Advances in Flywheel Energy- Storage Systems

Transcription

1 Advances in Flywheel Energy- Storage Systems Bryan B. Plater and James A. Andrews Active Power, Inc. Austin, Texas Introduction F lywheels seem to have inherent appeal as an alternative to traditional energystorage technologies. Part of this appeal must surely be due to the sheer simplicity of storing kinetic energy in a spinning mass. For decades, most engines have used this concept to smooth their operation. Prior to the development of cost-effective power-conversion electronics, the primary method of limiting power interruptions to critical loads was by adding inertia to a motor/generator set feeding a load. Over the last 20 years or so, the promise of a compact, safe, environmentally benign, low-maintenance, long-lasting and predictable source of energy has intrigued inventors and investors alike for applications such as electric vehicles, utility load-leveling and satellite control. More recently, the flywheel has regained consideration as a viable means of supporting a critical load during mains power interruption, due to the lower capital expense and extended run time now available from many systems, as well as continued customer dissatisfaction with traditional electrochemical energy storage. Interestingly, continuing PowerPulse is a registered trademark of, and published by, Darnell.Com Inc. Materials may not be reproduced or republished without written permission from Darnell.Com Inc.

2 advances in the power electronics field, which initially provided an alternative to flywheels for blip protection, have now enabled some flywheel designs to deliver a cost-effective alternative to the power quality market. Traditional Flywheels Until recently, the power quality industry limited the use of flywheels to steel wheels coupled with motor/generator sets, such that the increase in rotary inertia (and hence stored kinetic energy) allowed longer ride-through during utility power interruptions. The effective increase in run time for such systems rarely exceeded one second at rated load, corresponding to a delivery of less than 5 percent of the additional stored energy from the wheel. Delivery of more energy would result in reduced rotational speed, and hence, reduced electrical frequency, which was usually unacceptable. Although these systems provided adequate protection against a large segment of power sags and outages, they were unable to sustain power for a full reclosure event, or allow sufficient ridethrough for the start of a typical standby generator. One can extend the power delivery time of the system (shown in Figure 1) by making a few modifications to the design. The inherent frequency and voltage reduction that accompanies a decelerating generator is unacceptable for virtually all loads. By inserting a rectifier after the generator, the system is capable of delivering approximately 75 percent of the flywheel s energy as usable dc power for a substantial increase in ridethrough time. The dc power must then be filtered and inverted back to ac power at 60Hz (or other appropriate frequency). Adding a variable-speed drive to the system allows efficient motoring of a large inertia from lower rotational speeds, enabling a smaller motor to be used for this standby power source. As is evident in Figure 2, the effective increase in ride-through offered by the modified flywheel configuration offers substantially more protection than the older version of the traditional flywheel. However, this added run time comes with a higher price, and also requires several additional components and more floor space. Regardless of configuration, traditional flywheel systems have several advantages over their modern high-tech counterparts. The use of steel, one of the most abundant and predictable materials in engineering, is nearly universal. Steel allows designers to keep manufacturing costs down while maintaining adequate factors of safety. Because steel flywheels generally weigh more and have lower material strengths than composite wheels (see Table 1), they must spin at relatively low rotational speeds, which enables the use of conventional bearing systems. A related disadvantage to steel flywheels is that they typically have lower energy (and power) densities than modern composite wheels. Traditional flywheels usually operate in air, which causes increased aerodynamic drag losses as well as a higher operating noise level. In addition, the integration of an external flywheel, regardless of material choice and system configuration, requires multiple bearing sets, which can reduce overall system reliability and increase maintenance costs.

3 Figure 1. Traditional integration of a steel flywheel with a motor generator set. Figure 2. System with variable-speed drive and rectification/inversion electronics. Table 1. Advantages and disadvantages of the traditional flywheel. Advantages Disadvantages Steel safe, predictable Low RPM makes design simple Inexpensive materials keep costs down Low energy/power density Multiple bearing sets High aerodynamic noise and drag High-Speed Flywheel In an effort to achieve higher energy and power densities and take advantage of modern composite material and power electronics technologies, many designers have attempted to develop compact flywheel batteries capable of extremely high tip speeds (linear velocity at outside radius of the flywheel). Until recently, the targeted applications for these composite flywheel systems have been electric or hybrid-electric vehicles and satellite momentum control. Invariably, these applications demand maximum stored energy and delivered power with extreme constraints on system weight and volume. Since the stored energy in a flywheel is proportional to the square of its rotational speed, the obvious method for maximizing stored energy is to push the speed of the flywheel. Of course, all designs have a limiting speed, which is set by the stresses developed within the wheel due to inertial loads that are also proportional to the square of rotational speed. Composite wheels weigh less, and hence, develop lower inertial loads at a given speed. In addition, high-tech composites are often stronger than conventional engineering

4 metals. This combination of high strength and low weight enables extremely high tip speeds related to conventional wheels. For a given geometry, the limiting energy density (energy-per-unit-mass) of a flywheel is proportional to the ratio of material strength to weight density, otherwise known as the specific strength. The following table illustrates the advantages that composite materials offer in this respect. Recent advances in composite materials technology may allow nearly an order of magnitude advantage in specific strength of composites when compared to even the best engineering metals. The result of this continuous research in composites has been flywheels that operate at rotational speeds in excess of 100,000rpm, with tip speeds in excess of 1,000m/sec. The incredible advantages afforded by modern composite materials don t come without cost. The ultra-high rotational speeds that are required to store significant kinetic energy in these systems virtually rule out the use of conventional mechanical bearings. Instead, most systems run on magnetic bearings. This relatively recent innovation uses magnetic forces to levitate a rotor, eliminating the frictional losses inherent in rolling element and fluid film bearings. Unfortunately, aerodynamic drag losses force most high-speed flywheels to operate in a partial vacuum, which complicates the task of dissipating the heat generated by ohmic losses in the bearing electromagnets and rotor. In addition, active magnetic bearings are inherently unstable, and require sophisticated computer control to maintain levitation. The integrated generator of these systems is usually a rotating-field design, with the magnetic field supplied by rare-earth permanent magnets. Since the specific strength of these magnets is typically just fractions of that of the composite flywheel, they must spin at much lower tip speeds; in other words, they must be placed very near the hub of the flywheel. This compromises the power density of the generator. Table 2. Specific strength of several candidate flywheel materials. Material Specific Strength (in 3 ) Graphite/epoxy 3,509,000 Boron/epoxy 2,740,000 Titanium & its alloys 1,043,000 Wrought stainless steels 982,000 Wrought high-strength steels 931,000 7,000 Series aluminum alloys 892,000

5 Figure 3. Modern high-speed flywheel system. An alternative is to mount them closer to the outer radius of the wheel, but contain their inertial loads with the composite wheel itself. Obviously, this forces the designer to either de-rate the machine speed, or operate closer to the stress limit of the system, thus compromising safety. As in the case of the traditional flywheel that is coupled to an existing motor/generator set, these systems usually include rectification, filtering and inversion electronics to allow the delivery of a large percentage of stored energy. The resulting system promises to be a compact, light-weight flywheel battery that needs little maintenance, suffers insignificant degradation from multiple discharges and exhibits minimal sensitivity to operating temperature. However, current component costs drive system prices to levels that make competing with existing energy-storage technologies exceedingly difficult, except in perhaps the most esoteric applications. These costs practically eliminate the option of operating the wheels with reasonably large factors of safety. Thus, some sort of inertial containment system becomes necessary to minimize collateral damage from a failed flywheel. Such failure can occur for many reasons, including crack growth from material flaws undetected at manufacture, excessive shock loads in the installed environment and magnetic bearing failure. The cost and/or complexity of providing sufficient containment further reduces the competitiveness of this technology. Table 3. Advantages and disadvantages of the high-speed flywheel. Advantages Disadvantages Compact Safety concerns High efficiency High material costs Low maintenance Expensive magnetic bearings No aerodynamic noise

6 Figure 4. Active Power s flywheel energy storage. CleanSource The Best of Both Worlds The fact that the power quality market has inherently different requirements than do the electric vehicle, utility load-leveling and satellite-control markets prompted Active Power to investigate potential design hybrids that offer high performance at a competitive cost. The CleanSource family of flywheel batteries is the result of this effort. The primary difference between the power-quality market and other potential markets for flywheel batteries is the other markets high value placed on energy density, measured either in stored energy-per-unitweight or stored energy-per-unit-cost. Certainly, lower weight and higher stored energy have value in the power-quality market, but this value diminishes once certain minimum thresholds are met. By focusing on the requirements of the power-quality market, which really comes down to high-power density (measured in either delivered power-per-square-foot or delivered power-per-unit-cost, Active Power developed and delivered a competitive alternative to electrochemical energy storage, employing low-tech materials in a high-tech design to offer the benefits of both traditional and modern flywheel systems with the drawbacks of neither. This new flywheel system is a power-dense integrated motor/generator/flywheel that is both costeffective and safe. Ride-through requirements for the power quality industry fall into two main categories: enough time to power the load until standby generator startup (~ 10 to 45 seconds) enough time to power the load through the vast majority of events (~ 5 seconds) The design takes maximum advantage of a flywheel s natural capacity to produce high power in a compact space for a relatively short period of time. By designing the system so that it can deliver usable power over a broad speed range, the safety and cost issues that surround the energy overload required by more traditional low-speed flywheel energy storage systems has been eliminated. By designing the product for applications that require relatively short power delivery times, the system delivers what the application demands at the lowest possible cost. The product covers a broad range of power classificacopyright 2001, Darnell Group Inc.

7 tions with single or paralleled systems to extend power and/or runtime as needed. As shown in Figure 5, the shorter the ride-time required, the higher the constant power capabilities. The high-power, low-loss capability of this new system is the result of a proprietary generator technology, a vacuum enclosure, and a combination of magnetic and conventional mechanical bearings. The generator s unique design minimizes eddy current losses, and provides for complete control of developed voltage throughout the speed range of the machine. The partial vacuum reduces aerodynamic losses and noise, while the hybrid bearing system minimizes bearing losses and extends service life. Given the power-quality market s minimal concern with installed system weight for a particular level of stored energy, the product incorporates rotors machined from a solid block of forged high-strength steel, instead of dangerously pushing the envelope of rotational speed with esoteric materials. With operating speeds well under 10,000rpm, the rotors exhibit a high level of safety Figure 5. Power versus time for a single CleanSource flywheel system. Figure 6. Clean- Source exploded view.

8 Figure 6 shows the complete integration of the motor, generator and flywheel rotor functions into one single piece of solid forge steel. The integration and use of the full length of the rotor as an active generator results in lower costs and higher power densities. The field coil provides current to magnetize the teeth of the steel rotor that rotates past the copper coils imbedded in the armature to generate power. As the rotor slows during a discharge, the field is increased to raise the magnetism of the rotor teeth, thereby compensating for the speed loss, which in turn keeps the voltage constant until approximately 80 percent of the rotor energy is consumed. The field circuit also serves as the magnetic bearing, unloading a large portion of the rotor weight, which greatly extends the life of the mechanical bearings. Multiple accelerometers feed an onboard microprocessor to allow monitoring of shaft vibration. The microprocessor periodically performs spectral analysis of the acceleration signals to detect any unusual signal components. A number of thermistors detect local operating temperatures within the unit and the associated electronics cabinet. The potential for runaway motoring is addressed through automatic shutdown upon loss of any system-critical component. In system-critical component. In addition, conservative trip levels either issue caution or shutdown signals in the event of any unusual vibration or temperature readings. All discharge events are logged with high precision to enable trending and evaluation by the customer or engineers. The flywheel energy-storage system is designed for long service intervals and extreme operating environments, and are the first competitive alternative to batteries in cost, footprint and efficiency. The technology can be applied to larger systems, and the current systems can be placed in arbitrarily large arrays of parallel units to offer flexibility in energy storage, peak-power rating and ridethrough time. Applications There are three primary markets for a flywheel system: continuous power power-quality improvement battery isolation and redundancy Figure 7. Continuous power solution.

9 Figure 8. Glitch protection configuration. Figure 9. Battery hardening and isolation. Continuous Power. In many circumstances, the end user of critical power has a strong desire to eliminate the requirement for electrochemical batteries due to environmental restrictions, maintenance concerns and/or limited space. If the power quality configuration includes a standby engine/generator for long-term protection, a flywheel energystorage system may be well suited for providing power until the start and synchronization of the genset. Power-Quality Improvement. Due to the phenomenon of the vast majority of power quality events having a duration of only a few seconds, some power users have the opportunity to improve the quality of their power in all but long-term outage situations with minimal cost and space outlays. Batch or process manufacturing sites with a history of short-term power glitches or sags (which have remained unprotected due to the high costs or space requirements of traditional energy storage) are ideal applications for high-power flywheel systems. Battery Isolation and Redundancy. One of the primary determinants of electrochemical battery life is the number of times the cell is discharged, the battery life being inversely proportional to the number of discharge events. A flywheel battery is an effective means of protecting a chemical battery string with a small and durable package. The flywheel not only easily handles the vast majority of power-quality events, but acts as a

10 redundant dc power source in times of chemical battery maintenance or recharge. Conclusion Due to a number of well-documented weaknesses, the power-quality market needs a costeffective alternative to electrochemical energy-storage systems. The ideal alternative to conventional batteries must be cost-effective, compact, low maintenance and have a long installed life. DG Copyright 2001 PowerPulse is a registered trademark of, and published by, Darnell.Com Inc. Materials may not be reproduced or republished without written permission from Darnell.Com Inc. Darnell.Com Inc. shall not be responsible for any representations, statements of fact, or individual opinions made by advertisers or authors in PowerPulse. The views expressed here are those of the author(s) and do not necessarily reflect the opinion or agreement of Darnell.Com Inc. Inquiries should be addressed to Darnell.Com Inc., 1159-B Pomona Road, Corona, California