- Email: [email protected]
A photovoltaic powered electrolysis converter system with maximum power point tracking control
international journal of hydrogen energy xxx (xxxx) xxx
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
A photovoltaic powered electrolysis converter system with maximum power point tracking control Mustafa Ergin S‚ahi_n Recep Tayyip Erdogan University, Department of Electrical and Electronics Engineering, Rize, Turkey
highlights Photovoltaic
graphical abstract
houses
and
cars
storing their energy as hydrogen to benefit
from
solar
energy
efficiently. Photovoltaic energy is converted to the desired voltage level using the buck converter. Hydrogen generating is occurred with electrolysis process. The photovoltaic sources are used with a maximum power point tracking algorithm. The system is simulated using a perturb and observe algorithm and proportianal integral controller.
article info
abstract
Article history:
One of the main problems for renewable and other innovative energy sources is the storage
Received 27 October 2019
of energy for sustainability. This study focuses on two different scenarios to benefit from
Received in revised form
solar energy more efficiently. Photovoltaic (PV) energy is converted to the desired voltage
3 January 2020
level using a buck converter for generating hydrogen with electrolysis process. A maximum
Accepted 23 January 2020
power point tracking (MPPT) algorithm is used to benefit from the photovoltaic sources
Available online xxx
more efficiently. The basic electrolysis load for hydrogen production needs low voltage and high current and controlled sensitively to supply these conditions. The photovoltaic
Keywords:
powered buck converter for electrolysis load was simulated in MATLAB/Simulink software
Hydrogen energy generation from
using a perturb and observe (P and O) MPPT algorithm and PI controller. The simulation
PV
results show that in normal, short circuit and open circuit working conditions the PV and
Electrolysis process
load voltages are stabilized. The efficiency of the proposed system is reached more than
DC-DC buck converter
90% for high irradiance levels.
Proportional integral control
© 2020 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
Maximum power point tracking Perturb and observe method
E-mail address: [email protected]. https://doi.org/10.1016/j.ijhydene.2020.01.162 0360-3199/© 2020 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved. Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
2
international journal of hydrogen energy xxx (xxxx) xxx
Introduction The fossil fuels used today are limited sources and assumed to be run out soon. Some of the sustainable energy sources are developed and implemented to get over this problem but most of them have detrimental effects on nature such as electrochemical reactions [1,2]. On the other hand, the sun continues to heat the world as a renewable energy source. The main problem of renewable energy sources is the storage of energy uninterruptable. The storage of energy in batteries is not an efficiently conventional method. Therefore, storage of the energy as hydrogen is proposed as a new solution. However, the conversion of electricity to hydrogen energy and connect it to the grid, and the necessary technology investments complicate this transformation [3,4]. The sun, our endless power source, solar energy is continuing to heat and illuminate our world on the other side. Solar collectors and solar cells are two different methods to absorb solar energy today [5]. This energy can absorb only 20e25% efficiently with different solar cells today [4]. The solar radiation changes with time and not stable every time [5,7]. These disadvantages make the use of storage systems compulsory. The conventional storage systems are inadequate and inefficient for storage of so high energy [28,33,35]. This makes attractive the conversion of solar energy to hydrogen with these conversion systems [4,8,29]. For energy conversion, the DC-DC converters are getting important. A solution is suggested to obtain hydrogen gas from water electrolysis and storage of this gas by compression [9,10]. The electrolysis of water needs a minimum of 1.2 V DC voltage for hydrogen generation theoretically [11]. Also, electrocatalytic water splitting offers a solution to both challenges, in which it converts electrical energy into oxygen (O2) and hydrogen (H2), which serves as fuel and feedstock for the chemical industry. Porous materials, chemicals, and nanomaterials are used as a solution in some studies [37e39]. The photovoltaic systems are designed for high voltages for common bus voltages 12 V or 24 V for battery charge especially [12]. Therefore, a DC-DC buck converter must decrease the voltage and increase the current to desired values. Some different studies are proposed and tried in literature, but they are not tried for photovoltaic sources and electrolysis loads. Moreover, an efficient controller for nonlinear electrolysis loads is not investigated in these studies [13,14]. For the switching power supplies, and an effective controller is necessary every time. As a solution, the proportional-integral (PI) controller is used in this study. This easy applicate control method has some advantages for instantaneous load changes and effective error control, dynamic response [15,16]. The electrolysis process required low voltage and high currents with a DC-DC buck converter. This process with this controller is proposed to reduce the power losses and to increase the stability of the dynamic system [17,18]. The technological developments are decreasing the material costs in ever past days. For example, the electrolyzes system capital cost decreases from 430 $/kW to 300 $/kW in the last 8 years. Hydrogen production cost is decreases from 4.10 $/kg H2 to 2.30 $/kg H2 in last 8 years [40]. On the other
side, PV panel prices decreased less than 400 $/kW in the last years. For the PV powered electrolyzes system capital cost is can be calculated at nearly 750 $/kW with the other complementary components [24]. This subject is studied very rarely by the researchers in the literature. Some of related theoretical and specific studies are given in related parts above. The other one investigates similar a problem but it is very restricted and not include more details about the methods and solutions [42]. Another one is focused on the optimization of such a system and not include such a simulation and control method and not give more detailed simulation results and efficiency analyses [43]. Another one suggested a different topology and control method to solve such a problem [44]. Generally, not anyone such a comprehensive study to propose a detailed solution and solve this problem. This paper focuses on two different hydrogen production process applications. The first one is photovoltaic houses and storage of their energy as hydrogen energy, the second one is photovoltaic cars and storage of their energy as hydrogen energy. The photovoltaic energy is converted to the required voltage level for two photovoltaic sourced scenarios using the DC-DC buck converter for generating hydrogen. An MPPT algorithm is necessary for photovoltaic sources to work with the PV panel at the maximum power point (MPP). The electrolysis load for hydrogen production needs low voltages and high currents especially. For this aim, the P and O MPPT algorithm with a PI controller is proposed as a solution for a sensitive controller in this study. The photovoltaic powered DC-DC buck converter for electrolysis load was simulated in MATLAB/Simulink software using the proposed converter system for normal, short and open circuit conditions. The efficiency comparison is made for different solar irradiation and temperature levels.
Proposed converter system for hydrogen production Some different scenarios are proposed by scientists, towards the hydrogen energy in the future [4,19e22,30,31,36]. This paper focuses on two different applications to convert and use solar energy more efficiently. The solar houses and cars are two different applications to the storage of their energy as hydrogen instead of conventional storage technologies. The hydrogen conversion cycle is a common point in these two photovoltaic sourced scenarios without load parts. While the electrical vehicles use the generated electrical energy to drive brushless direct current (BLDC) motors, the other direct current (DC) and alternating current (AC) loads are used in solar houses. The combination of these two main applications is given in Fig. 1. At-home applications need to store solar energy for times when the sun is not shining. Hydrogen provides a safe, efficient, clean way to do this [20]. The solar energy is absorbed by photovoltaic arrays and converted to required voltages by the buck converter for the electrolysis loads at homes. The electrolysis process needs only water and converts it to hydrogen and oxygen gases with applied voltage. Generated hydrogen gas can be stored physically as either a gas or liquid and can
Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
3
international journal of hydrogen energy xxx (xxxx) xxx
Fig. 1 e The combination of two main scenarios. also be stored materially on the surfaces of solids or within solids. Storage of hydrogen as a gas typically requires 350e700 bar tank pressure and energy for storage, but storage of hydrogen on the surface of solids (by adsorption) or within solids (by absorption) not required such energy [45]. The hydrogen gas is stored in hydrogen tanks especially in the underground for security and efficiency. This gas can be burned directly or converted to electrical energy by fuel cells again. The generated oxygen can be used for fuel cells as an input gas as shown in Fig. 1. So the inputs and outputs of such a system can be used as a complementary material of this system [30]. The electrical vehicle application is similar to the first application. The solar energy is absorbed by photovoltaic arrays and converted to required voltages by the buck converter for electrolyzes loads [41]. The electrolysis process needs only aqueous electrolyte and converts it to hydrogen and oxygen gases with applied voltage. The hydrogen gas is stored in hydrogen tanks for cars appropriately. This gas is burned
the voltage level to the range of 1.23 Ve2.06 V which is required for electrolysis [9]. A buck type DC-DC converter is required for this reason. The buck type DC-DC converter decreases the voltage value of about 2 V and increases the current value at the same ratio. The chemical reactions in the electrolysis are given in chemical Equation (1). The increasing current increases the electron flow and increases hydrogen production at the same time. A robust controller is necessary under constant voltage and current with small fluctuations for nonlinear electrolysis load for this reason. The PI controller is proposed in this study for the MPP controller to control the electrolysis load voltage. The PI controller is applied to the P and O MPPT controller output error signals and generates a pulse width modulation (PWM) signal to switch the metal oxide semiconductor fieldeffect transistor (MOSFET) devices. The general view of the proposed photovoltaic-electrolysis system with converter and controller is shown in Fig. 2.
Anode (oxidaon) : 2 H2O(l) →O2(g) + 4H+(aq) + 4e− Eo = +1.23 V + − Eo = Cathode (reducon): 2 H (aq) + 2e →H2(g) (1) →2 H2(g) + O2(g) Eocell= -1.23 V Overall reacon : 2 H2O(l)
directly or converted to electrical energy by fuel cells again for cars similarly. The Toyota Corporation succeeded to produce such a car which is called ‘hydrogen car’, but it used batteries instead of photovoltaic for energy sources [23,32]. Moreover, it is very expensive and does not work efficiently. The common point in these two applications is how the energy will be converted more efficiently from solar energy to hydrogen energy. A buck type DC-DC converter is proposed as a solution for this problem in this paper. The DC-DC buck converter is designed for a 100 W commercial photovoltaic power input. These commercial photovoltaic panels generate 0e25 V output voltages for different loads and currents generally [24]. On the other hand, it is necessary to decrease
0.00
V
The proposed system components and modelling Photovoltaic panel model and control of MPP The photovoltaic sources are not behaving as infinite or linear sources in the application. They behave as finite and nonlinear sources as shown in their characteristic curves. At least 250 W power source is necessary for the designed hydrogen production system, and the necessary photovoltaic
Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
4
international journal of hydrogen energy xxx (xxxx) xxx
Fig. 2 e The proposed photovoltaic-electrolysis converter system.
Table 1 e Electrical characteristics of photovoltaic sources. Specifications Maximum power (Pm) Maximum power voltage (Vpm) Maximum power current (Ipm) Short circuit current (Isc) Open circuit voltage (Voc) Temperature coefficient of Voc Temperature coefficient of Isc Nominal operation cell temperature Cell per module (Ncell)
Values 249.86 W 31 V 8.06A 8.55 A 37.6 V 0.35%/K 0.06%/K 48 C ± 2 C 60
The photovoltaic cells are connected in series and parallel to increase the voltage and current of the photovoltaic module necessary values. Using this equation photovoltaic module model is designed in MATLAB/Simulink and simulated for necessary parameter values. Fig. 3 shows the current-voltage (IeV) characteristic of the photovoltaic source model and the power voltage (PeV) characteristic of the photovoltaic source model. The simulation results of this photovoltaic source model are suitable for the theory as expected. VPV ¼
ISC þ KI T Tref :G þ NP I0 IPV NS :n:k:T NS Rs ln : :IPV q I0 :NP NP Rsh (2)
panel is selected from the manufacturer catalogs [24]. This photovoltaic panel (Trina Solar TSM-250PA05.08) electrical characteristic values are given in Table 1. The photovoltaic panel equation which is derived from the basic Shockley diode equation for all parameters and used in previous studies is given in Equation (2) [6]. The solar cell current depends on cellular operating temperature (T) and the absorption of sunlight (G) are combined with this equation.
The other component of the PV powered electrolyzes system is MPPT, and it is used to obtain maximum power from PV due to its nonlinear characteristics. Perturb and observe (P and O) is one of the MPPT methods used in PV systems [17,18]. The P and O method algorithm and MATLAB/Simulink model are shown in Fig. 4 (a, b). The method measures the PV voltage and currents to calculate the maximum power. If the power last
Fig. 3 e Photovoltaic panels IeV and PeV curves. Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
international journal of hydrogen energy xxx (xxxx) xxx
5
Fig. 4 e (a) The P&O method algorithm, (b) MATLAB/Simulink model of MPPT.
Fig. 5 e Block diagram of the DC-DC converter with a PI controller [16].
Fig. 6 e The buck converter study states (a), current-voltage, and switching signals (b). Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
6
international journal of hydrogen energy xxx (xxxx) xxx
Fig. 7 e The basic experimental set up for alkaline electrolysis system (a), MATLAB/Simulink model of nonlinear electrolyzes model (b).
value is bigger than the previous value compared to the voltage last value with the previous value. If the voltage last value is bigger than the previous value, the duty ratio (D) is increased one step otherwise is decreased one step. In this way, an optimum duty cycle is obtained to drive the buck converter and to work on MPP.
Proportional integral controller (PID) The most commonly used control method is PID for DC-DC converters. It is the most preferred because it can be designed easily and suitable for linear systems and industrial applications [15]. The control signal which is generated by the PID controller for continuous mode operation of the converter depends on the error function given in Equation (3).
Zt uðtÞ ¼ KP eðtÞ þ KI
eðtÞdt þ KD
deðtÞ dt
(3)
0
The general block diagram of the converter with the PI controller is shown in Fig. 5. The error signal which is obtained from the difference of output and reference voltage applied to the PI controller to implement the PWM process for generating the switching signal to drive the converter MOSFETs [16].
DC-DC buck converter The DC-DC buck converter is the main part of the proposed PV sourced electrolysis system to increase the obtained voltage
Fig. 8 e The simulation model of the proposed converter system in MATLAB/Simulink. Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
international journal of hydrogen energy xxx (xxxx) xxx
from the PV panel at electrolyzes voltage level. This converter reduces the 31 V photovoltaic voltage to the range of 2e3 V which is necessary for the electrolysis process. Depend on the voltage value decrease, the current value increases in the same ratio. These high currents occurred some loses on the components naturally. Some solutions are proposed to reduce loses in different papers [12,14].
DIL ¼ Dð1 DÞ
7
A 250 W power DC-DC buck converter is designed to obtain a 2 V output voltage from 31 V DC input voltage. For the converter, circuit capacitor and inductor values were calculated for 100 kHz switching frequency and 5% ripple at an output voltage as Equations (3) and (4). The DC-DC buck converter is designed for electrolyzes load. Inductance maximum ripple value and duty ratio are used to calculate inductance critical value as in Equation (4);
VPV VPV 31 / L ¼ Dð1 DÞ /L > 2:79 mH ¼ 0:1ð1 0:1Þ 100x103 x10 f :L f :DIL
(4)
Output voltage ripple value and inductance ripple values are used to calculate capacitor critical value as in Equation (5); Table 2 e Simulation parameters of the proposed system. Parameters DC input voltage DC output voltage Nominal load current Nominal power Switching frequency Filter inductance (L) Filter capacitance (C) DC Link capacitance Electrolysis initial resistance (Re0) PI parameters PV solar irradiation (G) PV operation temperature (T)
Values 0e37.6 V 2e2.8 V 80 A 250 W 100 kHz 500 mH 10,000 mF 3000 mF 0.08 U Kp ¼ 0.7, Kı ¼ 0.001 1000 W/m2 25 Co
DVC ¼
DIL DIL 10 /C ¼ /C > 1250 mF ¼ 8:f :C 8:f :DVC 8x100x103 x0:1
(5)
The DC-DC buck converter works depending on two states of switching MOSFET. In the first state, the MOSFET switch is turned on and the input voltage is applied to the output and current flows through the inductance, capacitor, and load. In the second state, the MOSFET switch is turned off, the inductance voltage continues to be applied to the output as a voltage source and currently continues to flow through the inductance, capacitor, and load by the freewheeling diode. These states are repeated by the switching cycle and current flows continuously [25e27]. The DC-DC buck converter study
Fig. 9 e (a) The simulation results for PV source voltage, current, and power with the P&O MPPT algorithm. (b) Simulation results for electrolysis load voltage, current, and power. Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
8
international journal of hydrogen energy xxx (xxxx) xxx
Fig. 10 e The inductance current and voltage variation depending on switching PWM signal.
states, voltage and current signals depend on switching signals are given in Fig. 6 (a, b).
Electrolysis load model The electrolysis chemical reaction needs to be perfectly understood before analysing the behaviour of the electrolysis load. The suitable voltage is applied to electrodes placed in the water or liquid solution to start a chemical reaction which is called electrolyze. Hydrogen gas occurred at the cathode and oxygen gas occurred at the anode when the electrolysis
chemical reaction is started. For the water decomposition, 1.23 V voltage is enough theoretically as shown in Equation (1). However, in practical applications depend on electrodes type it can be changed. During the reaction, 1-mol oxygen gas occurs at the anode and 2-mol hydrogen gas occurs at the cathode as shown in Equation (1). The basic experimental set up for the alkaline electrolysis system is shown in Fig. 7(a). The electrolyze behaves like a nonlinear resistance during the chemical reaction in reality. This nonlinearity depends on the chemical composition of the solution, the pressure of electrolyzes stack, temperature and the other parameters
Fig. 11 e The input (a), and the output (b) voltage, current and power variations for load short circuit. The input (c), and the output (d) voltage, current and power variations for load open circuit. Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
9
international journal of hydrogen energy xxx (xxxx) xxx
Fig. 11 e (continued).
which exist during the chemical reaction [34]. The derived nonlinear mathematical model of the electrolyze depends on these parameters is given in Equation (5) [12,14]. b Ie Re ¼ Re0 1 þ a Iebase
(6)
In Equation (5), Reo is initial resistance and its value is 0.08 U. The a and b are constant values that depend on nonlinear characteristics of the electrolyze model. These values are selected between 0.35 and 0.65 for a, and between 2 and 4 for b in typical applications. After these values have been written in Equation (6), the electrolysis load model is obtained as in Equation (7). The electrolysis resistance depends on the electrolysis current is given in Equation (7). The electrolysis nonlinear load model is simulated in MATLAB/ Simulink using this equation as in Fig. 6(b). 3 Ie Re ¼ 0:08 1 þ 0:5 50
(7)
General system modelling The proposed photovoltaic sourced DC-DC buck converter for electrolyzes load with MPPT and PI controller was simulated in MATLAB/Simulink software. The simulation model of the proposed converter system is shown in Fig. 8. The photovoltaic source model and electrolysis load model was used in this converter model instead of the linear source and load. The PV voltage and current are applied to a P&O method for the MPPT algorithm and the obtained a duty signal. The PI control uses this signal and generates a PWM signal to drive the MOSFET switching. The simulation parameters of the proposed system model in Fig. 8 are given in Table 2. The simulation results are obtained for these parameters.
Simulation results Simulation results of the proposed photovoltaic sourced DCDC buck converter with electrolysis load are given in this
Table 3 e The efficiency comparison table for different solar irradiation and temperature levels. 0C
Temperature
25C
50C
75C
Solar Irradiation PPV (W) Pload(W) Efficn. PPV (W) Pload(W) Efficn. PPV (W) Pload (W) Efficn. PPV (W) Pload (W) Efficn. 100 W/m2 200 W/m2 300 W/m2 400 W/m2 500 W/m2 600 W/m2 700 W/m2 800 W/m2 900 W/m2 1000 W/m2
11.72 30.36 52.18 76.38 102.6 130.4 159.7 190.2 222 254.4
9.4 25.68 45.09 66.85 90.55 115.9 142.7 170.6 199.7 229.4
80% 84.6% 87.9% 87.5% 88.3% 88.9% 89.4% 89.7% 89.9% 90.2%
11.98 30.99 53.23 77.90 104.5 132.7 161.9 191.7 221.1 248.8
9.61 26.23 46.04 68.23 92.33 118 144.7 172 199 224.8
80.2% 84.6% 86.5% 87.6% 88.4% 88.9% 89.4% 89.7% 90% 90.4%
12.23 31.6 54.19 79 105.1 131.3 156.3 179 201.4 223.6
9.83 26.77 46.9 69.23 92.9 116.8 139.7 160.1 183 201.3
80.4% 84.7% 86.5% 87.6% 88.4% 88.9% 89.4% 89.4% 90.9% 90%
12.47 32.01 54.09 76.59 97.53 117.6 138.4 157.9 177.8 197.9
10.4 27.13 46.82 67.11 86.63 103.6 123 138.3 165.6 177.8
83.4% 84.7% 86.6% 87.6% 88.8% 88.1% 88.9% 87.6% 93.2% 89.9%
Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
10
international journal of hydrogen energy xxx (xxxx) xxx
Fig. 12 e The efficiency comparison with solar irradiation in different temperatures. chapter. The simulation results were tried for different photovoltaic source and load conditions. The photovoltaic array is simulated for 250 W power photovoltaic panel characteristics which are given in Table 1. The simulation results were given in Fig. 3. The main confidant of this figure how to convert the 8 A current to 80 A and 31 V voltage to 2e3 V at the output of the converter. Using the P and O MPPT algorithm and buck converter this request is achieved. So, the simulation studies are focus on this subject. The simulation results for PV source with P&O MPPT algorithm and PI control are given in Fig. 9 (a). The simulation results show that the MPPT algorithm works correctly and the power, current and voltage values are stable around the MPP as expected in Table 1 and Fig. 3. These provide the maximum power from the PV panel and work efficiently. On the other side, simulation results for electrolysis load voltage, current, and power are given in Fig. 9 (b). The electrolysis load voltage reaches 2 V in 1 s with small ripples. The electrolysis load current and power reaches 80 A and 230 W respectively in 2 s. This shows that the system works more than 90% efficiency. The power losses on the system are originated from losses on the buck converter, and small shifts on the MPP. Nevertheless, this output voltage and power are enough for the efficient electrolysis chemical reaction. Different solutions can be proposed and studied to increase efficiency. The inductance current and voltage variation show the converter working conditions depend on the generated switching PWM signal from the controller. The inductance current, voltage, and PWM signal variation are given in Fig. 10. When the MOSFET switch is turn-on with switching signal, the current flows through the inductance and the voltage is positive, when the MOSFET switch is a turn-off, the current continuous flow decreasing and the inductance voltage is negative. Therefore, the converter works in continuous mode as expected in theoretical results in Fig. 6. The converter system is simulated for short and open circuit inrush conditions. The PV input and electrolysis load output voltage, current and power variations are observed for
the 1-s short circuit at firstly. The results are reaching the first expected values in a few seconds after the short circuit. These results are seen in Fig. 11(a and b). The output voltage, current and power variations are observed for the 1-s open circuit at the electrolysis load secondly. The results are reaching the first expected values in a few seconds. Depending on energy storage components on the converter, the first response shows an overshoot after the open circuit. These results are shown in Fig. 11(c and d). These results show that the simulation model with the PI controller is works correctly and stable for nonlinear input and output components. According to these results, it is possible to say the amount of the hydrogen gas produced from 250 W power PV panels in 1 h is 43.2 cm3 or 43.2 L theoretically through DC-DC buck converter and suitable electrolyzes devices. These results are calculated using the previous studies in Ref. [12]. Also, a detailed simulation made on system efficiency investigation for ten different solar irradiation levels and four temperature levels. The simulation results are given numerically and the efficiency is calculated in Table 3 forever weather conditions. These numerical results are compared in Fig. 12. These results show that the efficiency is more than 90% percent for high solar radiation because of the converter works full capacity in these conditions. However, the hightemperature effects the efficiency negatively, but increasing the solar irradiation balanced it.
Conclusions This paper focuses on two different applications to use PV energy and storage of this energy as hydrogen. Photovoltaic panels' electrical energy was converted to the desired voltage level using the DC-DC buck converter for generating hydrogen with electrolysis process. The P&O MPPT algorithm used with the PI controller in this study to observe the PV energy more efficiently and to convert its desired voltage level for the electrolysis process. The photovoltaic powered DC-DC buck
Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
international journal of hydrogen energy xxx (xxxx) xxx
converter for electrolysis load was simulated in MATLAB/ Simulink software more detailed using this controller. The simulation results are investigated for load voltage, current, and power with photovoltaic sources. As well as, this result was repeated for inrush conditions such as short circuit load and open-circuit load and the system stabilization is observed. These simulation results are suitable for theoretical results as expected. The efficiency comparison is made for different solar irradiation and temperature levels. The efficiency of the proposed system is reached more than 90% for high irradiance levels. The simulation results can be compared for different controllers with MPPT systems using more efficient converters as a future study. Also, this study aims to realized experimentally with research project support to improve the idea.
Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.ijhydene.2020.01.162.
references
[1] Guo Y, Tang J, Henzie J, Jiang B, Qian H, Wang Z, Yamauchi Y. Assembly of hollow mesoporous nanoarchitectures composed of ultrafine Mo 2 C nanoparticles on N-doped carbon nanosheets for efficient electrocatalytic reduction of oxygen. Materials Horizons 2017;4(6):1171e7. [2] Tan H, Tang J, Henzie J, Li Y, Xu X, Chen T, Yamauchi Y. Assembly of hollow carbon nanospheres on graphene nanosheets and creation of ironenitrogen-doped porous carbon for oxygen reduction. ACS Nano 2018;12(6):5674e83. [3] Melaina M, Eichman J. Hydrogen energy storage: grid and transportation services, technical report NREL/TP-540062518. 2015. Internet, http://www.nrel.gov/docs/fy15osti/ 62518.pdf. [4] Jain IP. Hydrogen the fuel for 21st century. Int J Hydrogen Energy 2009;34(17):7368e78. [5] Panwar NL, Kaushik SC, Kothari S. Role of renewable energy sources in environmental protection: a review. Renew Sustain Energy Rev 2011;15(3):1513e24. _ Physical structure, electrical design, [6] S‚ahin ME, Okumus‚ HI. mathematical modeling and simulation of solar cell modules. Turkish Journal of Electromechanics & Energy 2015;1(1):1e8. [7] Wild M, et al. From dimming to brightening: decadal changes in solar radiation at Earth's surface. Science 2005;308(5723):847e50. [8] Winter C-J. Hydrogen energydabundant, efficient, clean: a debate over the energy-system-of-change. Int J Hydrogen Energy 2009;34(14):1e52. [9] Laurencelle F, Chahine R, Hamelin J, Agbossou B, Fournier M, Bose TK. Characterization of Ballard MK5-E proton Exchange membrane fuel cell stack. J. Fuel Cells e Fundam Syst 2001;1:66e71. [10] Fahmy FH, Abdel-Rehim ZS. Hydrogen gas production, and utilization as electricity using a renewable energy source. Energy Sources 1999;21(7):629e41. [11] Khaselev O, Bansal A, Turner JA. High-efficiency integrated multijunction photovoltaic/electrolysis systems for hydrogen production. Int J Hydrogen Energy 2001;26(2):127e32.
11
_ Aydemir MT. Implementation of an [12] S‚ahin ME, Okumus‚ HI, electrolysis system with DC/DC synchronous buck converter. Int J Hydrogen Energy 2014;39(13):6802e12. s A, et al. Combined maximum power point tracking [13] Garrigo and output current control for a photovoltaic-electrolyzer DC/DC converter. Int J Hydrogen Energy 2014;39(36):20907e19. [14] Sahin ME, Okumus HI. Fuzzy logic controlled parallel connected synchronous buck DC-DC converter for water electrolysis. IETE J Res 2013;59(3):280e8. [15] Mohan BM, Sinha A. Mathematical models of the simplest Fuzzy PI/PD controllers with skewed input output fuzzy sets. ISA Trans 2008;47:300e10. _ Comparison of different controllers [16] S‚ahin ME, Okumus‚ HI. and stability analysis for photovoltaic powered buck-boost DC-DC converter. Elec Power Compon Syst 2018;46(2):149e61. [17] Haque A. A fast and reliable perturb and observe maximum power point tracker for solar PV system. International Journal of System Assurance Engineering and Management 2017;8(2):773e87. [18] Ebrahim MS, Sharaf AM, Atallah AM, Emarah AS. Modified P&O technique for hybrid PV-battery smart grid integrated scheme. Turk. J. Electrom. Energy 2018;3(2):1e7. [19] Sherif SA, Barbir F, Veziroglu N. Towards a hydrogen economy. Electr J 2005:62e76. [20] Gregorio M, Teresa V-S. Towards the hydrogen economy? Int J Hydrogen Energy 2007;32(12):1625e37. [21] Balat M, Ozdemir N. New, and renewable hydrogen production processes. Energy Sources 2005;27(13):1285e98. [22] The solar hydrogen cycle. Schatz Energy Research Centre; 2017. Internet, http://www.schatzlab.org/projects/hydrogen/ solarh2cycle.html. [23] Walters G. The car that runs on FRESH AIR, Mail-Online, 9 May 2015. 2017. Internet, http://www.dailymail.co.uk/ sciencetech/article-3074298. [24] 100W PV module datasheet. Solartech Power Inc., SPM 100PTS-N MODULE; 2016. Internet, http://www.solartechpower. com/SPM100PN.html. [25] Erickson RW, Maksimovic D. Chapter 8: converter transfer functions, fundamentals of power electronics. Springer Science & Business Media; 2007. [26] Mohan N, Undeland TM. Power electronics: converters, applications, and design. John Wiley & Sons; 2007. [27] Rashid MH. Power electronics handbook: devices, circuits, and applications. Academic Press; 2010. [28] Smaoui M, Lotfi K. Design and energy control of stand-alone hybrid wind/photovoltaic/fuel cell power system supplying a desalination unit. J Renew Sustain Energy 2014;6(4):043111. [29] Nikolaidis P, Andreas P. A comparative overview of hydrogen production processes. Renew Sustain Energy Rev 2017;67:597e611. [30] Hollmuller P, et al. Evaluation of a 5 kW photovoltaic hydrogen production and storage installation for a residential home in Switzerland. Int J Hydrogen Energy 2000;25(2):97e109. [31] Ni M, et al. A review and recent developments in photocatalytic water-splitting using TiO2 for hydrogen production. Renew Sustain Energy Rev 2007;11(3):401e25. [32] Firat Y. Utility-scale solar photovoltaic hybrid system and performance analysis for eco-friendly electric vehicle charging and sustainable home. Energy Sources, Part A Recovery, Utilization, and Environmental Effects 2018;41(6):734e45. [33] Lin MY, Hourng LW, Wu CH. The effectiveness of a magnetic field in increasing hydrogen production by water electrolysis. Energy Sources, Part A Recovery, Util Environ Eff 2017;39(2):140e7.
Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
12
international journal of hydrogen energy xxx (xxxx) xxx
[34] Shen M, et al. A concise model for evaluating water electrolysis. Int J Hydrogen Energy 2011;36(22):14335e41. [35] Abdel-Aal HK, Mohamed MA. Potentials of storing solar energy in the form of hydrogen for Egypt. Energy Sources 1989;11(2):95e103. lu TN. S‚ahin S., 21st Century's energy: hydrogen [36] Vezirog energy system. Energy Convers Manag 2008;49(7):1820e31. [37] Guo Y, Park T, Yi JW, Henzie J, Kim J, Wang Z, Yamauchi Y. Nanoarchitectonics for transition metal sulfide based electrocatalysts for water splitting. Adv Mater 2019;31(17):1807134. [38] Guo Y, Tang J, Qian H, Wang Z, Yamauchi Y. One-pot synthesis of zeolitic imidazolate framework 67-derived hollow Co3S4@ MoS2 heterostructures as efficient bifunctional catalysts. Chem Mater 2017;29(13):5566e73. [39] Guo Y, Tang J, Wang Z, Kang YM, Bando Y, Yamauchi Y. Elaborately assembled core-shell structured metal sulfides as a bifunctional catalyst for highly efficient electrochemical overall water splitting. Nanomater Energy 2018;47:494e502.
[40] National Renewable Energy Laboratory (Nrel). Current state of the art hydrogen production cost estimate using water electrolysis. Independent Review; September 2009. Document Number: NREL/BK-6A1-46676. [41] Sahin ME. An efficient solar-hydrogen DC-DC buck converter system with sliding mode control. El-Cezerıˆ Journal of Science and Engineering 2019;6(3):558e70. [42] Muhida R, et al. A maximum power point tracking for photovoltaic-SPE system using a maximum current controller. Sol Energy Mater Sol Cell 2003;75(3e4):697e706. [43] Dahbi S, et al. Optimised hydrogen production by a photovoltaic-electrolysis system DC/DC converter and water flow controller. Int J Hydrogen Energy 2016;41(45):20858e66. n DB, Ricardo JM. New concept in [44] Inthamoussou FA, Herna maximum power tracking for the control of a photovoltaic/ hydrogen system. Int J Hydrogen Energy 2012;37(19):14951e8. [45] US Department of Energy. Office of energy efficiency & renewable energy, hydrogen storage. 2019. Internet, www. energy.gov/eere/fuelcells/hydrogen-storage.
Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
Available online at www.sciencedirect.com
ScienceDirect journal homepage: www.elsevier.com/locate/he
A photovoltaic powered electrolysis converter system with maximum power point tracking control Mustafa Ergin S‚ahi_n Recep Tayyip Erdogan University, Department of Electrical and Electronics Engineering, Rize, Turkey
highlights Photovoltaic
graphical abstract
houses
and
cars
storing their energy as hydrogen to benefit
from
solar
energy
efficiently. Photovoltaic energy is converted to the desired voltage level using the buck converter. Hydrogen generating is occurred with electrolysis process. The photovoltaic sources are used with a maximum power point tracking algorithm. The system is simulated using a perturb and observe algorithm and proportianal integral controller.
article info
abstract
Article history:
One of the main problems for renewable and other innovative energy sources is the storage
Received 27 October 2019
of energy for sustainability. This study focuses on two different scenarios to benefit from
Received in revised form
solar energy more efficiently. Photovoltaic (PV) energy is converted to the desired voltage
3 January 2020
level using a buck converter for generating hydrogen with electrolysis process. A maximum
Accepted 23 January 2020
power point tracking (MPPT) algorithm is used to benefit from the photovoltaic sources
Available online xxx
more efficiently. The basic electrolysis load for hydrogen production needs low voltage and high current and controlled sensitively to supply these conditions. The photovoltaic
Keywords:
powered buck converter for electrolysis load was simulated in MATLAB/Simulink software
Hydrogen energy generation from
using a perturb and observe (P and O) MPPT algorithm and PI controller. The simulation
PV
results show that in normal, short circuit and open circuit working conditions the PV and
Electrolysis process
load voltages are stabilized. The efficiency of the proposed system is reached more than
DC-DC buck converter
90% for high irradiance levels.
Proportional integral control
© 2020 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
Maximum power point tracking Perturb and observe method
E-mail address: [email protected]. https://doi.org/10.1016/j.ijhydene.2020.01.162 0360-3199/© 2020 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved. Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
2
international journal of hydrogen energy xxx (xxxx) xxx
Introduction The fossil fuels used today are limited sources and assumed to be run out soon. Some of the sustainable energy sources are developed and implemented to get over this problem but most of them have detrimental effects on nature such as electrochemical reactions [1,2]. On the other hand, the sun continues to heat the world as a renewable energy source. The main problem of renewable energy sources is the storage of energy uninterruptable. The storage of energy in batteries is not an efficiently conventional method. Therefore, storage of the energy as hydrogen is proposed as a new solution. However, the conversion of electricity to hydrogen energy and connect it to the grid, and the necessary technology investments complicate this transformation [3,4]. The sun, our endless power source, solar energy is continuing to heat and illuminate our world on the other side. Solar collectors and solar cells are two different methods to absorb solar energy today [5]. This energy can absorb only 20e25% efficiently with different solar cells today [4]. The solar radiation changes with time and not stable every time [5,7]. These disadvantages make the use of storage systems compulsory. The conventional storage systems are inadequate and inefficient for storage of so high energy [28,33,35]. This makes attractive the conversion of solar energy to hydrogen with these conversion systems [4,8,29]. For energy conversion, the DC-DC converters are getting important. A solution is suggested to obtain hydrogen gas from water electrolysis and storage of this gas by compression [9,10]. The electrolysis of water needs a minimum of 1.2 V DC voltage for hydrogen generation theoretically [11]. Also, electrocatalytic water splitting offers a solution to both challenges, in which it converts electrical energy into oxygen (O2) and hydrogen (H2), which serves as fuel and feedstock for the chemical industry. Porous materials, chemicals, and nanomaterials are used as a solution in some studies [37e39]. The photovoltaic systems are designed for high voltages for common bus voltages 12 V or 24 V for battery charge especially [12]. Therefore, a DC-DC buck converter must decrease the voltage and increase the current to desired values. Some different studies are proposed and tried in literature, but they are not tried for photovoltaic sources and electrolysis loads. Moreover, an efficient controller for nonlinear electrolysis loads is not investigated in these studies [13,14]. For the switching power supplies, and an effective controller is necessary every time. As a solution, the proportional-integral (PI) controller is used in this study. This easy applicate control method has some advantages for instantaneous load changes and effective error control, dynamic response [15,16]. The electrolysis process required low voltage and high currents with a DC-DC buck converter. This process with this controller is proposed to reduce the power losses and to increase the stability of the dynamic system [17,18]. The technological developments are decreasing the material costs in ever past days. For example, the electrolyzes system capital cost decreases from 430 $/kW to 300 $/kW in the last 8 years. Hydrogen production cost is decreases from 4.10 $/kg H2 to 2.30 $/kg H2 in last 8 years [40]. On the other
side, PV panel prices decreased less than 400 $/kW in the last years. For the PV powered electrolyzes system capital cost is can be calculated at nearly 750 $/kW with the other complementary components [24]. This subject is studied very rarely by the researchers in the literature. Some of related theoretical and specific studies are given in related parts above. The other one investigates similar a problem but it is very restricted and not include more details about the methods and solutions [42]. Another one is focused on the optimization of such a system and not include such a simulation and control method and not give more detailed simulation results and efficiency analyses [43]. Another one suggested a different topology and control method to solve such a problem [44]. Generally, not anyone such a comprehensive study to propose a detailed solution and solve this problem. This paper focuses on two different hydrogen production process applications. The first one is photovoltaic houses and storage of their energy as hydrogen energy, the second one is photovoltaic cars and storage of their energy as hydrogen energy. The photovoltaic energy is converted to the required voltage level for two photovoltaic sourced scenarios using the DC-DC buck converter for generating hydrogen. An MPPT algorithm is necessary for photovoltaic sources to work with the PV panel at the maximum power point (MPP). The electrolysis load for hydrogen production needs low voltages and high currents especially. For this aim, the P and O MPPT algorithm with a PI controller is proposed as a solution for a sensitive controller in this study. The photovoltaic powered DC-DC buck converter for electrolysis load was simulated in MATLAB/Simulink software using the proposed converter system for normal, short and open circuit conditions. The efficiency comparison is made for different solar irradiation and temperature levels.
Proposed converter system for hydrogen production Some different scenarios are proposed by scientists, towards the hydrogen energy in the future [4,19e22,30,31,36]. This paper focuses on two different applications to convert and use solar energy more efficiently. The solar houses and cars are two different applications to the storage of their energy as hydrogen instead of conventional storage technologies. The hydrogen conversion cycle is a common point in these two photovoltaic sourced scenarios without load parts. While the electrical vehicles use the generated electrical energy to drive brushless direct current (BLDC) motors, the other direct current (DC) and alternating current (AC) loads are used in solar houses. The combination of these two main applications is given in Fig. 1. At-home applications need to store solar energy for times when the sun is not shining. Hydrogen provides a safe, efficient, clean way to do this [20]. The solar energy is absorbed by photovoltaic arrays and converted to required voltages by the buck converter for the electrolysis loads at homes. The electrolysis process needs only water and converts it to hydrogen and oxygen gases with applied voltage. Generated hydrogen gas can be stored physically as either a gas or liquid and can
Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
3
international journal of hydrogen energy xxx (xxxx) xxx
Fig. 1 e The combination of two main scenarios. also be stored materially on the surfaces of solids or within solids. Storage of hydrogen as a gas typically requires 350e700 bar tank pressure and energy for storage, but storage of hydrogen on the surface of solids (by adsorption) or within solids (by absorption) not required such energy [45]. The hydrogen gas is stored in hydrogen tanks especially in the underground for security and efficiency. This gas can be burned directly or converted to electrical energy by fuel cells again. The generated oxygen can be used for fuel cells as an input gas as shown in Fig. 1. So the inputs and outputs of such a system can be used as a complementary material of this system [30]. The electrical vehicle application is similar to the first application. The solar energy is absorbed by photovoltaic arrays and converted to required voltages by the buck converter for electrolyzes loads [41]. The electrolysis process needs only aqueous electrolyte and converts it to hydrogen and oxygen gases with applied voltage. The hydrogen gas is stored in hydrogen tanks for cars appropriately. This gas is burned
the voltage level to the range of 1.23 Ve2.06 V which is required for electrolysis [9]. A buck type DC-DC converter is required for this reason. The buck type DC-DC converter decreases the voltage value of about 2 V and increases the current value at the same ratio. The chemical reactions in the electrolysis are given in chemical Equation (1). The increasing current increases the electron flow and increases hydrogen production at the same time. A robust controller is necessary under constant voltage and current with small fluctuations for nonlinear electrolysis load for this reason. The PI controller is proposed in this study for the MPP controller to control the electrolysis load voltage. The PI controller is applied to the P and O MPPT controller output error signals and generates a pulse width modulation (PWM) signal to switch the metal oxide semiconductor fieldeffect transistor (MOSFET) devices. The general view of the proposed photovoltaic-electrolysis system with converter and controller is shown in Fig. 2.
Anode (oxidaon) : 2 H2O(l) →O2(g) + 4H+(aq) + 4e− Eo = +1.23 V + − Eo = Cathode (reducon): 2 H (aq) + 2e →H2(g) (1) →2 H2(g) + O2(g) Eocell= -1.23 V Overall reacon : 2 H2O(l)
directly or converted to electrical energy by fuel cells again for cars similarly. The Toyota Corporation succeeded to produce such a car which is called ‘hydrogen car’, but it used batteries instead of photovoltaic for energy sources [23,32]. Moreover, it is very expensive and does not work efficiently. The common point in these two applications is how the energy will be converted more efficiently from solar energy to hydrogen energy. A buck type DC-DC converter is proposed as a solution for this problem in this paper. The DC-DC buck converter is designed for a 100 W commercial photovoltaic power input. These commercial photovoltaic panels generate 0e25 V output voltages for different loads and currents generally [24]. On the other hand, it is necessary to decrease
0.00
V
The proposed system components and modelling Photovoltaic panel model and control of MPP The photovoltaic sources are not behaving as infinite or linear sources in the application. They behave as finite and nonlinear sources as shown in their characteristic curves. At least 250 W power source is necessary for the designed hydrogen production system, and the necessary photovoltaic
Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
4
international journal of hydrogen energy xxx (xxxx) xxx
Fig. 2 e The proposed photovoltaic-electrolysis converter system.
Table 1 e Electrical characteristics of photovoltaic sources. Specifications Maximum power (Pm) Maximum power voltage (Vpm) Maximum power current (Ipm) Short circuit current (Isc) Open circuit voltage (Voc) Temperature coefficient of Voc Temperature coefficient of Isc Nominal operation cell temperature Cell per module (Ncell)
Values 249.86 W 31 V 8.06A 8.55 A 37.6 V 0.35%/K 0.06%/K 48 C ± 2 C 60
The photovoltaic cells are connected in series and parallel to increase the voltage and current of the photovoltaic module necessary values. Using this equation photovoltaic module model is designed in MATLAB/Simulink and simulated for necessary parameter values. Fig. 3 shows the current-voltage (IeV) characteristic of the photovoltaic source model and the power voltage (PeV) characteristic of the photovoltaic source model. The simulation results of this photovoltaic source model are suitable for the theory as expected. VPV ¼
ISC þ KI T Tref :G þ NP I0 IPV NS :n:k:T NS Rs ln : :IPV q I0 :NP NP Rsh (2)
panel is selected from the manufacturer catalogs [24]. This photovoltaic panel (Trina Solar TSM-250PA05.08) electrical characteristic values are given in Table 1. The photovoltaic panel equation which is derived from the basic Shockley diode equation for all parameters and used in previous studies is given in Equation (2) [6]. The solar cell current depends on cellular operating temperature (T) and the absorption of sunlight (G) are combined with this equation.
The other component of the PV powered electrolyzes system is MPPT, and it is used to obtain maximum power from PV due to its nonlinear characteristics. Perturb and observe (P and O) is one of the MPPT methods used in PV systems [17,18]. The P and O method algorithm and MATLAB/Simulink model are shown in Fig. 4 (a, b). The method measures the PV voltage and currents to calculate the maximum power. If the power last
Fig. 3 e Photovoltaic panels IeV and PeV curves. Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
international journal of hydrogen energy xxx (xxxx) xxx
5
Fig. 4 e (a) The P&O method algorithm, (b) MATLAB/Simulink model of MPPT.
Fig. 5 e Block diagram of the DC-DC converter with a PI controller [16].
Fig. 6 e The buck converter study states (a), current-voltage, and switching signals (b). Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
6
international journal of hydrogen energy xxx (xxxx) xxx
Fig. 7 e The basic experimental set up for alkaline electrolysis system (a), MATLAB/Simulink model of nonlinear electrolyzes model (b).
value is bigger than the previous value compared to the voltage last value with the previous value. If the voltage last value is bigger than the previous value, the duty ratio (D) is increased one step otherwise is decreased one step. In this way, an optimum duty cycle is obtained to drive the buck converter and to work on MPP.
Proportional integral controller (PID) The most commonly used control method is PID for DC-DC converters. It is the most preferred because it can be designed easily and suitable for linear systems and industrial applications [15]. The control signal which is generated by the PID controller for continuous mode operation of the converter depends on the error function given in Equation (3).
Zt uðtÞ ¼ KP eðtÞ þ KI
eðtÞdt þ KD
deðtÞ dt
(3)
0
The general block diagram of the converter with the PI controller is shown in Fig. 5. The error signal which is obtained from the difference of output and reference voltage applied to the PI controller to implement the PWM process for generating the switching signal to drive the converter MOSFETs [16].
DC-DC buck converter The DC-DC buck converter is the main part of the proposed PV sourced electrolysis system to increase the obtained voltage
Fig. 8 e The simulation model of the proposed converter system in MATLAB/Simulink. Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
international journal of hydrogen energy xxx (xxxx) xxx
from the PV panel at electrolyzes voltage level. This converter reduces the 31 V photovoltaic voltage to the range of 2e3 V which is necessary for the electrolysis process. Depend on the voltage value decrease, the current value increases in the same ratio. These high currents occurred some loses on the components naturally. Some solutions are proposed to reduce loses in different papers [12,14].
DIL ¼ Dð1 DÞ
7
A 250 W power DC-DC buck converter is designed to obtain a 2 V output voltage from 31 V DC input voltage. For the converter, circuit capacitor and inductor values were calculated for 100 kHz switching frequency and 5% ripple at an output voltage as Equations (3) and (4). The DC-DC buck converter is designed for electrolyzes load. Inductance maximum ripple value and duty ratio are used to calculate inductance critical value as in Equation (4);
VPV VPV 31 / L ¼ Dð1 DÞ /L > 2:79 mH ¼ 0:1ð1 0:1Þ 100x103 x10 f :L f :DIL
(4)
Output voltage ripple value and inductance ripple values are used to calculate capacitor critical value as in Equation (5); Table 2 e Simulation parameters of the proposed system. Parameters DC input voltage DC output voltage Nominal load current Nominal power Switching frequency Filter inductance (L) Filter capacitance (C) DC Link capacitance Electrolysis initial resistance (Re0) PI parameters PV solar irradiation (G) PV operation temperature (T)
Values 0e37.6 V 2e2.8 V 80 A 250 W 100 kHz 500 mH 10,000 mF 3000 mF 0.08 U Kp ¼ 0.7, Kı ¼ 0.001 1000 W/m2 25 Co
DVC ¼
DIL DIL 10 /C ¼ /C > 1250 mF ¼ 8:f :C 8:f :DVC 8x100x103 x0:1
(5)
The DC-DC buck converter works depending on two states of switching MOSFET. In the first state, the MOSFET switch is turned on and the input voltage is applied to the output and current flows through the inductance, capacitor, and load. In the second state, the MOSFET switch is turned off, the inductance voltage continues to be applied to the output as a voltage source and currently continues to flow through the inductance, capacitor, and load by the freewheeling diode. These states are repeated by the switching cycle and current flows continuously [25e27]. The DC-DC buck converter study
Fig. 9 e (a) The simulation results for PV source voltage, current, and power with the P&O MPPT algorithm. (b) Simulation results for electrolysis load voltage, current, and power. Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
8
international journal of hydrogen energy xxx (xxxx) xxx
Fig. 10 e The inductance current and voltage variation depending on switching PWM signal.
states, voltage and current signals depend on switching signals are given in Fig. 6 (a, b).
Electrolysis load model The electrolysis chemical reaction needs to be perfectly understood before analysing the behaviour of the electrolysis load. The suitable voltage is applied to electrodes placed in the water or liquid solution to start a chemical reaction which is called electrolyze. Hydrogen gas occurred at the cathode and oxygen gas occurred at the anode when the electrolysis
chemical reaction is started. For the water decomposition, 1.23 V voltage is enough theoretically as shown in Equation (1). However, in practical applications depend on electrodes type it can be changed. During the reaction, 1-mol oxygen gas occurs at the anode and 2-mol hydrogen gas occurs at the cathode as shown in Equation (1). The basic experimental set up for the alkaline electrolysis system is shown in Fig. 7(a). The electrolyze behaves like a nonlinear resistance during the chemical reaction in reality. This nonlinearity depends on the chemical composition of the solution, the pressure of electrolyzes stack, temperature and the other parameters
Fig. 11 e The input (a), and the output (b) voltage, current and power variations for load short circuit. The input (c), and the output (d) voltage, current and power variations for load open circuit. Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
9
international journal of hydrogen energy xxx (xxxx) xxx
Fig. 11 e (continued).
which exist during the chemical reaction [34]. The derived nonlinear mathematical model of the electrolyze depends on these parameters is given in Equation (5) [12,14]. b Ie Re ¼ Re0 1 þ a Iebase
(6)
In Equation (5), Reo is initial resistance and its value is 0.08 U. The a and b are constant values that depend on nonlinear characteristics of the electrolyze model. These values are selected between 0.35 and 0.65 for a, and between 2 and 4 for b in typical applications. After these values have been written in Equation (6), the electrolysis load model is obtained as in Equation (7). The electrolysis resistance depends on the electrolysis current is given in Equation (7). The electrolysis nonlinear load model is simulated in MATLAB/ Simulink using this equation as in Fig. 6(b). 3 Ie Re ¼ 0:08 1 þ 0:5 50
(7)
General system modelling The proposed photovoltaic sourced DC-DC buck converter for electrolyzes load with MPPT and PI controller was simulated in MATLAB/Simulink software. The simulation model of the proposed converter system is shown in Fig. 8. The photovoltaic source model and electrolysis load model was used in this converter model instead of the linear source and load. The PV voltage and current are applied to a P&O method for the MPPT algorithm and the obtained a duty signal. The PI control uses this signal and generates a PWM signal to drive the MOSFET switching. The simulation parameters of the proposed system model in Fig. 8 are given in Table 2. The simulation results are obtained for these parameters.
Simulation results Simulation results of the proposed photovoltaic sourced DCDC buck converter with electrolysis load are given in this
Table 3 e The efficiency comparison table for different solar irradiation and temperature levels. 0C
Temperature
25C
50C
75C
Solar Irradiation PPV (W) Pload(W) Efficn. PPV (W) Pload(W) Efficn. PPV (W) Pload (W) Efficn. PPV (W) Pload (W) Efficn. 100 W/m2 200 W/m2 300 W/m2 400 W/m2 500 W/m2 600 W/m2 700 W/m2 800 W/m2 900 W/m2 1000 W/m2
11.72 30.36 52.18 76.38 102.6 130.4 159.7 190.2 222 254.4
9.4 25.68 45.09 66.85 90.55 115.9 142.7 170.6 199.7 229.4
80% 84.6% 87.9% 87.5% 88.3% 88.9% 89.4% 89.7% 89.9% 90.2%
11.98 30.99 53.23 77.90 104.5 132.7 161.9 191.7 221.1 248.8
9.61 26.23 46.04 68.23 92.33 118 144.7 172 199 224.8
80.2% 84.6% 86.5% 87.6% 88.4% 88.9% 89.4% 89.7% 90% 90.4%
12.23 31.6 54.19 79 105.1 131.3 156.3 179 201.4 223.6
9.83 26.77 46.9 69.23 92.9 116.8 139.7 160.1 183 201.3
80.4% 84.7% 86.5% 87.6% 88.4% 88.9% 89.4% 89.4% 90.9% 90%
12.47 32.01 54.09 76.59 97.53 117.6 138.4 157.9 177.8 197.9
10.4 27.13 46.82 67.11 86.63 103.6 123 138.3 165.6 177.8
83.4% 84.7% 86.6% 87.6% 88.8% 88.1% 88.9% 87.6% 93.2% 89.9%
Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
10
international journal of hydrogen energy xxx (xxxx) xxx
Fig. 12 e The efficiency comparison with solar irradiation in different temperatures. chapter. The simulation results were tried for different photovoltaic source and load conditions. The photovoltaic array is simulated for 250 W power photovoltaic panel characteristics which are given in Table 1. The simulation results were given in Fig. 3. The main confidant of this figure how to convert the 8 A current to 80 A and 31 V voltage to 2e3 V at the output of the converter. Using the P and O MPPT algorithm and buck converter this request is achieved. So, the simulation studies are focus on this subject. The simulation results for PV source with P&O MPPT algorithm and PI control are given in Fig. 9 (a). The simulation results show that the MPPT algorithm works correctly and the power, current and voltage values are stable around the MPP as expected in Table 1 and Fig. 3. These provide the maximum power from the PV panel and work efficiently. On the other side, simulation results for electrolysis load voltage, current, and power are given in Fig. 9 (b). The electrolysis load voltage reaches 2 V in 1 s with small ripples. The electrolysis load current and power reaches 80 A and 230 W respectively in 2 s. This shows that the system works more than 90% efficiency. The power losses on the system are originated from losses on the buck converter, and small shifts on the MPP. Nevertheless, this output voltage and power are enough for the efficient electrolysis chemical reaction. Different solutions can be proposed and studied to increase efficiency. The inductance current and voltage variation show the converter working conditions depend on the generated switching PWM signal from the controller. The inductance current, voltage, and PWM signal variation are given in Fig. 10. When the MOSFET switch is turn-on with switching signal, the current flows through the inductance and the voltage is positive, when the MOSFET switch is a turn-off, the current continuous flow decreasing and the inductance voltage is negative. Therefore, the converter works in continuous mode as expected in theoretical results in Fig. 6. The converter system is simulated for short and open circuit inrush conditions. The PV input and electrolysis load output voltage, current and power variations are observed for
the 1-s short circuit at firstly. The results are reaching the first expected values in a few seconds after the short circuit. These results are seen in Fig. 11(a and b). The output voltage, current and power variations are observed for the 1-s open circuit at the electrolysis load secondly. The results are reaching the first expected values in a few seconds. Depending on energy storage components on the converter, the first response shows an overshoot after the open circuit. These results are shown in Fig. 11(c and d). These results show that the simulation model with the PI controller is works correctly and stable for nonlinear input and output components. According to these results, it is possible to say the amount of the hydrogen gas produced from 250 W power PV panels in 1 h is 43.2 cm3 or 43.2 L theoretically through DC-DC buck converter and suitable electrolyzes devices. These results are calculated using the previous studies in Ref. [12]. Also, a detailed simulation made on system efficiency investigation for ten different solar irradiation levels and four temperature levels. The simulation results are given numerically and the efficiency is calculated in Table 3 forever weather conditions. These numerical results are compared in Fig. 12. These results show that the efficiency is more than 90% percent for high solar radiation because of the converter works full capacity in these conditions. However, the hightemperature effects the efficiency negatively, but increasing the solar irradiation balanced it.
Conclusions This paper focuses on two different applications to use PV energy and storage of this energy as hydrogen. Photovoltaic panels' electrical energy was converted to the desired voltage level using the DC-DC buck converter for generating hydrogen with electrolysis process. The P&O MPPT algorithm used with the PI controller in this study to observe the PV energy more efficiently and to convert its desired voltage level for the electrolysis process. The photovoltaic powered DC-DC buck
Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
international journal of hydrogen energy xxx (xxxx) xxx
converter for electrolysis load was simulated in MATLAB/ Simulink software more detailed using this controller. The simulation results are investigated for load voltage, current, and power with photovoltaic sources. As well as, this result was repeated for inrush conditions such as short circuit load and open-circuit load and the system stabilization is observed. These simulation results are suitable for theoretical results as expected. The efficiency comparison is made for different solar irradiation and temperature levels. The efficiency of the proposed system is reached more than 90% for high irradiance levels. The simulation results can be compared for different controllers with MPPT systems using more efficient converters as a future study. Also, this study aims to realized experimentally with research project support to improve the idea.
Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.ijhydene.2020.01.162.
references
[1] Guo Y, Tang J, Henzie J, Jiang B, Qian H, Wang Z, Yamauchi Y. Assembly of hollow mesoporous nanoarchitectures composed of ultrafine Mo 2 C nanoparticles on N-doped carbon nanosheets for efficient electrocatalytic reduction of oxygen. Materials Horizons 2017;4(6):1171e7. [2] Tan H, Tang J, Henzie J, Li Y, Xu X, Chen T, Yamauchi Y. Assembly of hollow carbon nanospheres on graphene nanosheets and creation of ironenitrogen-doped porous carbon for oxygen reduction. ACS Nano 2018;12(6):5674e83. [3] Melaina M, Eichman J. Hydrogen energy storage: grid and transportation services, technical report NREL/TP-540062518. 2015. Internet, http://www.nrel.gov/docs/fy15osti/ 62518.pdf. [4] Jain IP. Hydrogen the fuel for 21st century. Int J Hydrogen Energy 2009;34(17):7368e78. [5] Panwar NL, Kaushik SC, Kothari S. Role of renewable energy sources in environmental protection: a review. Renew Sustain Energy Rev 2011;15(3):1513e24. _ Physical structure, electrical design, [6] S‚ahin ME, Okumus‚ HI. mathematical modeling and simulation of solar cell modules. Turkish Journal of Electromechanics & Energy 2015;1(1):1e8. [7] Wild M, et al. From dimming to brightening: decadal changes in solar radiation at Earth's surface. Science 2005;308(5723):847e50. [8] Winter C-J. Hydrogen energydabundant, efficient, clean: a debate over the energy-system-of-change. Int J Hydrogen Energy 2009;34(14):1e52. [9] Laurencelle F, Chahine R, Hamelin J, Agbossou B, Fournier M, Bose TK. Characterization of Ballard MK5-E proton Exchange membrane fuel cell stack. J. Fuel Cells e Fundam Syst 2001;1:66e71. [10] Fahmy FH, Abdel-Rehim ZS. Hydrogen gas production, and utilization as electricity using a renewable energy source. Energy Sources 1999;21(7):629e41. [11] Khaselev O, Bansal A, Turner JA. High-efficiency integrated multijunction photovoltaic/electrolysis systems for hydrogen production. Int J Hydrogen Energy 2001;26(2):127e32.
11
_ Aydemir MT. Implementation of an [12] S‚ahin ME, Okumus‚ HI, electrolysis system with DC/DC synchronous buck converter. Int J Hydrogen Energy 2014;39(13):6802e12. s A, et al. Combined maximum power point tracking [13] Garrigo and output current control for a photovoltaic-electrolyzer DC/DC converter. Int J Hydrogen Energy 2014;39(36):20907e19. [14] Sahin ME, Okumus HI. Fuzzy logic controlled parallel connected synchronous buck DC-DC converter for water electrolysis. IETE J Res 2013;59(3):280e8. [15] Mohan BM, Sinha A. Mathematical models of the simplest Fuzzy PI/PD controllers with skewed input output fuzzy sets. ISA Trans 2008;47:300e10. _ Comparison of different controllers [16] S‚ahin ME, Okumus‚ HI. and stability analysis for photovoltaic powered buck-boost DC-DC converter. Elec Power Compon Syst 2018;46(2):149e61. [17] Haque A. A fast and reliable perturb and observe maximum power point tracker for solar PV system. International Journal of System Assurance Engineering and Management 2017;8(2):773e87. [18] Ebrahim MS, Sharaf AM, Atallah AM, Emarah AS. Modified P&O technique for hybrid PV-battery smart grid integrated scheme. Turk. J. Electrom. Energy 2018;3(2):1e7. [19] Sherif SA, Barbir F, Veziroglu N. Towards a hydrogen economy. Electr J 2005:62e76. [20] Gregorio M, Teresa V-S. Towards the hydrogen economy? Int J Hydrogen Energy 2007;32(12):1625e37. [21] Balat M, Ozdemir N. New, and renewable hydrogen production processes. Energy Sources 2005;27(13):1285e98. [22] The solar hydrogen cycle. Schatz Energy Research Centre; 2017. Internet, http://www.schatzlab.org/projects/hydrogen/ solarh2cycle.html. [23] Walters G. The car that runs on FRESH AIR, Mail-Online, 9 May 2015. 2017. Internet, http://www.dailymail.co.uk/ sciencetech/article-3074298. [24] 100W PV module datasheet. Solartech Power Inc., SPM 100PTS-N MODULE; 2016. Internet, http://www.solartechpower. com/SPM100PN.html. [25] Erickson RW, Maksimovic D. Chapter 8: converter transfer functions, fundamentals of power electronics. Springer Science & Business Media; 2007. [26] Mohan N, Undeland TM. Power electronics: converters, applications, and design. John Wiley & Sons; 2007. [27] Rashid MH. Power electronics handbook: devices, circuits, and applications. Academic Press; 2010. [28] Smaoui M, Lotfi K. Design and energy control of stand-alone hybrid wind/photovoltaic/fuel cell power system supplying a desalination unit. J Renew Sustain Energy 2014;6(4):043111. [29] Nikolaidis P, Andreas P. A comparative overview of hydrogen production processes. Renew Sustain Energy Rev 2017;67:597e611. [30] Hollmuller P, et al. Evaluation of a 5 kW photovoltaic hydrogen production and storage installation for a residential home in Switzerland. Int J Hydrogen Energy 2000;25(2):97e109. [31] Ni M, et al. A review and recent developments in photocatalytic water-splitting using TiO2 for hydrogen production. Renew Sustain Energy Rev 2007;11(3):401e25. [32] Firat Y. Utility-scale solar photovoltaic hybrid system and performance analysis for eco-friendly electric vehicle charging and sustainable home. Energy Sources, Part A Recovery, Utilization, and Environmental Effects 2018;41(6):734e45. [33] Lin MY, Hourng LW, Wu CH. The effectiveness of a magnetic field in increasing hydrogen production by water electrolysis. Energy Sources, Part A Recovery, Util Environ Eff 2017;39(2):140e7.
Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162
12
international journal of hydrogen energy xxx (xxxx) xxx
[34] Shen M, et al. A concise model for evaluating water electrolysis. Int J Hydrogen Energy 2011;36(22):14335e41. [35] Abdel-Aal HK, Mohamed MA. Potentials of storing solar energy in the form of hydrogen for Egypt. Energy Sources 1989;11(2):95e103. lu TN. S‚ahin S., 21st Century's energy: hydrogen [36] Vezirog energy system. Energy Convers Manag 2008;49(7):1820e31. [37] Guo Y, Park T, Yi JW, Henzie J, Kim J, Wang Z, Yamauchi Y. Nanoarchitectonics for transition metal sulfide based electrocatalysts for water splitting. Adv Mater 2019;31(17):1807134. [38] Guo Y, Tang J, Qian H, Wang Z, Yamauchi Y. One-pot synthesis of zeolitic imidazolate framework 67-derived hollow Co3S4@ MoS2 heterostructures as efficient bifunctional catalysts. Chem Mater 2017;29(13):5566e73. [39] Guo Y, Tang J, Wang Z, Kang YM, Bando Y, Yamauchi Y. Elaborately assembled core-shell structured metal sulfides as a bifunctional catalyst for highly efficient electrochemical overall water splitting. Nanomater Energy 2018;47:494e502.
[40] National Renewable Energy Laboratory (Nrel). Current state of the art hydrogen production cost estimate using water electrolysis. Independent Review; September 2009. Document Number: NREL/BK-6A1-46676. [41] Sahin ME. An efficient solar-hydrogen DC-DC buck converter system with sliding mode control. El-Cezerıˆ Journal of Science and Engineering 2019;6(3):558e70. [42] Muhida R, et al. A maximum power point tracking for photovoltaic-SPE system using a maximum current controller. Sol Energy Mater Sol Cell 2003;75(3e4):697e706. [43] Dahbi S, et al. Optimised hydrogen production by a photovoltaic-electrolysis system DC/DC converter and water flow controller. Int J Hydrogen Energy 2016;41(45):20858e66. n DB, Ricardo JM. New concept in [44] Inthamoussou FA, Herna maximum power tracking for the control of a photovoltaic/ hydrogen system. Int J Hydrogen Energy 2012;37(19):14951e8. [45] US Department of Energy. Office of energy efficiency & renewable energy, hydrogen storage. 2019. Internet, www. energy.gov/eere/fuelcells/hydrogen-storage.
Please cite this article as: Ergin S‚ahi_n M, A photovoltaic powered electrolysis converter system with maximum power point tracking control, International Journal of Hydrogen Energy, https://doi.org/10.1016/j.ijhydene.2020.01.162