Modeling and Control of a Battery Connected

t 1s IEEE International Conference on Power Electronics. Intelligent Control and Energy Systems (ICPEICES-2016)

Modeling and Control of a Battery Connected Standalone Photovoltaic System Priyabrata Shawl, Pradeep Kurnar Sahu2, Sornnath Maitl and Punit Kurnar4

1,4IIT Delhi, New Delhi, India 2 NIST, Berhampur 3NIT Rourkela I 2 E-mail: [email protected]. [email protected]. [email protected], [email protected] cycle of solar geometry and c1ouds. The BSS improves the reliability of these systems because the excess energy is stored in the battery bank, and this energy is provided to load when there is increase in load power or reduction in source power. This paper deals with ac standalone PV system, as most of the loads requires ac supply for its operation. Generally, standalone PV systems with battery backup mainly consist of a boost dc-dc converter, a bidirectional buck-boost dc-dc converter and a dc-ac inverter. These three converters work together to provide desired high quality power for local load. To increase the operating and conversion efficiency of solar cells a maximum power point tracking (MPPT) algorithm is used [3]. This algorithm extracts the maximum power available at some specified voltage and current irrespective of operating conditions. There are various MPPT control techniques reported in [4], [5], [6]. BSS is included in standalone PV system to serve as an energy regulator of the system by supplying continuous power to the load connected. Among all the available storage devices, batteries are mostly used for renewable energy sources like PV due to their high energy density and high performance. Battery storage devices composed of lead-acid, lithium-ion, nickel cadmium, nickel metal hydride, and sodium sulphur are available. Lead-acid batteries are employed here because of their characteristics suitable for renewable energy applications. In addition, they are reliable and efficient [7]. But we have to trade-off the efficiency for longer charging and discharging cycles. The inverter shown in Fig. l supplies non-linear and critical step changing loads, which are responsible for distorted sinusoidal output voltage waveform [8]. The total harmonic distortion (THD) of inverter output voltage as per the IEEE standard 1547 should be below 5%. Moreover, a c1osed-loop control system is designed to achieve fast dynamic response and robust performance under abrupt load tluctuations. Several inverter control strategies proposed recently to improve the robustness and dynamic response are dead beat control, repetitive control, conventional PI control, and adaptive control [9], [10].

Abstract-This paper presents modeling and control of a standalone photovoltaic (PV) system in which a battery is used as a backup source for power management between the source and the load. Lead-acid battery is commonly used in high

power

PV

applications

due

to

its

low

cost

and

availability in large size. The modeling of PV system and lead-acid battery by using the

corresponding

equivalent

circuits are discussed here. Three independent control loops are proposed to control the standalone PV system; MPPT control loop for extracting maximum power from PV module under different solar irradiation, battery control loop for bidirectional

power

flow

between

battery

and

dc-link

through buck-boost converter to keep the input dc voltage constant, and inverter control loop for maintaining good voltage

regulation

and

achieving fast dynamic response

under sudden load fluctuations. The stability of the above control loops are verified by using Bode diagram. Finally the

proposed method is applied to 2 kW, 110 V, 50 Hz, two- stage single-phase standalone PV system. The simulation and the experimental results are presented to validate the theoretical analysis, effectiveness and feasibility of the proposed control strategy. Keywords-Bidirectional Boost

Converter;

Buck-boost

Lead-acid

Battery;

Converter; MPPT

Dc-dc

Control;

Photovoltaic (PV); Voltage Source Inverter (VSJ)

I.

INTRODUCTION

In recent years, due to environmental pollution, fossil fuel exhaustion and increase in global energy demand, the development of alternative energy sources has become a global priority. Among all available renewable energy sources (such as solar, hydro, wind, and biomass), solar energy has highest importance as it is freely available, noise free and most promising renewable source of energy. Therefore in near future PV systems are expected to play an important role of energy source in meeting electricity demand. PV systems may be operated in grid tied, standalone or hybrid mode. Standalone photovoltaic systems have wide application in remote isolated areas and islands, where utility grid is not available to meet the essential electric load [1]. These standalone PV systems should include battery storage system (BSS) [2], as solar energy generation tends to be unsteady due to diurnal 978-1-4673-8587-9/16/$31.00 ©2016 IEEE

[1]


damping resistance is generally connected at the output of the VSI to minirnize the high frequency switching harmonics. In order to analyze the operation of the system, both modeling and control are important issues that are discussed here.

However all techniques are based on voltage mode control that leads to output voltage waveform sensitive to load variations. The objective of this work is to develop a PV system along with a BSS for standalone residential applications and to control the charging and discharging of BSS, which depends on the PV power and load power demand. In this paper three converters, namely boost converter, bidirectional buck-boost converter and the inverter are used for MPPT control, dc-link voltage control and ac output voltage control respectively. So here three independent control loops are designed to control the standalone PV system. An analog MPPT control technique [6] is used to extract the maximum power available at PV module, which shows fast and robust behavior even in changing environmental and load condition compared to conventional MPPT control techniques. A control technique for battery charging and discharging is presented to keep the dc-link voltage constant during changes in load demand or source power. This control strategy for battery maintains the power balance in the system by regulating the dc bus voltage across the dc-Iink capacitor. PV module is interfaced with battery by a common dc-link along with two dc-dc converters. Here a current-controlled single phase VSI with bipolar pulse width modulation (PWM) [11] is used to maintain stable voltage and frequency at local load. A step changing ac load is considered to evaluate the system dynamic behavior. To discuss these issues, the paper is organized as folIows. The system configuration and mathematical modeling of proposed system is presented in section 11, and the control strategies involved in the system are described in section III. Small signal modeling and stability analysis with the help of Bode diagram will be presented in section IV. Based on this concept, the simulation and experimental results are then presented in Section V to highlight the features and effectiveness of the proposed control schemes. 11.

A.

PV Modeling

For modeling of PV array, the mathematical characteristic equations are utilized [12], [13]. The electrical equivalent circuit of a PV cell is shown in Fig. 2(a), which is required for its modeling and simulation. It inc1udes a current source (Iph) in parallel with a diode (D), a shunt resistance (Rs1,) and a resistance (Rs) in series. The current output of PV module is given by

Ipv Np *Iph -Np *I [ (q*(Vpv+Ipv*Rp)j ] =

exp

D

N,AKT

-

1

and the photo current (Iph) and diode current (ID) are described in [12]. R J...... � + == ,---,---,-��

�+

10 D

R,h

1

Vpv

L-__�__L-______�

I

E -..

+

-

Controlled voltage source

Fig. 2: Equivalent Circuit of (a) Solar Cell and (b) Battery

B.

Battery Modeling

Out of all available models, the equivalent circuit model is commonly used for dynamic simulation [14], [15]. Battery possesses high energy density and if charging/discharging cyc1es are properly controlled, it can provide power at almost constant voltage. In this work a lead-acid battery is used as it is more convenient for renewable systems because of its low cost and availability in large size [14]. A generic battery model of lead-acid battery is used for dynamic simulation in which it is composed of a controlled-voltage source along with a series resistance (R) as shown in Fig. 2(b). Here control input E depends not only on the current but also on the battery state of charge (SOC). By knowing the characteristics of the battery, given in manufacturers' datasheet, it is very easy to find the values of parameters like Eo and K. The battery terminal voltage,

SYSTEM CONFIGURATION AND MODELING

�at

=E-R·i

For discharging:

E =Eo -K �. Q-lt it-K �. Q-lt t +Exp(t)

Fig. I: Battery Connected Standalone PV System

The PV/battery system, considered in this work is shown in Fig. l. This system mainly consists of a PV module, dc-dc boost converter with MPPT controller, BSS with bidirectional buck-boost converter, inverter with its controller, and local loads. A LC low-pass filter with a

Q ·t +Exp(t) E =Eo -K�.it-K it-O.1.Q Q-it

For charging:

[2]

{+1


Where, R is the internal resistance of the battery (D), Eo is the open circuit potential (V), i is the

where

chargingldischarging current of battery (A), K is the polarization voltage (V), and Q is the battery capacity (Ah). III.

A.

=

ifX>O -lifX�O

Practically it is more convenient to use the boolean 0/1 instead of -1/+ 1 for representing the sign of a value. Thus, comparators are then used for evaluating its sign by producing binary signals. And finally an XOR gate is used to multiply the two signs, which are now expressed as booleans or binary form. Then the exclusive-ORed output is sampled by SR-flipflop with a clock frequency of 1/T to operate the boost converter at constant frequency.

CONTROL STRATEGIES

Analog MPPT Contral

The maximum power of the PV module changes with extern al climate conditions. At any operating condition there is only one value of current (IIIlPP) and one value of voltage ( VIIlPP)' that defines the maximum power point (MPP) at which power is maximum [16]. The MPPT technique is implemented

B.

Buck-Boost Converter Control

The lead-acid battery is mainly used to provide a steady power to the local load irrespective of source or load power variation [2]. In this standalone PV system, lead-acid battery operates in two modes, in which sometimes it must be charged to store excess power from solar source or discharged to supply local load when solar power is not sufficient. The primary objective of the battery converter is to maintain a stable common dc bus voltage. A bidirectional buck-boost dc-dc converter is used to maintain continuous power flow between the dc bus and BSS with a constant dc-link voltage [18], [2]. A constant dc-link voltage is maintained by charging or discharging the lead-acid battery depending on the load change or solar irradiance change. So no matter the battery is charging or discharging, the dc-link voltage should be stable and regulated throughout the operation. For this system the reference dc-link voltage is considered as 200V, which is determined by the knowledge the boost MPPT control. The bidirectional buck-boost converter is controlled in such a way that the dc bus voltage will remain constant or stable during changes in the solar power or load demand variations.

DC-DC Boost converter

DL c::::::> �� Cdc

Sgn(X)

� ...,

Fig. 3: PV Module with Analog MPPT Control

by using a boost dc-dc converter, as it is required to increase the voltage level at the input of the inverter, across dc-link capacitor. Fig.3 shows the PV system block diagram with analog MPPT controller, which consists of a PV module, a boost converter comprising of an input filter capacitor C, an output low-pass filter with inductor Land capacitor Cdc, a diode D and a controllable switch S, regulated by the MPPT controller. Denoting v and i as the module voltage and current respectively, the power p will be maximum atv v'IlPP' =

where op /ov

op /ov

0 when v