Article of Flow Energy after Glenn one space 5 15 04
AIAA-2004-5705, AIAA-2004-5756
Utilization
of Wind Energy at High Altitude*
Alexander
Bolonkin
V.P. of
Consulting and Research Co.
1310 Avenue R, #6-F,
аT/F 718-339-4563.а aBolonkin@juno.comаа http://Bolonkin.narod.ru
а Ground based, wind energy extraction systems have reached their maximum capability. The limitations of current designs are: wind instability, high cost of installations, and small power output of a single unit. The wind energy industry needs of revolutionary ideas to increase the capabilities of wind installations. This article suggests a revolutionary innovation which produces a dramatic increase in power per unit and is independent of prevailing weather and at a lower cost per unit of energy extracted. The main innovation consists of large free-flying air rotors positioned at high altitude for power and air stream stability, and an energy cable transmission system between the air rotor and a ground based electric generator. The air rotor system flies at high altitude up to 14 km. A stability and control is provided and systems enableаа the changing of altitude.
а This article includes six examples having a high unit power output (up to 100 MW). The proposed examples provide the following main advantages: 1. Large power production capacity per unit Ц up to 5,000-10,000 times more than conventional ground-based rotor designs; 2. The rotor operates at high altitude of 1-14 km, where the wind flow is strong and steady; 3. Installation cost per unit energy is low. 4. The installation is environmentally friendly (no propeller noise).
---------------------
* Presented in International Energy
Conversion Engineering Conference at
Keyword: wind energy, cable energy transmission, utilization of wind energy at high altitude, air rotor, windmills, Bolonkin.
A - front area of rotor [m2];
a = 0.1 - 0.25 exponent of wind coefficient. One depends from EarthТs surface roughness;
Aa - wing area is served by aileron for balance of rotor (propeller) torque moment [m2];
Aw - аarea of the support
wing [m2];
C - retail price of 1 kWh [$];
c - production cost of 1 kWh [$];
CL - lift
coefficient (maximum CL ╗ 2.5);
CD Ц drag
coefficient;
DCL,a - difference of
lift coefficient between left and right ailerons;
D Ц drag force [N];
Dr - drag of rotor [N];
E -а annual energy
produced by flow installation [J];
F Ц annual profit [$];
Ho = 10 m - standard altitude of ground wind installation [m];
H - altitude [m];
I - cost of
Installation [$];
K1 - life time
(years);
L - length of cable [m];
Ly Ц lift force of wing [N];
M Ц annual maintenance [$];
NЦ power [W, joule/sec];
No - power at Ho ;
r - distance from center of wing to center of aileron [m];
R - radius of rotor (turbine)[m];
S - cross-section area of energy transmission cable [m2];
V - annual average wind speed [m/s];
Vo - wind speed at standard altitude 10 m [m/s](Vo= 6 m/s);
W - weight of installation (rotor + cables)[kg];
Wy Ц weight of
cable [kg];
g - specific
density of cable [kg/m3];
h - efficiency coefficient;
q - angle
between main (transmission) cable and horizontal surface;
l - ratio of blade tip
speed to wind speed;
v - speed of transmission cable [m/s];
r - density of flow, r =1.225 kg/m3 for air at sea level altitude H = 0; r =0.736 at altitude H =5 km; r = 0.413 at H =10
ааааа km;
s - tensile stress of cable [N/m2].
Introduction
ааа Wind is a clean and
inexhaustible source of energy that has been used for many centuries to grind
grain, pump water, propel sailing ships, and perform other work.
аа Wind farm is the term used for
a large number of wind machines clustered at a site with persistent favorable
winds, generally near mountain passes. Wind farms have been erected in
а A program of the United States
Department of Energy encouraged the development of new machines, the
construction of wind farms, and an evaluation of the economic effect of
large-scale use of wind power.
а The utilization of renewable
energy (СgreenТ energy) is currently on the increase. For example, a lot of
wind turbines are being installed along the British coast. In addition, the
British government has plans to develop off-shore wind farms along their coast
in an attempt to increase the use of renewable energy sources. A total of $2.4
billion was injected into renewable energy projects over the last three years
in an attempt to meet the government's target of using renewable energy to
generate 10% of the country's energy needs by 2010.
а This British program saves the emission of
almost a millions tons of carbon dioxide.
аUnfortunately, current wind
energy systems have deficiencies which limit their commercial applications:
1. Wind energy is
unevenly distributed and has relatively low energy density. Huge turbines
cannot be placed on the ground, many small turbines
must be used instead. In
2. аWind power is a function of the cube of wind
velocity. At surface level, wind has low speed and it is non-steady. If wind
velocity decreases in half, the wind power decreases by a factor of 8 times.
3. The productivity
of a wind-power system depends heavily on the prevailing weather.
4. Wind turbines
produce noise and visually detract from the landscape.аааа
аа There are many research programs and
proposals for the wind driven power generation systems, however, all of them
are ground or tower based. System proposed in this article is located at high
altitude (up to the stratosphere), where strong permanent and steady streams
are located.а The also proposes a
solution to the main technologist challenge of this system; the transfer of
energy to the ground via a mechanical transmission made from closed loop,
modern composite fiber cable.ааа
аааа The reader can find the
information about this idea in [1], the wind energy in references [2]-[3], a
detailed description of the innovation in [4]-[5], and new material used in the
proposed innovation in [6]-[9]. The application of this innovation and energy
transfer concept to other fields can be found in [10]-[19].
аа
аа Fig.2 shows other design of
the proposed high altitude wind installation. This rotor has blades, 10, connected
to closed-loop cables. The forward blades have a positive angle and lift force.
When they are in a back position the lift force equals zero. The rotor is
supported at the high altitude by the blades and the wing 2 and stabilizer 5.
That design also has energy transmission 3 connected to the ground electric
generator 4.
а Fig.3. shows a parachute wind
high altitude installation. Here the blades are changed by parachutes. The
parachutes have a large air drag and rotate the cable rotor 1. The wind 2 supports
the installation in high altitude. The cable transmission 3 passes the rotor
rotation to the ground electric generator 4.
а A system of fig.4 uses a large Darries air turbine located at high altitude. This turbine
has four blades. The other components are same with previous projects.
аа Wind turbine of fig.5 is a
wind ground installation. Its peculiarity is a gigantic cable-blade rotor. That
has a large power for low ground wind speed. It has four columns with rollers
and closed-loop cable rotor with blades 10. The wind moves the blades, the
blades move the cable, and the cable rotates an electric generator 4.
Fig.1
(left). Propeller high
altitude wind energy installation and cable energy transport system. Notation:
a Ц side view;аа 1 Ц wind rotor; 2 Ц wing
with ailerons; 3 Ц cable energy transport system; 4 Ц electric generator; 5 Ц
stabilizer; b Ц front view; c Ц side view with a support dirigible 9, vertical
cable 6, and wind speed sensors 7; d - keeping of the installation at a high
altitude by rotate propeller; e Ц three lines of the transmission - keeper
system. That includes: main (central) cable and two mobile transmission cables;
f Ц energy transport system with variable altitude; 8 Ц mobile roller.
Fig.2(right). High altitude wind energy installation with the cable turbine. Notation: 10 Ц blades; 11 Ц tensile elements (bracing)(option).
Fig.3 (left). High altitude wind energy installation with the parachute turbine.
Fig.4 (right). High altitude wind energy installation with Darrieus turbine.
Problems
of launch, start, guidance, control, stability, and others
а Launching. It is not
difficult to launch the installations having support wing or blades as
described in fig.1-4. If the wind speed is more than the minimum required speed
(>2-3 m/s), the support wing lifts the installation to the desired altitude.
Fig.5. Ground wind
cable rotor of a large power.
а Starting. All
low-speed rotors are self-starting. All high-speed rotors (include the ground
rotor of fig.5) require an initial starting rotation from the ground
motor-generator 4 (figs.1,5).
ааGuidance and Control.
The control of power, revolutions per minute, and torque moment are operated by
the turning of blades around the blade longitudinal axis. The control of
altitude may be manual or automatic when the wind speed is normal and over
admissible minimum. Control is effected by wing flaps and stabilizer
(elevator), fin, and ailerons (figs. 1,2,4).
а Stability.
Stability of altitude is produced by the length of the cable. Stability around
the blade longitudinal axis is made by stabilizer (see figs.1,2,4).
Rotor directional stability in line with the flow can be provided by fins
(figs. 1). When the installation has the support wing rigidly connected to the
rotor, the stability is also attained by the correct location of the center of
gravity of the installation (system rotor-wing) and the point of connection of
the main cable and the tension elements. The center-of-gravity and connection
point must be located within a relatively narrow range 0.2-0.4 of the average
aerodynamic chord of the support wing (for example, see fig. 1). There is the
same requirement for the additional support wings such as fig.2-4.а
а Torque moment is balanced by transmission
and wing ailerons (see figs.1-4).
а The wing lift force, stress
of main cable are all regulated
automatic by the wing flap or blade stabilizer.
а The location of the installation
of fig.2 at a given point in the atmosphere may be provided by tension
elements shown on fig.2. These tension elements provide a turning capability
for the installation of approximately ▒ 450
degrees in the direction of flow (see. Fig.2.).
аа Minimum
wind speed. The required minimum wind-speed for most of the
suggested installation designs is about 2 m/s. The probability of this low wing
speed at high altitude is very small (less 0.001). This minimum may be
decreased still further by using the turning propeller in an autogiro mode. If
the wind speed is approximately zero, the rotor can be supported in the
atmosphere by a balloon (dirigible) as is shown on fig.1c or a propeller
rotated by the ground power station as is shown on fig.1d. The rotor system may
also land on the ground and start again when the wind speed attains the minimum
speed for flight.
а A Gusty winds. Large
pulsations of wind (aerodynamic energy) can be smoothed out by inertial
fly-wheels.
а The suggested Method and
Installations for utilization of wind energy has following peculiarities from
current conventional methods and installations:
1.
Proposed installation allows the collection of
energy from a large area Ц tens and even hundreds of times more than
conventional wind turbines. This is possible because an expensive tower is not
needed to fix our rotor in space. Our installation allows the use of a rotor
with a very large diameter, for example 100-200 meters or more.
2.
The proposed wind installations can be located at
high altitude 100 m - 14 km. The wind speeds are 2-4 times faster and more
stable at high altitude compared to ground surface winds used by the altitude
of conventional windmills (10-70 meters of height). In certain geographic areas
high altitude wind flows have a continuous or permanent nature.а Since wind power increases at the cube of
wind speed, wind rotor power increases by 27 times when wind speed increases by
3 times.
3.
In proposed wind installation the electric generator
is located at ground. There are proposals where electric generator located near
a wind rotor and sends electric current to a ground by electric wares. However,
our rotor and power are very large (see projects below).аа Proposed installations produce more power by
thousands of times compared to the typical current wind ground installation
(see point 1, 2 above). The electric generator of 20 MW weighs about 100 tons
(specific weigh of the conventional electric generator is about 3-10 kg/kW). It
is impossible to keep this weigh by wing at high altitude for wind speed lesser
then 150 m/s.
4.
One of the main innovations of the given invention
is the cable transfer (transmission)
of energy from the wind rotor located at high altitude to the electric
generator located on ground. In proposed Installation it is used a new cable
transmission made from artificial fibers. This transmission has less a weigh in
thousands times then copper electric wires of equal power. The wire having
diameter more 5 mm passes 1-2 ampere/sq.mm. If the
electric generator produces 20 MW with voltage 1000 Volts, the wire
cross-section area must be 20,000 mm2, (wire diameter is160 mm). The
cross-section area of the cable transmission of equal power is only 37 mm2
(cable diameter 6.8 mm2 for cable speed 300 m/s and admissible
stress 200 kg/mm2, see Project 1). The specific weight of copper is
8930 kg/m3, the specific weight of artificial fibers is 1800 kg/m3.
If the cable length for altitude 10 km is 25 km the double copper wire weighs
8930 tons (!!), the fiber transmission cable weighs only 3.33 tons.а It means the offered cable transferor energy
of equal length is easier in 2682 times, then copper
wire.а The copper wires is very
expensive, the artificial fiber is cheap.
аа All previous attempts to place
the generator near the rotor and connect it to ground by electric transmission
wires were not successful because the generator and wires are heavy.
Some
information about wind energy
а The power of a
wind engine strongly depends on the wind speed (to the third power). Low
altitude wind (H = 10 m) has the
standard average speed V = 6 m/s. High altitude wind is
powerful and that has another important advantage, it is stable and
constant.а This is true practically
everywhere.
аа Wind in the troposphere and
stratosphere are powerful and permanent.а
For example, at an altitude of 5 km, the average wind speed is about 20
M/s, at an altitude 10-12 km the wind may reach 40 m/s (at latitude of about
20-350N).
а There are permanent jet streams
at high altitude. For example, at H =
12-13 km and about 250N latitude. The average wind speed at its core
is about 148 km/hа (41
m/s). The most intensive portion, with a maximum speed 185
km/h (51 m/s) latitude 220, and 151 km/h (42 m/s) at latitude 350
in
а The wind speed of V = 40 m/s at an altitude H = 13 km provides 64 times more energy
than surface wind speeds ofа 6 m/s at an
altitude ofа 10 m.
а This is a gigantic renewable
and free energy source. (See reference: Science
and Technolody,v.2, p.265).
а The primary innovations
presented in this paper are locating the rotor at high altitude, and an energy
transfer system using a cable to transfer mechanical energy from the rotor to a
ground power station.аа The critical
factor for this transfer system is the weight of the cable, and its air drag.
а Twenty years ago, the mass and
air drag of the required cable would not allow this proposal to be
possible.а However, artificial fibers are
currently being manufactured, which have tensile strengths of 3-5 times more
than steel and densities 4-5 times less then steel.а There are also experimental fibers (whiskers)
which have tensile strengths 30-100 times more than a steel and densities 2 to
5 times less than steel.а For example, in
the book [6] p.158 (1989), there is a
fiber (whisker) CD, which
has a tensile strength of s = 8000 kg/mm2
and density (specific gravity) of g = 3.5 g/cm3.а If we use an estimated strength of 3500 kg/mm2
(s =7.1010
N/m2, g = 3500 kg/m3),
then the ratio is g/s = 0.1´10-6 or s/g = 10´106. Although the
described (1989) graphite fibers are strong (s/g = 10´106), they are at
least still ten times weaker than theory predicts. A steel fiber has a tensile
strength of 5000 MPA (500 kg/sq.mm), the theoretical
limit is 22,000 MPA (2200 kg/mm2)(1987);
the polyethylene fiber has a tensile strength 20,000 MPA with a theoretical
limit of 35,000 MPA (1987). The very high tensile strength is due to its
nanotubes structure.
а Apart from unique electronic properties, the mechanical behavior of nanotubes also has provided interest because nanotubes are seen as the ultimate carbon fiber, which can be used as reinforcements in advanced composite technology. Early theoretical work and recent experiments on individual nanotubes (mostly MWNTТs, Multi Wall Nano Tubes) have confirmed that nanotubes are one of the stiffest materials ever made. Whereas carbon-carbon covalent bonds are one of the strongest in nature, a structure based on a perfect arrangement of these bonds oriented along the axis of nanotubes would produce an exceedingly strong material. Traditional carbon fibers show high strength and stiffness, but fall far short of the theoretical, in-plane strength of graphite layers by an order of magnitude. Nanotubes come close to being the best fiber that can be made from graphite.
а For example, whiskers of Carbon
nanotube (CNT) material have a tensile strength of
200 Giga-Pascals and a YoungТs modulus over 1 Tera Pascals (1999).а The theory predicts 1 Tera
Pascals and a YoungТs modules of 1-5 Tera Pascals. The hollow
structure of nanotubes makes them very light (the specific density varies from
0.8 g/cc for SWNTТs (Single Wall Nano
Tubes) up to 1.8 g/cc for MWNTТs, compared to 2.26
g/cc for graphite or 7.8 g/cc for steel).
а Specific strength
(strength/density) is important in the design of the systems presented in this
paper; nanotubes have values at least 2 orders of magnitude greater than steel.
Traditional carbon fibers have a specific strength 40 times that of steel.
Since nanotubes are made of graphitic carbon, they have good resistance to
chemical attack and have high thermal stability. Oxidation studies have shown
that the onset of oxidation shifts by about 1000 C or higher in
nanotubes compared to high modulus graphite fibers. In a vacuum, or reducing
atmosphere, nanotube structures will be stable to any
practical service temperature.
а The artificial fibers are cheap and widely
used in tires and everywhere.а The price
of SiC whiskers produced by Carborundum
Co. with s=20,690 MPa
and g=3.22 g/cc was $440 /kg in 1989. The
market price of nanotubes is too high presently (~$200 per gram)(2000). In the last 2-3 years, there have been several
companies that were organized in the
аа Below, the author provides a
brief overview of recent research information regarding the proposed
experimental (tested) fibers. In addition, the author also addresses additional
examples, which appear in these projects and which can appear as difficult as
the proposed technology itself. The author is prepared to discuss the problems
with organizations which are interested in research and development related
projects.
Table # 1. Material properties.
----------------------------------------------------------------------------------------------------------------------------
Materialааааааааааа аааааааааааа Tensileаааааааааааа ааааа Densityаааааааааааааааааа а Fibersаааааааааааа Tensileааааааааааааа Density
ааааааааааааааааааааааа strengthааааааааааа ааааааа ааааааааааааg/ccааааааааааааааааааааааааааааааааааааааааааааа strengthааааааааааааааааааааааа g/cc
Whiskersааааааааа kg/mm2 ааааааааааааааааааааааааааааааааааааааааааааааааааааааааааааааааааааааа kg/mm2
AlB12ааааааааааааааа 2650аааааааааааааааааааааааааааа 2.6ааааааааааааааааааа QC-8805аааааааааа 620аааааааааааааааааа 1.95
Bааааааааааааааааааааа 2500аааааааааааааааааааааааааааа 2.3ааааааааааааааааааа TM9аааааааааааааааа 600аааааааааааааааааа 1.79
B4Cааааааааааааааааа 2800аааааааааааааааааааааааааааа 2.5ааааааааааааааааааа Thorael ааааааааааа
ааааааааааа 565аааааааааааааааааа 1.81
TiB2аааааааааааааааа 3370аааааааааааааааааааааааааааа 4.5ааааааааааааааааааа Allien
1 ааааааааааа 580аааааааааааааааааа 1.56
SiCаааааааааааааааааа 1380-4140аа ааааааааааааааааа 3.22ааааааааааааааааа Allien
2 ааааааааааа ааааааааааа 300аааааааааааааааааа 0.97
Reference [6]-[9].
аа Industrial fibers with s = 500-600 kg/mm2,
g = -1800 kg/m3,
and sдg = 2,78x106 are
used in all our projects (admissible s =200-250 kg/mm2)(see below).
Power of a wind energy N [Watt,
Joule/sec]
N=0.5hrAV3 аааа[W]а .аааааааааааааааааа аааааааааааааааа (1)
а The coefficient of efficiency, h, equalsа 0.15-0.35 for low speed rotors (ratio of
blade tip speed to wind speed equals l ╗ 1); h = 0.35-0.5 for high speed rotors (l = 5-7). The Darrieus rotor
has h = 0.35 - 0.4. The
propeller rotor has h = 0.45-0.50. The
theoretical maximum equals h = 0.67.
а The energy is produced in one
year is (1 year ╗ 30.2´106 аwork sec) [J]
E=3600´24´350 ╗ 30´106Nа аааа[J].ааааааааааааааааааааааааааааааа аа (1Т)
а Wind speed increases with
altitude as follows
V=(H/Ho)aVoа
,а ааааааааааааааааааааааааааааааааааааааааааааа а(2)
whereа a = 0.1 - 0.25
exponent coefficient depends from surfaceа
roughness.а When the surface is
water,а a = 0.1; when surface is
shrubs and woodlandsа a = 0.25 .
а Power increases with altitude
as the cube of wind speed
N=(H/Ho)3aNoа ,аааааааааааааааааааааааааааааааааааааааааааааа (3)
where No is power at Ho.
аThe drag of the rotor equals
Dr=N/Vа .аааааааа ааааааааааааааааааааааааааааааааааааааааааааааа а (4)
а The lift force of the wing, Ly , is
Ly=0.5CLrV2Awаа ,аа Ly╗W ,аааа ааааааааааааааааааааааааа аа(5)
where CL is lift coefficient (maximum CL ╗ 2.5), Aw is area of the wing, W is
weight of installation + 0.5 weight of all cables.
а The drag of the wing is
ааааааааааааааааааааааааааааааааааааааааааааааааааааааааааааааааааааааа D = 0.5CDrV2Aw ,ааааааааааааааааааааааааааааааааааааааааааа (6)
where CD is the
drag coefficient (maximum CD ╗1.2) .
а The optimal speed of the
parachute rotor equals 1/3V and the
theoretical maximum of efficiency coefficient is 0.5 .
а The annual energy produced by
the wind energy extraction installation equals
E=8.33Nаа аа[kWh] .аааа ааааааааааааааааааааааааааааааааааа (7)
Cable
Energy Transfer, Wing Area, and other Parameters
Cross-section area of transmission cable, S , is
S=N/vs ,а аааааа ааааааааааааааааааааааа ааааааааааааааааааааааа (8)
Cross-section area of main cable, Sm , is
Sm=(Dr+D)/s ,а ааааааааа ааааааааааааааааааааааа ааааааааааа (8Т)
а Weight of cable is
Wr=SLg а,ааааааааааааааааааааааааааааааааааааааааааа ааааа (9)
The production cost, c, in
kWh is
а,ааааааааааааааааааааааааааааааааааааааааааааа (10)
The annual profit
аF= (C-c)E .ааааааааааааааааааааааааааааааааааааааааааааааааааа (11)
аThe required area of the support
wing is
а,аааааааааааааааааааааааааааааааааааааааааааааа (12)
where q is the angle between the
support cable and horizontal surface.
а The wing area is served by ailerons for
balancing of the rotor (propeller) torque moment
а ,аааааааааааааааааааааааааааааааааааааааааааа (13)
а The minimum wind speed for
installation support by the wing aloneа
а,аааааааааааааааааааааааааааааааааааааа (14)
where W is the total weight of the airborne system including transmission. If a propeller rotor is used in a gyroplane mode, minimal speed will decrease by 2-2,5 times.а If wind speed equals zero, the required power for driving the propeller in a propulsion (helicopter) mode is
Ns = W/K2ааааааааааааа [kW],ааааааааааааааааааааааааааааааааааааааааааааааааааааа (15)
а The specific weight of energy
storage (flywheel) can be estimated byааа
Es=s/2gааааааааааа [J/kg].аааааааааааааааааааааааааааааааааааааа (16)ааааа
а For example, if s =200 kg/mm2,
g =1800 kg/m3,
then Es=0.56
MJ/kg or Es=0.15 kWh/kg.
а For comparison of the different
ground wind installations their efficiency and parameters are computed for the
standard wind conditions: the wind speed equals V=6 m/s at the altitude H=10
m.
High-speed
air propeller rotor (fig.1)
а For example, let us consider a
rotor diameter of 100 m (A = 7850 m2),
at an altitude H = 10 km (r = 0.4135 kg/m3), wind speed of V = 30 m/s , an efficiency
coefficient of h = 0.5, and a
cable tensile stress of s = 200 kg/mm2
.
аThen the power produced is N = 22 MW [Eq. (1)], which
is sufficient for city with a population of 250,000. The rotor drag is Dr = 73 tons [Eq.(4)], the cross-section of the
main cable area is S =1.4Dr/σ
=l.35´73/0.2 ╗ 500 mm2, the
cable diameter equals d =25 mm; and
the cable weight is W =22.5 tons [Eq.(9)] (for L=25
km). The cross-section of the transmission cable is S = 36.5 mm2 [Eq.(8)], d = 6.8 mm,
weight of two transmission cablesа is W = 3.33 tons for cable speed v =300 m/s [Eq.(9)].
аThe required wing size is 20´100 m (CL=0.8) [Eq.(12)], wing area served by
ailerons is 820 sq.m [Eq.(13)].
If CL=2, the minimum
speed is 2 m/s [Eq.(14)].
а The installation will produce
an annual energy E =190 GWh [Eq.(7)].а If the
installation cost is $200K, has a useful life of 10 years, and requires
maintenance of $50K per year, the production cost is c = 0.37 cent per kWh [Eq(10)]. If retail price is $0.15 per kWh, profit $0.1 per
kWh, the total annual profit is $19 millions per year [Eq.(11)].а
Large
air propeller at altitude H = 1 km (fig.1)
аа Let us consider a propeller
diameter of 300 m, with an area A = 7´104 m2,
at an altitude H = 1 km, and a wind
speed of 13 m/s. The average blade tip speed is 78 m/s.
аThe full potential power of the
wind streamer flow is 94.2 MW. If the coefficient of efficiency is 0.5 the
useful power is N = 47.1 MW. For other wind speed. the useful
power is: V = 5 m/s, аN =
23.3 MW; V = 6 m/s,а N =
47.1 MW; V = 7 m/s,а N =
74.9 MW; V = 8 m/s,а N =
111.6 MW;а V = 9 m/s,а N = 159 MW; V = 10 m/s,аа N =218 MW.
Estimation
of economical efficiency
аLet us assume that the cost of
the Installation is $3 million, a useful life of 10 years, and request
maintenance of $100,000/year. The energy produced in one year is E = 407 GWh [Eq.(7)]. The basic cost of energy
is $0.01 /kWh.
The
some technical parameters
Altitude
H = 1 km
а The drag is about 360 tons.
Ground connection (main) cable has cross-section area of 1800 sq,mm [Eq.(8Т)],
d =
48 mm, and has a weight of 6480 kg. The need wing area is 60x300 m. The aileron
area requested for turbine balance is 6740 sq.m.
а If the transmission cable speed
is 300 m/s, the cross-section area of transmission cable is 76
sq.mm and the cable weight is 684 kg
(composite fiber).
аAt an altitude of H =13 km. the air density is ρ=0.2666, and the wind speed is V = 40 m/s. The power for efficiency
coefficient 0.5 is 301.4 MgW. The drag of the
propeller is approximately 754 tons. The connection cable has a cross-sectional
area of 3770 sq.mm, a diameter is d =70 mm and a weight of 176 tons. The
transmission cable has a sectional area 5 sq.c and a
weight of 60 tons (vertical transmission only 12 tons).
а The installation will produce
energy E=2604 GWh
per year. If the installation costs $5 million, maintenance is $200,000/year,
and the cost of 1 kWh will be $0.0097/kWh.аа
Air
low speed wind engine with free flying cable flexible rotor (fig.2)
аа Let us consider
the size of cable rotor of width 50 m, a rotor diameter of 1000 m, then the
rotor area is A=50´1000=50,000 sq.m.а The angle rope
to a horizon is 70o. The angle of ratio lift/drag is about 2.5o.
а The average conventional wind
speed at an altitude H = 10 m is V = 6 m/s. It means that the speed at
the altitude 1000 m is 11.4 - 15 m/s. Let us take average wind speed V=13 m/s at an altitude H = 1 km.
аThe power of flow is
N=0.5.rV3Acos200=0.5´1.225´133´1000´50´0.94=63 MW .
If the coefficient efficiency is h = 0.2 the power
of installation is
h = 0.2´63 = 12.5 MW .
а The energy 12.5 MW is enough
for a city with a population at 150,000.
If we decrease our Installation to a 100x2000 m the power decreases
approximately by 6 times (because the area decreases by 4 times, wind speed
reaches more 15 m/s at this altitude. Power will be 75 MW. This is enough for a
city with a population about 1 million of people.
аа If the average wind speed is
different for given location the power for the basis installation will be: V =5
m/s,а N =7.25
MW; V
=6 m/s,а N =12.5 MW; V=7
m/s,а N = 19.9 MW; V =
8 m/s,а N = 29,6 MW; V =
9 m/s,а N = 42.2 MW; V =
10 m/s,а N = 57.9 MW.
Economical
efficiency
а Let us assume that the cost of
our installation is $1 million. According to the book УWind PowerФ by P. Gipe [2], the conventional wind installation with the rotor
diameter 7 m costs $20,000 and for average wind speeds of 6 m/s has power 2.28
kW, producing 20,000 kWh per year. To produce the same amount of power as our
installation using by conventional methods, we would need 5482 (12500/2.28)
conventional rotors, costing $110 million.аа
Let us assume that our installation has a useful life of 10 years and a
maintenance cost is $50,000/year. Our installation produces 109,500,000
kWh energy per year. Production costs of energy will be approximately
150,000/109,500,000 = 0.14а
cent/kWh. The retail price of 1 kWh of energy in
Estimation
some technical parameters
а The cross-section of main cable
for an admissible fiber tensile strange s = 200kg/sq.mm is S =2000/0.2 = 10,000 mm2.а That is two cable of diameter d =80 mm. The weight of the cable for
density 1800 kg/m3 is
W = SLg = 0.01´.2000´.1800 = 36 tons .
а Let us assume that the weight
of 1 sq.m of blade is 0.2 kg/m2 and the
weight of 1 m of bulk is 2 kg. The weight of the 1 blade will be 0.2 x 500 =
100 kg, and 200 blades are 20 tons. If the weight of one bulk is 0.1 ton, the
weight of 200 bulks is 20 tons.
аThe total weight of main parts
of the installation will be 94 tons. We assume 100 tons for purposes of our
calculations.
аThe minimum wind speed when the
flying rotor can supported in the air is (for Cy = 2)
V=(2W/CyrS)0.5=(2´100´104/2´1.225´200´500)0.5 = 2.86
m/s
а The probability of the wind
speed falling below 3 m/s when the average speed is 12 m/s, is zero, and for 10
m/s is 0.0003. This equals 2.5 hours in one year, or less than one time per
year. The wind at high altitude has greater speed and stability than near
ground surface. There is a strong wind at high altitude even when wind near the
ground is absent. This can be seen when the clouds move in a sky on a calm day.
Low
speed air drag rotor (fig.3)
аа Let us consider a parachute
with a diameter of 100 m, length of rope 1500 m, distance between the
parachutes 300 m, number of parachute 3000/300 = 10, number of worked parachute
5, the area of one parachute is 7850 sq.m, the total
work area is A = 5 x 7850 = 3925 sq.m.а The full power
of the flow is 5.3 MW for V=6 m/s. If
coefficient of efficiency is 0.2 the useful power is N = 1 MW. For other wind speed the useful power is: V =5 m/s,а N =0.58 MW; V =6 m/s,а N = 1 MW; V =7 m/s,а N =1.59 MW; V = 8 m/s,а N=2.37 MW; V = 9 m/s, N =3.375 MW; V = 10 m/s, N = 4.63 MW.
ааааааааааааааааааааааааааааааааааааааааааааааа Estimation of economical efficiency
аа Let us take the cost of the
installation $0.5 million, a useful life of 10 years and maintenance of
$20,000/year. The energy produced in one year (when the wind has standard speed
6 m/s) is E = 1000x24x360 = 8.64
million kWh. The basic cost of energy is 70,000/8640,000
= 0.81 cent/kWh.
The
some technical parameters
а If the thrust is 23 tons, the
tensile stress is 200 kg/sq.mm (composed fiber), then
the parachute cable diameter is 12 mm, The full weight of the installation is
4.5 tons. The support wing has size 25x4 m.
High
speed air Darreus rotor at an altitude 1 km (fig.4)
а Let us consider a rotor having
the diameter of 100 m, a length of 200 m (work area is 20,000 sq.m). When the wind speed at an altitude H=10 m is V =6 m/s, then at an altitude H
= 1000 m it is 13 m/s. The full wind power is 13,46
MW. Let us take the efficiency coefficient 0.35, then the power of the
Installation will be N = 4.7 MW. The
change of power from wind speed is: V =
5 m/s,а N = 2.73 MW; V = 6
m/s,а N
= 4.7 MW; V = 7 m/s,а N =
7.5 MW; V = 8 m/s,а N =
11.4 MW; V = 9m/s,а N =
15.9 MW; V = 10 m/s, N = 21.8 MW.а
а At an altitude of H = 13 km with an air density 0.267 and
wind speed V = 40 m/s, the given
installation will produce power N =
300 MW.
Estimation
of economical efficiency
ааа Let us take the cost of the
Installation at $1 million, a useful life of 10 years, and maintenance of
$50,000 /year. Our installation will produce E = 41 millions kWh per yearа (when the wind
speed equals 6 m/s at an altitude 10 m). The prime cost will be
150,000/41,000,000 = 0.37 cent/kWh. If the customer price is $0.15/kWh and
profit from 1 kWh is $0.10 /kWh the profit will be $4.1 million per year.
Estimation
of technical parameters
аа The blade speed is 78 m/s.
Numbers of blade is 4. Number of revolution is 0.25 revolutions per second. The
size of blade is 200x0.67 m. The weight of 1 blade is 1.34 tons. The total
weight of the Installation is about 8 tons. The internal wing has size 200x2.3
m. The additional wing has size 200x14.5 m and weight 870 kg. The cross-section
area of the cable transmission having an altitude of H = 1 km is 300 sq.mm, the weight is 1350
kg.
Project
#6
Ground
Wind High Speed Engine (fig.5)
а Let us consider the ground wind
installation (fig.5) with size 500x500x50 meters. The work area is 500x50x2=
50,000 sq.m. Theа tower is 60 meter tall, the flexible
rotor located from 10 m to 60 m. If the wind speed at
altitude 10 m is 6 m/s, that equals 7.3 m/s at altitude 40 m.
аThe theoretical power is
Nt =
0.5rV3A
=
0.5´1.225´7.33´5´104 =11.9а MgW.
а For coefficient of the
efficiency equals 0.45 the useful power is
N = 0.45´11.9=5.36а MW.
For other wind speed at an altitude 6 m/s the useful power is: V =
5 m/s,а N =
3.1 MW; V = 6 m/s,а N =
5.36 MW; V = 7 m/s,а N =
8.52 MW; V = 8 m/s,а N =
12.7 MW; V = 9 m/s,а N =
18.1 MW; V = 10 m/s,а N =
24.8 MW.
Economic
estimation
а In this installation the rotor
will be less expensive than previous installations because the high-speed rotor
has a smaller number of blades and smaller blades (see technical data below).
However this installation needs 4 high (60 m) columns. Take the cost of the
installation at $1 million with a useful life of 10 years. The maintenance is
projected at about $50,000 /year.
а This installation will produce E =
5360 kW x 8760 hours = 46.95 MWh energy (for the
annual average wind-speedа
V = 6 m/s at H = 10 m).
The cost of 1 kWh is 150,000/46,950,000 = 0.4 cent/kWh. If the retail price is
$0.15/kWh and delivery cost 30%, the profit is $0.10 per kWh, or $4.7 million
per year.
Estimation
of some technical parameters
а The blade speed is 6 x 7.3 = 44
m/s. The distance between blades is 44 m. The number of blade is 4000/44 = 92.
Discussion
and Conclusion
аа Conventional windmills are
approached their maximum energy extraction potential relative to their
installation cost. No relatively progress has been made in windmill technology
in the last 50 years. The wind energy is free, but its production more
expensive then its production in heat electric stations. Current wind
installations cannot essential decrease a cost of kWh, stability of energy
production. They cannot increase of power of single energy unit. The renewable
energy industry needs revolutionary ideas that improve performance parameters
(installation cost and power per unit) and that significantly decreases (in
10-20 times) the cost of energy production. This paper offers ideas that can
move the wind energy industry from stagnation to revolutionary potential.
а The
following is a list of benefits provided by the proposed system compared to
current installations:
1. The
produced energy at least in 10 times cheaper then energy received of all
conventional electric stations includes current wind installation.
2. The
proposed system is relatively inexpensive (no expensive tower), it can be made
with a very large thus capturing wind energy from an enormous area (hundreds of
times more than typical wind turbines).
3. The
power per unit of proposed system in some hundreds times more of typical
current wind installations.
4. The
proposed installation not requires large ground space.
5. The
installation may be located near customers and not require expensive high
voltage equipment. It is not necessary to have long, expensive, high-voltage transmission
lines and substations. Ocean going vessels can use this installation for its
primary propulsion source.
6. No
noise and bad views.
7. The energy
production is more stability because the wind is steadier at
high altitude. The wind may be zero near the surface but it is typically strong
and steady at higher altitudes. This can be observed when it is calmа on the ground,
but clouds are moving in the sky. There are a strong permanent air streams at a
high altitude at many regions of the
8. The installation
can be easy relocated in other place.а
ааа As with any new idea, the
suggested concept is in need of research and development. The theoretical
problems do not require fundamental breakthroughs. It is necessary to design
small, free flying installations to study and get an experience in the design,
launch, stability, and the cable energy transmission from a flying wind turbine
to a ground electric generator.
а This paper has suggested some
design solutions from patent application [4]. The author has many detailed
analysis in addition to these presented projects. Organizations interested in
these projects can address the author (http://Bolonkin.narod.ru
,а aBolonkin@juno.com ).
References
1.
Bolonkin A.A., Utilization of Wind Energy at High
Altitude, AIAA-2004-5756, AIAA-2004-5705. International Energy Conversion
Engineering Conference at
2.
Gipe P., Wind Power,
Chelsea Green Publishing Co.,
3.
Thresher R.W. and etc, Wind Technology Development:
Large and Small Turbines, NRFL, 1999.
4.
Bolonkin, A.A., ФMethod of
Utilization a Flow Energy and Power Installation for ItФ,
5.
Bolonkin, A.A., Transmission Mechanical Energy to
Long Distance.AIAA-2004-5660.
6.
Galasso F.S., Advanced Fibers and Composite, Gordon and Branch
Scientific Publisher, 1989.
7.
Carbon and High Performance Fibers Directory and
Data Book, London-New.
8.
Concise Encyclopedia of Polymer Science and
Engineering, Ed. J.I.Kroschwitz, N. Y: Wiley, 1990,
1341 p.
9.
Dresselhaus, M.S., Carbon
Nanotubes, by, Springer, 2000.
10.
Bolonkin, A.A., УInexpensive Cable Space Launcher of
High CapabilityФ, IAC-02-V.P.07, 53rd International Astronautical Congress. The World Space Congress Ц 2002,
10-19 Oct.ааа 2002/Houston,
JBIS, Vol.56,
pp.394-404, 2003.
11.
Bolonkin, A.A, УNon-Rocket Missile Rope LauncherФ,
IAC-02-IAA.S.P.14, 53rd Internationalаа Astronautical
Congress. The World Space Congress Ц 2002, 10-19 Oct 2002/Houston,
аJBIS, Vol.56, pp.394-404,
2003.
12.
Bolonkin, A.A.,а УHypersonic Launch System of
Capability up 500 tons per day and Delivery Cost $1 per LbФ. IAC-02-S.P.15, 53rd
International Astronautical Congress. The World Space
Congress Ц 2002, 10-19 Oct 2002/Houston,
13.
Bolonkin, A.A., УEmployment Asteroids for Movement
of Space Ship and ProbesФ. IAC-02-S.6.04, 53rd International Astronautical Congress. The World Space Congress Ц 2002,
10-19 Oct. 2002/Houston,а
JBIS, Vol.56,
pp.98-197, 2003.
14.
Bolonkin, A.A., УNon-Rocket Space Rope Launcher for
PeopleФ, IAC-02-V.P.06, 53rd International Astronautical
Congress. The World Space Congress Ц 2002, 10-19 Oct 2002/Houston,
JBIS, Vol.56,
pp.231-249, 2003.
15.
Bolonkin, A.A., УOptimal
JBIS, Vol.56,
pp.87-97, 2003.
16.
Bolonkin, A.A., УNon-Rocket Earth-Moon Transport
SystemФ, COSPAR-02 B0.3-F3.3-0032-02, 34th Scientific Assembly of
the Committee on Space Research (COSPAR). The World Space Congress Ц 2002,
10-19 Oct 2002/Houston,
17.
Bolonkin, A.A., УNon-Rocket Earth-Mars Transport
SystemФ, COSPAR-02B0.4-C3.4-0036-02, 34th Scientific Assembly of the
Committee on Space Research (COSPAR). The World Space Congress Ц 2002, 10-19
Oct 2002/Houston,
18.
Bolonkin, A.A., УTransport System for delivery
Tourists at Altitude 140 kmФ. IAC-02-IAA.1.3.03, 53rd International Astronautical Congress. The World Space Congress Ц 2002,
10-19 Oct. 2002/Houston,
19.
Bolonkin, A.A., ФHypersonic
Gas-Rocket Launch System.Ф, AIAA-2002-3927, 38th AIAA/ASME/SAE/ ASEE
Joint Propulsion Conference and Exhibit, 7-10 July, 2002.
20.
Bolonkin, A.A., Multi-Reflex Propulsion Systems for
Space and Air Vehicles and Energy Transfer for Long Distance, JBIS, Vol, 57, pp.379-390, 2004.
21.
Bolonkin A.A., Electrostatic Solar Wind Propulsion
System, AIAA-2005-3653. 41 Propulsion Conference,
10-12 July, 2005,
22.
Bolonkin A.A., Electrostatic Utilization of
Asteroids for Space Flight, AIAA-2005-4032. 41 Propulsion Conference,
10-12 July, 2005,
23.
Bolonkin A.A., Kinetic Anti-Gravitator,
AIAA-2005-4504. 41 Propulsion Conference, 10-12 July,
2005,
24.
Bolonkin A.A., Sling Rotary Space Launcher,
AIAA-2005-4035. 41 Propulsion Conference, 10-12 July,
2005,
25.
Bolonkin A.A., Radioisotope Space Sail and Electric
Generator, AIAA-2005-4225. 41 Propulsion Conference,
10-12 July, 2005,
26.
Bolonkin A.A., Guided Solar
Sail and Electric Generator, AIAA-2005-3857. 41 Propulsion Conference,
10-12 July, 2005,
27.
Bolonkin A.A., Problems of Electrostatic Levitation
and Artificial Gravity, AIAA-2005-4465. 41 Propulsion Conference,
10-12 July, 2005,
28.
A.A. Bolonkin, Space Propulsion using Solar Wing andа Installation for
It. Russian patent applicationа
#3635955/23 126453,
29.
A.A, Bolonkin, Installation for Open Electrostatic
Field. Russian patent application #3467270/21а 116676,
30.
A.A.Bolonkin, Gettingа of Electric
Energy from Space and Installation for It. Russian patent applicationа #3638699/25 126303,
31.
A.A.Bolonkin, Protection from
Charged Particles in Space and Installation for It. Russian patent application
#3644168а 136270
of
32.
A.A.Bolonkin, Method of
Transformation of Plasma Energy in Electric Current and Installation for It.
Russian patent application #3647344а 136681 of
33.
аA.A.Bolonkin, Method of Propulsion using Radioisotope
Energy and Installation for It.а of Plasma Energy in Electric Current and Installation for
it. Russian patent application #3601164/25а 086973а
of
34.
A.A.Bolonkin, Transformation
of Energy of Rarefaction Plasma in Electric Current and Installation for it.
Russian patent application #3663911/25а 159775 of
35.
A.A.Bolonkin, Method of a
Keeping of a Neutral Plasma and Installation for it. Russian patent application
#3600272/25а 086993
of
36.
A.A.Bolonkin, Radioisotope
Propulsion. Russian patent application #3467762/25а 116952 of
37.
A.A.Bolonkin, Radioisotope
Electric Generator. Russian patent application #3469511/25а 116927 of
38.
A.A.Bolonkin, Radioisotope
Electric Generator. Russian patent application #3620051/25а 108943 of
39.
A.A.Bolonkin, Method of
Energy Transformation of Radioisotope Matter in Electricity and Installation
for it. Russian patent application #3647343/25а 136692 of
40.
A.A.Bolonkin, Method of
stretching of thin film. Russian patent application #3646689/10а 138085 of