April 2010

The world’s first electric power generating wind turbine was manufactured by Charles Brush in 1888, and it delivered 12 kW of power reliably for 20 years (Fig. 1). The battle to make wind energy economically viable has been going on ever since, and it must be fought on many fronts. One key element in the struggle for increased reliability and efficiency is achieving excellent shaft alignment of the wind turbine generator train. Is this an uphill battle for you?

The Challenges

Many challenges face the wind turbine operator and alignment technician in achieving the goal of having in-tolerance alignment under operating conditions. A good laser alignment system with the correct feature set is an essential weapon in your arsenal to win this war.

One concern involves compliance with safety regulations. In the United States, several states do not permit the presence of a person in the nacelle when the blades are turning, yet to perform the alignment it is necessary to turn the shafts. In these cases a special safety exemption must be obtained, or if your laser alignment system allows it, accurate readings can be taken without the need to rotate the blades by more than the clearance allowed by the blades’ safety tether. Because of the enormous input to output ratio of the gearbox, at a normal operating input speed of 20 RPM to an output speed of 1800 RPM, the turbine blades would only need to be rotated 1 degree to achieve a 90-degree rotation of the gearbox’s output shaft. The best laser systems allow you to collect accurate and repeatable alignment readings with as little as 70 degrees of shaft rotation, starting anywhere and stopping anywhere. Moreover, this safety requirement in turn demands that your laser system have the ability to take readings automatically and continuously over this small angle of rotation so that enough data can be collected to be statistically meaningful. More importantly, it must allow the shafts to be rotated in either direction, particularly if the turbine blades are tethered or if the law does not permit a full rotation of the turbine with the technician present in the nacelle.

Another concern is that, in those types of turbines where stopping rotation by applying the brakes can itself affect the alignment, due to the flexibility of the anchoring structures, a true continuous sweep measurement mode is essential so that readings can be automatically collected while the shaft is in motion. This eliminates the need to have to stop the rotation by engaging the brake to take measurement points that may otherwise be influenced by this external force acting on the rotor. In these cases simply allowing the tether to stop the rotation is of benefit, as enough unloaded data will be collected before the tether stops the shafts.

Once in the nacelle, lack of space is a major concern. Very often the technician is only able to access the machines from one side. Does your laser system allow you to switch the viewpoint of the machine train? Does it let you move the generator into alignment with the gearbox even if you have to place it at the left end of the machine train on your alignment computer’s screen?

Not least of the challenges facing the alignment technician is ascending the turbine mast to the nacelle with the equipment necessary to perform the alignment. Unless the technician is athletically inclined and believes he or she can arrive at the nacelle with enough energy left to do the job (Fig. 2), valuable crane or helicopter time must be scheduled (Fig. 3). Thus, careful planning is needed to keep costs from escalating out of control.

Obtaining correct target specifications for a good alignment under running conditions to feed into your laser system is extremely important. Wind turbines are notoriously flexible structures. Thus, changes in wind speed and direction, generator load, and temperature can have a very significant effect on the shaft alignment between the gearbox and generator. A good quality coupling is essential to withstand these variations but does not excuse a poor alignment, because although the coupling may be able to take it, the bearings and the gearbox won’t. Excessive vibration and the resulting radial and axial loading on the rotating components will decrease efficiency and shorten operating life drastically. Therefore, your laser system must be able to handle target specs so that you can compensate for the positional changes that occur between the stopped (unloaded) condition at time of alignment and the running condition. One system successfully tested for monitoring positional change in the turbine generator train is Permalign®. An extensive study performed by Pruftechnik and Nordex in Germany demonstrated its viability. Immensely valuable data was collected that permitted the generator and gearbox to be misaligned in such a way that they would grow into alignment with each other under average operating conditions (Fig. 4). Vibration was significantly reduced, and operating efficiency and longevity increased.

The Alignment

Mounting your brackets can be a challenge in tight spaces. Does your system offer compact magnetic brackets with offset adapters as well as chain brackets? In Fig. 5the laser emitter is mounted on a magnetic bracket with offset adapters directly to the brake disc on the gearbox output shaft, behind the flex plate of the fiber spool piece coupling. The disc brake mechanism impedes a full rotation of the shafts.

The vertical adjustment of the gearbox is usually undertaken with hydraulic jacks or with conical counter-rotating adjustment pads, rather than with shims (Fig. 6). This requires your laser system to be able to simultaneously monitor vertical and horizontal moves, in real time.

Another critically important feature your laser system should possess is “standard deviation.” Since time is of the essence and shaft rotation may be limited, it is very important for the technician to be able to assess the quality of his or her readings immediately, while simultaneously eliminating needless repeatability checks. The ability to individually disable aberrant measurement points within a set of readings is imperative. For a greater understanding of the subject of measurement quality and standard deviation in shaft alignment readings, we refer the reader to a PowerPoint presentation, as well as an article on turbine alignment, which can be found on LUDECAwind’s Web site, listed at the end of this article.

Only the best laser alignment systems unite all of the necessary features and functionality required to perform accurate and timely wind turbine shaft alignment. This is one critical aspect of wind turbine maintenance where skimping with a cheap laser system will cost you dearly. Buying the right system pays off immediately in labor savings, operational efficiency and, crucially, in unnecessary repairs and their concomitant unscheduled downtime.

With another record year to the industry’s credit, wind energy deployment both worldwide and in the United States continues to show the effectiveness of policies and incentives coupled with a clean, affordable, and reliable energy supply. By the end of 2009 some 9,922 MW had been installed in the U.S. alone, for a cumulative amount of nearly 35,000 MWs. For the first time wind energy was tied with natural gas on annual installations, representing approximately 2 percent of our energy consumption. It is important to recognize that effective programs triggered by the American Recovery and Reinvestment Act (ARRA), such as loan guarantees and incentive options, played a key role in the large installation numbers during the recent economic downturn.

There are many existing and ongoing studies targeted at evaluating and calculating the feasibility of large penetrations of renewable in the future energy mix, with all studies showing that wind energy will represent a significant percentage of the clean energy portfolio. Additionally, with current administration goals—which aggressively suggest up to 10 percent renewable energy by 2012, and 25 percent by 2025—it is important to understand that in addition to clean technology viability and feasibility, a robust supply chain and manufacturing sector is imperative in meeting these goals.

With no offshore wind installations, land based utility-scale machines range from 1 to 2.5 MWs. These machines are typically installed on a 60 to 80-plus meter towers and have rotor diameters that can range between 60 and 100 meters, depending on the turbine rating and the wind speed class of the site. Manufacturers offer the same basic-rated machine with different rotor diameters and tower heights to accommodate the certified resource characterization of the specific site. These options are targeted at delivering a turbine that balances both energy capture and reliability in order to ensure cost-effective energy production over the designed lifetime, which is typically 20 years.

The manufacturing capacity of the components needed to support these large machines must have the needed flexibility to support the various options offered by the manufacturers. Unlike other key components such as generators and gearboxes, wind blades can require plant modularity, given the size of the articles and the capital intensive molds that must be available to support projected orders.

Focusing on 2009, when nearly 10GW were installed, the raw materials that were needed to manufacture the diverse set of components can sometimes strain the system and obligate manufacturers to lock in large amounts of supply (~175M lbs of fiberglass). This favors large manufacturers, of course, who have both the capital backing and reputation to be able to lock in both prices and large quantities of materials or components.

Independent of supply, the economic downturn has increased the focus on reliability. Companies and research institutions focus on technology improvements that not only improve the efficiency of the machine, but also focus on improving the reliability and availability of the asset. As the industry has matured over the past 30-plus years, both the technology and methodology used to design and evaluate components has increased substantially, but the materials and manufacturing processes have improved as well.

As an example, several manufacturers have developed blade designs that require carbon fiber for lightweight and stiffness constraints, while others continue to use lower-cost fiberglass given experience and manufacturing infrastructure. Although both methodologies have proven to be successful, balancing the economic benefit of the innovations while preserving or enhancing reliability will be the innovation filter for years to come.

There is an immense set of technology options that can potentially improve wind systems such as condition health monitoring systems, distributed sensor networks, and advance materials options such as nano particles to strengthen local areas, etc., but evaluating and balancing these technologies on an economic basis is key. It is difficult to predict what the next generation of technologies will bring to this industry, but we can be certain that as it continues to mature and leverage technologies from other sectors the resulting turbines will be smarter, more efficient, and they will represent a significant percentage of our energy mix.

The AWEA WINDPOWER show always provides impressive demonstrations of the latest and greatest achievements in all the areas of science and technology that are related to wind-power generation. Wind turbine gearbox suppliers present the very best gear alloys, forging and cutting processes, the mirror-like tooth surfaces achieved by the isotropic superfinish technology, and the best bearings and lubrication systems, etc.

In contrast with these innovations, the gear tooth geometry of almost all gearboxes has become outdated. In most cases it presents century-old tooth proportions produced by the standard 20o pressure angle generating racks with some addendum modification. This tooth geometry is universal and good for general gear applications because of gear interchangeability and the availability of standard tooling. It does not fit to the specific operating conditions and requirements of wind turbine gearboxes, however, which include a long lifetime of some 20 years, relatively low speeds, high static loads, size and weight constraints, high efficiency, noise and vibration limitations, and low cost of fabrication and maintenance, etc. The gear tooth geometry must be customized to satisfy all of these conditions.

Unlike the traditional standard gear design, the proposed alternative Direct Gear Design method does not use the basic rack parameters. It uses desired performance parameters and operating conditions to define the optimized gear tooth shape. This design approach is developed for involute gears and based on the theory of generalized parameters created by E.B. Vulgakov [1], and it can be defined as an application-driven gear drive development process with primary emphasis on performance maximization and cost efficiency without concern for any predefined tooling parameters.

Gear Tooth and Mesh Synthesis
There is no need for a basic gear rack to define the gear tooth profile. Two involutes of one base circle—or, in case of asymmetric gears, two different base circles—the arc distance between them and the tooth tip circle describe the gear tooth (see Fig. 1 A-B, Fig. 1 C-D) and the equally spaced teeth form the gear. The fillet between teeth is not in contact with the mating gear teeth. This portion of the tooth profile is critical, however, because this is the area of maximum bending stress concentration.

In a wind turbine gearbox the tooth load on one drive flank is significantly higher and is applied for longer periods of time than for the opposite coast one. An asymmetric tooth shape reflects this functional difference. The design intent of asymmetric gear teeth is to improve performance of the primary drive profiles at the expense of the performance for the opposite coast profiles. The coast profiles are unloaded or lightly loaded during relatively short work period. Asymmetric tooth profiles also make it possible to simultaneously increase the contact ratio and operating pressure angle beyond conventional gear limits. The main advantage of asymmetric gears is contact stress reduction on the drive flanks, resulting in higher torque density, or load capacity per gear size. Another important advantage is the possibility to design the coast flanks and fillet independently from the drive flanks, managing tooth stiffness and load sharing while keeping a desirable pressure angle and contact ratio on the drive profiles. This allows reducing gear noise and vibration level. Asymmetric gears make it possible to simultaneously increase the transverse contact ratio and operating pressure angle far beyond conventional gear limits [2].

Tooth Fillet Profile Optimization
The tooth fillet design begins when the involute flank parameters are completely defined. The initial fillet profile is a trajectory of the mating gear tooth tip in the tight, zero-backlash mesh. The fillet optimization process [3] utilizes three methods: random search method locating fillet points; trigonometric functions for fillet profile approximation; and FEA for stress calculation.

The first and the last fillet profile points of the initial fillet profile lay on the form diameter circles (Fig. 2A, Fig. 2B) and can’t be moved during an optimization process. The random search method is used to move the fillet finite element nodes along the beams that connect the fillet center and the nodes of the initial fillet profile.

The bending stresses are calculated for every new fillet profile points’ combination. If the maximum bending stress is reduced, the program continues searching in the same direction. If not, it steps back and starts searching in a different direction. After the given number of iteration steps the optimization process is stopped, resulting in the optimized fillet profile that provides minimum bending stress concentration. Table 1 illustrates fillet profile optimization and the achievable maximum bending stress reduction for standard AGMA 201.2 gears.

The 20-25 percent bending stress reduction provided by the fillet profile optimization can be used directly to increase the tooth bending strength if it limits the gear drive load capacity. However, the gear drive load capacity and its size and weight are typically defined by the contact stress, or tooth surface durability. In this case the benefit of the bending stress reduction can be converted to reduce the contact stress and simultaneously increase gear mesh efficiency [4]. This is possible by increasing the number of teeth and applying the finer diametral pitch (lesser module) for the teeth with optimized filler, keeping the given center distance.

Table 2 presents a comparison of the traditionally designed gear pair with the 27-tooth pinion and the 49-tooth gear, the 3 mm module, and the 32 and 30 mm face widths for the pinion and gear accordingly with similar gear pairs created by Direct Gear Design with the symmetric and asymmetric tooth profiles. This table illustrates the contact and bending stress reduction and increased mesh efficiency for direct design gears. Fig. 3 presents the carrier assembly of a two-stage planetary turboprop engine [5]. Application of asymmetric gears allowed significant weight-output torque reduction in comparison with similar gearboxes with conventional symmetric gears.

Optimized gear profiles require custom tooling. For profile machining process (Fig. 4) the tool profile is the same as a space profile between the neighboring teeth. For generating machining process like gear hobbing (Fig. 4) the tool profile is defined by reverse generation when the designed gear forms the tooling rack profile. In this case the rack profile (pressure) angles are selected to provide the best machining conditions.

Summary
Unlike traditional gear design, which is based on the rack generating process and driven by manufacturing convenience, Direct Gear Design is driven entirely by application requirements, when technical and market performance of product is critical. It provides the ultimate in optimized gear geometry solutions for custom gear transmissions such as wind turbine gearboxes. The benefits of this approach include higher load capacity, reduced size and weight, extended lifetime, reduced noise and vibration, higher efficiency and reliability, and reduced cost. AKGears has implemented direct gear design in many high-performance gear drives, and we welcome your inquiries. 

References:
1) E.B. Vulgakov, Gears with Improved Characteristics, Mashinostroenie, Moscow, 1974 (in Russian).
2) A.L. Kapelevich, “Geometry and design of involute spur gears with asymmetric teeth”, Mechanism and Machine Theory, 2000, Issue 35, pp. 117-130.
3) A.L. Kapelevich, Y.V. Shekhtman, Tooth Fillet Profile Optimization for Gears with Symmetric and Asymmetric Teeth, AGMA Fall Technical Meeting, San Antonio, Texas, October 12-14, 2008, (08FTM06).
4) A.L. Kapelevich. Direct Design for High-Performance Gear Transmissions, Gear Solutions, December 2008, 22-31.
5) A.S. Novikov, A.G. Paikin, V.L. Dorofeyev, V.M. Ananiev, A.L. Kapelevich, Application of Gears with Asymmetric Teeth in Turboprop Engine Gearbox, Gear Technology, January/February 2008, 60-65.

It’s a fact of life that all electromechanical components will fail at some point. Having the right strategy in place to manage that failure can be the difference between success and… well, failure. This is especially true for wind turbine owners and operators. Without an effective strategy at the ready failure can be devastatingly expensive in terms of lost output, time, and most importantly, revenue.

Do you have a coherent strategy in place when failure strikes? When a problem arises, finding a fast and cost-effective solution is vital. Can the part be repaired? Does it need to be replaced? Is there inventory available? There is also a strong environmental argument in reusing existing parts instead of scrapping them and buying new.

Expanding Too Fast?
The rapid expansion of wind energy in the last 20 years has often resulted in a rush to get new technologies to market, and this has inevitably led to reliability issues. These issues can directly affect a wind farm’s revenue stream through downtime, plus there is a cost associated with additional operations and maintenance events. Some of these new technologies have a propensity for a higher failure rate within the control systems and power electronics compared with other components within the system. Approximately 54 percent of all wind turbine malfunctions are due to failures within the control electronics, electrical systems, and sensors. Unlike many of the mechanical components in a wind turbine, the majority of the electronics/electrical systems only have a single supply source—the OEM. Since the parts generally have a high unit cost, the cost of a new replacement part can be as much as three times that of repairing an existing part. In addition, gaps in the supply chain can result in exceedingly long and costly lead times.

Obstacles and Opportunities
As our industry continues to mature, many wind turbines are now coming out of their manufacturers warranty period. Numerous owners and operators throughout the United States and Europe have already experienced that some parts in their wind turbines have become obsolete, and there is no accompanying documentation to support their repair. Now if one of these parts malfunctions it is no longer just a case of replacing it, and engineers must find out how the part works and carry out detailed root cause analysis of why it failed. The complex electronics contained in today’s wind turbine assemblies requires a unique combination of skills and repair capabilities. The repair process must be carefully managed to ensure that the parts are repaired within appropriate time scales and in a cost-effective manner. This involves the use of detailed workflow and quality procedures so that each stage in the repair process is monitored and certified according to ISO standards. Fig. 1

Often the most difficult part of this repair process/recertification involves ensuring that parts are exercised to their full design specifications. The majority of electronic control parts are custom made for the wind turbine environment, therefore it is not practical or reliable to test these parts using standard off-the-shelf equipment. Nor is it practical to use a wind turbine to exercise the parts. This situation has necessitated the development of complex new software, tooling, and test jigs in order to carry out full testing and validation procedures. To replicate an often highly intelligent series of electronic interfaces and inputs/outputs signals need to be analyzed in great detail, and software and hardware emulation techniques are used to design tooling equipment capable of ensuring full test coverage on all areas of functionality of the parts. Finding a depot repair facility with expertise in this arena can be difficult, but not impossible. At DEX, for instance, we have facilities in the U.S. and Europe already in full operation. At every stage in the repair or refurbishment process it is crucial that the highest quality assurance standards are adhered to, and that a full audit trail can be provided showing the progress of the repair and certifying that the quality standards are in full compliance.

In all, a holistic engineering solution is necessary to report failure symptom trends, epidemic failures, third party module issues, and evidence of design quality and reliability problems. Understanding the “why” and “how” of component failures is paramount to prevent future disruptions in production.

Proactive Solutions
To maximize efficiency and effectiveness in this new environment, a coherent multi-faceted solution strategy must be in place before failure strikes. Specifically the need for a component repair or replacement strategy, asset management strategy, and an inventory management system need to be addressed proactively. Fig. 2

A coherent asset management strategy will mean that repairs are carried out quickly and efficiently, minimizing downtime. Once a part becomes defective it should be swapped out and sent out immediately for repair to ensure that a stock of working components are always available. Parts pooling schemes between operators can also help ensure that contingency stocks are kept high. The location, storage, and transportation of spare parts are crucial. Our research and experience has found that parts are frequently stored in warehouses close to the wind turbine site without adequate protection from the elements and are often in poor condition due to the adverse environment. In many cases the parts are unusable and need to be refurbished or repaired before they can be fitted. This can be avoided by storing parts in a central location under strictly controlled environmental conditions. Sensitive electronics are particularly vulnerable to temperature changes, humidity, and so forth, and can rapidly deteriorate if not stored properly.

Being able to track spare parts is equally important. Having an efficient asset management system with 24/7 visibility of where all parts are located and their condition—good inventory, WIP/defective—can save valuable time by identifying where spare part deficiencies may occur. DEX has developed a unique Web portal that clients can use to track and monitor their inventory providing them with real-time status of their operation. Fig. 3

Case Study
To help illustrate the necessity of having a comprehensive strategy in place I’d like to share what occurred with one of our clients. A major European wind farm operator whose turbines had come out of the warranty period had a series of rotor current controller parts failures, significantly reducing their production. With no advance strategy in place they simply turned to the OEM to purchase replacement parts. To their shock they learned that the lead time for their replacement parts was three to six months. In addition, the pricing structure from the OEM was going to significantly increase their costs. Compounding a worsening situation, with their reduced production the wind farmer was forced to purchase additional power from the grid to meet contractual obligations. Production loss on these turbines alone was in excess of five months, costing them an estimated $370,000. Not surprisingly, they began searching for alternative solutions to resolve this situation and alleviate any reoccurrences in the future. Fig. 4

With the wind farmer turning to our company, our DEX engineering team evaluated the defective parts and developed a full test and repair solution for the entire rotor current controller systems. A proprietary diagnostic software package was developed to ensure test reliability and repeatability during the product test phase. Upgrades were also carried out on the turbines, dramatically improving overall reliability. As a result the turbines were brought back online to a position of full production resolving their immediate need, and their material cost was reduced by a factor of three. A comprehensive solution has been set in place to prevent a repeat of long-term outages and disruptions in production.

In these challenging times wind turbine operators need to utilize any and all opportunities for maintaining the output of the operation at its highest capacity, cutting costs, and maximizing revenue and profits. Having a comprehensive strategy in place before failures occur is critical. Having the proper component repair or replacement strategy, asset management strategy, and an inventory management system in place can make a major contribution to achieving these goals. 

Nearly all utility-scale wind farm management systems have state of the art automatically controlled wind turbines, but at the electric grid intertie the substation is often managed manually. Some substations may have remote control capabilities, to be sure, but they are still completely manually controlled and require operator intervention for restoration. The real irony is that the cost to implement a full auto-restoration scheme at a modern substation with intelligent electronic protection devices is likely to cost no more than the lost revenue of one extended substation outage, and so it’s truly being “pennywise and pound foolish” to avoid implementation of such a system.

The modern wind farm grid intertie substation uses what is termed as intelligent electronic devices, or “IEDs.” An IED is any microprocessor-based unit that receives voltage, current, status, or other data from sensors and other devices and can issue commands via hardwire contacts or communication ports based upon internal programmed setpoints and logic. Substation IEDs are typically a variety of digital relays, digital logic processors, communication processors, and real-time controllers. The signals from inputs are interpreted in a systematic and predetermined way so that the system power devices are energized, de-energized, or otherwise controlled appropriately. The digitized analog values, status bits, timing functions, and logic functions that are integral elements of the substation IEDs can be leveraged to add auto-restoration capability to the substation. Although few if any IEDs must be added to perform auto-restoration, there is necessary effort to program the IEDs for auto-restoration. Standardization of the IED schemes minimizes the effort to programming, test, and commission the IEDs, and therefore the cost to implement auto-restoration. A simplified distributed wind generation substation diagram is shown in Fig .1.

A manually restored substation will have the protection relays indicated on the left side of fig. 2 under the heading “basic.” The scheme status switches are feedback indicating the system status of breakers and relays, etc. The relay IED scheme buttons are manually actuated locally or remotely to set the scheme into the desired mode. An auto-restoration may then be initiated manually using proper timing and the devices on the “basic” side of the diagram. A substation is fully automated by the addition of the microprocessor-based logic processor IED indicated on the right side of Fig. 2 under the heading “advanced.” The logic processor IED communicates with the substation relay IEDs to determine the status of all the power devices in the substation, and then upon pre-programmed logic stored in the memory of the logic processor IED the auto-restoration will be executed automatically at the appropriate time using measured data rather than timing to restore the substation.

The restoration sequence for a fully automated wind project follows the state flow diagram in Fig. 3. The normal state in which turbines are generating and providing electrical energy to the utility grid is the “normal” state at the top right of the diagram. When there is line fault or system disturbance the EPS breaker (breaker “U2” in Fig .1) will open and the system changes to the “EPS disconnected” state. The flow of energy to the grid ceases and electrical energy is no longer marketable. During the transition to the next state, two timed sequences are activated simultaneously. First a “power fail trip” status (labeled “PF trip” in Fig. 3) is declared three or more seconds after the loss of three-phase voltage. All relays for the line, feeder, and capacitor breakers are programmed with this status and all line, feeder, and capacitor breakers are opened at this time. The EPS breaker will close after the utility has cleared the fault or disturbance and the power fail trip operations have completed. The system state is now “ready for restoration.” The line breaker (breaker “U1” in Fig .1) will close if the appropriate conditions are all true after a set time delay. After the line breaker is closed, the feeder breakers will automatically close one at a time with predetermined time delays. Typically all the feeder breakers close in 10-second intervals so that all feeders are restored within one or two minutes. Once all feeder breakers are closed the collector substation returns to the “system available for generation state.” The total outage time for a fully automated collector project—turbines and substation—is typically 10 minutes or less. Any outage time due to the loss of the utility source is minimized and therefore provides optimized generation availability.

The value of automated substation restoration can be calculated by the economic advantage of reduced outage time relative to the longer time required for manual restoration. For a typical 100 MW wind farm with 33 percent generation capacity that is connected to the utility grid 70 percent of the time, the value per hour for lost delivery to the grid when generation would otherwise have been available is $3,500 per hour at 7.5 cents per kilowatt-hour. If manual restoration requires 12 hours before restoration can be accomplished, then the total energy value for this example is $42,000. It should be noted that this represents only the economic energy advantage and does not take into account the avoided cost for labor and vehicle expenses that are incurred with manual restoration. Manual restoration cost includes accounting for the time for an operator to be notified of the outage, the operator to travel to and from the collector substation, to analyze the reason for the outage, and then perform the restoration. The savings may be less than 12 hours if the manual restoration can be accomplished in good weather when an operator is readily available, but on a weekend night in inclement weather the time for manual restoration can be one or two days. After considering all these factors it is not difficult to recoup the investment for the auto-restoration investment after only one or two line outages.

The provision for auto-restoration on distributed generation, particularly for utility-scale wind farms, has been shown to be very feasible, with the microprocesser based relays and communication processors already being used to protect the collector substation that is part of every renewable energy project. With renewable energy very much dependent upon the uptime and delivery of energy to the grid, auto-restoration provides a very good way to accelerate the return on the large capital investment required for renewable generation. With only a marginal investment, the addition of auto-restoration may pay for itself in as few as one or two outages.