- Email: [email protected]
Soft switching maximum power point tracker with resonant switch in PV system
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Soft switching maximum power point tracker with resonant switch in PV system* Selim Oncu a,*, Salih Nacar b a b
Karabuk University Engineering Faculty, Karabuk, 78050, Turkey Kastamonu University Taskopru Vocational High School, Kastamonu, 37400, Turkey
article info
abstract
Article history:
Conventional boost converters are used in many applications such as power factor
Received 11 November 2015
correction, electronic ballasts, battery chargers, photovoltaic (PV) applications. PV systems
Received in revised form
need maximum power point tracker (MPPT) to have efficient power conversion. Maximum
13 January 2016
power point trackers are usually implemented by pulse width modulation (PWM)
Accepted 14 January 2016
controlled converters. In PWM control, the switching losses and the voltageecurrent
Available online xxx
stresses of the device increase. In the present paper, an MPPT with soft switching boost converter is designed, simulated and experimentally tested. The topology adds a resonant
Keywords:
network to the conventional boost converter and controls the PV power by changing the
Boost converter
switching frequency. In the proposed converter, Perturb & Observe method is used as
MPPT
power tracking algorithm that is implemented with dsPIC30F2020 and zero voltage
Resonant switch
switching (ZVS) is achieved in wide range. The variable frequency controlled MPPT is tested
Zero voltage switching
for different radiation levels and loads and switching characteristic is compared with PWM
Frequency control
controlled boost converter. The proposed single switch resonant converter is simulated in PSIM and experimental validation is also given for a 40 W PV generation system. Copyright © 2016, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
Introduction Over the last years, for improving the efficiency and the power density of renewable energy sources like photovoltaic (PV) energy conversion systems, the interest in power electronics applications has increased. The output power of the PV panel is related to the intensity of solar radiation, the temperature, and the load [1e5]. PV systems work in different environmental and electrical conditions. For this reason, PV generators are connected to loads by dcedc converters called MPPT. The MPPT changes the operation point of the PV system to find
the maximum possible power in any case [6]. Generally PWM type boost converters are used for transferring the maximum output power of the panel. The advantages of the boost converters in such applications are the low input current ripple, and the easy control [1,7]. Furthermore, in the continuous current mode, the boost converter has better characteristics than the buck converter [8e11]. For increasing the power density of the converter, it is necessary to increase the switching frequency, and in this way the size and weight of the converter decrease too. However, operating at high switching frequency can produce several problems, such as increased switching losses and
*
¨ -BAP-13/1-YL-034. This work was supported in part by the Karabuk University BAP, KBU * Corresponding author. E-mail address: [email protected] (S. Oncu). http://dx.doi.org/10.1016/j.ijhydene.2016.01.088 0360-3199/Copyright © 2016, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Oncu S, Nacar S, Soft switching maximum power point tracker with resonant switch in PV system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.088
2
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
vgs
stress [12]. For preventing such problems, a snubber circuit or a soft switching cell in the PWM converter [13,14] can be used. For decreasing the switching losses in PV applications, some researchers tried to use soft switching PWM converters [15e18]. But PWM converters with an auxiliary soft switching circuit are more complex and expensive since they need an additional power switch, a complex control algorithm, and a floating gate driver [14,17e22]. By using a resonant switch, the mentioned drawbacks of the PWM converters may be avoided [23e28]. In this study, for tracking the MPP a quasi-resonant boost converter is designed and experimentally tested; soft switching is achieved by help of a resonant switch. The ZVS of the power device is achieved by using an LC resonant circuit. Parasitic inductances and capacitances can be incorporated into this resonant circuit. No snubber circuit is used. In addition, neither auxiliary switch nor floating gate drive is required. Perturb and observe MPPT algorithm is applied with dsPIC30F2020 and implemented on the single switch resonant converter. MPP of the 40 W PV panel is tracked by changing both switching frequency and the duty cycle. Zero voltage switching operation is maintained while tracking the MPP. Soft switching MPPT is compared with PWM controlled MPPT at 35 kHz switching frequency. The proposed system has good start-up characteristics with low switching losses and, it tracks the maximum power value in different conditions by variable frequency operation.
v ZnIi
Vo i Ii t0 t1
t4
t2 t3 Ts
Fig. 2 e Theoretical waveforms of ZVS boost converter.
Mode I (t0 t < t1): When the switch is turned off, the input current flows through the resonant capacitor. Fig. 3 shows the equivalent circuit for Mode I. This mode continues until the capacitor voltage reaches Vo. During this stage, D2 is open circuit and the resonant capacitor is charged from zero to Vo. The capacitor voltage rises linearly, so its voltage can be derived as:
ZVS boost converter for PV systems
vðtÞ ¼
The soft switching boost converter with a resonant switch is shown in Fig. 1(a). The converter consists of a PV Panel, a power switch (M) with its anti-parallel diode (D1), a choke inductor and filter capacitors (Li and Ci, Co), a resonant circuit (L and C), a rectifier diode (D2), and a load (Ro). In steady state, the input inductor and the input voltage source of the converter can be represented by a current source and output circuit can be replaced by a voltage sink [25,27]. Fig. 1(b) shows the equivalent circuit of the resonant boost converter. The resonant circuit shapes the voltage of the switch from a square wave to a sinusoidal wave. Fig. 2 shows the switching waveforms of the quasi-resonant boost converter [25]. A short description of the functioning of the proposed MPPT is as follows. Assuming the input current and output voltage are constant and the converter is in steady state, the converter has four operation modes [25].
Ii t C
(1)
Mode II (t1 t < t2): When the capacitor voltage reaches Vo, the diode D2 is forward biased and the resonance occurs (Fig. 4). The capacitor voltage and the inductor current are expressed by: vðtÞ ¼ Vo þ Zn Ii sinur t
(2)
iðtÞ ¼ Ii ð1 cosur tÞ
(3)
where Zn is the characteristic impedance and ur is the resonant angular frequency.
1 ur ¼ pffiffiffiffiffiffi LC
(4)
PV Panel Ii=Ipv
Li
L
+ Ci
D2
M
D1 C
+ v -
-
(a)
D2
+
i
Vi=VPV
L
Io Ii Co
Ro Vo
M
C
Vo
-
(b)
Fig. 1 e (a) The proposed MPPT with quasi resonant boost converter, (b) the simplified circuit. Please cite this article in press as: Oncu S, Nacar S, Soft switching maximum power point tracker with resonant switch in PV system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.088
3
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
L Ii M
L
D2
i + Cv -
M
Vo
Fig. 5 e Equivalent circuit for Mode III.
Fig. 3 e Equivalent circuit for Mode I.
Zn ¼
i + C v -
Ii
Vo
D2
rffiffiffiffi L C
(5)
This mode ends when the capacitor voltage drops to zero, and it is clamped to this value by the diode D1. The reverse current due to the resonant inductor flows through the antiparallel diode D1. Hence, the zero voltage switching can be accomplished. Mode III (t2 t < t3): In this mode, the resonant inductor discharges through D2 as shown in Fig. 5; the inductor current decreases linearly. Hence, this mode can be considered as the inductor discharging stage. This mode ends when the inductor current reaches zero. During this stage, the capacitor voltage is kept at zero. Mode IV (t3 t < t4): In order to keep the zero voltage switching conditions, the switch should be turned on after v drops to zero and before the anti-parallel diode current drops to zero. When the switch is turned on (Fig. 6), the switch current rises up to the input current. The input current flows through the switch until it is turned off again and the quasi-square switch current is drawn. By increasing the switching frequency, the output voltage and the voltage conversion value decrease [25].
The DC output voltage of the quasi resonant boost converter is controlled by varying the switching frequency [24]. In order to maintain the soft switching, the switch should be turned on at t2 or after. Additionally, the theoretical waveforms show that, v has a DC component (Vo) and an AC component of ZnIi, which affects t01 and t12. In the designed converter, Ii is proportional to the radiation level and the load. The peak value of v will increase as the load current or the radiation level increase. There is a lower bound of the input current in order not to lose the ZVS conditions for any given output voltage or load condition. Since the PV system's environmental or load conditions are not constant, time intervals for tOFF should be adjustable. The capacitor charging (t01) and the resonance (t12) durations can be derived from Eqs. (1) and (2). Turn off time (tOFF) can be calculated by adding t01 and t12 time intervals. The minimum or maximum needed turn off time for ZVS conditions can be expressed as a function of the resonant parameters, the output voltage and the input current: 1
tOFFmin ¼ C
Vo Iimax
sin þ
Design of the soft switching MPPT
Vo Iimin
Vo Zn Iimax
þp (6)
ur 1
tOFFmax ¼ C
sin þ
Vo Zn Iimin
þp (7)
ur
The MPPT with ZVS boost converter is designed for a 40 W PV panel (VMPP ¼ 17.35 V, IMPP ¼ 2.3 A). Based on the solar radiation, upper and lower operation limits of the converter are defined. Radiation levels of 1000 W/m2 and 500 W/m2 are accepted as reference in the design criteria. Experimentally, the maximum power point currents for 1000 W/m2 and 500 W/ m2 radiations are measured as 2.3 A and 1.16 A respectively. The output voltage of the MPPT is chosen to be of 28 V for charging a 24 V battery. The switching frequency range is defined from 25 kHz to 40 kHz.
Fig. 7 shows the electrical scheme and the flowchart of the proposed MPPT system for achieving ZVS operation. The PV current and voltage are measured by using analog to digital converter (ADC) channels of the microcontroller. Perturb and Observe (P&O) algorithm is used to observe the power variation of the PV. According to the flowchart, MPPT starts to work with initial switching frequency (fsmin); determines the output power of the PV panel and increases the switching frequency. Output power is calculated again for the new switching frequency. Calculated powers are compared and
L
L
Ii M
i + C v -
D2 Vo
Fig. 4 e Equivalent circuit for Mode II.
Ii M
i + C v -
D2 Vo
Fig. 6 e Equivalent circuit for Mode IV.
Please cite this article in press as: Oncu S, Nacar S, Soft switching maximum power point tracker with resonant switch in PV system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.088
4
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
IPV
PV Panel
+
Li
L
Ii
i
D2 + +
Ci
VPV
MOSFET Driver
Co
M
D1
-
Ro V o
C v -
LCD
ADC1 P&O Algorithm ADC2 dsPIC30F2020
(a) fs = fsmin P(k) = V(k) x I(k) fs = fs++ P(k+1) = V(k+1) x I(k+1)
P(k+1) > P(k) fs = fs++
fs = fs--
update toff V(k) = V(k+1) I(k) = I(k+1) P(k) = P(k+1)
(b) Fig. 7 e Proposed system (a) electrical scheme, (b) flow chart of the P&O algorithm.
both the switching frequency and tOFF time is updated. Boundary conditions of tOFF are defined by using Eqs. (6) and (7).
Simulation and experimental results In the experimental studies, an IRFP260 power MOSFET, a 61 mH inductor, and a 104 nF capacitor are used as the resonant switch. The soft switching converter is implemented with a 40 W PV panel, a 2200 mF input capacitor, a 700 mH input inductor, and a 1000 mF output capacitor. A STTH1502 is used
as output diode. The power circuit is loaded with a resistive load, whose value can be adjusted from 20 U to 40 U. Fig. 8 shows the experimental set up of the ZVS boost converter. Experimental set up consists of the soft switching boost converter (ZVS MPPT) with the control circuit and the variable resistor group (Load). The system performance is tested by using an adjustable artificial lighting system with halogen lamps. The control circuit is built with a dsPIC30F2020. The Perturb & Observe method is adopted for the MPPT algorithm. Microcontroller changes the switching frequency to track the MPP and decides the suitable tOFF time.
Please cite this article in press as: Oncu S, Nacar S, Soft switching maximum power point tracker with resonant switch in PV system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.088
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
Fig. 8 e The ZVS MPPT.
Fig. 9 e Change of the capacitor voltage for Mode II depending on the input current.
Fig. 9 shows the change in the capacitor voltage based on the designed parameters depending on the input current. According to calculated results in Fig. 9, for the defined output voltage the peak capacitor voltage decreases as the input current decreases; and ZVS may be lost for lower input currents (0.8 A), which mean lower output currents. Therefore, MPPT should be designed to have a larger AC component than
5
the DC component to work with soft switching for any output voltage or load. As stated before; the capacitor charging stage (t01) and the resonance period (t12) affect the turn off time. This situation is clearly seen in Fig. 9. In proposed study, the maximum and the minimum turn off times are calculated as 14.17 ms and 10.51 ms respectively. As seen from Fig. 9, AC component of the capacitor voltage oscillates on the output voltage (Vo). So, for larger output voltages and currents, peak value of the capacitor voltage will increase. High voltage power switch will be needed. Hence, this converter is more suitable for low voltage low power PV systems. PV based traffic light systems, street lighting, hydrogen generation and battery chargers are some of the application areas of the ZVS MPPT converter. The PSIM simulation results of the soft switched MPPT is shown in Fig. 10. ZVS is achieved in a wide range. The ZVS boost converter with the MPPT algorithm is experimentally tested for various load and radiation levels. Test conditions for the load and the artificial lighting system are set up before experimental studies. The capacitor voltage and Vgs are given in Fig. 11 for 500 W/m2 and 1000 W/m2 (Ro ¼ 40 U). The simulation result properly matches the experimental result and the power MOSFET is switched ON and OFF with zero voltage. Fig. 12 shows the PV current, voltage and power waveforms for same conditions. Maximum possible power can be transferred from the PV panel for both conditions. To show the validity of the MPPT algorithm, the soft switching converter is experimentally tested for numerous load and radiation levels. Some of the test results are shown in Fig. 13. In Fig. 13(a) the converter starts to work with 20 U load at 800 W/m2; then the load is adjusted up to 40 U step by step. The maximum possible PV power (approximately 32 W) is caught up with at the beginning of the work. The generated power does not change with the variance in the load. Performance of the converter is tested the for different radiations levels. The converter works in different radiation levels (Ro ¼ 20 U) and tracks the maximum power with soft switching. Fig. 13(b) shows the measured results for four different radiation levels.
Fig. 10 e Soft switching simulation results. Please cite this article in press as: Oncu S, Nacar S, Soft switching maximum power point tracker with resonant switch in PV system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.088
6
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
Fig. 11 e Experimental results for different radiation levels.
Fig. 12 e Experimental results for PV current, voltage and power.
Finally, the proposed system is compared with PWM controlled MPPT. Both converter are tested at 35 kHz switching frequency at 1000 W/m2 (Ro ¼ 20 U). Efficiency of the proposed MPPT is measured as 0.911 while the PWM controlled one is 0.898. The switching waveforms for both converters are shown in Fig. 14. Maximum switching loss in PWM controlled converter is 18 W, but it is 3 W in soft switched MPPT. Comparison results
show that, ZVS conditions compress the switching losses. The peak power of the switching loss is reduced significantly.
Conclusion In this study, the boost converter based MPPT is implemented with the resonant switch by applying Perturb & Observe
Fig. 13 e PV current, voltage and power. Please cite this article in press as: Oncu S, Nacar S, Soft switching maximum power point tracker with resonant switch in PV system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.088
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
7
Fig. 14 e The switch voltage, current and power waveforms for proposed and PWM controlled MPPT.
method with dsPIC30F2020. Performance of the single switch converter is experimentally tested by changing the load and radiation step by step. The proposed system works with ZVS and tracks the MPP of the PV system in different load and radiation levels. In conventional PWM converters MPP is adjusted with the modulation of the duty ratio. In the designed topology, MPP is tracked by adjusting the switching frequency. 40 W prototype is tested in various conditions and compared with PSIM simulation results; soft switching waveforms, IeV and power measurements of the panel are given. The designed converter tracks the maximum possible PV power with low switching losses. Soft switching conditions are achieved without using any snubber or auxiliary switching cell. The soft switched MPPT is compared with conventional PWM MPPT at 35 kHz. The switching waveforms show that the designed system has better switching characteristics than PWM controlled converter and, the proposed design decreases the peak switching power losses, when compared to conventional design. Proposed topology is suitable for low voltage PV generation systems like as the battery chargers, the hydrogen generation and the traffic lighting systems.
references
[1] Won CY, Kim DH, Kim SC, Kim WS, Kim HS. A new maximum power point tracker of photovoltaic arrays using fuzzy controller. In: IEEE Power Electronics Specialists Conference; 1994. p. 396e403. [2] Scamman D, Bustamante H, Hallett S, Newborough M. Offgrid solar-hydrogen generation by passive electrolysis. Int J Hydrogen Energy December 2014;39(35):19855e68. [3] Kulaksız AA, Akkaya R. Training data optimization for ANNs using genetic algorithms to enhance MPPT efficiency of a stand-alone PV system. Turk J Electr Eng Comput Sci 2012;20(2):241e54. [4] Vishwas K, Suryanarayana K, Renukappa NM, Prabhu LV. Modeling of multiphase boost converter for solar battery charging system. In: IEEE Students' Conference on Electrical Electronics and Computer Science; 2014. p. 1e6. [5]] Kivrak S. Design of a low cost sun tracking controller system for photovoltaic panels. J Renew Sustain Energy May 2013;5(3):1e9.
[6] Bouzelata Y, Kurt E, Chenni R, Altın N. Design and simulation of a unified power quality conditioner fed by solar energy. Int J Hydrogen Energy November 2015;40(44):15267e77. [7] Xiao W, Ozog N, Dunford WG. Topology study of photovoltaic interface for maximum power point tracking. IEEE transactions on industrial electronics June 2007;54(3):1696e704. [8] Glasner I, Appelbaum J. Advantage of boost vs. buck topology for maximum power point tracker in photovoltaic systems. In: Nineteenth Convention of Electrical and Electronics Engineers in Israel; 1996. p. 355e8. [9] Mummadi V. Maximum power point tracking algorithm for non-linear DC sources. In: IEEE Region 10 Colloquium and the Third International Conference on Industrial and Information Systems; 2008. p. 1e6. [10] Siwakoti YP, Chhetri BB, Adhikary B, Bista D. Microcontroller based intelligent DC/DC converter to track maximum power point for solar photovoltaic module. In: Innovative Technologies for an Efficient and Reliable Electricity Supply Conference; 2010. p. 94e101. [11] Altin N. Interval type-2 fuzzy logic controller based maximum power point tracking in photovoltaic systems. Adv Electr Comput Eng August 2013;13(3):65e70. [12] Moschopoulos G, Jain P, Joos G. A novel zero-voltage switched PWM boost converter. In: Power Electronics Specialists Conference; 1995. p. 694e700. [13] Long DZ, Gegner JP, Lee CQ. PWM boost converter with zerovoltage-switching. In: Man and Cybernetics Conference; 1992. p. 441e6. [14] Mc Atee CJ, Agrawal JP, Moghbelli HA. 100 watt power supply using ZVS boost converter. In: Applied Power Electronics Conference and Exposition; 1995. p. 496e502. [15] Bal S, Anurag A, Babu BC. An improved soft switching DC-DC converter for low power PV applications. In: International Symposium on Electronic System Design; 2012. p. 276e80. [16] Song IB, Jung DY, Ji YH, Choi SC, Jung YC, Won CY. A soft switching boost converter using an auxiliary resonant circuit for a PV system. In: IEEE 8th International Conference Power Electronics and ECCE; 2011. p. 2838e43. [17] Lee YS, Chou ZH, Huang SS, Huang SY. Quadratic boost converter with switched capacitor and coupled inductor for PV system applications. In: IEEE 10th International Conference on Power Electronics and Drive Systems; 2013. p. 38e43. [18] Simon S, Das J. PV powered soft switched boost converter using MPPT control. In: Magnetic, Machines and Driver, 2014 Annual International Conference; July 2014. p. 1e6. [19] Han DW, Lee HJ, Shin SC, Kim JG, Jung YC, Won CY. A new soft switching ZVT boost converter using auxiliary resonant
Please cite this article in press as: Oncu S, Nacar S, Soft switching maximum power point tracker with resonant switch in PV system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.088
8
[20]
[21]
[22]
[23]
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
circuit. In: Vehicle Power and Propulsion Conference; 2012. p. 1250e5. Park SH, Cha GR, Jung YC, Won CY. Design and application for PV generation system using a soft-switching boost converter with SARC. Ind Electron February 2010;57(2):515e22. Amala JN, Rajan SE, Vengatesh RP. Design and analysis of high frequency soft-switching boost converter employing electronic PI-controller. In: International Conference on Emerging Trends in Electrical and Computer Technology; 2011. p. 392e7. Bo C, Congling W, Yao H. Research on soft switching boost converter. In: Second International Conference on Digital Manufacturing and Automation; 2011. p. 1015e8. Taufik PL, Anwari M. Digitally controlled ZVS quasi-resonant boost converter with M-type switch. In: Intelligent and Advanced Systems Conference; 2007. p. 823e8.
[24] Kazimierczuk MK. Design-oriented analysis of boost zerovoltage-switching resonant DC/DC converter. IEEE Trans Power Electron April 1988;3(2):126e36. [25] Liu KH, Lee FCY. Zero-voltage switching technique in DC/DC converters. IEEE Trans Power Electron July 1990;5(3):293e304. [26] Li Q, Wolfs P. An analysis of the ZVS two-inductor boost converter under variable frequency operation. IEEE Trans Power Electron January 2007;22(1):120e31. [27] Chuang YC, Ke YL, Chuang HS, Wang YS. A novel singleswitch resonant power converter for renewable energy generation applications. IEEE Trans Ind Appl March/April 2014;50(2):1322e30. [28] Soumiah S, Vengatesh RP, Rajan SE. Performance evaluation of single switch high frequency resonant power converter for alternative energy sources. In: 2015 International Conference on Circuit, Power and Computing Technologies; 2015. p. 1e10.
Please cite this article in press as: Oncu S, Nacar S, Soft switching maximum power point tracker with resonant switch in PV system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.088
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
Soft switching maximum power point tracker with resonant switch in PV system* Selim Oncu a,*, Salih Nacar b a b
Karabuk University Engineering Faculty, Karabuk, 78050, Turkey Kastamonu University Taskopru Vocational High School, Kastamonu, 37400, Turkey
article info
abstract
Article history:
Conventional boost converters are used in many applications such as power factor
Received 11 November 2015
correction, electronic ballasts, battery chargers, photovoltaic (PV) applications. PV systems
Received in revised form
need maximum power point tracker (MPPT) to have efficient power conversion. Maximum
13 January 2016
power point trackers are usually implemented by pulse width modulation (PWM)
Accepted 14 January 2016
controlled converters. In PWM control, the switching losses and the voltageecurrent
Available online xxx
stresses of the device increase. In the present paper, an MPPT with soft switching boost converter is designed, simulated and experimentally tested. The topology adds a resonant
Keywords:
network to the conventional boost converter and controls the PV power by changing the
Boost converter
switching frequency. In the proposed converter, Perturb & Observe method is used as
MPPT
power tracking algorithm that is implemented with dsPIC30F2020 and zero voltage
Resonant switch
switching (ZVS) is achieved in wide range. The variable frequency controlled MPPT is tested
Zero voltage switching
for different radiation levels and loads and switching characteristic is compared with PWM
Frequency control
controlled boost converter. The proposed single switch resonant converter is simulated in PSIM and experimental validation is also given for a 40 W PV generation system. Copyright © 2016, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
Introduction Over the last years, for improving the efficiency and the power density of renewable energy sources like photovoltaic (PV) energy conversion systems, the interest in power electronics applications has increased. The output power of the PV panel is related to the intensity of solar radiation, the temperature, and the load [1e5]. PV systems work in different environmental and electrical conditions. For this reason, PV generators are connected to loads by dcedc converters called MPPT. The MPPT changes the operation point of the PV system to find
the maximum possible power in any case [6]. Generally PWM type boost converters are used for transferring the maximum output power of the panel. The advantages of the boost converters in such applications are the low input current ripple, and the easy control [1,7]. Furthermore, in the continuous current mode, the boost converter has better characteristics than the buck converter [8e11]. For increasing the power density of the converter, it is necessary to increase the switching frequency, and in this way the size and weight of the converter decrease too. However, operating at high switching frequency can produce several problems, such as increased switching losses and
*
¨ -BAP-13/1-YL-034. This work was supported in part by the Karabuk University BAP, KBU * Corresponding author. E-mail address: [email protected] (S. Oncu). http://dx.doi.org/10.1016/j.ijhydene.2016.01.088 0360-3199/Copyright © 2016, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Oncu S, Nacar S, Soft switching maximum power point tracker with resonant switch in PV system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.088
2
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
vgs
stress [12]. For preventing such problems, a snubber circuit or a soft switching cell in the PWM converter [13,14] can be used. For decreasing the switching losses in PV applications, some researchers tried to use soft switching PWM converters [15e18]. But PWM converters with an auxiliary soft switching circuit are more complex and expensive since they need an additional power switch, a complex control algorithm, and a floating gate driver [14,17e22]. By using a resonant switch, the mentioned drawbacks of the PWM converters may be avoided [23e28]. In this study, for tracking the MPP a quasi-resonant boost converter is designed and experimentally tested; soft switching is achieved by help of a resonant switch. The ZVS of the power device is achieved by using an LC resonant circuit. Parasitic inductances and capacitances can be incorporated into this resonant circuit. No snubber circuit is used. In addition, neither auxiliary switch nor floating gate drive is required. Perturb and observe MPPT algorithm is applied with dsPIC30F2020 and implemented on the single switch resonant converter. MPP of the 40 W PV panel is tracked by changing both switching frequency and the duty cycle. Zero voltage switching operation is maintained while tracking the MPP. Soft switching MPPT is compared with PWM controlled MPPT at 35 kHz switching frequency. The proposed system has good start-up characteristics with low switching losses and, it tracks the maximum power value in different conditions by variable frequency operation.
v ZnIi
Vo i Ii t0 t1
t4
t2 t3 Ts
Fig. 2 e Theoretical waveforms of ZVS boost converter.
Mode I (t0 t < t1): When the switch is turned off, the input current flows through the resonant capacitor. Fig. 3 shows the equivalent circuit for Mode I. This mode continues until the capacitor voltage reaches Vo. During this stage, D2 is open circuit and the resonant capacitor is charged from zero to Vo. The capacitor voltage rises linearly, so its voltage can be derived as:
ZVS boost converter for PV systems
vðtÞ ¼
The soft switching boost converter with a resonant switch is shown in Fig. 1(a). The converter consists of a PV Panel, a power switch (M) with its anti-parallel diode (D1), a choke inductor and filter capacitors (Li and Ci, Co), a resonant circuit (L and C), a rectifier diode (D2), and a load (Ro). In steady state, the input inductor and the input voltage source of the converter can be represented by a current source and output circuit can be replaced by a voltage sink [25,27]. Fig. 1(b) shows the equivalent circuit of the resonant boost converter. The resonant circuit shapes the voltage of the switch from a square wave to a sinusoidal wave. Fig. 2 shows the switching waveforms of the quasi-resonant boost converter [25]. A short description of the functioning of the proposed MPPT is as follows. Assuming the input current and output voltage are constant and the converter is in steady state, the converter has four operation modes [25].
Ii t C
(1)
Mode II (t1 t < t2): When the capacitor voltage reaches Vo, the diode D2 is forward biased and the resonance occurs (Fig. 4). The capacitor voltage and the inductor current are expressed by: vðtÞ ¼ Vo þ Zn Ii sinur t
(2)
iðtÞ ¼ Ii ð1 cosur tÞ
(3)
where Zn is the characteristic impedance and ur is the resonant angular frequency.
1 ur ¼ pffiffiffiffiffiffi LC
(4)
PV Panel Ii=Ipv
Li
L
+ Ci
D2
M
D1 C
+ v -
-
(a)
D2
+
i
Vi=VPV
L
Io Ii Co
Ro Vo
M
C
Vo
-
(b)
Fig. 1 e (a) The proposed MPPT with quasi resonant boost converter, (b) the simplified circuit. Please cite this article in press as: Oncu S, Nacar S, Soft switching maximum power point tracker with resonant switch in PV system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.088
3
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
L Ii M
L
D2
i + Cv -
M
Vo
Fig. 5 e Equivalent circuit for Mode III.
Fig. 3 e Equivalent circuit for Mode I.
Zn ¼
i + C v -
Ii
Vo
D2
rffiffiffiffi L C
(5)
This mode ends when the capacitor voltage drops to zero, and it is clamped to this value by the diode D1. The reverse current due to the resonant inductor flows through the antiparallel diode D1. Hence, the zero voltage switching can be accomplished. Mode III (t2 t < t3): In this mode, the resonant inductor discharges through D2 as shown in Fig. 5; the inductor current decreases linearly. Hence, this mode can be considered as the inductor discharging stage. This mode ends when the inductor current reaches zero. During this stage, the capacitor voltage is kept at zero. Mode IV (t3 t < t4): In order to keep the zero voltage switching conditions, the switch should be turned on after v drops to zero and before the anti-parallel diode current drops to zero. When the switch is turned on (Fig. 6), the switch current rises up to the input current. The input current flows through the switch until it is turned off again and the quasi-square switch current is drawn. By increasing the switching frequency, the output voltage and the voltage conversion value decrease [25].
The DC output voltage of the quasi resonant boost converter is controlled by varying the switching frequency [24]. In order to maintain the soft switching, the switch should be turned on at t2 or after. Additionally, the theoretical waveforms show that, v has a DC component (Vo) and an AC component of ZnIi, which affects t01 and t12. In the designed converter, Ii is proportional to the radiation level and the load. The peak value of v will increase as the load current or the radiation level increase. There is a lower bound of the input current in order not to lose the ZVS conditions for any given output voltage or load condition. Since the PV system's environmental or load conditions are not constant, time intervals for tOFF should be adjustable. The capacitor charging (t01) and the resonance (t12) durations can be derived from Eqs. (1) and (2). Turn off time (tOFF) can be calculated by adding t01 and t12 time intervals. The minimum or maximum needed turn off time for ZVS conditions can be expressed as a function of the resonant parameters, the output voltage and the input current: 1
tOFFmin ¼ C
Vo Iimax
sin þ
Design of the soft switching MPPT
Vo Iimin
Vo Zn Iimax
þp (6)
ur 1
tOFFmax ¼ C
sin þ
Vo Zn Iimin
þp (7)
ur
The MPPT with ZVS boost converter is designed for a 40 W PV panel (VMPP ¼ 17.35 V, IMPP ¼ 2.3 A). Based on the solar radiation, upper and lower operation limits of the converter are defined. Radiation levels of 1000 W/m2 and 500 W/m2 are accepted as reference in the design criteria. Experimentally, the maximum power point currents for 1000 W/m2 and 500 W/ m2 radiations are measured as 2.3 A and 1.16 A respectively. The output voltage of the MPPT is chosen to be of 28 V for charging a 24 V battery. The switching frequency range is defined from 25 kHz to 40 kHz.
Fig. 7 shows the electrical scheme and the flowchart of the proposed MPPT system for achieving ZVS operation. The PV current and voltage are measured by using analog to digital converter (ADC) channels of the microcontroller. Perturb and Observe (P&O) algorithm is used to observe the power variation of the PV. According to the flowchart, MPPT starts to work with initial switching frequency (fsmin); determines the output power of the PV panel and increases the switching frequency. Output power is calculated again for the new switching frequency. Calculated powers are compared and
L
L
Ii M
i + C v -
D2 Vo
Fig. 4 e Equivalent circuit for Mode II.
Ii M
i + C v -
D2 Vo
Fig. 6 e Equivalent circuit for Mode IV.
Please cite this article in press as: Oncu S, Nacar S, Soft switching maximum power point tracker with resonant switch in PV system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.088
4
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
IPV
PV Panel
+
Li
L
Ii
i
D2 + +
Ci
VPV
MOSFET Driver
Co
M
D1
-
Ro V o
C v -
LCD
ADC1 P&O Algorithm ADC2 dsPIC30F2020
(a) fs = fsmin P(k) = V(k) x I(k) fs = fs++ P(k+1) = V(k+1) x I(k+1)
P(k+1) > P(k) fs = fs++
fs = fs--
update toff V(k) = V(k+1) I(k) = I(k+1) P(k) = P(k+1)
(b) Fig. 7 e Proposed system (a) electrical scheme, (b) flow chart of the P&O algorithm.
both the switching frequency and tOFF time is updated. Boundary conditions of tOFF are defined by using Eqs. (6) and (7).
Simulation and experimental results In the experimental studies, an IRFP260 power MOSFET, a 61 mH inductor, and a 104 nF capacitor are used as the resonant switch. The soft switching converter is implemented with a 40 W PV panel, a 2200 mF input capacitor, a 700 mH input inductor, and a 1000 mF output capacitor. A STTH1502 is used
as output diode. The power circuit is loaded with a resistive load, whose value can be adjusted from 20 U to 40 U. Fig. 8 shows the experimental set up of the ZVS boost converter. Experimental set up consists of the soft switching boost converter (ZVS MPPT) with the control circuit and the variable resistor group (Load). The system performance is tested by using an adjustable artificial lighting system with halogen lamps. The control circuit is built with a dsPIC30F2020. The Perturb & Observe method is adopted for the MPPT algorithm. Microcontroller changes the switching frequency to track the MPP and decides the suitable tOFF time.
Please cite this article in press as: Oncu S, Nacar S, Soft switching maximum power point tracker with resonant switch in PV system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.088
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
Fig. 8 e The ZVS MPPT.
Fig. 9 e Change of the capacitor voltage for Mode II depending on the input current.
Fig. 9 shows the change in the capacitor voltage based on the designed parameters depending on the input current. According to calculated results in Fig. 9, for the defined output voltage the peak capacitor voltage decreases as the input current decreases; and ZVS may be lost for lower input currents (0.8 A), which mean lower output currents. Therefore, MPPT should be designed to have a larger AC component than
5
the DC component to work with soft switching for any output voltage or load. As stated before; the capacitor charging stage (t01) and the resonance period (t12) affect the turn off time. This situation is clearly seen in Fig. 9. In proposed study, the maximum and the minimum turn off times are calculated as 14.17 ms and 10.51 ms respectively. As seen from Fig. 9, AC component of the capacitor voltage oscillates on the output voltage (Vo). So, for larger output voltages and currents, peak value of the capacitor voltage will increase. High voltage power switch will be needed. Hence, this converter is more suitable for low voltage low power PV systems. PV based traffic light systems, street lighting, hydrogen generation and battery chargers are some of the application areas of the ZVS MPPT converter. The PSIM simulation results of the soft switched MPPT is shown in Fig. 10. ZVS is achieved in a wide range. The ZVS boost converter with the MPPT algorithm is experimentally tested for various load and radiation levels. Test conditions for the load and the artificial lighting system are set up before experimental studies. The capacitor voltage and Vgs are given in Fig. 11 for 500 W/m2 and 1000 W/m2 (Ro ¼ 40 U). The simulation result properly matches the experimental result and the power MOSFET is switched ON and OFF with zero voltage. Fig. 12 shows the PV current, voltage and power waveforms for same conditions. Maximum possible power can be transferred from the PV panel for both conditions. To show the validity of the MPPT algorithm, the soft switching converter is experimentally tested for numerous load and radiation levels. Some of the test results are shown in Fig. 13. In Fig. 13(a) the converter starts to work with 20 U load at 800 W/m2; then the load is adjusted up to 40 U step by step. The maximum possible PV power (approximately 32 W) is caught up with at the beginning of the work. The generated power does not change with the variance in the load. Performance of the converter is tested the for different radiations levels. The converter works in different radiation levels (Ro ¼ 20 U) and tracks the maximum power with soft switching. Fig. 13(b) shows the measured results for four different radiation levels.
Fig. 10 e Soft switching simulation results. Please cite this article in press as: Oncu S, Nacar S, Soft switching maximum power point tracker with resonant switch in PV system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.088
6
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
Fig. 11 e Experimental results for different radiation levels.
Fig. 12 e Experimental results for PV current, voltage and power.
Finally, the proposed system is compared with PWM controlled MPPT. Both converter are tested at 35 kHz switching frequency at 1000 W/m2 (Ro ¼ 20 U). Efficiency of the proposed MPPT is measured as 0.911 while the PWM controlled one is 0.898. The switching waveforms for both converters are shown in Fig. 14. Maximum switching loss in PWM controlled converter is 18 W, but it is 3 W in soft switched MPPT. Comparison results
show that, ZVS conditions compress the switching losses. The peak power of the switching loss is reduced significantly.
Conclusion In this study, the boost converter based MPPT is implemented with the resonant switch by applying Perturb & Observe
Fig. 13 e PV current, voltage and power. Please cite this article in press as: Oncu S, Nacar S, Soft switching maximum power point tracker with resonant switch in PV system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.088
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
7
Fig. 14 e The switch voltage, current and power waveforms for proposed and PWM controlled MPPT.
method with dsPIC30F2020. Performance of the single switch converter is experimentally tested by changing the load and radiation step by step. The proposed system works with ZVS and tracks the MPP of the PV system in different load and radiation levels. In conventional PWM converters MPP is adjusted with the modulation of the duty ratio. In the designed topology, MPP is tracked by adjusting the switching frequency. 40 W prototype is tested in various conditions and compared with PSIM simulation results; soft switching waveforms, IeV and power measurements of the panel are given. The designed converter tracks the maximum possible PV power with low switching losses. Soft switching conditions are achieved without using any snubber or auxiliary switching cell. The soft switched MPPT is compared with conventional PWM MPPT at 35 kHz. The switching waveforms show that the designed system has better switching characteristics than PWM controlled converter and, the proposed design decreases the peak switching power losses, when compared to conventional design. Proposed topology is suitable for low voltage PV generation systems like as the battery chargers, the hydrogen generation and the traffic lighting systems.
references
[1] Won CY, Kim DH, Kim SC, Kim WS, Kim HS. A new maximum power point tracker of photovoltaic arrays using fuzzy controller. In: IEEE Power Electronics Specialists Conference; 1994. p. 396e403. [2] Scamman D, Bustamante H, Hallett S, Newborough M. Offgrid solar-hydrogen generation by passive electrolysis. Int J Hydrogen Energy December 2014;39(35):19855e68. [3] Kulaksız AA, Akkaya R. Training data optimization for ANNs using genetic algorithms to enhance MPPT efficiency of a stand-alone PV system. Turk J Electr Eng Comput Sci 2012;20(2):241e54. [4] Vishwas K, Suryanarayana K, Renukappa NM, Prabhu LV. Modeling of multiphase boost converter for solar battery charging system. In: IEEE Students' Conference on Electrical Electronics and Computer Science; 2014. p. 1e6. [5]] Kivrak S. Design of a low cost sun tracking controller system for photovoltaic panels. J Renew Sustain Energy May 2013;5(3):1e9.
[6] Bouzelata Y, Kurt E, Chenni R, Altın N. Design and simulation of a unified power quality conditioner fed by solar energy. Int J Hydrogen Energy November 2015;40(44):15267e77. [7] Xiao W, Ozog N, Dunford WG. Topology study of photovoltaic interface for maximum power point tracking. IEEE transactions on industrial electronics June 2007;54(3):1696e704. [8] Glasner I, Appelbaum J. Advantage of boost vs. buck topology for maximum power point tracker in photovoltaic systems. In: Nineteenth Convention of Electrical and Electronics Engineers in Israel; 1996. p. 355e8. [9] Mummadi V. Maximum power point tracking algorithm for non-linear DC sources. In: IEEE Region 10 Colloquium and the Third International Conference on Industrial and Information Systems; 2008. p. 1e6. [10] Siwakoti YP, Chhetri BB, Adhikary B, Bista D. Microcontroller based intelligent DC/DC converter to track maximum power point for solar photovoltaic module. In: Innovative Technologies for an Efficient and Reliable Electricity Supply Conference; 2010. p. 94e101. [11] Altin N. Interval type-2 fuzzy logic controller based maximum power point tracking in photovoltaic systems. Adv Electr Comput Eng August 2013;13(3):65e70. [12] Moschopoulos G, Jain P, Joos G. A novel zero-voltage switched PWM boost converter. In: Power Electronics Specialists Conference; 1995. p. 694e700. [13] Long DZ, Gegner JP, Lee CQ. PWM boost converter with zerovoltage-switching. In: Man and Cybernetics Conference; 1992. p. 441e6. [14] Mc Atee CJ, Agrawal JP, Moghbelli HA. 100 watt power supply using ZVS boost converter. In: Applied Power Electronics Conference and Exposition; 1995. p. 496e502. [15] Bal S, Anurag A, Babu BC. An improved soft switching DC-DC converter for low power PV applications. In: International Symposium on Electronic System Design; 2012. p. 276e80. [16] Song IB, Jung DY, Ji YH, Choi SC, Jung YC, Won CY. A soft switching boost converter using an auxiliary resonant circuit for a PV system. In: IEEE 8th International Conference Power Electronics and ECCE; 2011. p. 2838e43. [17] Lee YS, Chou ZH, Huang SS, Huang SY. Quadratic boost converter with switched capacitor and coupled inductor for PV system applications. In: IEEE 10th International Conference on Power Electronics and Drive Systems; 2013. p. 38e43. [18] Simon S, Das J. PV powered soft switched boost converter using MPPT control. In: Magnetic, Machines and Driver, 2014 Annual International Conference; July 2014. p. 1e6. [19] Han DW, Lee HJ, Shin SC, Kim JG, Jung YC, Won CY. A new soft switching ZVT boost converter using auxiliary resonant
Please cite this article in press as: Oncu S, Nacar S, Soft switching maximum power point tracker with resonant switch in PV system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.088
8
[20]
[21]
[22]
[23]
i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 6 ) 1 e8
circuit. In: Vehicle Power and Propulsion Conference; 2012. p. 1250e5. Park SH, Cha GR, Jung YC, Won CY. Design and application for PV generation system using a soft-switching boost converter with SARC. Ind Electron February 2010;57(2):515e22. Amala JN, Rajan SE, Vengatesh RP. Design and analysis of high frequency soft-switching boost converter employing electronic PI-controller. In: International Conference on Emerging Trends in Electrical and Computer Technology; 2011. p. 392e7. Bo C, Congling W, Yao H. Research on soft switching boost converter. In: Second International Conference on Digital Manufacturing and Automation; 2011. p. 1015e8. Taufik PL, Anwari M. Digitally controlled ZVS quasi-resonant boost converter with M-type switch. In: Intelligent and Advanced Systems Conference; 2007. p. 823e8.
[24] Kazimierczuk MK. Design-oriented analysis of boost zerovoltage-switching resonant DC/DC converter. IEEE Trans Power Electron April 1988;3(2):126e36. [25] Liu KH, Lee FCY. Zero-voltage switching technique in DC/DC converters. IEEE Trans Power Electron July 1990;5(3):293e304. [26] Li Q, Wolfs P. An analysis of the ZVS two-inductor boost converter under variable frequency operation. IEEE Trans Power Electron January 2007;22(1):120e31. [27] Chuang YC, Ke YL, Chuang HS, Wang YS. A novel singleswitch resonant power converter for renewable energy generation applications. IEEE Trans Ind Appl March/April 2014;50(2):1322e30. [28] Soumiah S, Vengatesh RP, Rajan SE. Performance evaluation of single switch high frequency resonant power converter for alternative energy sources. In: 2015 International Conference on Circuit, Power and Computing Technologies; 2015. p. 1e10.
Please cite this article in press as: Oncu S, Nacar S, Soft switching maximum power point tracker with resonant switch in PV system, International Journal of Hydrogen Energy (2016), http://dx.doi.org/10.1016/j.ijhydene.2016.01.088