Content uploaded by M. Tariq Iqbal
Author content
All content in this area was uploaded by M. Tariq Iqbal on Dec 17, 2020
Content may be subject to copyright.
IEEE 15th, NECEC conference
St. John’s NF, 2005
1
A Small Induction Generator Based Grid Connected Wind Turbine Simulator
M. T. Iqbal
Faculty of Engineering and Applied Science
Memorial University of Newfoundland
St. John's, Newfoundland, Canada, A1B 3X5
E-mail: tariq@engr.mun.ca
Abstract
A number of types of small wind turbines are available in the market. Such wind turbines
are primarily designed for isolated operation. The available small wind turbines are based
on permanent magnet type generators or field regulated alternators. A common
application of such wind turbines is battery charging. Such wind turbines may be
connected to the grid using commercially available grid tie inverters, in combination with
a charge controller, a battery bank and a dump load. This option is not cost effective
mainly because of the high cost of the power electronics, batteries and ongoing
maintenance cost of the battery bank. An induction generator based wind turbine that
meets all the requirements of the IEEE 1547 standard would be the best low cost option
for households. This paper describes the design and development of a drive train and a
controller for a small induction generator based wind turbine. The paper focuses on a
wind turbine simulator developed to help design and test the proposed induction
generator based wind turbine. The wind turbine simulator consists of a Labview based
computer controlled DC motor, a three-phase induction generator, instrumentation for
voltage, current, torque and speed measurements. The paper describes the first version of
the wind turbine controller that is based on speed measurement and a soft starter. System
design, initial test results and future directions are described in the paper.
Introduction
Wind energy development in the world is growing at a rate of almost 40% per year [1].
There are over 55000 large wind turbines installed today employing more than 70,000
people worldwide. As of July 2005 Canada’s installed wind energy capacity was 570
MW [2]. Canada has large utility-scale wind turbines installed in Alberta, Saskatchewan,
Ontario, Quebec, Prince Edward Island, Nova Scotia and the Yukon. Canada has a
national wind resource map accessible at www.windatlas.ca. Canada could reasonably
meet 20% of its total energy needs with wind power [2]. Canada has the ability to
manufacture utility-scale wind turbine components, such as blades, towers and nacelles.
However, within Canada there are currently no manufacturers of generators, gearboxes
and control systems, nor any comprehensive wind turbine manufacturing facilities.
Canadian Wind Energy Association has recently started a major initiative on small wind
energy (300W-300kW). Details may be found at www.smallwindenergy.ca
The wind turbine market place can be viewed in three tiers. The first tier involves large
utility grade machines, which are between 250kW and 5MW machines. Most wind power
is produced by these machines, which are typically installed in large farms. Utility grade
wind power accounts for most of the worldwide capacity. The second tier includes mid
sized machines from 20kW to 250kW and is typically grid connected machines in remote
IEEE 15th, NECEC conference
St. John’s NF, 2005
2
communities. This middle market has not matured into any particular use pattern. The
third tier includes machines smaller than 20kW, and goes down to around 300W. These
machines are used for individual residential use, remote site power, recreational marine
power and water pumping [3].
Canada has an immense wind energy resource that can be tapped. There is great interest
in wind energy amongst the general public especially people living in rural and remote
areas. People living in remote communities often pay rates as high as $0.20 per unit for
diesel generated electricity. Installation and use of medium and large systems need
considerable investment and is done by power producing utility companies using a
commercial utility interconnect. In contrast, small wind machines are installed by
homeowners, farmers and small businesses, typically using a grid tie inverter connected
to the smallest edges of the power grid. This generating architecture is known as
‘distributed’ or ‘micro’. A number of types of small wind turbines are available from
various manufactures [3,7]. Such wind turbines are primarily designed for isolated
operation. Small wind turbines are based on the permanent magnet type generators or
field regulated alternators. Common application of such wind turbines is battery
charging. Such wind turbines may be connected to the grid using commercially available
grid tie inverters, in combination with a charge controller, battery, and dump load
[8,9,10]. If a person living in a village or out of town wants to generate some electricity
using wind energy their only option is a combination of small wind turbine, battery
storage, charge controller, dump load and a grid tie inverter. This option is not cost
effective mainly because of high cost of power electronics and batteries and ongoing
maintenance of the battery. The Gridtek inverter from Bergey is a single exception,
where the inverter uses an ACDCAC architecture for connection to the grid, the
machine is still expensive and does not meet standards such as IEEE 1747 [4].
Small power production (Micro or Distributed) is being encouraged by the introduction
of net metering regulations. This option is already available in 34 states in US and in
parts of Ontario and Manitoba. New standards and laws regulating such micro production
are being developed and implemented. In the US, recently, the Institute of Electrical and
Electronics Engineers approved its new standard IEEE 1547 for interconnecting
distributed energy resources to the power system. IEEE 1547 supercedes the older IEEE
929 standard. An equivalent standard for Canada is being worked on and will be
implemented in the near future [4,5]. In the near future, net metering option will also
become available to every one in Canada. This will open a great opportunity to reduce
power bill for people living in villages and outside of towns.
The existing grid in rural communities is single phase. A transformer near the houses
converts single phase to 2-phase. A typical house has 2-phase input 120V, 200A each.
Most of the electrical load in the house is single phase. In a net metering situation if a
person wishes to produce small amount to electricity, say 5KW, using a wind turbine then
the best option is to produce as 1-phase 120/220V. Typical house energy consumption in
Canada is 2.8kW. A 3kW small grid wind turbine can bring down the power bill
significantly. If a wind turbine is suitably selected the net energy consumption in a house
IEEE 15th, NECEC conference
St. John’s NF, 2005
3
over a year may be zero [6]. Such an option would be a great attraction for people living
in villages and around small towns.
In this research we have developed a small wind turbine drive train based on a high slip
induction generator that is directly connected to the grid. A wind turbine controller being
developed make sure that this wind turbine meets all requirements of the IEEE 1547 and
Micro power connect interconnection guideline being developed for Canada. This paper
describes the system under development and some primary test results.
Wind Turbine Simulator and Controller
We are building a wind turbine drive train driven by a variable speed dc motor. The main
purpose of this drive train is to design and test of wind turbine controller. Figure 1 shows
a block diagram of the small wind turbines drive train. The drive train consists of an ac to
dc converter, a DC motor, an Induction generator, solid state relays and control
electronics. Presently a PC with Labview is being used for system control and testing.
Later we will design and build a microprocessor-based controller to control the system in
various situations. A rechargeable battery will power the system’s microprocessor based
controller, frequency circuit and sync circuit. The battery will be charged using generated
power from the system.
120/220VAC Line
L N
Wind Speed
Self-Excitation Capacitor
f V I
Figure 1.0 Block diagram of proposed small wind turbines Simulator
A typical control sequence of the wind turbine controller may occur as follows. Assume a
low wind, generator disconnected from the grid and turbine running at a low speed. As
the wind speed picks up, the generator speed will increase and its output signal frequency
Ac to dc
converter
DC
Motor
Induction
Generator
Solid
State
Relay
Main Winding
Solid
State
Relay
Auxiliary
Winding
Wind Turbine’s Microprocessor
Based Controller (to be developed)
Solid
State
Relay
Dump
Load
Frequency
Circuit
Sync
Circuit
Rechargeable
Battery
Instrumentation to
monitor grid
parameters and
transients
Control
PC
IEEE 15th, NECEC conference
St. John’s NF, 2005
4
will also increase. The generator is not self excited yet, therefore it will only produce a
small signal due to its remnant magnetism. The system controller will sense the generator
output via the frequency circuit and it will also receive a sync signal from the grid
through the sync circuit. When the generator output frequency is about same as the grid
frequency solid-state relay will operate and the generator will be connected to the grid.
After that, the controller will monitor the current flow between the grid and induction
generator. If current is positive i.e. power going to the grid, then controller will maintain
the grid connection. If current goes negative i.e. generator switch to motoring mode, then
controller will disconnect the generator from the grid. This may happen due to a drop in
the wind speed. The controller will keep monitoring the grid and generator output. If after
some time, the generator output frequency approaches the grid frequency the controller
will reconnect the system to the grid. This operation will repeat depending upon the wind
speed and availability of the grid. A hysteresis band will be used in the control algorithm
to reduce the number of switching. Dump load will be used when wind is available but
the grid is not available. On a windy day if grid is not available, then after observing the
frequency signal, the controller will attach the self-exciting capacitor to the auxiliary
winding of the induction generator. The controller will observe the voltage and current
output of the generator. If the voltage exceeds a certain limit (due to generator speeding
up) the controller will connect the dump load to the generator by operating a solid-state
relay. This will make sure that wind turbine does not over speed due to soft stall [11].
The controller will connect and disconnect the dump load as required on the basis of
current and voltage measurement to make sure the system operates safely. If grid
becomes available then controller will disconnect the dump load and connect the
generator back to the grid. The decision whether to maintain or not to maintain grid
connection will be based on the generator current from its main winding.
In the lab environment, as shown in the figure 1.0 the induction generator will be
connected to a variable speed DC motor. This setup is required to test and finalize the
controller design. During the testing phase induction generator will operate in the
variable speed mode using the DC motor and various system signals as well as grid
parameters will be recorded. Controller design will make sure the system meets the
Micropower Connect Interconnection Guidelines [4, 12, 13, 14]. For example, on a
windy day in the event of grid failure the system will disconnect within 2 cycles if
voltage is more than 137% or it will disconnect within 6 cycles if voltage is less than
50%. Wind turbine system induced grid voltage and current transients will be recorded in
all possible situations to make sure system meets micropower connect interconnection
guidelines and new IEEE 1547 standard. System based on 5hp induction machines will
be tested on a standard 60A utility outlet [13]. An off the shelf microprocessor and
development PCB will be used to build the controller for the generator system. This
approach will allow for rapid prototyping of the controller and low cost reconfiguration
of connections to allow for maximizing performance of the system. The controller
application will be coded according to a state transition map. As such, the application
can be coded in any language with sufficient semantic expression, such as C, C++,
assembly or others.
IEEE 15th, NECEC conference
St. John’s NF, 2005
5
Progress So Far
Small wind turbine simulator is based on Mawdslay’s multiform experimental setup. A
photograph of the system is shown in figure 2.0 It consist of a 3kW, 120V shunt type DC
motor, a tacho meter, a torque transducer and a experimental machine. The experimental
machine can be connected as 2/4 pole basic dc machine or 2/4/6/10/12/14 pole 3-phase
induction/synchronous machine or A.C. commutator machine or single phase induction
motor. Presently, it is connected as 4-pole, 4 coils in parallel, 3 phase, Y-connected
induction motor with the rotor winding shorted by a shorting ring (sheet 11). It was
connected in this way to achieve voltage levels of 120/208V while running at about
1800rpm so that it can be connected to the grid. The DC motor is armature controlled
with a fixed 76V field supply. Controlled variable DC voltage for the DC motor is
obtained using a phase controlled relay and a bridge rectifier. Coupled 3-phase induction
generator is connected to the grid using a two relays based soft-starter. Generator current,
voltage, speed and mechanical torque are measured. Figure 3 shows details of
instrumentation electronics. A PC controls the system with a NI 6024 data acquisition
card and Labview 5.5 PC measures system speed, current, torque and voltage and
controls two relays. When system speed is about 1800rpm PC first connects the generator
to the grid-using relay R2. Series power resistors limit the surge current. After a delay of
few seconds PC switches on relay R1 so that the generator is connected directly to the
grid. The generator is disconnected from the grid if current drops to zero and stays for
some time.
IEEE 15th, NECEC conference
St. John’s NF, 2005
6
Figure 2: A photograph of current wind turbine simulator
Figure 3: Instrumentation details of WT simulator
IEEE 15th, NECEC conference
St. John’s NF, 2005
7
The system calibration equations are also provided in figure 3. The system can be
controlled in manual mode or in auto mode. Figure 4 and figure 5 show Labview control
diagrams in manual and in auto mode. In manual mode DC motor speed is controlled
using a virtual knob and relays can be engaged or disengaged using mouse. All
calibration equations are embedded in the Labview code so that display window provides
a calibrated display. In auto mode DC motor speed is controlled based on the pre-
recorded wind speed data stored in a file. This is shown in figure 5. Based on the current
rotor speed and wind speed the tip speed ratio is calculated. Present tip speed ratio is used
to calculate the desired torque. When the system is connected to the grid its speed is
determined by the grid frequency. In this case the shaft torque is adjusted to make sure
that the system is running at an optimum tip speed ratio. Motor speed is regulated such
that the system is running at the optimum tip speed ratio. A PID controller ensures that
the system is at optimum tip speed ratio. Grid connection and disconnection decisions are
based on speed and current measurements. PID controller parameters can be adjusted on
the Labview main screen. Essentially in auto mode, based on the prerecorded wind speed
data and system model the DC motor is controlled in such a way that it represents a 3kW
wind turbine rotor. It will produce a shaft torque same as a real wind turbine rotor
running in the wind.
Figure 4: Labview control diagram for manual control of WT simulator
IEEE 15th, NECEC conference
St. John’s NF, 2005
8
Figure 5: Labview control diagram of WT simulator
In other words squirrel cage induction generator will behave as if a wind turbine drives it.
With such a setup at hand we can work on the wind turbine controller design and test that
in the lab arrangement. Presently, a PC function as the wind turbine simulator and the
controller. Figure 6 shows PC screen shot while system is under going a test. System
speed, voltage, current and torque are displayed on the screen. Grid connection and
disconnection transients can be studied using this setup. Eventually as shown in the figure
1 a microprocessor-based controller will be designed to control the system. That
controller will make sure that system meets the IEEE 1547 standard. i.e if its voltage is
below 50V or above 120V then it disconnects in 0.16s, if voltage is more than 50V but
less that 88V then it disconnect in 2s, if voltage is more that 110V and less than 120V
then it disconnect in 1s. Similarly controller will also look at the frequency and make sure
that it is with in the range specified in the IEEE 1547 standard. Wind turbine simulator
describe above will help design and test the proposed wind turbine controller.
IEEE 15th, NECEC conference
St. John’s NF, 2005
9
Figure 6: Labview output during testing
Primary Test Results
The wind turbine simulator setup described above has been used to test a wind turbine
controller. Figure 7 and 8 shows some primary test results. Figure 7 shows what happens
to the generator speed and current when the induction generator running at about
1800rpm is connected to the gird through a soft-starter. As soon as the generator is
connected to the grid through series resistors its speed decreases and there is a current
surge. The resistors limit current surge. After few seconds when it is directly connected to
the grid its speed becomes a constant. This test can be used to identify the suitable series
resistors and time delay between operations of two relays. Figure 8 shows what happen
to the generator speed and current when it is disconnected from the grid. As soon as
induction generator is disconnected from the grid its speed increases and current
decreases. Such test can be used to find out the expected over speed spikes in the wind
turbine. In the near future further testing will be done on the system to determine the
following:
1. Best PID controller parameters for the wind turbine simulator
2. Suitable value and power rating of soft starter resistors (Presently it is 50W, 15Ω)
3. What is the suitable time delay between the operation two relays
4. How to automatically make decision about grid connection on the basis of speed
and about grid disconnection on the basis of current
IEEE 15th, NECEC conference
St. John’s NF, 2005
10
5. Design a microcontroller based system controller
6. Study the impact of connection and disconnection of small induction generator
based wind turbine on the grid.
7. Test wind turbine simulator and newly designed controller in a variety of possible
real time situations.
Figure 7: System response at grid connection. Channel
A is speed while channel B is Current
Figure 8: System response at grid disconnection.
Channel A is speed while channel B is Current
Conclusions
This paper described design and development of a wind turbine simulator and a controller
for a small induction generator based wind turbine. The paper reports the progress made
so far. The designed wind turbine simulator consists of a Labview based computer
controlled DC motor, a three-phase induction generator, instrumentation for voltage,
current, torque and speed measurements. The paper also described the first version of the
wind turbine controller that is based on speed measurement and a soft starter. Primary
test results are presented. It appears that a small wind turbine will have no significant
effect on the grid while connecting or disconnecting from the grid. Simulator being
developed is a great tool to design and develop induction generator based grid connected
wind turbine.
IEEE 15th, NECEC conference
St. John’s NF, 2005
11
Acknowledgements
Author would like to thank National Science and Engineering Research Council
(NSERC) for providing support for this research. Author would also like to thank Dr. A.
Rahman for providing his Mawdsley’s machine for this research.
References
Zervos A. and Teske S., ‘Wind force 12: How wind could provide 12% of the
world’s electricity by 2020’, Renewable Energy World, pp 73-89, July-August
2002.
Canadian Wind Energy Association. http://www.canwea.ca
Manufacturers of Small wind turbines, American Wind Energy Association.
(http://www.awea.org/faq/smsyslst.html)
http://www.micropower-connect.org
Chang, L., ‘Wind energy conversion systems,’ IEEE Canadian review, spring
2002
Iqbal, M.T., ‘A feasibility study of a zero energy home in Newfoundland,’
Renewable Energy, Vol. 29, No.2, pp 277-289, 2004
Sagrillo, M., ‘Apples & Oranges 2002 - Choosing a Home-Sized Wind
Generator,’ Home power, August/Sept 2002
Rogers, A. L., Manwell, J. F, and McGowan, J. G., ‘Design requirements of
medium sized wind turbine for remote and hybrid power systems,’ Renewable
energy, Vol. 26, No. 2 pp 157-168, 2002.
Corbus, D., et al, ‘ Small wind turbine testing and applications development,’
Proceedings of Wind power 99, Burlington, Vermont, June 20-23, 1999.
Clausen, P.D., and Wood, D. H., ‘ Recent advances in small wind turbine
technology,’ Wind Engineering, Vol. 24, No. 3, pp 189-201, 2000
Muljadi, E., Forsyth, T., and Butterfield, C.P., ‘ Soft-stall control versus furling
control for small wind turbine power regulation,’ Proceedings of Wind power ’98,
Bakersville, CA April 27-30, 1998.
Machado , R. Q., Marra., and E. G., Pomilio., ‘Bi-directional Electronic interface
of induction generator connected to a single phase feeder,’ International Journal
of Renewable Energy Engineering, August 2002.
Dubois, M., ‘Comparison of generator topologies for direct-drive Wind Turbines,
IEEE Nordic Workshop on Power and Industrial Electronics (NORPIE/2000),
Aalborg, Denmark, June 13-16, 2000.
Sharma, H. Islam, S., Pryor, T. and Nayar, C.V, ‘Power quality issues in a wind
turbine driven induction generator and diesel hybrid autonomous grid,’ Journal of
electrical and electronics engineering, Australia, Vol. 21, No. 1, pp 19-25, 2001