Double-switch switched-inductor converter with minimal switch voltage stress for renewable energy conversion

https://doi.org/10.1016/j.compeleceng.2022.107682Get rights and content

Highlights

  • For renewable energy conversion, a novel DC-DC converter configuration known as the double-switch switched-inductor (DS-SI) converter and its extension are discussed.

  • The suggested topology is extended to attain a more significant voltage gain.

  • A 500 W experimental prototype is used to validate the proposed converter configuration.

  • The efficiency curve is shown for various power and input voltage conditions.

Abstract

In recent years, based on the notion of switched-inductor, DC-DC converters have been used in various applications, including DC microgrids, renewable energy integration, and high voltage light-emitting diodes. This publication presents a unique double-switch switched-inductor converter with minimal switch voltage stress for renewable energy conversion. The suggested converter is formed from different design arrangements of the traditional switched-inductor converter to reduce voltage stress and increase performance. The converter's non-ideal model and several operation modes depending on inductor current characteristics and output voltage ripple are described. The suggested topology is extended to attain a more significant voltage gain. The design and comparison of the proposed converters with recent converters are discussed. Finally, experimental prototype results are presented for a load power of 500 W to validate the proposed converter configuration. The efficiency curve is shown for various power and input voltage conditions. The loss breakdown characteristics for a load power of 500 W are provided.

Introduction

In recent years, with growing concerns about energy consumption and the depletion of fossil energy resources, there has been an increasing interest and attention towards efficient, reliable, low-cost solutions for microgrids and renewable energy sources and conversion [1]. DC distribution systems attract much attention due to renewable energy sources, e.g., photovoltaic and fuel cells [2]. Further, computing techniques can improve the power quality of integrated systems [3]. Regrettably, the voltage levels of these energy sources are insufficient for direct use. Additional sources must be linked in series; however, such resources are expensive, take up a large space, and significantly decrease overall efficiency and reliability [4]. As a result, front-end converters are required to accomplish the requisite voltage level for real-time applications [5]. In Fig. 1, a typical DC grid architecture is depicted, wherein renewables and batteries are connected to regular DC buses via DC-DC converters to boost the voltage from (12∼48 V) to the DC bus voltage level (200∼400 V). The generated 400 V voltage can be supplied directly or through a DC-DC converter to the DC loads based on the requirement. Also, inverter circuitry can be an intermediate stage between a 400 V DC bus and an AC load or utility grid. The classical boost converter can theoretically achieve significant voltage gain by providing a duty cycle greater than 0.9 [6]. Reverse recovery of a diode, high conduction losses, low efficiency, high current rating inductor, etc., are often considered practical limitations of the classical boost converter [7].

Multiple voltage-boosting approaches have been proposed to obtain high voltage gain, including cascade connection, switched reactive cells, i.e., inductors and capacitors [8], and multiplier [9] with a classical converter. It is also well known that the cascade connection of two (or several) converter modules can provide a high voltage. However, the main drawbacks are the high and different voltage ratings of components/devices. Furthermore, the voltage rating of elements increases as cascaded modules increase; for example, the voltage rating of devices in the final cascaded module is equal to the output voltage [10]. Moreover, the order of the converter is high, and the energy is transferred from one converter module to another converter module, which results in high conduction losses. For high voltage gain, the literature reports a connection of the two converters as a quadratic boost converter [11]. However, the switch's voltage rating is relatively high (equal to the output voltage), the voltage gain plot is very non-linear, and requires a precise control technique. Several capacitors’ charging and discharging configurations are suggested for switched-capacitor based converters to attain high voltage [12]. These converters are well-suited to low-power applications. Furthermore, the voltage gain of such converters depends on the number of capacitors, and the direct charging of the capacitor through the source results in a high input current spike. The voltage multiplier stages are connected to primary converters in [13] to generate a higher voltage. However, these configurations require many diodes and series/parallel connected capacitors, which yield low efficiency due to transferring the energy between the multiple capacitor loops [14]. In [15], the switched-inductor (SI) based converters are proposed to mitigate the limitations and obtain higher potential. These converters demand high-voltage switches and have limited voltage gains. The active switched-inductor (ASI) based converters are introduced in [16] to advance the configuration in reducing switch voltage rating and the number of diodes. Still, their voltage gains are limited despite using the inherent voltage lifting technique. In [17], the ASI technique is used with a switched-capacitor network for the same purpose. However, the intermediate stage of such converters requires many diodes and reactive elements. The energy is distributed throughout several loops, resulting in high cost and low performance. In [18], an active-passive inductor network is suggested to produce significant voltage gain with no capacitor and diode circuitry. In [19], the inductors of the converter, as presented in [18], are replaced by a switched-inductor to achieve a higher output voltage. However, in [18]–[19], the voltage/current rating devices are not uniform, and this issue increases when one moves from the input to the output side.

Furthermore, these converters can only handle floating loads because their output is not connected to a common ground input. The ASI circuitry is extended to obtain a higher voltage by operating the converter in three modes [20]. However, these converters require an additional series-connected switch-diode to work in three modes. The voltage gain is limited, two duty cycle functions are used, and it requires a complex control scheme. Besides, these power converters support loads where input and output do not have common ground. The converter with multiple legs and two duty functions is presented in [21]-[22]. However, they require multiple legs with different ratings and are suitable only for a floating load. In [23], the basic configuration of a quad switched-inductor with two switches is proposed. However, the detailed analysis, circuit extension, operation modes based on different inductor current conditions, the ripple in output voltage, and operation regions for various inductance levels are not discussed. Moreover, the converter is investigated only with a few simulation results. The main contributions of the presented research work are:

  • For renewable energy conversion, a novel DC-DC converter configuration known as the double-switch switched-inductor (DS-SI) converter and its extension are discussed.

  • The advantages of the proposed converter are low switch voltage stress, low inductor current rating, and high voltage at the output without multiple capacitor loops.

  • By using a common ground for input and output, the modular circuit structure is formed, and high voltage gain is achieved without the use of a multiplier stage.

  • Different modes of operation, non-ideal conditions of passive and active components, and voltage-current characteristics are investigated. The converter extension to enhance voltage gain is discussed.

  • The ripple content in output voltage and operation regions are discussed for various inductance levels.

  • The proposed converter comparison with the current configuration addressed in the literature shows the unique benefits.

    The experimental results demonstrated a clear validation of the theoretical study, functionality, and performance of the original development.

  • Efficiency is investigated at various power and input voltage conditions, demonstrating the proposed converter's capabilities for real-time renewable energy conversion.

The paper is organized as follows: Section II deals with the circuit and operation modes of the proposed converter. In section III, a detailed analysis of the converter is presented. Section IV deals with the extended version of the suggested converter to achieve high voltage gain. In sections V and VI, comparisons of converters, output voltage ripple, and operating regions are presented. Section VII discusses the design and experimental results of the suggested converter. Finally, the article is concluded in section VIII.

Section snippets

Double-switch switched-inductor (DS-SI) converter

Fig. 2 shows the proposed DS-SI converter power circuit. The circuit comprises two active-passive reactive networks (APRN), an intermediate diode Da, load R, and capacitor C. The output and input ports of the proposed circuit are both grounded. The capacitor C is linked in parallel with the load R. The APRN-1 comprises two inductors (L11 and L12), switch S1, and four diodes (D11, D12, D13, and D14). The APRN-2 consists of two inductors (L21 and L22), switch S2, and four diodes (D21, D22, D23,

Voltage and currents

From the characteristics of the waveforms, the peak, rms, and average voltage across inductors are obtained as,VL,pkL11,L12,L21,L22={VoutGV,t=0toDTVout(GV1)4GV,t=DTtoTVL,rmsL11,L12,L21,L22=Vout2GVGV2+4GV3GV+3;VL,avgL11,L12,L21,L22=0

From the characteristics of the waveforms, the peak, rms, and average current through inductors are obtained as,IL,pkL11,L12,L21,L22=Iin4(GV+3GV)+Vout2L11f(GV)(GV1GV+3)IL,rmsL11,L12,L21,L22=Iin216(GV+3GV)2+(Vout23L11f(GV)(GV1GV+3))2IL,avgL11,L12,L21,L22=Iin

Extension of DS-SI converter

The passive reactive networks (PRN) APRN-1 and APRN-2 can be extended to achieve high output voltage. The power circuit is shown in Fig. 10(a), where m and n number of inductors are shown in the APRN-1 and APRN-2, respectively.

The extended power circuit voltage gain is,GV,Ex=1+(m+n)d1d=GV,DSSI+(m+n4)d1d

The switches voltage stress are obtained as,VS1,Ex=1+md1dVin,VS2,Ex=nd1dVin

The normalized voltage stresses for switches are obtained as,VS1,Ex|norm=n+GV,Ex(m)GV,Ex(m+n),VS2,Ex|norm=nGV,Ex1

Comparison

Table I shows the comparison between the proposed and similarly recommended converters, with a maximum of six inductors in the power circuit. The voltage gain of the suggested converter is greater than that of the converters described in [15]-[16], [20], and is the same as that of the converters described in [18]-[19]. Compared to the converters proposed in [16, 18], and [20], the normalized voltage of the switch of the proposed converter is lower. Fig. 11(a) depicts the plots of total

Ripple in output voltage and operation regions

The ripple in output voltage for CCM-SIC, CCM-IIC, and DCM can be obtained by,ΔVout=[1RfC×Vout(VoutVin)(Vout+3Vin)]forCCMSICΔVout=2(VoutVin)LC[Vout4VinR+Vin2Lf(3Vin+Vout)]2forCCMIICΔVout=2LC(VoutVin)[VoutRVout(VoutVin)2LfR]2forDCM

From (64)-(66), for CCM-SIC, it is observed that the value of ripple in output voltage is independent of inductance value; however, in CCM-IIC and DCM, it is based on inductance value. The critical value of inductance between CCM-SIC and DCM, i.e., LC, and

Design and results

As illustrated in Fig. 2, the suggested DS-SI converter was developed in the lab, and the specifications are listed in Table III. The worst efficiency (say η=90%) is considered for designing the reactive elements. To achieve the required voltage, the typical duty cycle is calculated as,d(η=90%)=VG1(VG+3)η=6.671(6.67+3)0.9=65.14%

As mentioned in table III, the converter operates in CCM-IIC and DCM when L<LKminand L<LCmin, respectively. By using (67)-(68), the value of LC, LCmin, and LCmax are

Conclusion

A novel non-isolated double-switch switched-inductor configuration is suggested for high voltage gain and low switch voltage stress. The suggested converter provides a higher voltage at a lower duty cycle and has the advantage of minimal switch voltage stress (less than output voltage) and low current rating inductors. Low switch voltage allows low conduction resistance switches that further reduce the conduction losses. For various modes, mathematical analysis with characteristics waveform,

Author contribution

Mahajan Sagar Bhaskar developed the proposed concept and involve in the literature survey, design, mathematical analysis, writing, experimental work, and revision of the manuscript.

Sanjeevikumar Padmanaban involves in the validation of the mathematical analysis, validate the investigated results and supports to improve the quality of the manuscript.

Dhafer Almakhles involves in the validation of the mathematical analysis, writing draft, and simulation and experimental work of the proposed

Declaration of Competing Interest

We authors confirm that the article has no conflict of interest. The article solemnly presented to the special issue “Developments in Renewable Energy Generation and Automation (VSI-reg)” of Computers & Electrical Engineering, Elsevier Journal.

Acknowledgement

The authors would like to acknowledge the technical and financial support of Renewable Energy Lab, Department of Electrical Engineering, College of Engineering, Prince Sultan University (PSU), Riyadh 11586, Saudi Arabia.

M. S. BHASKAR was a post-doctoral researcher in the Department of Energy Technology, Aalborg University, Esbjerg, Denmark, Ph.D. in 2019. Currently, he is with Renewable Energy Lab, Prince Sultan University, Saudi Arabia. He has authored 150 plus scientific papers and received the Best Paper Research Paper Awards from IEEE-GPECOM’20, IEEE-CENCON’19. He is an Associate Editor of IET Power Electronics, UK.

References (23)

  • M. Veerachary et al.

    Analysis and design of quadratic following boost converter

    IEEE Trans Ind Appl

    (Nov. 2020)
  • Cited by (0)

    M. S. BHASKAR was a post-doctoral researcher in the Department of Energy Technology, Aalborg University, Esbjerg, Denmark, Ph.D. in 2019. Currently, he is with Renewable Energy Lab, Prince Sultan University, Saudi Arabia. He has authored 150 plus scientific papers and received the Best Paper Research Paper Awards from IEEE-GPECOM’20, IEEE-CENCON’19. He is an Associate Editor of IET Power Electronics, UK.

    SANJEEVIKUMAR PADMANABAN is with Department of Electrical and Electronics Engineering, Anna University, Chennai, India. He is a Fellow of the IE (India), the IETE (India), and the IET (UK). He is an Associate Editor for refereed journals, particularly the IEEE SYSTEMS JOURNAL, IEEE TIA, IEEE ACCESS, IET Power Electronics, IET Electronics Letters, etc.

    DHAFER J. ALMAKHLES received a Ph.D. degree from The University of Auckland, New Zealand, in 2016. Currently, he is with Renewable Energy Lab, Prince Sultan University, Saudi Arabia. His-research interests include power electronics, control theory, renewable energy systems. He is a senior member of IEEE.

    NIKITA GUPTA received Ph.D. degree from DTU, Delhi, India in 2018. Currently, she works as Assistant Professor at the University Institute of Technology, Himachal Pradesh University, India. Her research interests include power systems, power quality, electronics, renewable energy, and microgrids.

    UMASHANKAR SUBRAMANIAM is with Renewable Energy Lab, College of Engineering, Prince Sultan University, Saudi Arabia. He is a Senior Member-IEEE and has published over 250+ research papers on power electronics applications in renewable energy and allied areas. He is an Editor of Heliyon, Elsevier, and several collaborative research projects.

    View full text