Transcript
1
Performance of Grid Connected Photovoltaic Inverter with Maximum Power Point Tracker and Power Factor Control S. Mekhilef, M.E, Ahmed, and M. A. A. Younis Dept. of Electrical Engineering, University of Malaya 50603 Kula Lumpur, Malaysia
ABSTRACT Detailed analysis, simulation and hardware results of grid connected inverter with maximum power point tracker and power factor control in Malaysian climate are presented. A six-switch topology inverter with symmetrical Pulse Width Modulation (PWM) switching technique is used. A low pass filter is incorporated in the circuit to filter out unwanted harmonics and produce a sinusoidal AC current. Low total harmonic current distortion at the inverter output can be achieved. The three-phase PWM switching pattern was developed using Xilinx FPGA. The developed system is also capable of adjusting the power factor to unity. A 3kVA power transformer with a 1:2 ratio is constructed to provide galvanic isolation for better circuit performance and circuit protection. Hardware results for proposed solar photovoltaic inverter configuration interconnected to the grid are also presented. From the experimental results it is confirmed that the harmonic distortion of the inverter output waveform is within the limits laid down by the utility companies. Index Terms—grid connected inverter, maximum power point tracker, FPGA, and photovoltaic systems
Malaysia is located almost on the equator and is blessed with an abundance of sunlight almost all year round So obviously, with the right planning and strategies that are coupled to the right technology and development in the market, the potential for photovoltaic system as an alternative source of power in this country looks promising and is constantly gaining ground and popularity. To harvest the vast solar energy especially in tropical country like Malaysia, it would be desirable if the energy conversion units are simple, reliable, low cost and high efficiency. High efficiency can be achieved by the use of all the power generated for the unit and even contribute to the gird while the energy is not used. The output power induced in the photovoltaic modules is influenced by an intensity of solar cell radiation, temperature of the solar cells and so on. Therefore, to maximize the efficiency of the renewable energy system, it is necessary to track the maximum power point of the input source [6-14]. The prototype of the three-phase grid connected inverter with maximum power point tracker control and power factor correction was tested with and without the MPPT. The tests were carried out at a power greater than 3kW. The current performance proved to be satisfactory and complies with IEEE recommended practice on utility interface of PV system, IEEE Std 929-2000 and UL1741-1999.
1. INTRODUCTION In recent years, the increase of energy demand and the problems of fossil-fuel sources due to their environmental pollution and future shortages, have led to the development of technologies needed to use nonpolluting alternative energy sources such as solar and wind sources [1-4] It is beyond doubt that especially the high development of the power electronics has made the energy produced by the above alternative sources accessible and at the same time also at low cost [3-8]. Moreover, it has allowed the spreading of the Distributed Generation (DG), consisting in the use of a great number of small and medium generation systems connected to the distribution grid to feed a dedicated consumer or to be support of the grid itself [9-10] Power supply reliability and power quality have become important issues for all kind of power electronics systems including photovoltaic systems. Interconnecting a photovoltaic system with utility, it is necessary that the PV system should meet the harmonic standard and the active power supply requirement. Several utility connected photovoltaic systems have been proposed [13-15]. Among these systems, the most common type is the parallel running PV system with bi-directional power flow to provide unity power factor on the utility line.[16-17]
2. THE PROPOSED SYSTEM TOPOLOGY The first significant decision that an inverter designer must make is the choice of an overall circuit topology. The PV array voltage and utility grid interconnect voltage drive the topology selection [9-12]. There can be wide DC input voltage variations resulting from various combinations of array power, temperature and module configuration. The primary topology consideration is whether or not to use a DC-to-DC converter stage between the PV array and the DC to AC inverter block to pre-regulate the DC voltage. The inverter with DC-to-DC converter stage will operate over a wider DC input range but with a cost premium and lower conversion efficiencies at some operating points. The topology that provides the best energy yield under a given set of operating conditions will be determined by the remaining system components [15-18]. The overall circuit of the proposed three-phase grid connected inverter system consists of a solar array, DC-DC boost converter acting as maximum power point tracker, power conditioning unit, a low pass LC filter, a high power transformer, a phase-locked loop circuit, Xilinx FPGA as a main PWM generator that is capable of generating high frequency PWM as well as controllable displacement factor, and a personal computer as shown in figure 1.
CCECE/CCGEI May 5-7 2008 Niagara Falls. Canada 978-1-4244-1643-1/08/$25.00 2008 IEEE
001129
2 Solar Array
Three Phase Inverter
DC-DC Converter
Low Pass Filter
ADC
Magnetic Switch
Three Phase Power Transformer
Load
Phase Lock Loop
Isolation & Driver
Utility
PWM Generator (Xilinx)
Isolation Amplifier
Personal Computer
Fig. 1: Overall block diagram of three-phase grid connected inverter
A six power switching device bridge topology has been chosen to form a controlled bridge that simplified the circuit since fewer power switching devices are involved. High frequency three-phase PWM with less noise interference is generated using Xilinx FPGA. The closed loop system is formed by the feedback of the output voltage to a personal computer via an isolation amplifier. A program was developed using Genie software is used as a main feedback controller of the system that processes the reference input and the feedback signal to produce the required modulation index to the DC-DC converter and also the power-conditioning unit. The modulation index is then fed into the PWM generator circuit that consists of a FPGA and phase-locked loop. The pulses produced from the generator unit are used to drive the power switching devices via isolated driver circuits. A. PWM Pattern for Three-Phase Inverter The developed PWM is based on symmetrical sampling. The triangular carrier signal and the sinusoidal modulating signal are used. The principle of PWM generation is shown in figure 2.
(b) Fig. 2: (a) Principle of PWM generation, (b)Three 600 sine waveforms portions
Each phase is been divided into six 60 segments; the PWM is generated for the first 60 only by storing sixty samples of red phase (A) and sixty samples of blue phase (B) in a look-up table and the yellow phase (C) segment is derived by the addition of red and blue phases to form a three-phase modulating signal as shown in figure 2-b. This technique reduces the usage of Configurable Logic Blocks (CLB) in the Xilinx FPGA, and the memory requirements. The decoding of the look-up table to form the full wave of the three phase modulating signals is shown in table I. TABLE I DECODING OF LOOK-UP TABLE
(a)
Angle ()
Red phase
0-60 60-120 120-180 180-240 240-300 300-360
a c b a c b
Blue phase c b a c b a
Yellow phase b a c b a c
B. Inverter Circuit Configuration The proposed main circuit comprises of a capacitor, DCDC boost converter, three phase inverter with six power 001130
3 IGBTs (S1 to S6) types BUP314, six fast switching diodes (D1 to D6) type BYP101, a low pass LCR filter, a three phase power transformer, load, and utility supply as shown in figure 3. The low pass filter absorbs the high order of harmonic components produced by the PWM switching in the inverter and produces almost sinusoidal current at the output of the inverter within the limits of the power supply utility. The three-phase power transformer is not only meant to step-up the voltage level but also to provide ohmic isolation for better protection. Solar Arrays
ipv
DC/DC Converter
Ld
Power Inverter
is
Dd
S1
D1
S3
D3
S5
determines the angle of current lag in the system i.e. lagging power factor. When a negative cycle occurs, the counter remains at the reset state. Similarly, during the negative cycle another counter operates. However, this condition will determine the angle of lead of the current in the system i.e. leading power factor. Shifting of the PWM patterns is carried out by the delay or advance of the reset signal. The reset signal is connected to all the re-settable modules. A positive triggering edge of the positive and negative cycle is used as a reference by the reset signal. Advancing or delaying the reset signal by the external command will force the current in the main circuit to lead or lag the supply voltage. The principle operation of the reset unit is shown in figure 5.
D5
IA
Cd Vpv
Cs
Sd
A
Vs
IB IC
B C S4
Utility
Vr Vy Vb
Magnetic switch ir
ia1
ib2
iy
ib1
ic2
ib
ic1 i2
S6
Transformer
Load
ia2
i3
D4
i1
D6
S2
D2
Low pass filter
iL1
L1
iL2
L2
iL3
L3
r
r
r
c
c
c
Fig. 3: Schematic diagram of the proposed three-phase grid connected inverter
The operation of the three phase power inverter can be divided into six modes, termed Mode I, Mode 2, Mode 3, Mode 4, Mode 5, and Mode 6 as shown in table II. In this analysis, the inverter is connected to the Y-connected resistive load and the transformer core is unsaturated.
Mode
TABLE II MODES OF OPERATION Phase angle () S1 S3 S5
S2
S6
S4
Mode1 Mode2 Mode3 Mode4 Mode5 Mode6
0 /3 /3 2/3 2 /3 4/3 4/3 5 /3 5/3 2
On On On -
On On On
On On On
On On On -
On On On -
On On On
Fig. 5: Principle of operation of the reset unit
D.Maximum Power Point Tracker 1) Energy production from the system The total energy production by the PV system daily for five months has been collected; the amount of energy generated by the PV system is changing from one day to another this is due to weather changes i.e. during cloudy and raining day. The average monthly energy production by the PV system is shown in figure 6. During April the PV system generates the highest level of energy for the sun.
C.Power Factor Correction Unit The power factor correction unit consists of two eight-bit counters, two eight-bit comparators, a VHDL code block, and a few logic gates are used to form a reset unit. This unit is clocked by the signal derived from carrier unit. During the positive cycle, one counter starts counting from 0 to 180. During the counting process, the comparator compares the counter value with the external input data. If the values are equal, it produces a pulse output signal. This signal
Energy (kWh)
120 100 80 60 40 20 0 1
2
3
4
5
M o n th s
Fig. 6: Monthly energy production
001131
The output current of the PV system has been captured for a
4 period of four days from 7:15 in the morning till 19:30 in the evening it can be seen from the plots that the peaks values occurs between 13:00 and 16:30 as shown in figure 7.
time. The input (in parallel with PV) and output capacitors values are 4700f and 470F, respectively. The input inductor value is 10mH and is wound on ferrite core with 1mm air gap. The system efficiency is defined as:
K
Po Pin
Po Po Pd
Where Pin and Po are the DC-DC converter input and output power, respectively, while Pd is the power loss. The power loss consists of the IGBP and diode conduction and switching losses, the inductor core and copper losses and the control system power consumption. The theoretical values were calculated using data given by the manufacturers of the circuit elements. The theoretical and measured efficiency for various output power levels is shown in figure 9. It is seen that the efficiency is quite high approximately 90% and relatively constant for a wide output power range.
Fig. 7: PV output current
0.95
2)MPPT Implementation
Theoretical 0.9
Efficiency (%)
The MPP tracking process is shown in figure 8. The starting points vary, depending on the atmospheric condition, while the modulation index is changed continuously, resulting in the system steady-state operation around the maximum power point. The proposed MPPT was implemented in two different stages. The first stage senses the output voltage and current, digitise them using the ADC .The digital values are then used by the Genie software to generate the modulation index which is used as input to Xilinx FPGA to generate the required PWM in the second stage.
0.85
Measured 0.8 0.75 0.7 0.65 0.6 500
PV Output Power
Steady state operation
1000
1500
2000
2500
3000
Power (W) Fig. 9: System efficiency under PV MPPT conditions at 25 0C
Pmax
Possible starting points Time
Fig. 8: MPP tracking process
A prototype MPPT system has been developed using the above-described method and was tested in the laboratory. The PV array used with this system consists of 42 SP75 Siemens modules, producing a 3.15kW maximum power at an irradiation level of 1kW/m2 and temperature of 250C. In order to test the proposed system under various atmospheric conditions, the PV array was first simulated with a DC power supply by adjusting its output voltage and current limit settings. The power switch consists of one IGBT rated at 600V, 50A, while the diode has a 200ns reverse-recovery
The actual PV output power and corresponding theoretical maximum output power for various irradiation levels is shown in figure 10(a). It is seen that the proposed system always tracks the PV maximum power point. Figure 10(b) shows the PV output power for various irradiation levels, with the MPPT control disconnected and with the DC-DC converter modulation index set such that the PV array produces the maximum power at each at 1 kW/m2 at 250C. The theoretical maximum PV power at each irradiation level is also indicated in the figure. A comparison between figure 10 (a) and (b) shows that the use of the proposed MPPT control system increases the PV output power by as much as 18% for the irradiation in the range of 0.2-0.75kW/m2. Since both the sun irradiation and the air temperature change slowly during the daytime, the system is expected to track effectively the PV maximum power point, under normal changes of atmospheric conditions (e.g., cloud shadowing). This has been also verified experimentally by partially covering the PV modules and noting that the system tracked successfully.
001132
5 the voltage and has an inductance of 8mH, with 1:2 ratio. A pure resistive load of 300 ohm is connected in star configuration. The AC side filter is a passive low pass LCR filter connected in delta configuration. The low order unwanted harmonics can be filtered out by connecting a low pass filter at the output inverter as shown in figure 12.
PV Output Power (W)
3000 Theoretical Power
2500
Measured Power
2000 1500 1000
!
500 0
0
0.25
0.50
0.75
1 !
2
Irradiation (kW/ m )
(a) 3000 2500 2000
(a)
1500
1
1000 PV Output Power
500 0
x: 5ms/div, y: 5A/div
Maximum Power
0
0.25
0.50
0.75
Current 5A/div
PV Output Power (W)
!
1
Irradiation (kW/m2)
2
(b) Fig. 11: (a) Theoretical and measured PV output power under MPPT conditions at various irradiation levels and (b) the actual PV output power and the corresponding theoretical maximum PV power at various irradiation levels (PV maximum power at 1 kW/m2 at 250C)
3
0
20
40
60
80
100
Time 10ms/div
3. EXPERIMENTAL RESULTS
(b)
A prototype model of a three-phase grid connected inverter was constructed and tested to compare the performance of the inverter with the Pspice® simulated results. The block diagram of the experimental set-up is shown in figure 1. A window based graphical user interface software was developed using the Genie software program where the information on the output voltage, displacement angle and modulation index are displayed on the PC screen. The changes on the output voltage are also displayed on-line. The displacement angle and modulation index can be easily change by clicking the scroll icon provided on the screen. The AC output voltage, and DC input voltage are read on-line by the PC via a voltage isolation amplifier and display the value on the screen. A general purpose data acquisition card PCL816 is slotted in the PC to interface with the external signals. The system is also capable of shifting the PWM signal that will consequently change the power factor. The isolation transformer is connected in Delta/Wye connection to step up
Fig.12: Inverter output current (a) before filter (b) after filter
It can be seen clearly from table III that the 3rd to 9th are lower than 4.0%, 11th to 15th are lower than 2.0% the and THD is 4.2% which comply with IEEE Std 519-1992. TABLE III DISTORTION LEVEL OF THE OUTPUT CURRENT 5th 7th 9th 11th 13th Harmonics 3rd
15th
Distortion L
1.1%
3.2%
3.1%
3.7%
3.5%
1.8%
1.5%
Power factor adjustment is demonstrated in figure 13(a) for leading, figure 13(b) for lagging, The inverter could be forced to operate at unity power at any load condition. The power factor adjustment data is entered via personal computer using Genie software.
001133
6
T
Displacement angle
3.
Volt, A (5V, 5A /div)
Voltage Current
4. !
5.
6.
7.
time, mS
(a)
8.
T Displacement angle
Volt, A (5V, 5A /div)
Voltage Current
!
time, mS
(b) Fig. 13: Effect of shifting PWM pattern on operating power factor
4. CONCLUSION The prototype of the three-phase grid connected inverter with maximum power point tracker control and power factor correction was tested with and without the MPPT. The tests were carried out at a power greater than 3kW. The current performance proved to be satisfactory and complies with IEEE recommended practice on utility interface of PV system. The PV array output power delivered to the inverter can be maximized using MPPT control system, which consist of a power conditioner to interface the PV output to the inverter, and a control unit, which drives the power conditioner such that it extracts the maximum power from the PV array. Low cost and low power consumption MPPT system has been developed and tested. Experimental results show that the use of the proposed MPPT control increases the PV output power by as much as 18%. REFERENCES 1. Jain, S.; Agarwal, V.; “A Single-Stage Grid Connected Inverter Topology for Solar PV Systems With Maximum Power Point Tracking” IEEE Transactions on Power Electronics,Volume 22, Issue 5, Sept. 2007, pp.1928 – 1940. 2. Koizumi, H.; Mizuno, T.; Kaito, T.; Noda, Y.; Goshima, N.; Kawasaki, M.; Nagasaka, K.; Kurokawa, K.; ”A Novel Microcontroller for Grid-
001134
Connected Photovoltaic Systems” IEEE Transactions on Industrial Electronics, Volume 53, Issue 6, Dec. 2006, pp.1889 – 1897. Barbosa, P.G.; Braga, H.A.C.; Rodrigues, Md.C.B.; Teixeira, E.C.; “Boost current multilevel inverter and its application on single-phase grid-connected photovoltaic systems” IEEE Transactions on Power Electronics, Volume 21, Issue 4, July 2006 , pp.1116 – 1124. Yang Chen; Smedley, K.; “Three-phase boost-type grid-connected inverters” IEEE Applied Power Electronics Conference and Exposition, 2006. APEC '06, 19-23 March 2006, pp. 7. Wildenbeest, E.J.; de Haan, S.W.H.; van der Borg, N.C.; “Recent test results of 6 commercial inverters for grid connected PV systems” IEEE Photovoltaic Specialists Conference 1994, Volume 1, 5-9 Dec. 1994 , pp. 909 – 912. ter Reide, F.J.; Wanner, J.; ”High-performance inverters for gridconnected PV applications” IEEE Photovoltaic Specialists Conference 1994, Volume 1, 5-9 Dec. 1994, pp.913 – 916. Chen, Y.-M.; Liu, Y.-C.; Hung, S.-C.; Cheng, C.-S.; ”Multi-Input Inverter for Grid-Connected Hybrid PV/Wind Power System” IEEE Transactions on Power Electronics” Volume 22, Issue 3, May 2007, pp.1070 – 1077 Meza, Carlos; Negroni, Juan Jose; Guinjoan, Francesc; Biel, Domingo; ”Inverter Configurations Comparative for Residential PV-Grid Connected Systems” IEEE Conference on Industrial Electronics, IECON 2006, Nov. 2006 pp. 4361 - 4366
