Assessment of a proposed hybrid photovoltaic array maximum power point tracking method

Assessment of a proposed hybrid photovoltaic array maximum power point tracking method Water Science ScienceDirect Water Science 30 (2016) 108–119 journal homepage www elsevier com/locate/wsj Full Len[.]

Water Science ScienceDirect Water Science 30 (2016) 108–119 journal homepage: www.elsevier.com/locate/wsj Full Length Article Assessment of a proposed hybrid photovoltaic array maximum power point tracking method Yasmin Adel a,b , Rameen Abdelhady a,∗ , Ahmed M Ibrahim b a National Research Center, Ministry of Water Resources and Irrigation, Egypt b Electrical and Machine Power Department, Cairo University, Egypt Received 30 July 2016; received in revised form 14 October 2016; accepted 19 October 2016 Available online December 2016 Abstract Photovoltaic arrays have limited conversion efficiency and thus, a maximum power point tracking technique is essential This makes the maximum power point tracking (MPPT) require prior prediction of the mentioned point in spite of the undeniable changes in the environment In this manuscript an introduction and assessment of the different techniques of MPPT is presented The categorization scheme of the MPPT techniques is according to either the predefinition of operating points without system data update (offline methods) or continuous sampling of system variables, to update the PV module measurements (online methods) Whereas hybrid method is a combination of both A number of techniques from each class were simulated in MATLAB/Simulink environment in order to compare their performance Moreover, the hybrid method was simulated in two successive steps without pre-assumption of the output of the offline method The results demonstrated the relevance of the hybrid method when applied to a photovoltaic system due to its good performance, fast response and less fluctuations, when subjected to sudden climatic changes © 2016 National Water Research Center Production and hosting by Elsevier B.V This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) Keywords: PV systems; MPPT; Online methods; Offline methods; Hybrid methods Introduction Recently the use of solar energy has been emerging The main advantages of photovoltaic (PV) systems are zero greenhouse gas emission, low maintenance costs, fewer limitations with regard to site of installation and absence of mechanical noise arising from moving parts (Reisi et al., 2013) The global PV market reached 173 GW in 2014 However, there are three major limitations in photovoltaic generation systems: the conversion efficiency to electric power is low (9–17%), the variability of electric power generated due to the weather conditions and sunlight hours at ∗ Corresponding author E-mail addresses: yasmin.adel1982@hotmail.com (Y Adel), rameens@hotmail.com (R Abdelhady), drahmed@nahdetmisr.com (A.M Ibrahim) Peer review under responsibility of National Water Research Center http://dx.doi.org/10.1016/j.wsj.2016.10.004 1110-4929/© 2016 National Water Research Center Production and hosting by Elsevier B.V This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) Y Adel et al / Water Science 30 (2016) 108–119 109 Fig Single diode equivalent circuit of a solar cell daytime (Xiao et al., 2006), and the relatively higher cost (12.5 ¢/kWh) as compared to that produced by conventional power generation (7.4 ¢/kWh for natural gas combined cycle) systems or even to other renewable sources such as hydro electrical energy (8.4 ¢/kWh) (Energy Innovation, 2015) A solar cell (also called photovoltaic cell or photoelectric cell) is a solid state device that converts the energy of sunlight directly into electricity by the photovoltaic effect (Zhang et al., 2011) Assemblies of cells are used to make solar modules, also known as solar panels Photovoltaic system uses various materials and technologies such as crystalline Silicon (c-Si), Cadmium telluride (CdTe), Gallium arsenide (GaAs), chalcopyrite films of Copper-Indium-Selenide (CuInSe2 ) Solar cells exhibit a non-linear current–voltage characteristics that depend on solar radiations and temperatures There is a point on the characteristics curve where the output power from the array (a string of panels) has a maximum value Solar cells are usually assessed by measuring the current voltage characteristics of the device under specific conditions of illumination and then determining a set of parameters In order to ensure efficient operation of the solar array the maximum power point (MPP) of the array has to be tracked MPPT, in addition to rising the power delivered from the PV module to the load, is considered as a PV system lifetime booster (Bahgat et al., 2005) 1.1 PV system For simplicity in analyzing characteristics of solar cells, electrical equivalent circuits are used in representing them Researchers used numerous equivalent circuits to help in predicting the behavior under various environmental conditions, and further in obtaining (I–V) and (P–V) characteristic curves The most commonly used equivalent circuits are the single diode model (De Blas et al., 2002) (shown in Fig 1), the double diode model (Cabestany and Castaner, 1983) and the three diode model (Khanna et al., 2015) According to the above circuit the solar cell can be represented by: a current generator, a diode indicating the recombination losses, a shunt resistance symbolizing losses from currents that return across the junction and a series resistance denoting resistance losses By applying a simple Kirchhoff’s current law (KCL); the relation of the current I and the voltage V is given as: I = Iph − Io (e q(V +IRs ) nkT − 1) − V + IRs Rsh (1) where Iph is the photocurrent, Io is the saturation current of the diode, Rs is the series resistance, Rsh is the shunt resistance, n is the diode ideality factor, k is Boltzmann’s constant (1.4 × 10−23 ), q is the electron charge (1.6 × 10−19 ), T is the absolute temperature in Kelvin 110 Y Adel et al / Water Science 30 (2016) 108–119 Fig Typical IV/PV characteristic of a solar cell at kW/m2 and different temperatures Fig PV characteristic of a solar cell at 75 ◦ C and different illumination levels By plotting the I–V curve in the relation in Eq (1), the existence of a unique point is noticed; the spot near the knee of the I–V curve (Fig 2) This is the point at which the product of current and voltage achieves its maximum, which is noticed more on the P–V curves (Figs and 3) This point, as seen from the figures varies according to the environmental conditions (illumination and temperature) Many approaches have been introduced to deduce the MPP from Eq (1) which implies determining the major parameters: the diode saturation current, the series resistance, the ideality factor, the photocurrent and the shunt resistance Since the IV equation is implicit (current exists on both sides of the equation), extracting the parameters by the simple least square method is not possible Also the determination Y Adel et al / Water Science 30 (2016) 108–119 111 Fig System block diagram requires measured IV characteristics at the specified insolation and temperature Zhang et al (2011) used an explicit analytic expression for I with the help of Lambertω function to be able to utilize the conventional curve fitting methods (Eq (2))    b   ((ca) + (da) + V ) (2) I − (1/ (a + b)) V − (b (c + d) / (a + b)) + (e/a) × lambertω ((adb) / (e (ab))) exp e (a + b) where: a = Rs , b = Rsh , c = IL , d = Io and e = (nkT)/q After concluding these parameters the MPP can be deduced using the above equation and the fact that ∂P/∂VV =Vm = at Vmax A practical solution requires the introduction of a tracker which is inserted between the PV system and the load A DC–DC converter is used to manage the power and a controller is introduced within the tracker to compensate the parameter variation due to environmental conditions Various MPPT methods have been developed These methods can be grouped based on different features In essence, MPPT methods are categorized into: offline methods which are dependent on solar cell models, online methods which are usually referred to as the model-free method and hybrid method which represents a sequence of the offline and online methods In this manuscript a hybrid method was assessed by simulation in two successive steps without preassumption of the output of the offline method Also simulation of different techniques under online and offline methods and comparison with the hybrid method is presented Methodology 2.1 System overview In general, any PV system consists of: PV array, DC–DC converter, MPPT controller and a battery The battery is discarded as the main objective is the MPPT of the solar array Fig shows the block diagram of the system used in this simulation model The PV array used in this model is of type Sun Power SPR-305-WHT which consists of 20 parallel strings and series-connected solar modules per string The module characteristics under STC (1 kW/m2 and 25 ◦ C) are: 64.2 V VOC , 5.96 A ISC , 54.7 V VMPP and 5.58 A IMPP The DC–DC converters are used for matching the characteristics of the load with those of the solar panels i.e to balance the system In our model the simulated converter is a kHz–500 V boost converter and a 100 ohm resistance is used as load The control signal generated by several MPPT methods feed the boost converter switch 2.2 The MPPT control In this research comparison of different conventional tracking techniques with the introduced method and assessment of all routines under different climatic conditions, is carried out The radiation and rapid change in radiation (shadowing) were taken into account The temperature was discarded avoiding complexity For a given solar radiation, when the cell temperature increases, the Voc , drops slightly, while the Isc current increases considerably which makes the total effect marginal (Salmi et al., 2012) 112 Y Adel et al / Water Science 30 (2016) 108–119 The MPPT methods were simulated in MATLAB/Simulink software environment MATLAB/Simulink is selected, due to its reusability, extendibility, and flexibility in such systems Among the offline techniques simulated in this research were the open circuit voltage (OCV) and the short circuit current (SCC) techniques discussed below MPPT based on the online methods of perturb and observe (P&O) and incremental conductance (IncCond) also have been performed to compare between offline and online techniques in general The proposed hybrid method is then simulated, the control signals consists of two stages: set point figuring and tuning The first stage is the set point loop which approximates the maximum power by the VOC offline method The next stage is using the input from the VOC method as an initial condition to the P&O online method, which is the second phase of tracking the MPP (fine tuning loop) 2.2.1 Offline methods Offline methods generally require the knowledge of one or more of the solar panel parameters values, such as the open circuit voltage (VOC ), short circuit current (ISC ), temperature and radiation These values generate the control signal necessary for driving the solar cell to its MPP In the tracking operation, this control signal remains constant if ambient conditions can be regarded as fixed and there are no attempts to regulate the output power of the PV system (Reisi et al., 2013) Offline techniques include: open circuit voltage method (OCV), short circuit current method (SCC) and artificial intelligence (AI) The OCV technique is one of the most straightforward off line methods It uses the almost linear relationship between the open circuit voltage (VOC ) and the MPP voltage (VMPP ) under changed climatic circumstances as described by the following equation: VMPP = k × VOC (3) where k is a constant, depending on the solar cell characteristics The constant k is derived empirically after measuring several VOC and VMPP for any specific cell under different climatic conditions Values for the constant k, in general, vary between 0.73 to 0.80 (Schoeman and Wyk, 1982), here in our specific case, the constant k is assumed to be 0.8 Despite the simple implementation (has to with hardware used to implement this method) and low costs, this technique suffers from two main drawbacks: the MPP tracking is not precise and measurement of VOC necessitates periodic shedding of the load There can be no need to shed the load in order to measure the VOC as pilot cells (Brunelli et al., 2009), whose characteristics represent those of the original PV array, might be used Another offline technique is the SCC which depends on the fact that IMPP is approximately linearly related to the ISC of the PV array Although this method is more accurate than the OCV approach yet the implementation is more complex and the periodic shedding of the load or the exploitation of pilot cells, to measure the ISC has not been avoided The third offline technique is AI which has several disciplines under it including artificial neural networks (ANNs) and fuzzy logic (FL) 2.2.2 Online methods In online methods, also known as model-free methods, usually the instantaneous values of the PV output voltage or current are used to generate the control signals (Reisi et al., 2013) The techniques under the online methods are based on the principle of the optimal control theory These techniques include perturbation and observation technique (P&O), and the incremental conductance technique (IncCond) The P&O technique, is one of the most simple online methods which, has been taken into account by a number of researchers (Wasynczuk, 1983; Hua and Lin, 1996; Hua et al., 1998) P&O is implemented by applying small constant perturbations to the voltage or the current signal of the solar panel (Vref , Iref ) Its working principle depends on the sampling method The array voltage here is decreased by increasing the duty cycle of the boost converter and vice versa The duty cycle of the boost converter is between and and the increment used to decrease or increase the voltage is in the order of 10−4 from literature After each perturbation, the output power is measured If the output power is greater in value than the power of the preceding step, power will move toward the MPP (the left side of the PV curve), therefore a voltage perturbation of the same sign must be applied in the succeeding step A smaller value of power, on the contrary, infers that power has deviated from MPP (the right side of the PV curve), and a perturbation of opposite sign will have to be applied Y Adel et al / Water Science 30 (2016) 108–119 113 Fig Flowchart of P&O technique (Reisi et al., 2013) These continuous perturbations eventually cause the approach to the MPP This flowchart of this procedure is shown in Fig The P&O technique has two drawbacks: the continuous perturbations causes the oscillation around the MPP and never reaching it actually Also P&O technique can fail under varying atmospheric conditions (Reisi et al., 2013) In the P&O the value of the perturbations applied to the system is the main factor controlling the convergence of the output power to the MPP In the incremental conductance technique as the algorithm takes two samples of voltage and current to compute MPP This technique is based on the fact that the slope of the PV array power curve is zero at the MPP, positive for values of output power smaller than MPP, and negative for values of the output power greater than MPP (Irisawa et al., 2000; Hohm and Ropp, 2003; Koizumi and Kurokawa, 2005; Harada and Zhao, 1993) The main limitation of the IncCond method is that it obliges complex control circuitry which might have resulted in a high cost system previously 2.2.3 Hybrid method The hybrid method is a combination of both the above methods It is divided in two successive signals: the first depending on a not accurate offline prediction of the MPPT taking into consideration the atmospheric conditions then applying one of the online techniques, following the same workflow in Fig 5, to accurately deduce the MPPT This method eliminated the drawbacks of the other two 114 Y Adel et al / Water Science 30 (2016) 108–119 Fig Output generated power from PV system without MPPT control at different radiation levels In this manuscript the hybrid method is applied, two control signals are generated not just a pre-assumption of the output of the offline method The first signal is generated from the OCV technique and the second from the P&O technique following the workflow in Fig This combination eliminated the disadvantages of both techniques: the inaccuracy of the OCV technique, since the maximum point reached is not the final and misestimating the initial point (Vref ) in the P&O technique, as it is the output from the first technique Also, comparison of different conventional tracking techniques with the introduced method and assessment of all methods under different climatic conditions, is proposed (VOC technique) (P&O technique) Results MPPT is an essential component in the PV system since it allows an increase in the power delivered from the PV module to the load, and it also boosts the functioning lifetime of the PV system Fig shows the PV generated power performance when there is no MPPT control technique applied to the system In the figure the power generated diminishes to zero for an increment of a second (reaching 0.1 s for 1000 W/m2 ), the generated power from the array then rises to a level not more than one tenth of the MPP in high radiation levels Comparison of the efficiency of different MPPT methods under different techniques and the proposed technique was carried out at constant temperature (25 ◦ C) and three variations in radiation level (250, 750, 1000 W/m2 ) The efficiency was calculated by the following formula shown in Eq (4):   Pmax.Tracked × 100 η= (4) Pmax where η is the efficiency of the tracking method, Pmax Tracked is the MPP tracked and Pmax is the MPP as measured in data sheet of the array Simulation results of the OCV technique and the SCC technique (Figs and 8) show that the OCV method maintains a range of 85% efficiency at high illumination levels yet the efficiency drops slightly to 82% at low illumination levels On the other hand, SCC method exhibits an extremely low efficiency at low illumination levels For the online techniques (Figs and 10); the efficiency of the P&O reached 97% The P&O method’s efficiency showed direct proportionality to the illumination levels, reaching 92% at 250 W/m2 The IncCond method showed the Y Adel et al / Water Science 30 (2016) 108–119 115 Fig Output generated power from PV system with OCV MPPT control at various radiation levels Fig Output generated power from PV system with SCC MPPT control at various radiation levels highest efficiency which reached 99% at low illumination levels The introduced hybrid method showed the highest efficiency of all reaching 99%, since the tracking method is a mingle of both methods: the online and offline (Fig 11) Figs 7–9 show fluctuations since the simulation duration is s For some techniques it takes more time than others to reach steady state As for the P&O and IncCond techniques the fluctuations exist regardless their high efficiency 116 Y Adel et al / Water Science 30 (2016) 108–119 Fig Output generated power from PV system with P&O MPPT control at various radiation levels Fig 10 Output generated power from PV system with IncCond MPPT control at various radiation levels Y Adel et al / Water Science 30 (2016) 108–119 Fig 11 Output generated power from PV system with hybrid MPPT control at various radiation levels Fig 12 Output generated power from PV system with hybrid MPPT control at rapid change in radiation level 117 118 Y Adel et al / Water Science 30 (2016) 108–119 since in these techniques the maximum point is never reached but it is oscillated around The hybrid method is the most stable as shown in Fig 11 The above methods were also tested under rapid change in radiation (from 1000 to 200 W/m2 ); to depict shadowing for about 0.25 s The results showed that all the methods maintained the MPP reached at 1000 W/m2 except for the P&O technique, whose efficiency decreased to 92% Power loss is observed due to perturbation; this method also the fails to track the maximum power under fast changing atmospheric conditions Fig 12 shows the effect of shadowing when the introduced hybrid method is applied The system adapted almost instantaneously to the new radiation drop without time to reach steady state The figure shows the lack of noise in the curve, although no filter was used with any of the control methods Conclusion In this manuscript, a number of MPPT techniques have been modeled and their efficiencies were compared after simulating them under different climatic conditions These methods have been categorized into three groups: off-line, online and hybrid methods, which uses signals from both techniques The results indicate that the efficiency of the online methods (97%) exceeded that of the offline (85%) and the introduced technique surpassed them both (99%) Practically speaking: low cost, low hardware requirements and easy implementation have to be put into consideration However, the hybrid method has the best performance, it also showed to maintain its efficiency in case of shadowing conditions It is recommended to further study it taking temperature into consideration The proposed method shows good efficiency and better performance at startup and reacts well to fast changes in climatic conditions Conflict of interest There is no conflict of interest References Bahgat, G.B., Helwa, N.H., Ahmad, G.E., El Shenawy, E.T., 2005 Maximum power point tracking controller for PV systems using neural networks Renew Energy 30, 1257–1268 Brunelli, D., Moser, C., Thiele, L., 2009 Design of a solar harvesting circuit for battery less embedded systems IEEE Trans Circuits Syst I: Regul Pap 56 (11) Cabestany, J., Castaner, L., 1983 Evaluation of solar cell parameters by nonlinear algorithms J Phys D: Appl Phys 16, 2547–2558 De Blas, M.A., Torres, J.L., Prieto, E., Garcia, A., 2002 Selecting a suitable model for characterizing photovoltaic devices Renew Energy 25, 371–380 Energy Innovation, Policy & Technology, 2015 Comparing the Costs of Renewable and Conventional Energy Sources, Available at: http://energyinnovation.org/2015/02/07/levelized-cost-of-energy/ (Accessed July 2016) Harada, K., Zhao, G., 1993 Controlled power interface between solar cells and AC source IEEE Trans Power Electron 8, 654–662 Hohm, D.P., Ropp, M.E., 2003 Comparative study of maximum power point tracking algorithms Prog Photovolt 11, 47–62, http://dx.doi.org/10.1002/pip.459 Hua, C., Lin, J.R., 1996 DSP-based controller application in battery storage of photovoltaic system August In: Proceedings of the 1996 IEEE IECON 22nd International Conference on Industrial Electronics, Control, and Instrumentation, vol 3, pp 1705–1710, IEEE Hua, C., Lin, J., Schen, C., 1998 Implementation of a DSP-controlled photovoltaic system with peak power tracking IEEE Trans Ind Electron 45 (1), 99–107 Irisawa, K., Saito, T., Takano, L., Sawada, Y., 2000 Maximum power point tracking control of photovoltaic generation system under non-uniform insolation by means of monitoring cells Photovoltaic Specialists Conference, 2000 Conference Record of the Twenty-Eighth IEEE, 1707–1710, IEEE Khanna, V., Das, B.K., Bisht, D., Singh, P.K., 2015 A three diode model for industrial solar cells and estimation of solar cell parameters using PSO algorithm Renew Energy 78 (6), 105–113 Koizumi, H., Kurokawa, K., 2005 A novel maximum power point tracking method for PV module integrated converter Power Electronics Specialists Conference, 2005 PESC’05 IEEE 36th, 2081–2086, IEEE Reisi, A.R., Moradi, M.H., Jamasb, Sh., 2013 Classification and comparison of maximum power point tracking techniques for photovoltaic system: a review Renew Sustainable Energy Rev 19 (C), 433–443 Salmi, T., Bouzguenda, M., Gastli, A., Masmoudi, A., 2012 MATLAB/Simulink based modelling of solar photovoltaic cell Int J Renew Energy Res (2) Y Adel et al / Water Science 30 (2016) 108–119 119 Schoeman, J.J., Wyk, J.D., 1982 A simplified maximal power controller for terrestrial photovoltaic panel arrays In: Power Electronics Specialists Conference, June, pp 361–367, IEEE Wasynczuk, O., 1983 Dynamic behavior of a class of photovoltaic power systems IEEE Trans Power Appar Syst 102 (9), 3031–3037 Xiao, W., Lind, M.G.J., Dunford, W.G., Chapel, A., 2006 Real-time identification of optimal operating points in photovoltaic power systems IEEE Trans Ind Electron 53 (4), 1017–1026 Zhang, C., Zhang, J., Hao, Y., Lin, Z., Zhu, Ch., 2011 A simple and efficient solar cell parameter extraction method from a single current voltage curve J Appl Phys 110 (6), 0645041–0645047, http://dx.doi.org/10.1063/1.3632971, Available at: http://scitation.aip.org/content/aip/journal/jap/110/6/10.1063/1.3632971 (Accessed December 2013)  ... is a point on the characteristics curve where the output power from the array (a string of panels) has a maximum value Solar cells are usually assessed by measuring the current voltage characteristics... DSP-controlled photovoltaic system with peak power tracking IEEE Trans Ind Electron 45 (1), 99–107 Irisawa, K., Saito, T., Takano, L., Sawada, Y., 2000 Maximum power point tracking control of photovoltaic. .. as measured in data sheet of the array Simulation results of the OCV technique and the SCC technique (Figs and 8) show that the OCV method maintains a range of 85% efficiency at high illumination