When electricity production exceeds electricity consumption, energy storage becomes necessary. The current total worldwide energy production exceeds 3400 GW while storage capacity is only 90 GW.2 Electrical utilities commonly store the energy during non-peak hours in the form of hydroelectric storage by pumping water uphill when electricity is cheap and storing it in a reservoir. This water can then be discharged to spin a turbine and generate electricity during peak hours. The main disadvantage of this method is that the process is only 66% efficient, meaning two-thirds of the total electrical energy is lost in the process. Flywheels on the other hand offer a means of storing energy in a rotating disk with efficiencies higher than 90%.3
A more efficient way of storing energy would also allow for the expansion of renewable energy generating systems such as wind and solar. For example, wind turbines have great fluctuation and are not predictable. By storing the renewable energy, the energy will be made predictable because it can be used during peak hours. Also, a local supply of energy storage would be “compensating for load variations [that] would make it possible to operate transmission, subtransmission, and distribution networks with lighter designs”.2 In contrast to renewable energy, fossil fuel and nuclear plants run nearly 100% of the time at full throttle making them very predictable. These types of plants account for the base load, which is 45% of peak load. Intermediate load generators include older and less efficient generators that only operate during the day to meet demand. Peak load generators allow for peak consumption to be met with gas turbines or diesel engines.2 If the energy can be effectively stored using flywheels, the less efficient peak load generators would be obsolete since less predictable renewable energy can be stored for peak use. For this reason, electric utilities would want 5 MWh storage capacities. This is the “largest that can be factory-assembled and truck-mounted for delivery to substations or energy storage facilities”.4
Besides grid energy storage for electrical utilities, there is also application for energy storage in space onboard satellites that receive energy by solar power, where flywheels can be used for energy storage as well as attitude control instead of batteries. Flywheels can also be used in hybrid-electric cars and trains coupled with regenerative breaking. Another application of flywheels is uninterruptible power supplies for semiconductor fabrication factories and computer data centers. The goals of this study will be to explain how flywheels work, the advantages and disadvantages of flywheel energy storage, as well as technical and economic criteria that will ensure the flywheel’s success.
Flywheels consist of a rotating mass, a vacuum containment structure, bearings, and an electrical device that can function as a motor or generator. Energy is supplied to the flywheel when the device functions as a motor, and energy is discharged when the device functions as a generator.
Kinetic energy is given by
where KE is the kinetic energy, is the moment of inertia, and is the angular velocity of the flywheel.
A solid cylinder will have a moment of inertia equal to
Where r is the radius, a is the length of the cylinder, m is the mass, and is the density.
Figure 1 – University of Texas at Austin Composite Flywheel.5
A composite flywheel, such as the one shown in Figure 1, is usually a hollow circular cylinder, which can be approximated by a composite rim attached to the shaft via a web. In this case, the moment of inertia is given by
Where is the outside radius of the hoop, and is the inside radius of the hoop.6 Because kinetic energy is proportional to the square of angular speed, but only linearly proportional to mass, the best way to increase the kinetic energy is to increase the angular speed.
The tensile strength of the material is what limits the maximum speed. The hoop stress σθ is given by
Where v is the poisson ratio.6
The figure of merit for flywheels is the maximum specific energy density Esp, given by
where Ks is the shape factor, σm is the maximum tensile strength.5 Thus the specific energy density is inversely proportional to the density of the material and linearly proportional to the tensile strength. Composites are thus the ideal flywheel material because they possess very high specific energy densities due to their low densities and high tensile strengths. As shown in Figure 2, the composite material T-1000 graphite has a specific energy density about 16 times that of steel.
Figure 2: Specific energy density chart for different flywheel materials.5
Now that the workings and materials selection of flywheels is established, several applications will be discussed.
One application of flywheels is in space. Flywheels can be used in satellites because power is generated by solar panels only during the day, so energy needs to be stored when the satellite is in a dark orbit. For the International Space Station in particular, nickel-hydrogen batteries are currently being used with a rating of 38,000 cycles and 6.5 year lifetime.7 A flywheel can be used with the same space and weight, but the flywheel would last 3-10 times longer. The cost savings is estimated by NASA to be $200 million. The flywheel would be 93.7% efficient. Flywheels can also be used for 3-axis attitude control by creating a torque as the flywheel is charged or discharged. The US Airforce is leading the attitude control research that could also make its way to space applications. Current targets are a specific energy of 150 kJ/kg, and a torque at least 50 Nm on each axis.8
Flywheels can also be used for hybrid electric vehicles. The engine can be designed to always run at an optimized rpm and a constant speed. Sudden bursts of power needed to accelerate can come from a flywheel, which can be recharged by the engine. Regenerative breaking can also be used to recharge the flywheel instead of dissipating the braking energy to waste heat by friction.8 One example of flywheels being implemented in cars is the Porsche GT3 R Hybrid. A flywheel can deliver 120 kW (161 hp) in 8 seconds to two electric motors connected to the front wheels for a very powerful boost. Because it also has regenerative breaking, the GT3 is able to make fewer pit stops and win more races.9 A schematic of the flywheel implementation in the GT3 is shown in Figure 3.
Figure 3: Porsche GT3 R Hybrid.9
Another application of flywheels is for semiconductor fabrication facilities and computer data centers. AMD’s Dresden Germany semiconductor fabrication facility has a 30 MW flywheel that can absorb 5 MW for 5 seconds. The 5 second storage interval allows the AMD plant to switch from one power source to the next uninterrupted. Internet data centers also need a form of energy storage for power interruptions. Since 80% of power line disturbances last for less than a second, flywheels can be used to store energy for the next generation cloud computing server farms.8 For large internet companies like Google, “the cost of a power interruption can exceed $1 million per minute.” 10 With this in mind, flywheels may offer a cost-effective solution.
One of the primary advantages of flywheels is the long lifetime of 15 to 20 years. This is evidenced by the fact that a composite flywheel has 90,000 cycles even high speeds of 48,000 rpm with no evident degradation.8 Unlike batteries, flywheels exhibit no depth of discharge effects as shown in Figure 4. Depth of discharge means that the useable energy in a battery degrades over time. Laptop and cell-phone users are often victims of this effect. Flywheels also have the advantage that they can deliver large amounts of energy in a short span of time, and require little to no maintenance over their lifetime. They are also highly scalable, and have fewer environmental impacts compared to batteries.
Figure 4: Efficiency vs Lifetime at 80% Depth of Discharge.2
The primary disadvantages of flywheels are that they are often more than 10 times expensive than pumped hydroelectric and generally more expensive than compressed air energy storage, as seen in Figure 5.
Figure 5 – Capital Cost per Cycle of various energy storage methods.2
Another disadvantage of flywheels is that they have both low volumetric and weight energy densities compared to traditional batteries as seen in Figure 6.
Figure 6: Weight Energy Density vs Volume Energy Density.2
Lastly, poor flywheel containment design can lead to dangerous situations, since composite flywheels can fly apart at very high speeds.
The main obstacle to widespread flywheel usage is the low energy density that is currently achievable. This limitation can possibly be overcome in the future with a better choice of composite flywheel materials. US Flywheel Systems predicts that with better composites, specific energies of 200 Wh/kg (10 times higher than current specific energy density) and specific power of 30 kW/kg will be possible.11
Another obstacle that needs to be overcome is the frictional losses that flywheels incur over time. In order to reduce losses to viscous friction, flywheels can be housed in a high vacuum, low pressure containment. Magnetic bearings have come to be used over lubricated shafts to reduce friction. Magnetic bearing flywheels have high frictional losses of 1% per hour, however high temperature superconductor bearing flywheels have energy losses of 0.1% per hour because the superconductors stabilize the load well since they are perfect diamagnets.4 Boeing is currently working on a compact 250 kW flywheel as shown in Figure 7, but these types of flywheels remain in the research phase, and may prove to be ineffective because energy needs to be supplied to provide for cooling.12
Figure 7: Boeing flywheel with high temperature superconducting bearings. 12
The last obstacle to flywheels is cost. At the present time, flywheels are simply not economically competitive with pumped hydroelectric storage. A good example of this is the 20 MW flywheel storage system recently built by Beacon Power. The private company would have been unable to build the plant had it not received a $43 million Department of Energy loan guarantee.13
3.4 Technical Objectives for Success
The objectives for commercial success of flywheels are overcoming the obstacles described in the previous section, namely higher energy density, fewer losses to friction, and a lower cost.
Energy storage is necessary when electricity production exceeds consumption. Flywheel energy storage offers a convenient method of small-scale storage which can help expand the unpredictable renewable wind and solar power generation. Flywheel energy storage also has application in space onboard satellites, hybrid electric cars, and semiconductor factories and internet data centers. Composite flywheels have proven to last 90,000 cycles with no degradation, and operate at speeds of 48,000 rpm. But due to their inherently low energy densities, new composite materials must be developed to make flywheels competitive with batteries. Lastly, high temperature superconducting bearings are currently in the research phase, and may offer a means of reducing frictional losses by a factor of 10.
- Werfel, F.N., U. Floegel-Delor, T. Riedel, R. Rothfeld, D. Wippich, B. Goebel, G. Reiner, and N. Wehlau. “A Compact HTS 5 KWh/250 KW Flywheel Energy Storage System.” IEEE Transactions on Applied Superconductivity 17.2 (2007): 2138-141.
- Ibrahim, H., A. Ilinca, and J. Perron. “Energy Storage Systems—Characteristics and Comparisons.”Renewable and Sustainable Energy Reviews 12.5 (2008): 1221-250.
- Kirk, J. “Flywheel Energy Storage – I.” Int. J. Mech. Sci 19 (1977): 223-31.
- Hull, J., T. Mulcahp, K. Uherka, R. Erck, and R. Abboud. “Flywheel Energy Storage Using Superconducting Magnetic Bearings.” Applied Superconductivity 2.7-8 (1994): 449-55.
- Liu, H., and J. Jiang. “Flywheel Energy Storage—An Upswing Technology for Energy Sustainability.” Energy and Buildings 39.5 (2007): 599-604.
- Bolund, B., H. Bernhoff, and M. Leijon. “Flywheel Energy and Power Storage Systems.”Renewable and Sustainable Energy Reviews 11.2 (2007): 235-58.
- Hebner, R., J. Beno, and A. Walls. “Flywheel Batteries Come Around Again.” IEEE Spectrum(2002): 46-51.
- Bitterly, J. “Flywheel Technology – Past, Present, and 21st Century Projections.” IEEE AES Systems Magazine (1998): 13-16.
- Strasik, M., J. Hull, J. Mittleider, J. Gonder, P. Johnson, K. McCrary, and C. McIver. “An Overview of Boeing ﬂywheel Energy Storage Systems with High-temperature Superconducting Bearings.” Superconductor Science and Technology 23 (2010): 1-5.
Still not convinced flywheels are amazing? You can still dislike flywheels here.