University of Arkansas - Texas Instruments

4 downloads 8329 Views 2MB Size Report
May 12, 2009 - 3.8 FILTER DESIGN ....................................................................................................................................................... 27. 3.8.1 Low-pass Filter Topology .
University of Arkansas College of Engineering

Department of Electrical Engineering

Photovoltaic Array with Maximum Power Point Tracking

Submitted by John Damron Brady Delperdang Tristan Evans Jordan Greenlee Lauren MeGee

Date Submitted:

May 12, 2009

Course Instructor:

Dr. Juan Balda

Graduate Assistant:

Derik Trowler

Abstract Renewable energy is one of the fastest growing trends in post-industrialized societies as they face growing energy demands and actively seek cost effective solutions. Among these solutions includes solar energy, specifically photovoltaic arrays. Photovoltaic arrays allow societies to drastically reduce energy expenses and dependency on non-renewable energy sources. Given a reasonable location and a well-designed application, photovoltaic arrays can provide an excellent, cost saving solution for users requiring large amounts of power. This report details a system that can be used to implement a grid connected photovoltaic array with maximum power point tracking. The system consists of a dc-dc boost converter and H-bridge design with a passive filter and step-up transformer. The dc-dc converter utilizes an MPPT algorithm, charging a dc bus capacitor. The H-bridge maintains a constant voltage on the dc bus capacitor, and outputs a PWM signal which is then conditioned by a low-pass filter. The output voltage, current and power will be monitored, displayed on an LCD, and a USB drive will allow for the history of the system to be uploaded to a PC. The system hardware, software, and control systems are discussed in the following report. Simulations are included to demonstrate the overall operation of the entire system.

ELEG 4071 Senior Design II

University of Arkansas

Page |2

Table of Contents ABSTRACT ............................................................................................................................. 2 LIST OF FIGURES .................................................................................................................... 5 LIST OF TABLES ...................................................................................................................... 6 1. INTRODUCTION ................................................................................................................. 7 1.1 SYSTEM OVERVIEW ...................................................................................................................................................7 1.2 MAXIMUM POWER POINT TRACKING OVERVIEW ............................................................................................................8 1.3 MOTIVATION FOR PV ARRAY WITH MPPT ....................................................................................................................9

2. THEORETICAL BACKGROUND ........................................................................................... 10 3. HARDWARE DESIGN OVERVIEW ....................................................................................... 12 3.1 INTRODUCTION.......................................................................................................................................................12 3.2 PV ARRAY MODULES...............................................................................................................................................12 3.3 DSP .....................................................................................................................................................................14 3.4 ANALOG SIGNALS....................................................................................................................................................15 3.4.1 Current Sensors ...........................................................................................................................................15 3.4.2 Voltage Sensors ..........................................................................................................................................16 3.4.3 Dc Voltage Sensors .....................................................................................................................................17 3.4.4 Ac Voltage Sensor .......................................................................................................................................18 3.4.5 Digital Interface and Logic Circuitry ............................................................................................................20 3.6 POWER ELECTRONICS ..............................................................................................................................................21 3.6.1 Design of the Boost dc-dc Converter ...........................................................................................................21 3.6.2 Design of the H-Bridge Circuit .....................................................................................................................23 3.7 HEAT SINKS ...........................................................................................................................................................26 3.8 FILTER DESIGN .......................................................................................................................................................27 3.8.1 Low-pass Filter Topology ............................................................................................................................27 3.8.2 LC filter design ............................................................................................................................................28

4. SOFTWARE DESIGN OVERVIEW ........................................................................................ 29 4.1 INTRODUCTION.......................................................................................................................................................29 4.2 ANALOG TO DIGITAL CONVERTER ...............................................................................................................................30 4.3 GRID SENSOR .........................................................................................................................................................30 4.4 DC BUS VOLTAGE PI NETWORK .................................................................................................................................30 4.5 MPPT ..................................................................................................................................................................31

4.6 OVERALL SYSTEM SIMULATION ...................................................................................... 31 5. IMPLEMENTATION ........................................................................................................... 34 5.1 INTRODUCTION.......................................................................................................................................................34 5.2 POWER BOARD.......................................................................................................................................................35 5.3 SIGNAL BOARD .......................................................................................................................................................35 5.4 IGBT GATE DRIVER BOARDS .....................................................................................................................................36 5.5 TEXAS INSTRUMENTS PARTS......................................................................................................................................37

6. INSTRUMENTATION ......................................................................................................... 38 6.1 HARDWARE ...........................................................................................................................................................38 6.1.1 PIC Microcontroller .....................................................................................................................................38 6.1.2 Explorer 16 Development Board .................................................................................................................39

ELEG 4071 Senior Design II

University of Arkansas

Page |3

6.1.3 Prototype PICtail Plus Daughter Board .......................................................................................................40 6.1.4 USB PICtail Plus Daughter Board ................................................................................................................41 6.1.5 LCD Screen ..................................................................................................................................................42 6.2 SOFTWARE ............................................................................................................................................................43 6.2.1 PIC24F Programming ..................................................................................................................................43 6.2.2 Visual Basic Program ..................................................................................................................................45 6.3 INSTRUMENTATION TESTING .....................................................................................................................................47

7. RESULTS .......................................................................................................................... 49 7.1 INTRODUCTION.......................................................................................................................................................49 7.2 SYSTEM RESULTS ....................................................................................................................................................49 7.3 INSTRUMENTATION RESULTS .....................................................................................................................................55 7.4 CONCLUSIONS ........................................................................................................................................................57

8. ACKNOWLEDGEMENTS .................................................................................................... 58 9. REFERENCES .................................................................................................................... 59

ELEG 4071 Senior Design II

University of Arkansas

Page |4

List of Figures Figure 1: System Overview ...........................................................................................................................................7 Figure 2: MPPT Algorithm .......................................................................................................................................... 10 Figure 3: Single Module, Current vs. Voltage Relationship at Varying Insolation ..................................................... 12 Figure 4: Single Module, Power vs. Voltage Relationship at Varying Insolation ....................................................... 13 Figure 5: PV Array, Current vs. Voltage Relationship at Maximum Insolation .......................................................... 13 Figure 6: PV Array, Power vs. Voltage Relationship at Maximum Insolation ............................................................ 13 Figure 7: TI TMS320F2808 DSP ................................................................................................................................ 14 Figure 8: Current Sensor Network ............................................................................................................................... 15 Figure 9: Voltage Sensor Placement ............................................................................................................................ 16 Figure 10. dc Voltage Sensor....................................................................................................................................... 17 Figure 11. Ac Voltage Sensor. ..................................................................................................................................... 18 Figure 12: Output from ac Voltage Sensor .................................................................................................................. 19 Figure 13: Digital Control Circuitry ............................................................................................................................ 20 Figure 14: Open-loop Boost Converter ....................................................................................................................... 22 Figure 15: Open-loop Boost Converter Output Current and voltage ........................................................................... 22 Figure 16: H-Bridge Circuit......................................................................................................................................... 23 Figure 17: H-Bridge Simulation Output With Filter .................................................................................................... 24 Figure 18: Full System Schematic ............................................................................................................................... 24 Figure 19: Output Current of the Full System ............................................................................................................. 25 Figure 20: Output Current of the System and Output Voltage of the dc-dc Converter ............................................... 25 Figure 21: Heat Sink Calculator Figure 22: Heat sink ATS1195-ND ............................ 26 Figure 23: LC filter ..................................................................................................................................................... 27 Figure 24: Simulink System Model ............................................................................................................................. 29 Figure 25: MPPT Subsystem ....................................................................................................................................... 31 Figure 26: PV Array System Model ............................................................................................................................ 32 Figure 27: Current and Voltage Outputs of the H-Bridge from Figure 45 Zoomed in ................................................ 32 Figure 28: Voltage Across the DC Bus Capacitor ....................................................................................................... 33 Figure 29: Entire System Overview............................................................................................................................. 34 Figure 30: Populated Signal Board .............................................................................................................................. 35 Figure 31: Populated Gate Driver Board ..................................................................................................................... 36 Figure 32: PIC24FJ256GB110 Plug-In Module .......................................................................................................... 38 Figure 33: Microchip Explorer 16 Development Board .............................................................................................. 39 Figure 34: Prototype PICtail Plus Daughter Board ...................................................................................................... 40 Figure 35: USB PICtail Plus Daughter Board ............................................................................................................. 41 Figure 36: LCD Display .............................................................................................................................................. 42 Figure 37: Data Gathering Program............................................................................................................................. 44 Figure 38: VB Program Flow Diagram ....................................................................................................................... 45 Figure 39: Solar Array Data Acquisition VB Program ................................................................................................ 46 Figure 40: Test Results of the Solar Array Data Acquisition ...................................................................................... 48 Figure 41: Output of System (Yellow) Following the Grid (Blue) .............................................................................. 49 Figure 42: Output After Changing the Reference to the PI Controller ........................................................................ 50 Figure 43: Output Working in Real-Time ................................................................................................................... 51 Figure 44: System Voltage Output Before and After Transformer…………………………………………………...52 Figure 45: System Output Current and Grid Voltage………………………………………………………………...53 Figure 46: MPPT Charging of DC Bus Capacitor…………………………………………………………………....54 Figure 47: Output from the VB Program ..................................................................................................................... 56

ELEG 4071 Senior Design II

University of Arkansas

Page |5

List of Tables Table 1. Design Parameters ......................................................................................................................................... 21 Table 2: Filter Design Parameters ............................................................................................................................... 28 Table 3: Test .csv file for Solar Array Data Acquisition Testing ................................................................................ 47 Table 4: Resulting Array.csv ....................................................................................................................................... 55

ELEG 4071 Senior Design II

University of Arkansas

Page |6

1. Introduction 1.1 System Overview This system encompasses a single-phase, grid connected photovoltaic array with maximum power point tracking and can be broken down into the following constituent elements:

1. PV Array 2. Boost dc-dc Converter 3. Energy Storage Capacitor 4. H-Bridge dc-ac Inverter 5. Output Filter 6. Step Up Transformer 7. Utility Grid

An illustration of the general overview of the grid-connected PV array system follows in Figure 1. The voltage directly off of the PV array is stepped up to the dc bus voltage across the energy storage capacitor (57 V) by use of a boost converter. Using the MPPT algorithm (see MPPT section for details), the boost converter charges the dc bus capacitor. The H-bridge PI network maintains a steady output relative to the desired dc bus voltage. A Phase Lock Loop routine creates an ac reference signal that matches that of the utility grid . The output filter then shapes and filters the ac signal into a sinusoid. A 36 – 120 V step up transformer is used to step up the voltage to match that of the utility grid.

Each of the power switching devices found in the circuit utilizes an individual isolated gate driver (see section concerning gate drivers). These gate drivers receive their PWM input signal from the DSP (Texas Instruments 2808).

Figure 1: System Overview

ELEG 4071 Senior Design II

University of Arkansas

Page |7

1.2 Maximum Power Point Tracking Overview The DSP incorporated in this system is responsible for the MPPT algorithm and implementation. Using current and voltage sensors, the MPPT is able to adjust the duty cycle of the dc-dc boost switch to draw the maximum possible power out of the PV array at a given insolation. A general overview of the MPPT network is explored below (Figure 2). 

Current and voltage sensors are used to measure the current and voltage of the PV array to be read by the DSP.



A voltage sensor is used to measure the dc Bus voltage at the output of the boost stage and is read by the DSP.



The power from the PV array is calculated by the DSP.



From the power calculation a maximum power point duty cycle is obtained to deliver the proper amount of current necessary to charge the capacitor on the dc bus.



The inverter PI network controls the discharge of power from the capacitor necessary to keep the dc bus voltage above that of the utility grid.

The MPPT algorithm works by comparing the current PV power value it has read with the last one it recorded. Depending on whether or not the current power is greater or less than that of the old, the MPPT will perturb the inverter’s reference current by a discrete amount in either a positive or negative direction. As this is an iterative process, the maximum power will eventually be discovered and the current values will gravitate towards that point. By keeping the current at the MPP, the voltage follows given a constant insolation.

ELEG 4071 Senior Design II

University of Arkansas

Page |8

1.3 Motivation for PV Array with MPPT Renewable energy is one of the fastest growing trends in post-industrialized societies as they face growing energy demands and actively seek cost effective solutions. Among these solutions includes solar energy, specifically photovoltaic arrays. Photovoltaic arrays allow societies to drastically reduce energy expenses and dependency on non-renewable energy sources. Given a reasonable location and a well-designed application, photovoltaic arrays can provide an excellent, cost saving solution for users requiring large amounts of power. The goal of this project is to build and design a system that can be used to implement a grid connected photovoltaic array with maximum power point tracking. This system implements maximum power point tracking (MPPT) to ensure energy savings. By using an MPPT algorithm, this application will be able to extract the most power from the sun, given the limitations of today’s silicon-based photovoltaic cells, and implement said power into the local power grid. Due to its low latitude and relatively sparse cloud cover, the Northwest Arkansas region enjoys decent insolation. On average, the Fayetteville area receives 3.5 to 5 kW hr/sq. meter of insolation per day, which is more than enough to justify installation of a grid tied photovoltaic system.

ELEG 4071 Senior Design II

University of Arkansas

Page |9

2. Theoretical Background In order to understand the simulation, it is first necessary to be acquainted with the control system used in the project. The system employed is a modification of the system that was first introduced by Fangrui et al. [1]. In this control scheme, a TI 2808 DSP is used to control the duty cycle for the switching of the dc-dc converter and the gates of the h-bridge.

The algorithm programmed in the DSP to control the dc-dc converter is the perturb and observe method [1]. In this method, the DSP is consistently monitoring the voltage and current from the PV array. These are multiplied together to produce the power from the PV array. Then, this reading of the power is compared to the previously received power reading. If the power has decreased, then the sign of the multiplier reverses. So, if the multiplier were positive and the DSP detects that the power received most recently is lower than that previously received, the multiplier would then become negative. If the power has increased, the multiplier will stay the same. Then, this multiplier is multiplied by a constant value which is then added to the current duty cycle to produce the next iteration’s duty cycle. The duty cycle is then changed, and the power is then sampled again.

In this way, the DSP is consistently working to get to the maximum power point (MPPT) of the PV array for the given insolation and temperature. This is desirable because the temperature and sunlight received will constantly be changing throughout the day.

Figure 2: MPPT Algorithm

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 10

To control the H-bridge, a PI control network and a phase locked loop were employed to control the output current of the H-bridge system and synch the output current of the H-bridge with the voltage from the grid. For the PI controller, the desired voltage constant is the reference and the dc bus voltage across the capacitor is the input to the PI controller. The output of the PI controller is the amplitude of the current desired for the system.

As an input to the phase lock loop the grid voltage is fed into a DSP input after being shifted and scaled. The phase lock loop determines when the grid voltage crosses from positive to negative. It is the goal of the phase lock loop to follow a 60 Hz sine wave. From this fact, the PLL begins to count up from zero once a negative voltage crossing has occurred. Upon reaching 2*pi, it resets and constantly is in synch with the grid in this manner. The desired amplitude for the current (output of the PI controller) is multiplied by the output of the phase lock loop. This value is then sent into our PWM blocks using the switching scheme discussed above. The phase lock loop ensures that the output current from the h-bridge will be in phase with the voltage from the grid. This is important since real output power is the goal, and having the current and voltage at unity power factor will produce the desired results.

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 11

3. Hardware Design Overview 3.1 Introduction The design of the system required isolating components, effective grounds, and power supplies. Two grounds, digital and analog, were used to separate the digital circuitry from the analog circuitry. Also, circuit designs to interface the hardware with the DSP were implemented. Finally special precautions were taken to ensure voltage and current circuitry was isolated by using isolated components. All of these issues are addressed in the following sections along with the designs for the analog circuitry, the digital circuitry, and power electronics circuitry.

3.2 PV Array Modules For this application, two Kyocera photovoltaic modules (KD205GX-LP) were chosen. These modules were selected because of their low cost and high power output. At maximum insolation of 1000 2

W/m each module will produce 205 W at 26.6 V and 7.71 A [2]. These two modules will be connected in series to provide a maximum power of 410 W at full insolation.

Along with information obtained from the data sheet, a Simulink model [3] was used to determine the MPP for varying levels of insolation. These insolation levels start at 1000 W/m2 and fall in 200 W intervals until the final value of 200 W/m2. The results of the subsequent simulations are illustrated in the following figures.

Figure 3: Single Module, Current vs. Voltage Relationship at Varying Insolation

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 12

Figure 4: Single Module, Power vs. Voltage Relationship at Varying Insolation

Figure 5: PV Array, Current vs. Voltage Relationship at Maximum Insolation

Figure 6: PV Array, Power vs. Voltage Relationship at Maximum Insolation

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 13

3.3 DSP The DSP used in this project was the TMS320F2808 from Texas Instruments, Inc. The speed and options of this DSP set it apart, making it ideal for use in grid-connected power systems. The 2808 DSP boasts a 100MHz clock, 2 8-channel 12-bit analog-to-digital converters, 16 PWM outputs, and 35 individually programmable General-Purpose-Input/Output (GPIO) pins [4]. The 100MHz clock is helpful for this project as it will allow for quick responses required for the switching in the H-Bridge PI algorithm. The DSP also lends a hand in the area of programming. The programs for the DSP were written were encoded into MATLAB Simulink instead of the more tedious Assembly Programming Language. Additionally, the DSP was sold in an evaluation board which includes all necessary circuitry for the DSP. The evaluation board includes a USB connection which was used to connect to a computer for programming of the DSP. The evaluation board comes in two setups: one with the DSP soldered on the evaluation board, and another with a slot connection. Due to the possibility of errors resulting in permanently damaging the DSP, the socketed connection was selected as it allows for the ability to replace the DSP should it become damaged. Figure 7 shows the DSP on the evaluation board in the socket-connected configuration.

Figure 7: TI TMS320F2808 DSP

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 14

3.4 Analog Signals Current and voltage sensors were needed to communicate with the DSP. Voltage dividers were used for all of the sensors to ensure that the output voltage was between the allotted range for the DSP, 0V-3V. Dc-dc converters were used to boost the 5V power supplies used for analog signals, to power the isolating op-amps (Texas Instruments ISO122). Two current sensors and three voltage sensors were implemented. The operation and design of these sensors are discussed in the following subsections.

3.4.1 Current Sensors Hall-Effect current sensors from Allegro Microsystems, Inc were used in the PV and grid sensing circuitry. The ACS712 current sensor is capable of sensing ±30 A current which it converts into a 0-5V isolated signal [5]. The ACS712 current Sensor is ideal for its quick operation. The sensor’s output has a 5 µs rise time in response to changes in the input. It also has a low resistance allowing for less possible errors. The isolation is needed as the voltage reference of the array is not the same as that used by the DSP. This 0-5V signal is fed into a voltage divider to bring the max of 5V to below the max of 3.3V on the 2808 DSP. The PIC Microcontroller has a max input of 3.3V which is ideal to use the same signal for both devices. Below in Figure 8 is the described sensing network.

Test Current IN +5V U1 1

IP+1

Vcc

IP+2

Viout

8 R1

2

6.8k

C2 3 4

IP-1

Filter

IP-2

GND

Output to DSP/PIC

7 6 5

C1 0.1uF

R2 10k

1nF

ACS712 Test Current OUT

Gound

Figure 8: Current Sensor Network

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 15

3.4.2 Voltage Sensors This project required two different kinds of voltage sensors. The first type of voltage sensor required is a dc voltage sensor which was connected both on the output of the PV array and the output of the dc-dc converter as shown in Figure 9. This type of sensor will be visited in greater detail in the next paragraph. The second type of voltage sensor is an ac voltage sensor. The ac sensor was connected to the grid which was used to interface with the H-bridge and the grid sensing algorithm. This voltage sensor was also used to measure the power output from the entire system for data acquisition purposes.

Figure 9: Voltage Sensor Placement

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 16

3.4.3 Dc Voltage Sensors The dc voltage sensor is concerned with sensing a dc voltage source that is 90 V or less. This value was chosen for safety because there is potential for voltage across the capacitor to rise higher, due to the design of the boost converter. The dc voltage sensor begins with the dc voltage source being fed into a voltage divider circuit and then into an isolating amplifier. The isolating amplifier used in the dc voltage sensor is the Texas Instruments ISO122. It is used to isolate the control circuitry from the signal. The initial voltage divider circuit is necessary to scale the maximum of 90V input down to ±15V because the IS0122 only operates in this range. After the isolating amplifier, the signal is then fed into another voltage divider. In determining the resistor values to use in the voltage divider, the equation for a voltage divider was utilized: 𝑉out = 𝑉in ∗

R2 R 2 +R 1

The Vout will be no more than 3.3 V, and the Vin will be no more than 90 V. The sensor circuit is shown below in Figure 10.

Figure 10. dc Voltage Sensor.

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 17

3.4.4 Ac Voltage Sensor The ac voltage sensor circuit which will be used is shown below in Figure 11. The input to the circuit is grid voltage at 169.7 V amplitude signal and 60 Hz. This signal is then fed into an isolating amplifier, specifically the Texas Instruments ISO122. This isolating amplifier transfers a signal across a 2pF differential capacitive barrier. This serves to isolate the grid from the rest of our control circuitry.

After the isolation amplifier, the signal is fed through a resistive network in order to begin paring down the voltage levels that the DSP will be able to read. The signal is then fed into an inverting op-amp and sent to the DSP. This op-amp serves two purposes. Primarily, it serves to shift the voltage up so that it will be impossible for the DSP to receive a negative voltage which would damage the DSP. Additionally, the op-amp performs additional scaling down of the voltage. In the original design, there was an additional unity-gain inverting op-amp to invert the signal from the grid, so that when it was inverted by the shifting amplifier, it would then be the correct signal. This was unnecessary because the signal, once received by the DSP, can be inverted easily.

0

V2 15Vdc R4 1.8k V

4

U1 R5 2

OPA124/BB -

V-

R1 5.1k

OUT 3

7

VOFF = 0 VAMPL = 14.1899 FREQ = 60

+

V+

10k V35

6

R3 5.7k

V1 V34

15Vdc

1.8Vdc

0

0

Figure 11. Ac Voltage Sensor.

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 18

Figure 12 below is the transient response from the circuit in figure 11. The output is a sinusoid with a peak no more than 2.4V and a minimum no less than 1.2V if the input is a sinusoid with a peak at 169.7 V. The grid is known to vary slightly, but there is still plenty of room between 0 and 1.2V so the minimum of the grid should not be a concern. However, the signal to the DSP cannot go above 3.0 V meaning that there is a little more than .6V at the maximum of the sinusoid. While it may seem thin, it must be noted that the sinusoid, once shifted back down to the x-axis, has an amplitude of .6V. A .6V margin of error is actually around 100% of the amplitude of the signal. This means that the grid would have to have a huge voltage spike to send the signal out of the range of the DSP analog to digital converter which would not be likely. Consequently, the voltage range is a good one to send to our DSP. 2.4V

2.2V

2.0V

1.8V

1.6V

1.4V

1.2V 0s

5ms

10ms

15ms

20ms

25ms

30ms

35ms

40ms

45ms

50ms

V(R2:2) Time

Figure 12: Output from ac Voltage Sensor

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 19

3.4.5 Digital Interface and Logic Circuitry In order to interface the DSP with the gate driving circuits, digital interfacing circuitry must be implemented. The gate drivers require a +5 V logic level, but the DSP only provides up to +3.3V. To match up the incompatible logic levels, the 74LVX3245 translating transceiver will be utilized. The transceiver is powered by a 5V digital power signal and a 3V power signal received from the DSP. One preventative measure taken in anticipation of two switches on the same leg of the H-bridge being simultaneously turned on was to implement further logic circuitry. This circuitry shown in Figure 13 will prevent shorting of the system. This is a necessary precaution to ensure correct operation of the system.

VCC R1 3.3k Enable

U1A 1

2 7404

U2A 1 2 13

Signal_1

12 Vout1

Out_1

7411 U1B 3

4 7404 U2B 3 4 5

Signal_2 U1C 5

6

Vout2

Out_2

7411 6

7404

Figure 13: Digital Control Circuitry

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 20

3.6 Power Electronics A dc-dc boost converter, h-bridge, transformer, and filter make up the power electronics design of the system. The boost converter charges a capacitor to the selected operational voltage of 57V. The hbridge discharges the capacitor by drawing current from it and switches the dc signal to 60Hz. The output of the H-Bridge is sent to a filter to condition the output signal. This signal is then stepped up through a transformer from 57V to 169.7V. This value is the 120Vrms that the grid operates at. The method of designing the boost converter, h-bridge, and filter are outlined in the following sections.

3.6.1 Design of the Boost dc-dc Converter

The purpose of the boost converter in this design is simply to provide current at the MPPT to charge the dc bus capacitor. This is accomplished through a PWM output from the DSP and is controlled by the MPPT algorithm [1].

Open loop simulation setup and results are illustrated in figures 14 and 15. The closed-loop simulations and control scheme are discussed in the Software Overview. Based on these simulations, it can be seen that the converter stabilizes around 300ms and outputs steady voltage and current levels. This is critical in providing a usable dc Bus.

Table 1. Design Parameters Parameter

Initial

Final

Capacitance

620µF

5600µF

Inductance

1.2mH

1.2mH

Switching

20kHz

20kHz

Frequency

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 21

Figure 14: Open-loop Boost Converter

Figure 15: Open-loop Boost Converter Output Current and voltage

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 22

3.6.2 Design of the H-Bridge Circuit The H-bridge is composed of 4 switches as shown in Figure 16 below. The objective of an Hbridge is to take a dc signal and produce a sinusoid out of it. Specifically, in this application, the system will be capable of taking a 57V dc signal from the dc-dc converter and producing a PWM signal that can be filtered to ultimately generate a sinusoidal signal that is in synch with the grid. It is important that the signal be in phase with the grid so that there is no complex power.

This is achieved in simulation using PSPICE software. The schematic was built using different components in the software program as shown below. ABM blocks were used to perform comparisons in order to switch our four switches as seen below in Figure 16. The switching scheme used was unipolar as discussed in [6].

Next, the circuit from Figure 16 was simulated, and the output was recorded in Figure 17 below.

Figure 16: H-Bridge Circuit

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 23

Figure 17: H-Bridge Simulation Output With Filter

Finally, the dc-dc converter was added to the schematic as shown in Figure 18 below. Simulations are shown in Figures 19 and 20. Figure 19 illustrates the current output of the filter, and Figure 20 shows both the current output of the filter and the voltage across the capacitor of the dc-dc converter. The frequency of the signal in Figure 20 is very accurate: 𝑓=

1 = 59.930 𝐻𝑧 16.686𝑚𝑆

Figure 18: Full System Schematic

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 24

Figure 19: Output Current of the Full System

80

(60.065m,59.522) 60

(64.323m,57.717)

40

20

0

-20 0s V(C2:2)

10ms -I(R28)

20ms

30ms

40ms

50ms

60ms

70ms

80ms

90ms

100ms

Time

Figure 20: Output Current of the System and Output Voltage of the dc-dc Converter

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 25

3.7 Heat Sinks It will be necessary to have a heat sink for our gates (IGBT’s). The IRGS4B60K IGBT will be used in the system. This IGBT has a maximum power output of 63W. A heat sink calculator was located [7]. From this calculator, specifications from the ATS1195-ND heat sink were entered and an output from the calculator is shown below in Figure 21. It is shown that the heat sink chosen will make the IGBT run at a much cooler temperature. Figure 22 is the chosen heat sink.

Figure 21: Heat Sink Calculator

ELEG 4071 Senior Design II

Figure 22: Heat sink ATS1195-ND

University of Arkansas

P a g e | 26

3.8 Filter Design

3.8.1 Low-pass Filter Topology To avoid using a gain operation amplifier, a passive analog filter was used. Due to the switches in the H-bridge inverter operating at 20 kHz, unwanted harmonics appear at the output. In order to lessen the noise and present a 60 Hz sinusoidal signal, a low-pass filter was used to remove the high frequency components. The LC topology gives a 20 dB/decade drop-off. With only two elements in this filter, the cost is low. The LC filter is the most often used filter for single phase ac systems. With a large inductor facing the dc supply and the large capacitor providing low impedance for the switching frequency, the LC was a good choice for the needed type of operation.

Figure 23: LC filter

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 27

3.8.2 LC filter design

RD is a load used to simulate the grid for testing, it is defined as the minimum voltage (V min) divided by the maximum current (Imax). The measured needed loss is (20log(Vo)) with an added head room. With an LC filter, the loss per octave is 24 dB.

𝑅𝐷 =

𝑉𝑚𝑖𝑛

𝐼𝑚𝑎𝑥

=

36 = 4.5Ω 8

frequency at needs loss = 𝑓𝑠𝑤 = 20 𝑘𝐻𝑧 𝑑𝐵 = loss required = 20log(𝑉𝑜 2) = 34.14 + headroom of 3dB = 37.13 dB Cutoff Frequency = 𝑓𝑜 =

𝑓 2

𝑑𝐵

𝐿

=

𝐿= 𝐶=

20,000 237 .13/24

= 6.844 𝑘𝐻𝑧

𝑅𝐷 4.5 = = 0.1046 𝑚𝐻 2𝜋𝑓𝑜 2𝜋6,844

1 1 = = 5.1677 𝜇𝐹 2𝜋𝑓𝑜 𝑅𝐷 2𝜋6,844 × 4.5 Table 2: Filter Design Parameters Vmin (Vo)

36 V

Harmonic Peak (Vo rms)

50.91 Vrms

Imax

8A

RD

4.5 Ω

fsw

20 kHz

Measured Needed Loss

37.13 dB

Cut-off Frequency (fo)

6.844 kHz

L

0.1046 mH

C

5.1677μF

Realistic C

4.7 µF

Since 5.2µF is not a typical capacitor value, a common value is needed. The closest common capacitor value to 5.2µF is 4.7µF. Since the capacitor will be used for an ac voltage, a polypropylene dielectric will be used.

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 28

4. Software Design Overview 4.1 Introduction

This section covers the details of the entire software system as it is designed to run on the DSP. The system was designed using Simulink and then compiled using Code Composer Studio. An overview of the Simulink model can be found in the figure below. Each of the following subsystems is discussed in detail in their respective sections.

Figure 24: Simulink System Model

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 29

4.2 Analog to Digital Converter

The analog to digital converter block reads the values of the sensors implemented in our system. The sensors are utilized are: 

Boost inductor current



PV voltage



dc bus voltage



Grid voltage sensor



Filter inductor current

These values are read in as fixed point values and converted throughout the system to aid in computation.

4.3 Grid Sensor

The grid sensor operates by shifting the fixed point input values from the ADC to allow them to have both negative and positive values. These values are then scaled and used with a PLL algorithm to match the grid voltage that has been sensed. The result of these calculations are run through another algorithm with the output of the PI network used to produce a PWM signal which, in turn, operates the switches of the H-Bridge.

4.4 Dc Bus Voltage PI Network

The PI network used to regulate the amount of voltage drawn off of the dc bus capacitor receives its values from the ADC. These values are also converted and scaled appropriately to match the real world voltage value of the dc bus capacitor. A reference voltage along with these scaled values is the inputs to the PI network. This allows a particular bus voltage to be maintained. This also ensures a steady output voltage that can be used when interfacing with the power grid. The output is used with the grid sensor values to determine the PWM signal for the H-Bridge.

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 30

4.5 MPPT The MPPT algorithm is implemented using a combination of if/else action subroutines that operate in accordance with the ―perturb and observe‖ algorithm theory addressed in section 2. However in this instance, the PV voltage and boost inductor current values are left as fixed point values as they are multiplied to determine the current PV output power. The output of the MPPT subsystem creates a duty cycle as a percent value and is used to create a PWM signal that is output by the DSP. This subsystem is illustrated below.

Figure 15: MPPT Subsystem

4.6 Overall System Simulation To verify the design, the overall system was simulated in MATLAB to examine the output current and voltage of the system and to ensure successful operation of the closed-loop control systems. Figure 26 shows the system modeled in Simulink. The output of the overall system can be seen in Figure 27. One modification was made to the control system pictured below. Instead of using a PI network to synch the output current to the grid voltage, a phase lock loop was implemented due to speed and stability issues.

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 31

Figure 26: PV Array System Model

Figure 27: Current and Voltage Outputs of the H-Bridge from Figure 45 Zoomed in

Figure 28 verifies that the chosen control system will indeed reach a steady state on the dc bus voltage at the desired level.

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 32

Figure 28: Voltage Across the DC Bus Capacitor

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 33

5. Implementation 5.1 Introduction

To implement the system, a total of 7 PCBs were used, including 5 IGBT gate driver boards, a power board, and a signal board. All boards are connected directly to the power board through headers, which allows for the boards to be easily disconnected and tested separately. The DSP is connected to the signal board, and all signals from the power board must first go through the signal board before reaching the DSP. The transformer is connected by terminal blocks to the power board. Figure 29 below is a picture of the entire system together.

Figure 29: Entire System Overview

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 34

5.2 Power Board

The power board was a 10‖ by 14‖ two-layer PCB. All power devices needed for the system were placed on the power board. Both grid connections and the three dc power supplies were located on this board. Also the 5 IGBT gate driver boards and the signal conditioning board were connected to this board using headers. The main parts of the boost converter, the H-bridge and the filter were on this board, including the IGBTs. To avoid excessive current from entering the signal board, the current sensors were placed on the power board, as well as voltage dividers for voltage reading. Terminal blocks were used to separate the boost converter from the H-bridge which allows for simple troubleshooting and testing.

5.3 Signal Board

The signal board was a 4‖ by 7‖ four-layer PCB using two headers to connect to the power board. The DSP was connected to the signal board via headers. The signal board contained the digital circuitry used to condition the signals to be used by the DSP. One terminal block was used on the signal board, to be used by the PIC for voltage sensing. The populated signal board can be seen in Figure 30.

Figure 30: Populated Signal Board

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 35

5.4 IGBT Gate Driver Boards

The IGBT gate driver boards are 1.5‖ by 1.5‖ single layer PCBs using three separate headers each to connect to the power board. A separate gate driver board was used for each IGBT. Each gate driver could be easily removed and replaced by another, allowing for simple troubleshooting. A populated gate driver board can be seen in Figure 31.

Figure 31: Populated Gate Driver Board

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 36

5.5 Texas Instruments Parts A TMS320F2808 DSP was used to run a maximum power point tracking algorithm (to get the maximum power out of our solar panel for a given level of sunlight) which was then used to drive a boost converter using the duty cycle obtained from that algorithm. The DSP is also used to control the output current (to be in phase with voltage from the grid) from the system using a PI network, a software phase lock loop and the PWM outputs of the DSP

An ISO122JP isolation amplifier was used to isolate voltage sensor signals from the power board to be used by the DSP.

A REF3318 voltage reference was used as an input to the OPA2131 to shift the scaled sinusoidal grid voltage signal positively by 1.8 volts, so that no negative voltage would be entered into the DSP.

An OPA2131 was used as a voltage adder to shift the sinusoidal grid voltage input to be positive at all times. This is important because the DSP analog inputs cannot read a negative voltage value.

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 37

6. Instrumentation 6.1 Hardware 6.1.1 PIC Microcontroller The instrumentation portion of this project will run off of a PIC Microcontroller by Microchip Technology, Inc. The PIC24FJ256GB110 has many features which give it the spotlight on the instrumentation. The biggest design aspect of the instrumentation portion of this project is to record and provide data for the performance of the array. The PIC features a powerful clock capable of processing 16 MIPS. This will be more than sufficient for this portion of the design. The PIC also supports USB-onthe-Go (USB OTG), it houses a Real-Time Clock/Calendar (RTCC), a 10-bit Analog-to-Digital Converter, and 83 pins over 7 ports for general purpose input/output (GPIO) functions[8]. These are the features which are instrumental to use in this portion of the project. The GPIO pins will be configured for USB and LCD output. The LCD screen will be used for displaying current power characteristics of the PV Array. These characteristics will be obtained from the A/D converter which has its inputs coming from the voltage and current sensors. The Real-Time Clock/Calendar is extremely important to provide the time and day of each gathering of data. When dumped into a .csv file, it will be easily parsed through to enable the showing of historical data of the performance of the PV Array. Like the DSP, the PIC provides an easy way to program the device using the very powerful and easy to use C-Programming language. The PIC24FJ256GB110 will be used in the configuration as a Plug-In Module (PIM) which will be inserted on the Explorer 16 Development Board. This allows for easy swapping of the PIM should something disable or short the microcontroller.

Figure 32: PIC24FJ256GB110 Plug-In Module

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 38

6.1.2 Explorer 16 Development Board Microchip offers a development board to create an environment which is easy to use and allows for quick understanding of the PIC Microcontroller. The Explorer 16 Development Board has many features which make it the ideal way to use the PIC Microcontroller for this project. It offers a vertical and horizontal PICtail Plus Daughter card connection for which Microchip has developed an array of daughter cards which can be used to increase the abilities of the Explorer 16 Development Board. It also features a temperature sensor and potentiometer connected to the analog-to-digital converter ports of the PIM for quick learning of the Adc. The development board also features a 2-line dot matrix LCD display, multiple push-buttons and LEDs for easy user-configurable programming.

Figure 33: Microchip Explorer 16 Development Board

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 39

6.1.3 Prototype PICtail Plus Daughter Board The PICtail Plus expansion slot on the Explorer 16 Development Board is a very useful tool for user of understanding the PIC microcontrollers. The vertical slot will hold a prototyping daughter board. This board features an 8cm x 8cm section of protoboard for general wiring use. It also provides access to all 100 pins of the PIC24F microcontroller. This feature will be used for providing connections to the LCD screen and a busy LED.

Figure 34: Prototype PICtail Plus Daughter Board

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 40

6.1.4 USB PICtail Plus Daughter Board The USB PICtail Plus Daughter Board is an expansion card which can be used on either of the two PICtail Plus expansion slots on the Explorer 16. This expansion card allows for an easy accessible way to learn how USB functionality works. There are three possible applications of the USB daughter card: Embedded Host, Device Mode, and USB On-The-Go. Embedded Host is the functionality used for this project which gives the ability to host a USB Thumb Drive to open and edit a file. These modes are manually selected through the four on-board jumpers.

Figure 35: USB PICtail Plus Daughter Board

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 41

6.1.5 LCD Screen To display the current statistical data of the PV Array, an LCD screen from Purdy Electronics Corporation. The AND721GST/GST-LED Intelligent Character Display boasts 4 lines capable of displaying 20 characters each [9]. The AND display if ideal for this type of project as it will run off of only a 5V power supply which will be provided through a TI dc-dc power converter (Texas Instruments, Inc. Part Number: dcH010515DN7). The LCD display is also easy to program utilizing a single 8-bit bus coming from the PIC Microcontroller in the form of Assembly-style register programming.

Figure 36: LCD Display

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 42

6.2 Software

6.2.1 PIC24F Programming The program on the PIC Microcontroller will be programmed using the C-Programming Language inside Microchip’s MPLAB development environment. This program is designed to interrupt every 10 minutes. This interrupt starts by collection the values from the Real-Time Clock/Calendar and Voltage and Current sensors from the Analog-to-Digital Converter stored in the temporary memory of the PIC. The first event to check for is if the PV Array is on. This is done by checking if the Voltage and Current values are equal to zero. If this is true, the .csv file, and the LCD Screen are updated to show the most recent values and returns from the interrupt to wait another 10 minutes for the next interrupt period. If the PV Array is on, then the Voltage and Current values are scaled back up to reflect the actual values. These values were scaled down from the simple voltage dividers to a max of 3.3 volts. The power is calculated using the following power equation:𝑃 = 𝑉 ∗ 𝐼. This value is stored in a .csv file in the following format per line: YEAR,MONTH,DAY,HOUR,MINUTE,POWERVALUE. Each line will show this and the resulting file will be read by the Visual Basic program to parse through the file. The following is a diagram of the flow chart for the program just described.

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 43

10 Minute Interrupt

Grab Voltage, Current, & RTCC Values

Scale Voltage/Current Values

Calculate Power

Open USB Thumb Drive

Update array.csv

Update LCD Screen

Return from Interrupt Figure 37: Data Gathering Program

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 44

6.2.2 Visual Basic Program In order to view the historical data of the PV Array, a file by the name of pvarray.csv will need to be obtained from the PV Array. This file will be on a thumb drive and should be inserted into a computer where the Visual Basic program is residing. From the program, browse and open the pvarray.csv file. This file will be opened in the program and parsed through to gather and display the historical data. The following is a flow chart of the VB program which will parse through the pvarray.csv file.

Figure 38: VB Program Flow Diagram

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 45

Figure 39: Solar Array Data Acquisition VB Program

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 46

6.3 Instrumentation Testing To test the instrumentation, the sensors will be connected in a way that functionality can be determined. For the ac sensor, it will be connected to the grid and the microprocessor used in the instrumentation portion. In terms of dc sensing, a dc sensor will be connected to a dc voltage supply and then also to the microprocessor. From this, functionality of the instrumentation portion will be determined. A laptop computer will also be connected to the interface on the instrumentation to confirm the ability to get information from the past about the PV-array system.

To test the Solar Array Data Acquisition VB program, a mock .csv file will be created to simulate the output file from the PIC Microcontroller. The created file is shown below. This file is created in a way to illustrate that the program will work regardless of the year it is created in. The ensuing table displays one of the resulting outputs of the program.

Table 3: Test .csv file for Solar Array Data Acquisition Testing

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 47

Figure 40: Test Results of the Solar Array Data Acquisition

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 48

7. Results 7.1 Introduction This section will provide results of the system and conclusions about the product. 7.2 will display results from the complete system output. This section will also highlight the output from the system utilizing the PI controller in order to set the output voltage at a level keeping our dc bus voltage as a specified constant. Section 7.4 will include potential future modifications and conclusions regarding the project.

7.2 System Results The results from our system are very promising. For testing and verification, the input to the system was a dc power supply. Output was taken across a 80 ohm resistive load. An output waveform can be seen in Figure 38 below. This shows that our output (yellow) indeed follows the grid (blue) at 60 Hz. The output is being held constant which is a product of our PI controller leveling out as expected. The filter performed fairly well, although some higher frequencies were still passed.

Figure 41: Output of System (Yellow) Following the Grid (Blue)

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 49

Next, the constant reference input in the PI controller was modified, and the output voltage changed as a consequence. This proves that the input reference voltage is adjustable. Again, the output current (yellow) is a 60 Hz sine wave in synch with the grid voltage (blue).

Figure 42: Output After Changing the Reference to the PI Controller

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 50

Below is a screenshot of the output working in real-time, constantly adjusting the amplitude in order to match the reference constant in the PI controller.

It can be seen near the origin of the

oscilloscope graph that the system senses the rising voltage on the boost capacitor and proceeds to increase the output voltage into the grid.

Figure 43: Output Working in Real-Time

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 51

The system with the transformer was tested with the boost stage in open-loop with a duty cycle of 65% and the input at 12V and 1.5A, a resistive load of 80Ω after the transformer, the DC bus voltage was maintained at a constant 58V, which put the output from the filter at 42Vrms. After the transformer, the voltage across the load was 134Vrms. The figure below proves the system’s ability to reach the 120Vrms needed for grid-connection. The voltage before the transformer (yellow) is in phase with the grid voltage (blue). The filter’s performance can be seen in this figure. The noise and flattened peak could be explained by the selection of the filter inductor and capacitor based on availability; better performance could be expected with more accurate parts. Figure 45 shows the output current (blue) is in phase with the grid voltage (yellow), supplying real power. These two figures prove the system is capable of supplying completely real power in phase with the grid voltage.

Figure 44: System Voltage Output Before and After Transformer

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 52

Figure 45: System Output Current and Grid Voltage

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 53

Figure 46: MPPT Charging of DC Bus Capacitor Figure 46 shows the charging of the DC bus capacitor to the target voltage for MPPT. This was accomplished with an input from a DC power supply of 5V at 0.33 A. This shows how the duty cycle of the boost converter changes to charge the capacitor at the maximum power point for the system. Unfortunately this is not as effective with a DC power supply as it is with actual solar cells—mostly because the power supply will enter either current or voltage controlled mode to regulate its output. This sometimes interferes with the MPPT algorithm implemented.

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 54

7.3 Instrumentation Results The goal of the Instrumentation is to gather and process data from the ac Voltage and Current sensors. The resulting data is written to a .csv file on a USB Thumb Drive. The file from the thumb drive can be loaded onto a computer which holds the Visual Basic program which will parse through the file as described previously. Below in Table 4 is a resulting array.csv file gathered during the demonstration. Following is a figure of the results produced from the Visual Basic Program.

Table 4: Resulting Array.csv

2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009 2009

ELEG 4071 Senior Design II

April April April April April April April April April April April April April April April April April April April

28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28 28

University of Arkansas

14 14 14 14 14 14 15 15 15 15 15 15 15 15 15 15 15 15 15

81 64 53 54 43 45 83 64 57 54 42 68 70 42 74 124 42 51 21

P a g e | 55

Figure 47: Output from the VB Program

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 56

7.4 Conclusions

The goal of the project is to design and build a dc-dc converter driven by an MPPT algorithm that charges a capacitor. This capacitor’s voltage is then regulated by a PI controller connected to an H-bridge inverter used to create a pwm signal, which is then filtered by a low-pass filter to eliminate the highfrequency harmonics and components of the signal. The output of the system is kept in synchronization with the grid by the phase lock loop. A major limiting factor of the current system is the dc bus capacitor. It is rated to only 80V which limits the input to the system. Using a large enough capacitor, the system could even be used without the need for a step-up transformer (a one-to-one transformer would most likely be used to provide isolation). The results of the system show the closed loop performance, the grid voltage and current synchronization. Due to the lack of a decent solar array and the lack of a DC power supply with adequate capabilities, the MPPT of the system can be shown only through simulation and practice where applicable.

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 57

8. Acknowledgements Dennis Rogers, for his help in the design of the mechanical support for the PV array. Diogenes Molina, for his recommendations on the circuitry and control design. Dr. Balda and Derik Trowler, for their guidance throughout the project. Tanner Blair for his help in the implementation of the VB Project code. Art Barnes for his help with control implementation. Ricardo Castillo for his help with Simulink and the programmable power supply. Coilcraft for their generous donations

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 58

9. References [1] L. Fangrui, Y. Kang, Y. Zhang, and S. Duan. "Comparison of P&O and Hill Climbing MPPT Methods for Grid-Connected PV Converter," Industrial Electronics and Applications, 2008,ICIEA 2008. 3rd IEEE Conference on. pp. 804-7. 2008. [2] KYOCERA Corporation, ―KD205GX-LP Data Sheet,‖ kd_205gx_lp_081508_web, Oct. 2008. [3] D. Maksimovic, R. Erikson, R. Zane. (2008, Jan.). ―ECEN 2060 MATLAB/Simulink Materials.‖ University

of

Colorado.

Boulder,

CO.

[Online].

Available:

http://ece-

www.colorado.edu/~ecen2060/matlab.html [4] Texas Instruments, Inc., ―TMS320F280x Data Manual,‖ SPRS230J, Sept. 2007. [5] Allegro MicroSystems, Inc., ―AC712 Datasheet,‖ ACS712-DS, Rev. 7, 2007.

[6] N. Mohan, T. Undeland, and W. Robbins, Power Electronics: Converters, Applications, and Design. Hoboken, NJ: Wiley, 2003. [7] A. Malhammar, ―Natural Convection Heatsink.‖ Thermal Design of Electronics, Frigus Primore, 11/2008. [Internet]. Available: http://www.frigprim.com/online/natconv_heatsink.html. [Accessed: 11/2008]. [8] Microchip Technology, Inc., ―PIC24FJ256GB110 Family Data Sheet,‖ DS39897B, Jan. 2008. [9] Purdy Electronics Corporation, ―AND721GST/GST-LED,‖ www.purdyelectronics.com, July 1999.

[10] S. Ang, A. Oliva. Power-Switching Converters, 2

nd

ed. Boca Raton, FL: Taylor and Francis Group,

2005, pp. 32-38

[11] D. Trowler, "IGBT Gate Driver With an Integrated Power Supply," B.S. thesis, University of Arkansas, Fayetteville, AR, 2008.

[12] P. Khamphakdi, W.Khan-ngern, ―The Analysis of Output filter for Grid Connected Single Phase Full Bridge Inverter Based on PSpice Simulation Technique‖

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 59

ELEG 4071 Senior Design II

University of Arkansas

P a g e | 60