Design of Fuzzy Logic Based Photovoltaic Fed Battery Charging … · 2019. 9. 10. · proposed intelligent and robust control technique based battery charging circuit fed by photovoltaic

Journal of Green Engineering, Vol.9 2, 27 - 281, Alpha Publishers This is an Open Access publication. © 2019 the Author(s). All rights reserved.

Design of Fuzzy Logic Based Photovoltaic Fed Battery Charging System

1Polamraju.V.S.Sobhan, 2M.Subba Rao, 3N. Bharath Kumar, 4A. Sriharibabu

Abstract

The battery energy storage systems emerged as the best option among the different strategies developed for meeting energy demand, effective utilization of energy and its conservation. The PV system supply energy under consistent working condition and with aid of battery storage system it can supply under transient condition also. The proposed intelligent and robust control technique based battery charging circuit fed by photovoltaic (PV) system, consists two power electronic converters boost and buck converters. The intelligent MPPT controller using fuzzy logic is employed to extract maximum power through boost converter which increases the variable PV voltage level. The output of the boost converter is fed to buck converter to maintain constant voltage suitable for the battery charging using another fuzzy logic controller. Fuzzy control makes the controlled system robust under the presence of system uncertainties and external disturbances such as variable irradiance and temperature. The MATLAB based simulation results show the effectiveness of proposed scheme such as higher efficiency, lower cost, reduced battery losses and enhancing life cycle. Keywords:Fuzzy control, MPPT, Battery charging system, Boost converter, Buck converter. 1 Introduction Energy is universally perceived as one of the most critical inputs for economic development and human advancement. The progression of an economy with its

1Associate Professor, Department of Electrical and Electronics Engineering, Vignan’s Foundation for Science, Technology and Research,Vadlamudi, Andhra Pradesh, India. E-mail: [email protected] 2Assistant Professor, Department of Electrical and Electronics Engineering, Vignan’s Foundation for Science, Technology and Research,Vadlamudi, Andhra Pradesh, India. E-mail: [email protected] 3Assistant Professor, Department of Electrical and Electronics Engineering, Vignan’s Foundation for Science, Technology and Research,Vadlamudi, Andhra Pradesh, India. E-mail: [email protected] 4Assistant Professor, Department of Electrical and Electronics Engineering, Vignan’s Foundation for Science, Technology and Research,Vadlamudi, Andhra Pradesh, India. E-mail: [email protected]

0

Journal of Green Engineering (JGE)

Volume-9, Issue-2, August 2019

271 Polamraju.V.S.Sobhan et. al.

worldwide competitiveness, depend entirely on the availability of cost effective and ecologically pleasant energy sources. The negative features of fossil fuels such as scarcity, high prices and global warming, are the main driving force behind the search of other alternative energy sources. The solar power is the major source in the modern world because of its environmental friendly characteristics, abundant in volume and easily accessible nature. PV based energy storage systems are alternative to the applications operating the transient condition. In remote areas the batteries are used as energy storage devices find applications in home lighting, battery-powered electric vehicle, and as a backup source in home and offices in addition to applications in power distribution systems such as load leveling, frequency regulation and power quality [1]. In the energy conversion system from solar energy, the photovoltaic (PV) cell is the fundamental element and its voltage, current and power characteristics rely upon solar energy and atmospheric temperature. Basically the PV system consists of photovoltaic source, power electronic converter and load [2]. To extract maximum power continuously from PV source and transfer to load an algorithm is required, called as maximum power point tracking (MPPT) algorithm. The MPPT algorithm considers load changes, insolation variation and cell temperature, and always ensure maximum efficiency and smooth power transfer.[3] A broad range of MPPT methods is presented in literature, for example Perturb and Observe (P&O), Incremental Conductance (IC) and nonlinear control method, sliding mode control (SMC), ANN based methods [4]. The method P&O fall short to follow the MPP under fast environmental changes, and the IC method oscillates around the MPP [5-6]. The ANN based MPPT algorithm needs large samples of training data to train, which restricts the computation time. The SMC is used for MPP tracking in photovoltaic systems suffers from the drawback of chattering phenomenon which causes high frequency oscillations around MPP. Constant voltage and constant current methods are the widely employed in battery charging systems using controlled buck converter. The other methods are PID controller and neural network based battery-charging circuit [7-8]. The drawback of a majority of the NN based methods is that the operating point changes quickly under disturbance conditions. Fuzzy control is an intelligent control strategy utilizes the advantage of the nonlinear nature of the system and achieves the objective using the knowledge of human experience. This robust controller performs well for the systems with modeling uncertainties and external disturbances, and also overcomes the disadvantage of conventional control methods which only can be applied to systems with the mathematical model. This paper deals with the design and analysis of two fuzzy logic based systems, one for to extract as much power as possible from the PV array with boost converter and another for to maintain constant voltage suitable to charge the battery using buck converter [9-11]. The organization of the paper is as follows, the Section 2, focuses on modeling of the PV array. Section 3 explains the proposed battery charging system and in section 4, the design of fuzzy logic (FL) control based MPPT controller through boost converter. It also presents voltage controller for the buck converter to charge battery based on fuzzy control. Sections 5 demonstrate the results of simulation and are analyzed to validate the effectiveness of designed fuzzy controllers. The conclusions are given in the last section.

Design of Fuzzy Logic Based Photovoltaic Fed Battery Charging System 272

2 Photovoltaic Solar Energy System Photovoltaic (PV) technology converts the solar energy into electrical energy through a network of many PV cells. Series arrangement of cells increases the generated voltage level and parallel arrangement increases, the current supplied by PV source. Ideally the PV cell can be viewed as a circuit with parallel branches of p-n junction diode and a current source. The model of an ideal PV cell with parasitic resistances is depicted in figure 1.

Figure 1 Solar cell circuit representation

The dynamic model of PV array obtained from the circuit is given in (1) which explains the current – voltage characteristic with npparallel and ns series solar cells.

shs

PVsPVAkTn

IRVq

oPhpPVRn

IRVIInI e s

PVsPV )(1

)(

(1)

Where VPV and IPV are PV generated voltage and current, Rs and Rsh are

parasitic resistances, and Io is the diode dark current. Other quantities are q =1.6×10−19 C, A is unit less factor depends on junction material, k =1.38×10−23 J/K and T is cell temperature (K).

The variation in photo current, Iph in terms of the two important variables

insolation G and temperature T, is expressed as,

)(_ STCiSTCPh

STC

Ph TTKIG

GI

(2)

Where IPh_STC ,TSTC and GSTC are the photo-current, temperature and insolation at

STC. The Ki is a constant given by manufacturers. The ratings of the PV module considered (CENTSYS 120W) at 25oC, 1000W/m2

from the datasheet are Pmax= 120W, Voc= 22V, Isc= 7.06 A, Vmax = 18V and Imax= = 6.67 A.

Figure 2 demonstrate the impact of differing climate conditions on the MPP locations in P-V characteristic. Figure 2(a) demonstrates the variation between terminal voltage (VPV) and the PV power generated when there is an increase in the insolation.

273 Polamraju.V.S.Sobhan et. al.

(a)

(b)

Figure 2 P-V curves (a) Variable G (b) Variable T

It is can be understood that the reduction in G, causes a decrease in the generated PV power. As shown in the Figure 2(b), at G = 1000 W/m2 and rising temperature conditions, the Voc decreases. Figure 3 depicts the components of proposed PV based battery charging system, consisting of PV array, power electronic converters, battery and controllers.

Figure 3 Layout of Battery Charging system

The operation of proposed battery charging system fed by PV source requires duty ratio control of two power electronic converters, boost and buck converters. In the

Design of Fuzzy Logic Based Photovoltaic Fed Battery Charging System 274

boost converter shown in figure 4(a), output voltage (VDC) is always greater than or equal to PV generated voltage VPV. The voltage level conversion is controlled the switch S1 at a high frequency. The designed parameters are given in table 1

(a) (b) Figure 4 Circuit model of (a) Boost converter (b) Buck converter

Table 1 Parameters of the boost converter

Parameters Value Inductance (L) 8.256 mH Capacitance (CPV) 268 µF Capacitance (CDC) 820 µF Switching frequency (fsw) 10 KHz Input Voltage (VPV) 18 V Output Voltage(VDC) 55 V

In the buck converter shown in figure 4(b), output voltage (VB) which is the

voltage across the battery is always lower than or equal to the input voltage VDC. The voltage regulation is achieved through controlled switching of S2 at a high frequency. The designed parameters are given in table 2.

Table 2 Parameters of the buck converter

Parameters Value Inductance (LB) 469 µH Capacitance (CB) 1818 µF Switching frequency (fsw) 10 KHz Input Voltage (VDC) 55 V Output Voltage(VB) 12V

3 Fuzzy Controller Design In FL based controller the choice of two inputs are error and its change, and single output is the change in duty ratio required to drive boost and buck converters as shown in figure 5. The control mechanism involves four major operations as following: the fuzzification converts crisp values of inputs e(k) and Δe(k) into proper fuzzy values, next operation is continuous interaction with rule base. The fuzzy inference engine decides which rule has to fire based on the current values of inputs and in the final stage the defuzzification is carried out by converting the fuzzy values which are decided by the fuzzy inference engine in to crisp output value that is the duty ratio D of boost or buck converter.

275 Polamraju.V.S.Sobhan et. al.

Figure 5 Fuzzy logic control system

Each input and output variables based on their range are assigned to different fuzzy sets denoted by linguistic variables. Each linguistic variable is defined with a specific membership function. These functions convert the crisp values into fuzzy values. A set of membership functions is defined for five fuzzy variables, NB(Negative-Big) ,NS(Negative-Small), ZE(Zero-Equal), PS(Positive-Small) and PB (Positive-Big) respectively.

3.1 Fuzzy Logic Based MPPT Controller The objective of FL based MPPT controller is to extract as much power as possible from the PV source by maintaining the operating point always at MPP. The output of boost converter which is feeding buck converter is always equal to Vmax.

The two inputs are

)1()()(

)1()(

)1()()(

111

1

kekeke

kVkV

kPkPke

PVPV

PVPV

(3)

The output is change in duty ratio, ΔD1 The duty ratio (D1) is computed from the change in duty ratio (ΔD1) as, D1 (k+1) = D1 (k) +ΔD1 (k) (4) The error determines the operating point location at instant k, whereas the change in error determines the heading of this point. The operating range of two inputs e1(k) and Δ e1(k) is determined from the simulation studies of P-V characteristic curves at different irradiations and temperatures. From the results the ranges of e1(k) and Δ

e1(k) are considered as [−10,10] and [−20,20] respectively. The range of output

variable ΔD1 is considered as [−0.0002; 0.0002]. The membership functions for the fuzzy variables considered for the inputs e1(k), Δ e1(k) and output ΔD1 are given in figure 6.The design of fuzzy rules is explained from the study of the P-V curve. The slope (dP/dV) is zero at MPP, positive in the region 0 to Vmax and negative in the region Vmax to Voc of the curve. According to the desired performances, the curve is partitioned into five regions and each region represents one fuzzy linguistic variable as demonstrated in figure 7.

Design of Fuzzy Logic Based Photovoltaic Fed Battery Charging System 276

(a) (b)

(c)

Figure 6 Fuzzy membership functions (a) e1(k) (b) Δ e1(k) (c) ΔD1(k)

Figure 7 PV curve partition

In the region PB, the slope is positive and the operating point is far from MPP. When the buck converter is operating in this region large increase in the duty ratio is essential to arrive at MPP quickly. In the region PS, the slope is still positive and the operating point is slightly away from MPP. When the buck converter is operating in this region slight increase in the duty ratio is required to reach the MPP quickly without any oscillation. In the region ZE, the slope is nearly zero and operating point very close to MPP. When the buck converter is operating in this region the duty ratio is maintained at the same value. In the region PS, the slope is negative and operating point is slightly away from MPP. When the buck converter is operating in this region slight decrease in the duty ratio is required to reach back to the MPP quickly without any oscillation. In the region PB, the slope is still negative and operating point is far from MPP. When the buck converter is operating in this region large decrease in the duty ratio is required to reach back to the MPP quickly. The corresponding 25 fuzzy rules are shown in table 3.

Table 3 Fuzzy rule matrix for MPPT controller

277 Polamraju.V.S.Sobhan et. al.

4 Result Analysis Figures 9(a) and (b) show voltage and duty cycle of boost converter when irradiance varied from 600 to 800 W/m2 at 1 sec to 1000 W/m2 at 2 sec and finally to 400 W/m2 at 3 sec keeping the temperature at 25oC. Figure 9(a), demonstrates that the FL based MPPT controlled boost converter output voltage lies within the range of Vmax at corresponding irradiance under variable irradiance conditions. The generated duty ratio to drive boost converter is shown in figure 9(b) which demonstrates the satisfactory performance of the designed controller under disturbance conditions.

(a)

(b)

Figure 9 Boost converter (a) Output voltage (b) Duty cycle

The output of buck converter which is fed to battery for charging and

corresponding current is shown in figure 10(a). Irrespective of the varying irradiance the proposed fuzzy logic based voltage controller maintains constant voltage i.e 12 V suitable for the battery. The current drawn by battery depends on the power supplied by the PV Source. The figure 10(b) and (c) shows variation in the buck converter duty ratio and output power under variable irradiance conditions i.e the irradiance is varied from 600 to 800 W/m2 at 1 sec to 1000 W/m2 at 2 sec and finally to 400 W/m2 at 3 sec. From figure 10(b), the fuzzy voltage controller has reacted accurately by controlling the duty cycle of the buck converter to generate constant voltage under

Design of Fuzzy Logic Based Photovoltaic Fed Battery Charging System 278

conditions of sudden change in irradiance. The figure 10(c) shows variation in the power drawn by the battery which is equal to the generated PV power at MPP.

(a)

(b)

(c)

Figure 10 Buck converter (a) output voltage and current (b) Duty cycle (c) output power

279 Polamraju.V.S.Sobhan et. al.

5 Conclusion

The design and analysis of MPPT controller and voltage controllers based on fuzzy logic theory are presented in this paper. The fuzzy MPPT controller drives boost converter such that the PV output power is always at it’s the maximum under variable irradiations. The variable output voltage of boost converter is fed to the fuzzy controlled buck converter to regulate the voltage equal to 12 V which is suitable to charge the battery. The two controllers fuzzy MPPT and fuzzy voltage controllers continuously alters the duty ratios of boost and buck converters respectively. The simulation results are confirming the achievement of maximum power extraction and constant voltage regulation. References

[1] N.Chaturvedi, R.Klein, J.Christen, A. Kojic,“Algorithms for advanced battery

management systems,” IEEE Control Systems, vol.30, no.3, pp.49–68, 2010. [2] M. G. Villalva, J.R. Gazoli, E. R. Filho,“Comprehensive approach to modeling

and simulation of photovoltaic arrays,” IEEE Trans. on Power Electronics, vol. 24, no. 5, pp. 1198–1208, May 2009.

[3] P. Padmavathi and N. Sudhakar, “Solar powered LED lighting with high gain boost converter,” Journal of Green Engineering, vol. 8, no.3, pp. 411–430, 2018.

[4] Macaulay, J, Zhou, Z,“A Fuzzy logical based variable step size P&O MPPT algorithm for PV System,” Energies, 11, 1340. 2018.

[5] Fathabadi.H, “Novel high-efficiency DC/DC boost converter for using in PV systems,” Solar Energy, vol.125, pp. 22-31, 2016.

[6] Shailendra.K.Tiwari, Bhim Singh, Puneet.K.Goel, “Design and control of microgrid fed by renewable energy generating sources,” IEEE Trans. on Industry Applications,vol.54, no.3, pp. 2041-2050, 2018.

[7] Zar Ni, A.T. Naing, HlaMyoTun, “ Design and construction of microcontroller based solar battery charger,” International Journal of Scientific & Technology Research , vol.5, no.6, pp.117-120,2016.

[8] Battery and Energy Technologies,http://www.mpoweruk.com. [9] M.Neethu , R.Senthilkumar, “Soft computing based MPPT controller for solar

powered battery charger”International Conference on Electrical Energy Systems (ICEES), Chennai, India, 21-22, Feb. 2019.

[10] M.SubbaRao, Ch.SaiBabu, S.Satyanarayana, “Digital fuzzy current mode controlled integrated PFC converter with external ramp compensation,” Journal of Circuits, Systems, and Computers, vol. 27, no. 9, pp.1-23, 2018.

[11] Ravi Dharavath, I. Jacob Raglend, “Fuzzy controller based solar photovoltaic system, fuel cell integration for conditioning the electrical power,” Journal of Green Engineering, vol.8, no.3, pp. 301–318,2018.

Design of Fuzzy Logic Based Photovoltaic Fed Battery Charging System 280

Biographies

Polamraju.V.S.Sobhan received the B.Tech degree in Electrical and Electronics Engineering from SVHCE, Machilipatnam in 1999 and M.E degree from Andhra University, Visakhapatnam in 2002. He is presently working in the Department of EEE, VFSTR University, Guntur, India. His research interests include Intelligent Controllers Design for PV Systems, Active Magnetic Bearings and Bearingless drives.

Mopidevi Subbarao received BTech from JNTUH in 2000, MTech from JNTUA in 2007 and Ph.D. from JNTUK in 2019. He is presently working in the Department of EEE, VFSTR University, Guntur, India His research interests include Power Electronics and Drives.

Bharath Kumar Narukullapati received his bachelor’s degree from VEC, JNTUH in 2005 and master’s degree from NEC, JNTUK in 2011. Currently he is working as AssistantProfessor in the department of EEE, VFSTR, Guntur, India.His research interests include Electromagnetic analysis,electrodynamics, Power electronics converters, and Distributed energy systems.

281 Polamraju.V.S.Sobhan et. al.

A Sriharibabu, completed his M.Tech. from NIT Calicut, Calicut in 2011. Currently he is working as Assistant Professor and pursuing Ph.D. in Vignan's Foundation for Science, Technology & Research (Deemed to be University), Valamudi. He has more than 7 years of experience in teaching. His research area of interest is Power electronics in Renewable Energy systems.