and the narrow pulse output PWM voltage U_bg is obtained from U_ag - U_ab as shown in Fig. 11(b). Therefore, in each switching period T_s=100㎲, the narrow pulse output PWM voltage U_bg is double the actual switching frequency. The inductor is charged when U_bg is at zero, and it is discharged when U_bg is at 200V (the narrow pulse), as shown in Fig. 11(b). Although a high step-down ratio is obtained, all of the active power switches operate with the proper duty cycles, where d₄≈0.44 is close to 0.5, taking the active power switch Q4 as an example as shown in Fig. 11(c). In addition, similar to the step-down mode, the inductor of the proposed converter is also charged and discharged twice during each switching period.

Fig. 11. Output PWM voltages, inductor current i_L and the corresponding gate signal-voltage stress in the step-up mode. (a) U_ag and U_ab. (b) U_bg and the inductor current i_L. (c) S₄ and U_Q4.

(a)

(b)

(c)

Experimental results of the bidirectional operation of the proposed converter between the step-down and the step-up modes are shown in Fig. 12. The high voltage side is connected to a DC-link bus (at a constant 200V), and the low voltage side is connected to battery stacks (nominal 48V). In order to easily control the power flow between the battery stacks and the DC-link bus, the proposed converter, which is controlled as a current source converter, acts as an interface between them. In Fig. 12(a), the converter steps down with the reference inductor current I_L= - 4A, and the power flow is from the DC-link bus to the batter stacks (U_l=54V). When the converter is controlled to operate continuously from the step-down mode to the step-up mode, the converter works in the step-up mode with the reference inductor current I_L=4A, and the power flow is from the battery stacks (U_l=52V) to the DC-link bus. The detailed transient process of the step-down to the step-up mode (I_L= - 4A to 4A) is shown in Fig. 12(b). The current of the battery stacks changes from - 4A to 4A over 3.2ms. In other words, the battery stacks are charged with a constant current of 4A before the controlled power flow command comes. Then, the battery current falls to zero (i_L=0) quickly, and the battery stacks switch to be discharged with a constant current of 4A over 3ms. The detailed transient process of the step-up to the step-down mode (I_L=4A to - 4A) is shown in Fig. 12(c). The battery current falls rapidly to zero (i_L=0) to switch the operating mode after the new power flow command comes. Then, the charged current of the battery stacks stabilizes at a constant current of 4A over 8ms. From Fig. 12, it can be seen that the proposed converter can respond quickly to the control signal under the current control loop. This is due to the effective modulation method and the reduced inductance.

Fig. 12. Experimental results of bidirectional operation between the step-down and the step-up modes. (a) Processes of step-up to step-down and step-down to step-up. (b) Transient process of the step-down to step-up mode (I_L=－4A to 4A). (c) Transient process of the step-up to step-down mode (I_L=4A to -4A).

(a)

(b)

(c)

The conversion efficiency of the proposed bidirectional converter at different voltages on the low voltage side (24V to 48V) and at different load powers (100W to 300W) has been measured using a Yokogawa-WT3000 Power Analyzer, as shown in Fig. 13. In the step-down mode, the input voltage is 200V. The minimum efficiency is 90.7% when the output voltage is 24V and the load power is 300W, while the maximum efficiency is 94.7% when the output voltage is 48V and the load power is 300W. In the step-up mode, the output voltage is set as 200V. The minimum efficiency is 83% when the input voltage is 24V and the load power is 300W, while the maximum efficiency is 93.5% when the input voltage is 48V and the load power is 200W. Therefore, the conversion efficiency improves by increasing voltage on the low voltage side.

Fig. 13. Conversion efficiency of the proposed bidirectional converter at different voltages of low voltage side and load powers (U_l=24~48V, U_h=200V).

The calculated power loss distributions for the experiment when U_l=48V, U_h=200V and P=300W are shown in Fig. 14. In the step-down mode, the total losses of the converter are 12.09W, and the loss distribution is shown in Fig. 14(a). By analyzing the power losses distribution, it can be concluded that the major losses come from the power switches Q₁ and Q₃. The turn-on, turn-off and conduction losses of Q₁ and Q₃ account for 38.627% of the total losses. The conduction losses of the power switches Q₁~Q₄ account for 25.475% of the total losses. The conduction losses and the core loss of the inductor account for 20.595% and 15.303% of the total losses, respectively. In the step-up mode, the total losses of the converter are 13.3W, and Fig. 14(b) shows the loss distribution. The largest power losses are the turn-on and turn-off losses of Q₂ and Q₄ and the conduction losses of Q₁~Q₄, which account for 58.271% of the total losses. The losses of the capacitors and the copper loss of the inductor account for 27.819% of the losses, and the remaining 13.91% of the total losses is the core loss of the inductor.

Fig. 14. Calculated power loss distributions for the experiment when U_l=48V, U_h=200V, P=300W. (a) In the step-down mode. (b) In the step-up mode.

(a)

(b)

Ⅴ. CONCLUSIONS

A wide-voltage-conversion range bidirectional DC-DC converter is proposed in this paper. A high step-down/step-up ratio can be achieved by using one typical LC energy storage component and a special common grounded asymmetric H-bridge comprised of four active power switches with anti-parallel diodes, instead of a transformer, a switched- capacitor, and a coupled-inductor or a Z-source net. In addition, its slave active power switches can achieve ZVS turn-on and turn-off without any extra hardware, and the equivalent frequency of the output PWM voltage is twice as high as the actual switching frequency. A 300W prototype obtains a maximum conversion efficiency of 94.7% in the step-down mode and 93.5% in the step-up mode. Therefore, the proposed bidirectional topology with a common ground is suitable for the energy storage systems of renewable applications and hybrid energy source electric vehicle applications.

ACKNOWLEDGMENT

This work was supported in part by the National Natural Science Foundation of China under Grant 51577130, and in part by the Research Program of Application Foundation and Advanced Technology of Tianjin, China under Grant 15JCQNJC03900.

REFERENCES

[1] B. Singh and R. Kumar, “Solar photovoltaic array fed water pump driven by brushless DC motor using Landsman converter,” IET Renew. Power Gener., Vol. 10, No. 4, pp. 474-484, Apr. 2016.

[2] M. S. Mahmoud, M. S. U. Rahman, and F. M. A.L.-Sunni, “Review of microgrid architectures–a system of systems perspective,” IET Renew. Power Gener., Vol. 9, No. 8, pp. 1064-1078, Nov. 2015.

[3] J. d. Santiago, H. Bernhoff, B. Ekergard, S. Eriksson, S. Ferhatovic, R. Waters, and M. Leijon, “Electrical motor drivelines in commercial all-electric vehicles: A review,” IEEE Trans. Veh. Technol., Vol. 61, No. 2, pp. 475-484, Feb. 2012.

[4] C. C. Chan, “The state of the art of electric and hybrid vehicles,” Proceedings of the IEEE, Vol. 90, No. 2, pp. 247-275, Feb. 2002.

[5] D. Wu, F. Tang, T. Dragicevic, J. C. Vasquez, and J. M. Guerrero, “A control architecture to coordinate renewable energy sources and energy storage systems in islanded microgrids,” IEEE Trans. Smart Grid, Vol. 6, No. 3, pp. 1156-1166, May 2015.

[6] N. L. Diaz, T. Dragicevic, J. C. Vasquez, and J. M. Guerrero, “Intelligent distributed generation and storage units for DC microgrids – A new concept on cooperative control without communications beyond droop control,” IEEE Trans. Smart Grid, Vol. 5, No. 5, pp. 2476-2485, Sep. 2014.

[7] H. Yoo, S. Sul, Y. Park, and J. Jeong, “System integration and power-flow management for a series hybrid electric vehicle using supercapacitors and batteries,” IEEE Trans. Ind. Appl., Vol. 44, No. 1, pp. 108-114, Jan./Feb. 2008.

[8] K. Hu, P. Yi, and C. Liaw, “An EV SRM drive powered by battery/supercapacitor with G2V and V2H/V2G capabilities,” IEEE Trans. Ind. Electron., Vol. 62, No. 8, pp. 4714-4727, Aug. 2015.

[9] K. Wu, C. W. d. Silva, and W. G. Dunford, “Stability analysis of isolated bidirectional dual active full-bridge DC–DC converter with triple phase-shift control,” IEEE Trans. Power Electron., Vol. 27, No. 4, pp. 2007-2017, Apr. 2012.

[10] N. Denniston, A. M. Massoud, S. Ahmed, and P. N. Enjeti, “Multiple-module high-gain high-voltage DC–DC transformers for offshore wind energy systems,” IEEE Trans. Ind. Electron., Vol. 58, No. 5, pp.1877-1886, May 2011.

[11] A. Rodriguez, A. Vazquez, D. G. Lamar, M. M. Hernando, and J. Sebastian, “Different purpose design strategies and techniques to improve the performance of a dual active bridge with phase-shift control,” IEEE Trans. Power Electron., Vol. 30, No. 2, pp. 790-804, Feb. 2015.

[12] A. Ajami, H. Ardi, and A. Farakhor, “A novel high step-up DC/DC converter based on integrating coupled inductor and switched-capacitor techniques for renewable energy applications,” IEEE Trans. Power Electron., Vol. 30, No. 8, pp. 4255-4263, Aug. 2015.

[13] R. J. Wai and R. Y. Duan, “High-efficiency bidirectional converter for power sources with great voltage diversity,” IEEE Trans. Power Electron., Vol. 22, No. 5, pp. 1986-1996, Sep. 2007.

[14] H. Wu, K. Sun, L. Zhu, L. Chen, and Y. Xing, “High step-up/step-down soft-switching bidirectional DC-DC converter with coupled-inductor and voltage matching control for energy storage systems,” IEEE Trans. Ind. Electron, Vol. 63, No. 5, pp. 2892-2903, May 2016.

[15] C.-C. Lin, L.-S. Yang, and G. W. Wu, “Study of a non-isolated bidirectional DC-DC converter,” IET Power Electron., Vol. 6, No. 1, pp. 30-37, Jan. 2013.

[16] H. Ardi, A. Ajami, F. Kardan, and S. N. Avilagh, “Analysis and implementation of a nonisolated bidirectional DC-DC converter with high voltage gain,” IEEE Trans. Ind. Electron., Vol. 63, No. 8, pp. 4878-4888, Aug. 2016.

[17] M. Kwon, S. Oh, and S. Choi, “High gain soft-switching bidirectional dc-dc converter for eco-friendly vehicles,” IEEE Trans. Power Electron., Vol. 29, No. 4, pp. 1659-1666, Apr. 2014.

[18] P. Jose and N. Mohan, “A novel ZVS bidirectional cuk converter for dual voltage systems in automobiles,” in Proc. IEEE IECON Conf. Rec., pp. 117-122, 2003.

[19] I.-D. Kim, S.-H. Paeng, J.-W. Ahn, E.-C. Nho, and J.-S. Ko, “New bidirectional ZVS PWM Sepic/Zeta DC–DC converter,” in Proc. IEEE ISIE Conf. Rec., pp. 555-560, 2007.

[20] R. J. Wai and R. Y. Duan, “High-efficiency bidirectional converter for power sources with great voltage diversity,” IEEE Trans. Power Electron., Vol. 22, No. 5, pp. 1986-1996, Sep. 2007.

[21] P. U R, and A. K. Rathore, “Extended range ZVS active-clamped current-fed full-bridge isolated DC/DC converter for fuel cell applications: analysis, design, and experimental results,” IEEE Trans. Ind. Electron., Vol. 60, No. 7, pp.2661-2672, Jul. 2013.

[22] L. Jiang, C. C. Mi, S. Li, M. Zhang, X. Zhang, and C. Yin, “A novel soft-switching bidirectional DC–DC converter with coupled inductors,” IEEE Trans. Ind. Appl., Vol. 49, No. 6, pp. 2730-2740, Nov./Dec. 2013.

[23] Y.-F. Wang, L.-K. Xue, C.-S. Wang, P. Wang, and W. Li, “Interleaved high-conversion-ratio bidirectional DC-DC converter for distributed energy-storage systems – Circuit generation, analysis, and design,” IEEE Trans. Power Electron., Vol. 31, No. 8, pp. 5547-5561, Aug. 2016.

[24] S.-C. Tan, S. Kiratipongvoot, S. Bronstein, A. Ioinovici, Y. M. Lai, and C. K. Tse, “Adaptive mixed on-time and switching frequency control of a system of interleaved switched-capacitor converters,” IEEE Trans. Power Electron., Vol. 26, No. 2, pp. 364-380, Feb. 2011.

[25] B. Wu, K. Smedley, and S. Sigmond, “A new 3X interleaved bidirectional switched capacitor converter,” in Proc. IEEE Appl. Power Electron. Conf. Expo., pp. 1433- 1439, 2014.

[26] O. Cornea, G. Andreescu, N. Muntean and D. Hulea, “Bidirectional power flow control in a dc microgrid through a switched-capacitor cell hybrid DC-DC converter,” IEEE Trans. Ind. Electron., Vol. 64, No. 4, pp. 3012-3022, Apr. 2017.

[27] J. Zhang and J. Ge, “Analysis of Z-source DC-DC converter in discontinuous current mode,” in Proc. Power Energy Eng. Conf. (APPEEC), pp. 1-4, 2010.

[28] H. Shen, B. Zhang, D. Qiu, and L. Zhou, “A common grounded Z-source DC-DC converter with high voltage gain,” IEEE Trans. Ind. Electron., Vol. 63, No. 5, pp.2925-2935, May 2016.

[29] Y. Zhang, Y. Gao, J. Li, M. Sumner, P. Wang, and L. Zhou, “High ratio bidirectional DC-DC converter with synchronous rectification H-bridge for hybrid energy sources electric vehicles,” J. Power Electron., Vol. 16, No. 6, pp. 2035-2044, Nov. 2016.

[30] B. Wang, J. Xu, R. J. Wai, and B. Cao, “Adaptive sliding-mode with hysteresis control strategy for simple multimode hybrid energy storage system in electric vehicles,” IEEE Trans. Ind. Electron., Vol. 64, No. 2, pp.1404-1414, Feb. 2017.

Yun Zhang (M’13) was born in Jiangsu, China, in 1980. He received his B.S. and M.S. degrees in Electrical Engineering from the Harbin University of Science and Technology, Harbin, China, in 2003 and 2006, respectively; and his Ph.D. degree in Electrical Engineering from the Harbin Institute of Technology, Harbin, China, in 2010. In 2010, he joined Tianjin University, Tianjin, China, as a Lecturer in the School of Electrical and Information Engineering, where he is presently working as an Associate Professor. His current research interests include topologies, modulation, and control strategies for the power converters of electric vehicles and microgrids. Dr. Zhang is an Associate Editor of the Journal of Power Electronics.

Yongping Gao was born in Shanxi, China. He received his B.S. degree in Electrical Engineering from the China University of Mining and Technology, Xuzhou, China, in 2015. He is presently working towards his M.S. degree in Electrical Engineering from Tianjin University, Tianjin, China. His current research interests include power electronics converters and energy management.

Jing Li (M'15) was born in Beijing, China. She received her B.S. (with Honors) and M.S. (with Distinction) degrees from the Beijing Institute of Technology, Beijing, China, in 1999 and 2002, respectively; and her Ph.D. degree from the University of Nottingham, Nottingham, ENG, UK, in 2010. She subsequently worked as a Research Fellow within the Power Electronic, Machine and Control Group (PEMC), University of Nottingham. She is presently working as a Lecturer in the Department of Electrical and Electronic Engineering, University of Nottingham Ningbo China, Ningbo, China. Her current research interests include condition monitoring for motor drive systems and power distribution systems, and the advanced control and design of motor drive systems.

Mark Sumner (SM’05) received his B.S. degree in Electrical and Electronic Engineering from Leeds University, Leeds, ENG, UK, in 1986; and his Ph.D. degree in Induction Motor Drives from the University of Nottingham, Nottingham, ENG, UK, in 1990. After graduating, he worked as a Research Assistant and he became a Lecturer in 1992. He is presently working as a Professor of Electrical Energy Systems. His current research interests include the control of power electronic systems including sensorless motor drives, diagnostics and prognostics for drive systems, power electronics for enhanced power quality and novel power system fault location strategies.