Power management of Battery-assisted PV power

generation
Gaurav Lamba1, Parth Rai1, Shashank1, Kundan Anand1*, Bhavnesh Kumar2
1 Electrical Engineering Department
2 Instrumentation and Control Engineering Department
1, 2 Netaji Subhas University of Technology, Dwarka, Delhi 110078

[Link].ee19@[Link], parth.ee19@[Link], shashank.ee19@[Link], Kundan.ee19@[Link], Kumar_bhavnesh@[Link]

Abstract— This paper illustrates a power management scheme maximum available power from PV array, second challenge is
for a hybrid PV-battery power generation system. A Solar PV to decide whether the battery will be charged or discharged.
module, a bidirectional DC-DC converter, a boost converter, and Several battery power management has been reported in recent
an energy storage device (battery) comprise the system. The
literature [5]–[8]. Another problem was that the old controllers
primary goal of this work is to manage the power flow, battery
like SCADA were too costly for the power management and the
charging and discharging. Excess energy generated by the PV
array is stored in the battery during periods of low load demand.
low cost controller implementation for battery management
During periods of high load demand, the battery functions as a system and power management system was not available.
backup power supply unit, delivering stored power to the load In this paper, an implementation of a PV-Battery hybrid
alongside the PV module. Two DC-DC converters serve as a link
power generation system has been presented. The main goal of
between the PV panel, battery, and load. While a boost converter
this research is to manage battery charging and discharging in a
connects PV and load, A bidirectional buck-boost converter
connects the battery and the load to the PV. The designed system PV-Battery hybrid power generation system using DC-DC
is modelled in MATLAB Simulink environment. The results converters and MPPT controller to extract the maximum power
obtained are satisfactory and fulfilling the required goals. from PV array using MPPT. The design system can be
implemented as a totally independent system in reality.
Index Terms— Battery, Boost Converter, Buck-boost converter,
MPPT, Photovoltaics II. RESEARCH METHODOLOGY
I. INTRODUCTION The mathematical modelling is basically a simplified way of
representing the system which is used to understand the
Energy is the key factor for the development of any nation.
functioning of the system and analyzing and predicting some
It is generated from fossil fuels, the traditional way of energy aspects.
generation and limited in supply, and the renewable sources
which are clean sources of energy generation and are in A. Design of Solar-PV generation
abundant supply. Among these sources of clean energy, the A component made up of semiconductors, which generates dc
photovoltaic (PV) generating system is most commonly used. current, in presence of solar irradiance, is known as PV Cell.
The characteristics of PV depend upon atmospheric conditions Fig. 1 represents PV cell’s equivalent circuit diagram [1]. It’s
mainly on irradiance and temperature. However, like every mathematically modeled IV characteristic is given by equation
renwable energy source, the solar energy is discontinuous in (1).
nature and hence an energy storage system is used to get +VE
uninterrupted and continuous energy supply. Battery is used, as Id Ish Rs

an energy storage device, with renewable sources as it is the

most efficient and long lasting energy storage device [1]. IL Rsh V

When battery is included in the system, the battery also

becomes a load for PV array. Battery takes the excess power to -VE
reach in a state where it can work as source in case of deficiency
in power from PV array [2]. To extract maximum power from Fig. 1. Equivalent circuit diagram of PV Cell.
PV array, a maximum power point tracking (MPPT) circuit is
used to track maximum power point (MPP) for a specific  q  V  IR s   V  IR
I  I L  I  e  KT   1  s
(1)
irradiance [3]. In the recent literature, many MPPT methods   R
  s
have been proposed. The MPPT methods include various
methods like Pertub and Observe (P&O), Incremental where, I is the cell current (A), IL is the light generated current
conductance (InC), Modified P&O, Modified InC, Meta- (A), Io is the diode saturation current, q is the charge of an
heuristic MPPT, Fuzzy logic MPPT, and Artificial neural electron = 1.6x10-19 Coulombs, K is the Boltzmann constant
network (ANN)-MPPT etc [3], [4]. After extraction of
(j/K), T is the cell temperature (K), Rs, Rsh is cell series and Buck-boost converter has been used for battery power
shunt resistance (Ω), V is the cell output voltage (V) [1], [9]. management as shown in Fig. 3 [4].
B. InC MPPT algorithm

Load Side

Load Side
PV Side

Battery
The MPPT circuit is a combination of a DC-DC converter S_P

Side
connected to PV array and a controller with an embedded SBoost S_N
MPPT algorithm. The PWM generated by controller is fed to
switching device of the DC-DC converter. Here, in this case a (a.) (b.)
Boost converter has been used as DC-DC converter Fig. 3. Equivalent circuit diagram (a.) Boost converter (b.)
During uniform irradiance, the InC MPPT algorithm has Bidirectional buck-boost converter
been identified as an efficient algorithm. The InC algorithm is
based on the observation that the slope of the characteristic is E. Power Management
zero at the MPP. The InC MPPT algorithm has been used in this The power extraction from PV arrays has been maximized
case to obtain minimum steady-state oscillations and quick using the MPPT circuit. But the generated power must be
convergence to MPP [10]. managed. Here, the power management in the circuit is
completely dependent on the battery power management
C. Battery
system. The basis of battery power management is the power
When there is a downfall in power from PV Arrays, Battery availability of power from PV array. If the power generated
provides the required amount of power to the DC bus. Rating
from PV arrays is higher compared to the load power
of the battery used in the designed system is 12V and 1.3Ah.
An equivalent circuit diagram of battery is presented in Fig. 2. requirement, then the bidirectional Buck-boost converter
According to Kirchhoff’s law [1], [5]: operates in buck mode and battery is charged. However, if the
power generated from PV arrays is lower compared to the load
VBat  E   R Bat  i   Vc (2) power requirement, then the bidirectional Buck-boost converter
State of charge (SOC) of battery is calculated by operates in boost mode and battery is discharged .
1 The battery power management happens in 2 stages [1], [3],
CBat 
SoC(t)  SoC(0)  i(t).dt (3) [11].
 Stage 1:
i. Comparison between reference DC voltage (Vdc_ref)
CBat RBat and output DC voltage (Vdc).
E0 ii. Output of above comparison is given in PID controller
VBat
as input.
iii. A reference signal is generated by PID controller.
 Stage 2:
i. Reference signal from PID controller is compared with
Fig. 2. equivalent circuit diagram of battery
Battery current (Ibat).
D. DC-DC Converter ii. Again a reference signal is generated after above
In the designed system, two DC-DC converters has been mentioned comparison.
iii. According to newly produced reference signal, the
implemented i.e. for MPPT and battery power management.
PWM signal for operation of Bidirectional DC-DC
Boost converter has been used for MPPT and Bidirectional
converter is generated.

Fig. 4. Designed system on MATLAB Simulink

 Fig. 5. Battery power management

The complete designed circuit has been presented in Fig. 4. scenarios: when the irradiance was 1000W/m2 and 100W/m2.
Whereas, the battery power management system has been Due to this the power, voltage and current varied. The
presented in Fig. 5. generated power at any irradiance must remain constant even
E. Component Ratings if the load varied. The MPPT algorithm adjusts the duty cycle
of the Boost converter such that the impedance is matched
The PV array consists of two PV modules of 10W each
for maximum power transfer. Whereas, due to integration of
connected in parallel. The rating of PV modules has been
battery in the system, the power at load must remain
presented in TABLE I below.
constant. Here in this case, a constant load of 5W has been
TABLE I. used. The maximum power from PV is 20W.
SPECIFICATION OF PV PANELS USED IN THE DESIGN
Characteristics Values
Rated Maximum Power 10W
Cell Technology Poly-Si
MPP Voltage (VMPP) 18.10V
MPP Current (IMPP) 0.55A
Open-circuit voltage (VOC) 21.70V
Nominal Operating Cell Temperature 47±2 ℃
Max. Series Fuse Rating 15A
Short-circuit current (ISC) 0.63A
Max. System Voltage 1000V DC
Operating Temperature - 40℃ to + 65℃ Fig. 6. PV voltage profile under variable irradiance
Application Class A
Tolerance 0-3% The MPPT circuit keeps the PV voltage around the MPP
Weight (kg) 1.30 voltage (VMPP) i.e. around 18.5 V. So, that the maximum
Dimension (mm) 360*275*18 power can be extracted in every irradiance condition. Here,
The ratings of boost converter for MPPT is presented in Fig. 6, it can be observed that the PV voltage remain around
TABLE II below. VMPP. For irradiance 1000W/m2 and 100W/m2, the PV
voltage is 18.5V and 16.2V respectively.
TABLE II.
SPECIFICATION OF COMPONENTS USED IN THE DESIGN OF BOOST
CONVERTER
Component Value
Inductor 27mH
Input capacitance value 220μF
Output capacitance value 470μF
Switching Frequency 5 kHz

The ratings of bidirectional buck-boost converter for battery

power managment is presented in TABLE III below. Fig. 7. PV current profile under variable irradiance

TABLE III.
SPECIFICATION OF COMPONENTS USED IN THE DESIGN OF BOOST
CONVERTER
Component Value
Inductor 27mH
Load side capacitance value 1000μF, 63V
Battery capacitance value 1000μF, 63V
Switching Frequency 5 kHz

III. RESULTS AND DISCUSSION Fig. 8. PV and battery power under variable irradiance
In this section, The experimental study's findings are The influence of MPPT circuit keeps the PV voltage around
presented and discussed. The system has been analysed for the MPP voltage (VMPP) but the current varies with variation
two cases, when the measurements were taken in two in irradiance. Here, Fig. 7, it can be observed that the PV
current is directly proportional to irradiance. Higher
irradiance corresponds to higher current. For irradiance
1000W/m2 and 100W/m2, the PV current is 1.1A and 0.11A
respectively.
The influence of MPPT circuit keeps the PV power around
the MPP. However, it varies with variation in irradiance. Due
to variation in PV power, the battery power also varies. Here,
Fig. 8, It's great to note that the PV power increases as the
Fig. 11. Battery SOC profile under variable irradiance
irradiance increases. Increasing the irradiance leads to an
increase in power output. When the irradiance is 1000W/m2, A constant 5W load has been considered here for the
the PV power is efficiently fulfilling the Load power analysis. The ratings of load is 5W, 20V DC. From Fig. 12,
requirement and charging the battery at the same time. Even it can be observed that the power remain around 5W
though the PV power is not enough to fulfill the load power throughout the operation. From Fig. 13, it can be observed
requirement at 100W/m2 irradiance, the battery can that the current remain around 0.25A throughout the
discharge and supply power to the load. When the battery operation. From Fig. 14, The voltage remains constant at
power is positive, it is successfully supplying power to the around 20V throughout the operation. As a result, it is
load and when it is negative, the battery is efficiently confirmed that the hybrid PV-Battery power generation is
charging. efficient because it successfully met the power requirements
of the load.
The voltage of the DC bus is set to 20V. As a result of
varying irradiance, the DC bus voltage fluctuates around
20V as depicted in the Fig. 9.

Fig. 12. Load power profile under variable irradiance

Fig. 9. DC bus voltage profile under variable irradiance

Fig. 13. Load current profile under variable irradiance

Fig. 10. Battery current profile under variable irradiance

The charging and discharging of battery can be verified by

observing the current flow direction of the battery. If the
positive terminal of current sensor is connected to positive
terminal of battery and the current is negative, then the
battery is charging mode. Whereas, if the current is positive,
the battery is in discharging mode. For irradiance 1000W/m2,
the battery current is negative (-1A) that means the battery is Fig. 14. Load voltage profile under variable irradiance
in charging mode. When the irradiance is 100W/m2, the Also, the similar results have been observed on the hardware
battery current is positive (0.11A), that means the battery is implementation of the designed system.
supplying power to the load. Similarly, the battery charging
and discharging can understood by observing SOC graph
presented in Fig. 11.
 IV. CONCLUSION Technol., vol. 9, no. 4, 2018, doi:
10.1504/ijret.2018.10016793.
This study presents an efficient and independent PV-
[5] B. R. Naidu, G. Panda, and B. Chitti Babu,
Battery hybrid power generation system. The PV power “Dynamic energy management and control of a grid-
generation was successfully operated at maximum power interactive DC microgrid system,” in Smart Power
point by an MPPT algorithm. Whereas, battery power Distribution Systems, Elsevier, 2019, pp. 41–67. doi:
management system successfully managed battery power by 10.1016/B978-0-12-812154-2.00003-1.
controlling a bidirectional buck-boost converter for battery [6] V. Suresh, M. Muralidhar, and R. Kiranmayi,
charging and discharging. Based on the findings, it is “Modelling and optimization of an off-grid hybrid
renewable energy system for electrification in a rural
possible to conclude that the designed system can be areas,” Energy Reports, vol. 6, 2020, doi:
implemented in real low-power independent circuits for a 10.1016/[Link].2020.01.013.
variety of purposes. Furthermore, the results obtained from [7] E. Dursun and O. Kilic, “Comparative evaluation of
designed circuit are fulfilling the design criteria and goals. different power management strategies of a stand-
alone PV/Wind/PEMFC hybrid power system,” Int.
IV. REFERENCES J. Electr. Power Energy Syst., vol. 34, no. 1, 2012,
[1] K. Anand, N. K. Singh, A. K. Singh, D. K. Singh, doi: 10.1016/[Link].2011.08.025.
and K. Ask, “Standalone Hybrid Power Generation [8] R. Sedaghati and M. R. Shakarami, “A novel control
using Photovoltaics, Battery and Bio-gas,” in 2018 strategy and power management of hybrid
5th IEEE Uttar Pradesh Section International PV/FC/SC/battery renewable power system-based
Conference on Electrical, Electronics and Computer grid-connected microgrid,” Sustain. Cities Soc., vol.
Engineering (UPCON), Nov. 2018, pp. 1–6. doi: 44, 2019, doi: 10.1016/[Link].2018.11.014.
10.1109/UPCON.2018.8596966. [9] L. Kong, J. Yu, and G. Cai, “Modeling, control and
[2] M. S. Behzadi and M. Niasati, “Comparative simulation of a photovoltaic /hydrogen/
performance analysis of a hybrid PV/FC/battery supercapacitor hybrid power generation system for
stand-alone system using different power grid-connected applications,” Int. J. Hydrogen
management strategies and sizing approaches,” Int. Energy, vol. 44, no. 46, 2019, doi:
J. Hydrogen Energy, vol. 40, no. 1, 2015, doi: 10.1016/[Link].2019.05.097.
10.1016/[Link].2014.10.097. [10] S. Motahhir, A. Chalh, A. El Ghzizal, and A.
[3] K. Yadav, B. Kumar, and D. Swaroop, “Mitigation Derouich, “Development of a low-cost PV system
of Mismatch Power Losses of PV Array under using an improved INC algorithm and a PV panel
Partial Shading Condition using novel Odd Even Proteus model,” J. Clean. Prod., vol. 204, 2018, doi:
Configuration,” Energy Reports, vol. 6, 2020, doi: 10.1016/[Link].2018.08.246.
10.1016/[Link].2020.01.012. [11] A. Aguilera-Gonzalez, R. Lopez-Rodriguez, I.
[4] S. K. Jha, T. Kumar, and B. Kumar, “Review of Vechiu, and S. Bacha, “Model predictive control for
maximum power point tracking techniques for optimal energy management of an island wind
photovoltaic arrays working under uniform/non- storage hybrid power plant,” in IFAC-
uniform insolation level,” Int. J. Renew. Energy PapersOnLine, 2020, vol. 53. doi:
10.1016/[Link].2020.12.1982.