A novel hybrid boost converter with extended duty cycles range for tracking the maximum power point in photovoltaic system applications

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 8 ) 1 e1 2

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.elsevier.com/locate/he

A novel hybrid boost converter with extended duty cycles range for tracking the maximum power point in photovoltaic system applications S. Belhimer a, M. Haddadi a, A. Mellit b,c,* a

Laboratoire de Dispositifs de Communication et de Conversion Photovoltaı¨que (LDCCP), Ecole Nationale Polytechnique, 10 Avenue Hassen Badi, El Harrach, 16200, Algiers, Algeria b Renewable Energy Laboratory (LER), Jijel University, Jijel, 18000, Algeria c The Abdus Salam International Centre for Theoretical Physics (ICTP), Starada Costiera, 11-34151, Trieste, Italy

article info

abstract

Article history:

Nowadays, a large number of power conversion applications is commonly based on DC/DC

Received 10 January 2018

converters with high voltage boost capability. Different voltage-boosting techniques have

Received in revised form

been reported in the literature. Each technique has its own merits and demerits depending

13 February 2018

on the application, cost, complexity, power density, reliability and efficiency. To meet the

Accepted 20 February 2018

growing demand for such applications, new power converter topologies are continuously

Available online xxx

being proposed. This paper focuses on a novel hybrid boost converter, which combines the conventional boost (CB) and the quadratic boost (QB). This new topology allows the

Keywords:

extension of the output voltage gain and the duty cycle range regarding to the original

Photovoltaic

topologies. Thus, it ensures high conversion voltage ratio for almost duty cycle values.

Hybrid boost

Consequently, it has two working modes, one as QB mode and the other one as CB mode. In

DC/DC converter

order to verify the performance of the proposed topology, several simulations have been

MPPT

carried out under Matlab/Simulink environment for both QB and CB modes. The well-

FPGA

known P&O algorithm was implemented into a FPGA (Field Programmable Gate Array) board in order to verify experimentally the designed hybrid boost. Experimental results confirm the convenience of the proposed topology for tracking the maximum power point in photovoltaic systems. © 2018 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

Introduction Industrial development around the world increased global energy demand in recent decades, along with the problem of energy shortage; this also caused environmental problems, especially from the overuse of fossil energy. This situation has led to the exploitation of renewable energy sources. Among

these renewable energy sources, photovoltaic (PV) energy is a very promising and has been gaining popularity. PV has many advantages: clean, quiet, and maintenance-free [1]. However, it also has some shortcomings related to the efficiency of PV cells, the sun intermittence and the dependence on the atmospheric conditions mainly solar irradiance and air temperature. Usually, PV modules provide large output voltage and current; however its maximum power is delivered for

* Corresponding author. Renewable Energy Laboratory (LER), Jijel University, Jijel, 18000, Algeria. E-mail addresses: [email protected], [email protected] (A. Mellit). https://doi.org/10.1016/j.ijhydene.2018.02.136 0360-3199/© 2018 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved. Please cite this article in press as: Belhimer S, et al., A novel hybrid boost converter with extended duty cycles range for tracking the maximum power point in photovoltaic system applications, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/ j.ijhydene.2018.02.136

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 8 ) 1 e1 2

only one particular value of current and voltage. The IeV and PeV characteristics of PV modules depend mainly on three parameters: solar irradiance, temperature and the cells aging [2]. To track the maximum power point (MPP) at any time, a DC/DC converter is often set between the PV modules and the load, as shown in Fig. 1, in order to ensure an impedance matching between the load impedance and the output DC/DC converter impedance at the MPP [3]. Global efficiency improvement of PV systems is possible by adding Maximum Power Point Tracking (MPPT) controllers in association with DC/DC or DC/AC converters [4]. In the last decades, power electronics knew such development which has contributed to the improvements of DC/DC and DC/AC converters. It has played an essential role in the improvement of PV systems. The main purpose of using DC/ DC converters is to increase or reduce the input DC voltage from the PV arrays and to track the MPP. Besides, there are two categories of DC/DC converters: isolated and non-isolated. It depends on the usage of a galvanic isolation between the input and the output by transformers. For the majority investigations, the non-isolated DC/DC converters have been mostly used presenting a suitable solution because of their capabilities to achieve a high level of efficiency and lower cost when compared with isolated ones. Each converter topology has its own modes of operation and its relevant advantages and disadvantages [5]. Usually in conventional PWM converters, the power switching devices should operate ideally at maximum switching frequencies to ensure wider conversion range. This operation consists to provide the lowest or the highest possible duty ratios of the converter; however it is limited by the finite commutation time of the power switching devices. Another approach to increase the conversion range is the use of a step-down or step-up transformer with the corresponding difficulties in switching surges and operating frequencies [6]. A solution to this problem is proposed in Ref. [7], it consists to use n-stages connected in cascade and use only one active switch, thus reducing the huge switching losses at each stage and avoiding complex control circuitry [8e10]. In this paper, considering the low cost implementation and the relatively high performance of boost converters which are

Fig. 1 e MPPT Control system.

suitable for many PV applications [11], a new boost topology, combining two n-stages connected in cascade using one active switch (n2f1; 2g), is performed. This topology has two working modes: one mode as 1-stage which is known by the conventional boost (CB), and the second one as 2-stages which is known by quadratic boost (QB). It benefits from both their advantages and will extend the duty cycle range for higher output voltage gain, because each mode ensures high output voltage for specific interval of duty cycle values: The quadratic boost works better for small duty cycles, which is opposite for the conventional boost. So, the extension of duty ratio range given by the hybrid boost and its ability to provide two different output voltages for the same duty cycle value make this hybrid topology advantageous and very interesting. By the way, this new topology is attractive especially when it is applied for tracking the MPP in PV systems. This hybrid boost DC/DC converter is powered directly by the PV module; it is controlled by a command block, including current/voltage sensors and a Field Programmable Gate Array (FPGA) board where the MPPT algorithm is implemented. The use of FPGA for designing the MPPT controller increases the robustness, provides high performance and makes the hardware implementation more flexible [12e14]. So, after reading the instantaneous PV voltage and current, the corresponding instantaneous power is computed and compared to its previous value. Then according to the result of this comparison, the duty cycle ratio is increased or decreased that means the perturbation and observation (P&O) of the system. By repeating this operation, the command system will track the MPP of the PV array. The P&O MPPT algorithm is used herein for its relatively high performance and accuracy in uniform environmental conditions and mainly for its ease of implementation [13,15e18]. Based on the value of the duty cycle ratio, the FPGA controller generates a PWM command signal which manages directly the on/off states of the main switch of the hybrid boost DC/DC converter. The proposed MPPT hybrid boost has been simulated using Matlab/Simulink, and then verified experimentally. A comparable research was established in Ref. [19], where a quadratic boost converter with MPPT ability for high step-up ratio application was proposed, but with a more advanced MPPT algorithm based on fuzzy logic. The authors have also used Matlab/Simulink to simulate their proposed converter but they have not performed their tests in a real PV system. However, their experimental verifications were conducted using a PV system emulator. Our proposed MPPT hybrid boost has been experimentally tested using a real PV system and the important parameters have been measured in real time and visualized by oscilloscope. Consequently, the hybrid boost presents very good performance in term of reaching the MPP for both working modes (QB and CB). This paper is structured as follows: a brief overview on the MPPT algorithms is highlighted in Section MPPT algorithms. Section The proposed hybrid boost presents a detailed description of the proposed DC/DC converter topology. Simulation procedure followed by a discussion and interpretation of results are given in Section Simulation results. Experimental verification is of the proposed hybrid boost is provided in Section Experimental verification.

Please cite this article in press as: Belhimer S, et al., A novel hybrid boost converter with extended duty cycles range for tracking the maximum power point in photovoltaic system applications, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/ j.ijhydene.2018.02.136

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 8 ) 1 e1 2

MPPT algorithms The sudden variation of the solar irradiance (G) induces the fluctuation of the MPP. Fig. 2 represents the PeV characteristics of a PV module for different solar irradiance and temperature variations. It is clear that the PV power provided increases proportionally with insulation and decreases with temperature [20]. The criteria to evaluate MPPT algorithms are known, as mentioned in Ref. [4], by their ability to locate the MPP, their fast tracking of the MPP, also their fast tracking under a change in environmental conditions, in addition to generating minimal oscillations in steady state, and ensuring a minimal complexity and cost. In literature, many papers discussed different kind of MPPT algorithms for solar energy [21e23]. A review of stand-alone PV systems can be found in Refs. [3,24]. A method to reduce the steady state oscillation and to mitigate the probability of losing the tracking direction of the P&O-based MPPT for PV system was proposed in Ref. [25]. The Incremental conductance algorithm was discussed in Ref. [26] using a variable step to track MPP quickly. Moreover, many algorithms based on fuzzy logic and artificial intelligent can be found in Refs. [12,27e29]. An efficient adaptive neuro-fuzzy inference system (ANFIS)-based PI controller for MPPT of PV systems was proposed [30]. Moreover, a novel Sliding Mode Control (SMC) based algorithm was exposed to be implemented in a DC/DC converter in order to make an autonomous photovoltaic system to work at the MPP [31].

Fig. 2 e Environment influence on the P-V characteristic: (a) under temperature variation at G ¼ 1000 W/m2, (b) under irradiance variation at T ¼ 25  C.

3

As mentioned above, the measure of the instantaneous PV voltage and current will be converted to numerical data and stored using FPGA, and then the corresponding instantaneous power is computed and compared to its previous value. After that, and according to the result of comparison, the duty cycle ratio will be increased or decreased by the P&O algorithm, schematized in Fig. 3, which also compares the instantaneous voltage with its previous value in order to know the perturbation direction. The value of the duty cycle ratio is generated for the PWM command signal to control the main switch of the DC/DC converter.

The proposed hybrid boost Circuit description The novel hybrid boost is considered as a step-up non-isolated DC/DC converter topology. It combines the conventional and the quadratic boost in one circuit using the same components as shown in Fig. 4. Therefore, it minimizes the electrical components cost and number by avoiding complex circuitry. It benefits from both their advantages in terms of extending the duty cycle range for higher output voltage gain. Moreover, the electrical circuit of this proposed topology is composed by one main switch (S), two boost inductors (L1) and (L2), three diodes (D), (D1), and (D2), and capacitors (Cin), (Cout) and (C1). It is closely similar to the quadratic boost circuit but it has the (S1), (S2) switches and the (D1) diode as additional components. The placement of these additional components has been chosen empirically in order to protect the main transistor (S) from dangerous peak values when switching from mode to mode. It is controlled via one main switch (S) using a PWM signal as shown in Fig. 5a. In addition, the continuous conduction mode (CCM) is required which means that the current of inductors never falls to zero as shown in Fig. 5b. This mode is known for high efficiency and the good utilization of semiconductor switches and passive components [32]. It is used herein, to ensure the transfer of the maximum energy from PV modules to the load with small losses.

Fig. 3 e Flowchart of the P&O MPPT algorithm.

Please cite this article in press as: Belhimer S, et al., A novel hybrid boost converter with extended duty cycles range for tracking the maximum power point in photovoltaic system applications, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/ j.ijhydene.2018.02.136

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 8 ) 1 e1 2

Vout ¼

T T Vin ¼ Vin toff T  ton

(6)

which gives Vout ¼

1 Vin 1D

(7)

For the quadratic boost, in order to find the relation between the voltage Vout and the first stage output VC1 Fig. 4 e The hybrid boost DC/DC converter.

VC1 ton ¼ ðVout  VC1 Þtoff

(8)

  VC1 ton þ toff ¼ Vout toff

(9)

so, Vout ¼

T T VC1 ¼ VC1 toff T  ton

(10)

1 1 VC1 ¼ Vin 1D ð1  DÞ2

(11)

or, Vout ¼

which gives: Vout ¼

Vin ð1  DÞ2

(12)

So to conclude the gain output voltage equation of an ideal hybrid boost is given by: 8 Vin > > > < Vout ¼ 1  D; for CB Mode Vin > > > ; for QB Mode : Vout ¼ ð1  DÞ2

Fig. 5 e Signal waveform: a) PWM Command signal, b) inductors current waveform.

(13)

where

Formula derivation D¼ For conventional boost, in the periodic switching scheme with period T, the average voltage across the inductor must be zero. The relationship of voltage and current for an inductor is: vL ¼ L

diL dt

(1)

And the inductor current is: iL ðtÞ ¼

1 L

Zt vL ðtÞdt þ iL ð0Þ

(2)

0

where iL(0) is the initial inductor current at time t ¼ 0. The average value of the voltage across the inductor in steady state and under the periodic switching operation is: VLðAVÞ ¼

1 T

ZT vL dt ¼ 0

(3)

0

According to the voltage second product, Vin ton ¼ ðVout  Vin Þtoff

(4)

or Vin T ¼ Vout toff

(5)

tON tON ¼ tON þ tOFF T

(14)

However, the non-idealities or parasitics of practical devices and electronic components affect greatly some performance of DC-DC converter and implies that the real Vout(D) characteristic is different from the ideal one (Eq. (13)). For instance, those non-idealities effects are considered because, in practical experiments, each component has a series resistor: Rs for inductors, an equivalent series resistance (ESR) for capacitors, rD for diodes and the Rds_on for the ON state of the MOSFETs [6].

Principe of operation The equivalent circuit models according to the PWM signals (ON and OFF states) of both QB and CB converter are shown in Fig. 6. It is known that when the main switch is ON, the inductors will be connected directly with the source power, at this moment they will store an electromagnetic energy, in their spires, which in its role will be restituted in the OFF state of the main switch, the diode (D) is closed and the load is supplied by the output capacitor. However, during the OFF state, the main switch can be considered as an open circuit, and at this period the output energy provided to the load is equal to the source power energy plus the stored energy in inductors.

Please cite this article in press as: Belhimer S, et al., A novel hybrid boost converter with extended duty cycles range for tracking the maximum power point in photovoltaic system applications, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/ j.ijhydene.2018.02.136

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 8 ) 1 e1 2

5

Fig. 6 e Hybrid boost modes equivalent circuit during the ON/OFF state of the main switch (S).

Besides, the idea of combining both boost topologies into one hybrid topology was observed from the simulation result of each topology. After that, the comparison between the variations output voltages according to the duty cycle was done as shown in Fig. 7a. It can be concluded that the quadratic boost is more suitable for small duty cycle values, because it can provide a bigger output voltage then the conventional boost, on the other hand the conventional boost is more suitable for big ones. Fig. 7b presents the operation of both converter modes during working mode and transition from mode to mode. The response time of the output voltage signal is very important, especially at the beginning and during working states (transition periods). The output ripples are considered mainly when transition from the two working modes. Two different values of duty cycle (D ¼ 40% and D ¼ 75%) have been taken as example to show both small and big duty cycle range (40% for QB and 75% for CB). This, small rippled output signal means the converter is more stable. Measuring the maximum voltage of limit conditions allow us to take into account the necessary precautions for making the right choices of electronic components, in order to ensure the safety and the stability of converters.

Control technique To summarize, our hybrid converter has two working modes: one as quadratic boost and the other one as conventional boost. Both working states of the hybrid boost are presented in Table 1, showing the state of the switches S1 and S2.

Fig. 7 e Hybrid boost output voltage: a) Variation according to the duty cycle of both modes: QB and CB, b) Transition from mode to mode with two values of duty cycle (D ¼ 40% and D ¼ 75%).

Please cite this article in press as: Belhimer S, et al., A novel hybrid boost converter with extended duty cycles range for tracking the maximum power point in photovoltaic system applications, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/ j.ijhydene.2018.02.136

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 8 ) 1 e1 2

Table 1 e Hybrid boost working states. S1

S2

Hybrid boost working state

0 0 1 1

0 1 0 1

OFF Quadratic boost (QB) Conventional boost (CB) Quadratic boost

When the command board provides small values of duty cycle, it is better for the hybrid boost to work as QB: S2 ¼ ON, S1 ¼ OFF (Section QB mode). But once it exceeds the of intersection value of the output voltage variation (Fig. 7a), it should work as CB: S2 ¼ OFF, S1 ¼ ON (Section CB mode). It can be noted that when both switches (S1) and (S2) are ON, the diodes D1 and D2 become in parallel, and at this situation all QB components are connected, so the hybrid boost works in QB mode. The mode to mode switch (from QB to CB or from CB to QB) can be held manually or automatically according to its application. The automatic switch depends on the intersection point which allow us to know the exact duty cycle boundary value in order to switch from one working mode to another. Each working mode depends on the value of the PWM. This control technique consists on using the QB mode for low duty cycle values and the CB for high values. In order to compare this hybrid boost with other topologies, each mode (CB and QB) should be compared with its analogous topologies. For instance during QB mode, the hybrid boost have to be compared to the conventional quadratic boost or with the cascaded boost topologies. It will have similar efficiency as the conventional quadratic boost, which is usually higher than a cascaded boost using the same

devices. Also, output ripple is better and electromagnetic interference (EMI) is less in QB based on phase shifting of power [19]. In addition, the same power of the cascaded converter will be processed twice (in both boosting stages) causing a lot of power loss. Nonetheless, the control will be more complicated in hybrid boost than conventional and quadratic ones; because the control of both modes switches has to be taken into consideration. However considering the extension advantage of duty ratio range given by the hybrid boost and its ability to provide two different output voltages for the same duty cycle value will distract the drawbacks of control complexity and cost.

Simulation results As mentioned above, the passage of clouds across the sky decreases rapidly the solar irradiance (G) and involves changing in the IeV characteristic of the PV module. So, considering the environment within which PV systems operate is very important. In order to assess the performance of the MPPT algorithms, many simulations was held using MATLAB/Simulink which is known for its ability to simulate such PV systems [26]. The Simulink model is presented in Fig. 8. It consists on a mathematical model of a PV module, a block function where the MPPT algorithm is implemented, a PWM generator (25 KHz frequency) with a duty cycle value provided by the MPPT algorithm, and the hybrid boost (with its both working modes as explained in Section The proposed hybrid boost). Both working modes are tested with the same system configuration; Table 2 presents the simulated parameters.

Fig. 8 e The Simulink block system. Please cite this article in press as: Belhimer S, et al., A novel hybrid boost converter with extended duty cycles range for tracking the maximum power point in photovoltaic system applications, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/ j.ijhydene.2018.02.136

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 8 ) 1 e1 2

Table 2 e Simulated parameters. Parameter PV module

Environment

DC/DC converter

Electrical load Simulation

Open circuit voltage (Voc) Short circuit Current (Isc) Voltage at MPP (Vmpp) Current at MPP (Impp) Maximum Power (Pmax) Irradiance (G)

Temperature (T) Inductances L1 and L2 Capacitors C1,Cin, and Cout MOSFET RDS_ON Switching Frequency (F) Resistor load (RL) Initial duty cycle Simulation time Solar irradiance variation step time

Value 21.1 V 7.6 A 17.1 V 7A 120 W 1000 W/m2 800 W/m2 600 W/m2 25  C 100 mH 220 mF 0.027 Ohm 25 KHz 100 Ohm 40% From 0 s to 0.4 s 0.1 s

Simulation test is performed at standard operating temperature of 25  C and a variable solar irradiance level ðG2½1000; 800; 600W=m2 Þ, to test the fast sunlight transition when clouds or obstacles pass in front of PV modules. Simulation results include the output power generated from the PV array using MPPT controller, output voltage provided by the DC/DC converter (supplying the load), and the duty cycle value at which the MPP is reached. Each mode has been simulated separately, and after that the results are compared and discussed. The analysis of the hybrid boost modes can are reported in the following subsections:

QB mode The quadratic boost mode exposed in Fig. 9 is a 2-stages cascaded boost. This mode is activated once the switch (S2) is ON. It is based on the main switch (S), two boost inductors (L1) and (L2), three diodes (D), (D1), and (D2), and three capacitors (Cin), (Cout) and (C1). The voltage is boosted up through two inductor charge up paths. The first path is through the boost inductor (L1), diode (D1), and active power switch (S). The second charge-up path is through the capacitor (C1), boost inductor (L2), and switch (S). Both charge-up paths charge the capacitor (Cout) across the load [7,11,33]. As explained in Section The proposed hybrid boost, the QB presents better performance when duty cycle values are small.

Fig. 9 e Hybrid boost circuit working in QB mode.

7

According to the simulation configuration, each 0.1 s the solar irradiance decreases immediately to 800 W/m2 and after that to 600 W/m2. This abrupt variation involves the loss of the MPP, so the MPPT algorithm, in its turn, begins to track the new MPP. Therefore, the duty cycle value will be adjusted according to the instantaneous measurements of current and voltage until the new MPP is attained. Fig. 10 shows the simulation results during the QB mode. The simulation work for QB mode shows that the first MPP, for G ¼ 1000 W/m2, is reached in a time of about 28 ms. In this case, the duty cycle oscillates between 57% and 60% and the power delivered by the PV module is about 119 W. The second MPP, when G ¼ 800 W/m2, is obtained after a time of about 5 ms for a range of duty cycle from 51% to 56%. The reached PV power for this irradiance is about 95 W. The output voltage is about 79.5 V. Furthermore, the third MPP, when G falls to 600 W/m2, is reached for a time of 7.2 ms and a range of duty cycle from 48% to 50% for a maximum power of 70 W. The output voltage is about 70.7 V. Finally, the solar irradiance comes back to its initial value of 1000 W/m2. The MPP is tracked again providing the same first results of duty ratio, power and output voltage but in a shorter time of about 10.4 ms. All results are reported in Table 3.

CB mode The conventional boost mode is theoretically considered as 1stage boost [34]. The hybrid boost mode is activated when the switch (S1) is ON and (S2) is OFF as shown in Table 1. It is based on the main switch (S), the boost inductor (L1), two diodes (D), (D1), and capacitors (Cin) and (Cout) as exposed in Fig. 11. The same analysis as QB mode was performed for this CB mode. It shows for G ¼ 1000 W/m2, that the MPP is reached in a time of about 90 ms which is longer than the QB time. In addition, the duty cycle oscillates between 84% and 86% and the power delivered by the PV module is about 120 W. Also for CB mode, when the solar irradiance decreases immediately, the PV system starts to track the new MPP according to the P&O algorithm. Even the CB mode reaches the MPP in keeps oscillating on it for all simulated irradiance steps. The simulation results during this mode are shown in Fig. 12 and recorded also in Table 3. The obtained simulation results are shown for both modes as presented in Fig. 13 and Table 3. It can be seen that for QB mode of the first MPP is reached faster than the CB mode. This can be understood because the QB presents better performance for low duty cycles, and in this system the MPPT scan process starts with an initial duty cycle value of D ¼ 40%. Although after reaching MPP and once the irradiance varies, the CB makes a shorter time to attain the new MPP than the QB. The hybrid boost provides a high gain voltage with good stability. The output voltage ripples can be seen when the zoom is used as done for the output voltage in Fig. 13, the pick to pick ripple value is about 0.6 V for CB and about 0.4 V for QB. As mentioned above, less rippled output signal means the converter is more stable. It seems that the QB is more stable then CB. As can be observed in both figures Figs. 10 and 12, the scanning process of P&O algorithm to reach the first MPP takes

Please cite this article in press as: Belhimer S, et al., A novel hybrid boost converter with extended duty cycles range for tracking the maximum power point in photovoltaic system applications, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/ j.ijhydene.2018.02.136

8

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 8 ) 1 e1 2

Fig. 10 e Quadratic boost mode with P&O MPPT.

Table 3 e MPPT Result parameters for both hybrid boost modes. Result parameters

PV power (W) Output voltage (V) Duty cycle of MPP (%) Time to reach the MPP (ms)

1000 W/m2

800 W/m2

600 W/m2

1000 W/m2

QB

CB

QB

CB

QB

CB

QB

CB

119.6 86.3 57e60 28

119.4 88.8 84e86 48.6

95 79.5 51e56 5

94.5 80.2 82e84 1.5

70 70.7 48e50 7.2

70.45 70.8 75e78 6.3

119.6 86.3 57e60 10.4

119.4 88.8 84e86 5.6

than the CB, but it has been observed that when the solar irradiance condition varies quickly, the CB reaches the new MPP before the QB.

Experimental verification Fig. 11 e Hybrid boost circuit working in CB mode.

about 28 ms for QB and about 48.6 ms for CB. Table 3 presents result parameters that can be concluded from all achieved simulations for both working modes. The case of G ¼ 1000 W/m2 is tested twice at the beginning and at the end of the simulation step. The first one takes a longer time to attain the MPP because of the P&O scanning process which starts from the initial duty cycle value (D ¼ 40%); however the second one starts its scanning process from the last MPP duty cycle value (in this work, the last MPP was for G ¼ 600 W/m2 where D ¼ 49% for QB and D ¼ 75%). It also shows the transition impact on the tracking system when the system comes back to the initial solar irradiance conditions. It can be noted that for the first MPP; the QB is faster

The designed system is divided mainly into two principal parts: the command part working with low power level and the power part usually working with higher value of current and voltage. The Command part aims to control the power block by measuring the different current and voltage variations using a Hall Effect current sensor (LA55-P) and a voltage Transducer (LV25-P). Nevertheless, according to the P&O MPPT algorithm, the main switch ON/OFF states of the power block is controlled. In the experiment phase, the hybrid boost working mode is selected manually by sending command to the switches (S1) and (S2). The automatic working mode selection can be the subject of future research when using more advanced MPPT algorithm or when the hybrid boost output will be connected to an inverter for DC/AC conversion. The P&O MPPT algorithm is implemented into a FPGA board (XC5VLX50-1FFG676 of Virtex5 family). This circuit is built around an ML501 development board. The codes are

Please cite this article in press as: Belhimer S, et al., A novel hybrid boost converter with extended duty cycles range for tracking the maximum power point in photovoltaic system applications, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/ j.ijhydene.2018.02.136

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 8 ) 1 e1 2

9

Fig. 12 e Conventional boost mode with P&O MPPT.

Fig. 13 e Hybrid boost modes with P&O MPPT. written using a description hardware language (VHDL) and are synthesized with ISE 10.1 software of Xilinx. The use of FPGA board for the implementation of MPPT control algorithms offers many advantages. It is confirmed that FPGA offers real hardware implementation of MPPT algorithm. It takes

advantage of hardware parallelism by overtaking the computing performance of digital signal processors (DSPs), Microcontroller and performing more operations per clock cycle. Therefore, these circuits are known by their possibility to implement complex control algorithms with short

Please cite this article in press as: Belhimer S, et al., A novel hybrid boost converter with extended duty cycles range for tracking the maximum power point in photovoltaic system applications, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/ j.ijhydene.2018.02.136

10

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 8 ) 1 e1 2

computing time [35,36]. Also, their speed allows better temporal resolution and improves the performance of MPPT control algorithms [12,37]. The power block is composed by inductors, capacitors, diodes, and semiconductor switches (MOSFET, IGBT). It ensures the energy transfer o from the input to the output. The crucial step to build this power block is the choice of components which can support the power system requirements. Table 4 lists the reference of the components chosen in this project for experimental setup. To protect the command block from current leakage, a galvanic isolation is required using optocouplers (6N136) and drivers (IR2121). The practical use of this MPPT hybrid boost converter was tested using Siemens (SM55) photovoltaic modules, its specifications at 1000 W/m2 solar irradiance and at 25  C cell temperature are listed in Table 5. Each PV module delivers a power about 55 W, with a MPP voltage and current of: Vmpp ¼ 17.4 V and Impp ¼ 3.15 A. In order to supply incandescent lamps, two PV modules mounted in cascade are used. Fig. 14 gives the schematic diagram of the designed hybrid DC/DC boost connected with PV modules and the control system. A photo of the experimental setup is shown in Fig. 15. Furthermore, the measurement outputs of the current sensor and the voltage transducer are visualized by the oscilloscope which is endowed by the multiplication operation. So, it can trace the power curve (Power ¼ Current  Voltage) as scoped in Fig. 16. By the way, the tracking progression of the MPP is clearly shown from the scanning process until the MPP is reached and kept.

In addition, once the PV system reaches the MPP for a specified mode, the switching to the other hybrid boost working mode has been experimentally tested. The transitions have been performed successfully, every time the MPP

Fig. 14 e The schematic diagram of the experimental setup.

Fig. 15 e Experimental setup photo.

Table 4 e Bill of materials. Component MOSFET Current sensor Voltage transducer Drivers Optocoupler Capacitors Inductors Diodes Command board Other components

Reference

Quantity

IRFP460 (ID ¼ 20 A RDS(ON)0.27 U, VDS ¼ 500 V) LA55 eP LV25 eP IR2121 6N135/6N136 Electrolytic 220 mF, 400 V 100 mH, 5 A U1640 ML501 development board (FPGA) Resistors, Switches, Connectors, PCB…

3 1 1 3 3 3 2 4 1 /

Fig. 16 e Measured PV current, voltage and power waveforms during the MPPT process (CB mode).

Table 5 e Siemens SM55 photovoltaic module specifications. Simens solar module SM55 Maximum power rating Pmax (W) Rated current Impp (A) Rated voltage Vmpp (V) Short circuit current Isc(A) Open circuit voltage Voc (V) NOCT ( C) Temp. coefficient: short-circuit current Temp. coefficient: open-circuit voltage Weight [Kg]

55 3.15 17.4 3.45 21.7 45 ± 2 1.2 mA/ C 0.77 V/ C 5.5

Fig. 17 e MPPT process and switching hybrid boost modes waveforms.

Please cite this article in press as: Belhimer S, et al., A novel hybrid boost converter with extended duty cycles range for tracking the maximum power point in photovoltaic system applications, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/ j.ijhydene.2018.02.136

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 8 ) 1 e1 2

was quickly attained for the selected working mode. The obtained curves are shown in Fig. 17. This last figure proves experimentally the performance of the proposed hybrid boost, with its two working modes, to reach and keep the MPP. The experimental results are very similar to the simulation ones. The difference was only in the response time of the real system which was late comparing to the simulation one. This is due to various experimental parameters like the sensing time, MPPT computing time, some components response time and the changed environment parameter (solar irradiance and temperature). Although, the hybrid boost is very convenient for PV applications.

[6]

[7]

[8]

[9]

Conclusion [10]

In this paper, a novel hybrid boost topology of DC/DC converter was simulated and tested experimentally. It presents the combination fruit of quadratic and conventional boost. The well-known P&O MPPT algorithm has been evaluated with both hybrid boost working modes (QB and CB). The simulated system has been tested under fast solar irradiance variations. The maximum PV power was always tracked and provided in real time to supply the electrical load. The simulation and experimental verification of both modes have been performed successfully. The results seem very satisfying, and prove that the hybrid boost is ultimately convenient with photovoltaic applications. Besides, further research should focus on the application of this new hybrid topology for inverters or with using some advanced MPPT algorithms for shaded conditions, where the PV modules has more than one MPP, in order to determine and track the global maximum power point.

[11]

[12]

[13]

[14]

[15]

Acknowledgements [16]

The authors would like to thank Dr. Karim KACED from the National Polytechnic School of Algiers for his help. The authors would like also to thank anonymous reviewers for their help, useful comments, and contribution to this paper's improvement.

references

[1] Belhachat F, Larbes C. Modeling, analysis and comparison of solar photovoltaic array configurations under partial shading conditions. Sol Energy 2015;120:399e418. [2] Mosa M, Shadmand MB, Balog RS, Rub HA. Efficient maximum power point tracking using model predictive control for photovoltaic systems under dynamic weather condition. IET Renew Power Gener 2017;11:1401e9.  lu C. MIMO nonlinear modeling and [3] Deveci O, Kasnakog robust control of standalone dc photovoltaic systems. Int J Hydrogen Energy 2017;42:18064e80. [4] Islam MR, Rahman F, Xu W. Advances in solar photovoltaic power plants. 1st ed. Springer-Verlag Berlin Heidelberg; 2016. [5] Forouzesh M, Siwakoti YP, Gorji SA, Blaabjerg F, Lehman B. Step-up DCeDC converters: a comprehensive review of

[17]

[18]

[19]

[20]

[21]

[22]

11

voltage-boosting techniques, topologies, and applications. IEEE Trans Power Electron 2017;32:9143e78. Erickson RW. DCeDC power converters. Wiley encyclopedia of electrical and electronics engineering. John Wiley & Sons, Inc; 2001. Morales-Saldana Ja, Galarza-Quirino R, Leyva-Ramos J, Carbajal-Gutierrez Ee, Ortiz-Lopez Mg. Multiloop controller design for a quadratic boost converter. IET Electr Power Appl 2007;1:362e7. Perez-Pinal FJ, Cervantes I. Multi-objective control for cascade boost converter with single active switch. In: Electric machines and drives conference, IEMDC '09 IEEE international; 2009. p. 1858e62. Ramos JL, Ortiz-Lopez MG, Morales-Saldana JA, L H. Control of a cascade boost converter with a single active switch. In: IEEE power electronics specialists conference; 2008. p. 2383e8. Rosas-Caro JC, Sanchez VM, Valdez-Resendiz JE, MayoMaldonado JC, Beltran-Carbajal F, Valderrabano-Gonzalez A. Quadratic buckeboost converter with positive output voltage and continuous input current for PEMFC systems. Int J Hydrogen Energy 2017;42:30400e6. Patidar K, Umarikar AC. High step-up converters based on quadratic boost converter for micro-inverter. Elec Power Syst Res 2015;119:168e77. Kaced K, Larbes C, Ramzan N, Bounabi M, Dahmane Ze. Bat algorithm based maximum power point tracking for photovoltaic system under partial shading conditions. Sol Energy 2017;158:490e503. Chekired F, Larbes C, Mellit A. Comparative study between two intelligent MPPT-controllers implemented on FPGA: application for photovoltaic systems. Int J Sustain Energy 2014;33:483e99. Guilbert D, Guarisco M, Gaillard A, N'Diaye A, Djerdir A. FPGA based fault-tolerant control on an interleaved DC/DC boost converter for fuel cell electric vehicle applications. Int J Hydrogen Energy 2015;40:15815e22. Quamruzzaman M, Mohammad N, Matin MA, Alam MR. Highly efficient maximum power point tracking using DCeDC coupled inductor single-ended primary inductance converter for photovoltaic power systems. Int J Sustain Energy 2016;35:914e32. Belkaid A, Colak I, Kayisli K. Implementation of a modified P&O-MPPT algorithm adapted for varying solar radiation conditions. Electr Eng 2016:1e8. Sahnoun MA, Ugalde HMR, Carmona J-C, Gomand J. Maximum power point tracking using P&O control optimized by a neural network approach: a good compromise between accuracy and complexity. Energy Procedia 2013;42:650e9. Bradai R, Boukenoui R, Kheldoun A, Salhi H, Ghanes M, Barbot JP, et al. Experimental assessment of new fast MPPT algorithm for PV systems under non-uniform irradiance conditions. Appl Energy 2017;199:416e29. Ozdemir S, Altin N, Sefa I. Fuzzy logic based MPPT controller for high conversion ratio quadratic boost converter. Int J Hydrogen Energy 2017;42:17748e59. Boukenoui R, Salhi H, Bradai R, Mellit A. A new intelligent MPPT method for stand-alone photovoltaic systems operating under fast transient variations of shading patterns. Sol Energy 2016;124:124e42. Harrag A, Bahri H. Novel neural network IC-based variable step size fuel cell MPPT controller: performance, efficiency and lifetime improvement. Int J Hydrogen Energy 2017;42:3549e63. Harrag A, Messalti S. How fuzzy logic can improve PEM fuel cell MPPT performances? Int J Hydrogen Energy 2018;43:537e50.

Please cite this article in press as: Belhimer S, et al., A novel hybrid boost converter with extended duty cycles range for tracking the maximum power point in photovoltaic system applications, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/ j.ijhydene.2018.02.136

12

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 8 ) 1 e1 2

[23] Benabda A, Kelaiaia MS, Labar H, Logerais P-O, Durastanti J-F. Boost chopper MPP assessment based on solar irradiance and predictive duty cycle applied to a PV system. Int J Hydrogen Energy 2017;42:19403e10.  zaro A. Review of the [24] Salas V, Olı´as E, Barrado A, La maximum power point tracking algorithms for stand-alone photovoltaic systems. Sol Energy Mater Sol Cell 2006;90:1555e78. [25] Ahmed J, Salam Z. A modified P&O maximum power point tracking method with reduced steady-state oscillation and improved tracking efficiency. IEEE Trans Sustain Energy 2016;7:1506e15. [26] Loukriz A, Haddadi M, Messalti S. Simulation and experimental design of a new advanced variable step size Incremental Conductance MPPT algorithm for PV systems. ISA Trans 2016;62:30e8. [27] Ahmed J, Salam Z. An accurate method for MPPT to detect the partial shading occurrence in a PV system. IEEE Trans Ind Inf 2017;13:2151e61. [28] Titri S, Larbes C, Toumi KY, Benatchba K. A new MPPT controller based on the Ant colony optimization algorithm for Photovoltaic systems under partial shading conditions. Appl Soft Comput 2017;58:465e79. [29] Messai A, Mellit A, Guessoum A, Kalogirou SA. Maximum power point tracking using a GA optimized fuzzy logic controller and its FPGA implementation. Sol Energy 2011;85:265e77. [30] Abido MA, Khalid MS, Worku MY. An efficient ANFIS-based PI controller for maximum power point tracking of PV systems. Arabian J Sci Eng 2015;40:2641e51.

[31] Ramos-Hernanz J, Lopez-Guede JM, Barambones O, Zulueta E, Fernandez-Gamiz U. Novel control algorithm for MPPT with Boost converters in photovoltaic systems. Int J Hydrogen Energy 2017;42:17831e55. [32] Czarkowski D. 13-DCeDC converters. In: Rashid A2, Muhammad H, editors. Power electronics handbook. 3rd ed. Boston: Butterworth-Heinemann; 2011. p. 249e63. [33] Sayed K, Abdel-Salam M, Ahmed A, Ahmed M. New high voltage gain dual-boost DC-DC converter for photovoltaic power systems. Elec Power Compon Syst 2012;40:711e28. [34] Ortiz-Lopez MG, Leyva-Ramos J, Diaz-Saldierna LH, CarbajalGutierrez EE. Multiloop controller for N-Stage cascade boost converter. In: IEEE international conference on control applications; 2007. p. 587e92. [35] Uzun S, Canal MR, Topaloglu N. Design and realization of a precise and flexible control card for a dental prosthesis baking furnace by using FPGA. J Facul Eng Architect Gazi Univ 2012;27:175e82.  s A. An interleaved, FPGA[36] Sobrino-Manzanares F, Garrigo controlled, multi-phase and multi-switch synchronous boost converter for fuel cell applications. Int J Hydrogen Energy 2015;40:12447e56. [37] Mellit A, Kalogirou SA. MPPT-based artificial intelligence techniques for photovoltaic systems and its implementation into field programmable gate array chips: review of current status and future perspectives. Energy 2014;70:1e21.

Please cite this article in press as: Belhimer S, et al., A novel hybrid boost converter with extended duty cycles range for tracking the maximum power point in photovoltaic system applications, International Journal of Hydrogen Energy (2018), https://doi.org/10.1016/ j.ijhydene.2018.02.136