Reactive power is becoming a hot issue in many regions of the country. Regulators and grid operators are grappling with ways to account fairly for reactive power supplies, and to encourage such resources to come online where they are needed. These analyses, however, are largely ignoring a vast fleet of infrastructure already installed on the network—namely, on-site generators and motors.
West Point military academy, for example, has four small synchronous generators that are used for combined heat and power or emergency power applications. If these generators also were used as synchronous condensers, they might supply additional revenue to pay for the distributed energy investment.
These engine generators already have been purchased, so the only costs in supplying reactive power would be the internal losses and some nominal equipment expenses.
Other options to control power factor is use of microturbines or adjustable speed drives. The payback on such solutions can be quite attractive, and merit a closer look as policy-makers consider how to account for reactive power.
Independent system operators (ISOs) are responsible for the operation of some of our large high-voltage transmission systems. ISOs are committed to reliability, the nondiscriminatory operation of the bulk power system, and to working with all stakeholders to create cost-effective and innovative solutions for the electric industry.
The New York Independent System Operator (NY-ISO) uses large, conventional generators connected to the transmission grid for providing reactive power. Payments are made from a pool consisting of total costs incurred by generators that provide voltage support service. The 2004 compensation rate was roughly $4,000 per MVAR per year. In addition, NY-ISO will pay a lost-opportunity cost when a generator is directed to reduce its real power output below its schedule in order to produce reactive power. NY-ISO is interested in sources for reactive power. In an August 2004 draft for discussion, the ISO staff proposed to market participants a set of modifications, including that:
This document points out that synchronous condensers are not effectively compensated for voltage support service (VSS). “When a synchronous condenser is operating it requires an amount of real power; the cost of this energy, if acquired at market rates, could easily exceed the potential revenue from the VSS payment. Another limit to the potential revenue is that a synchronous condenser is not eligible for an ICAP (installed capacity) contract and would be compensated only for the actual hours of operation,” the document states.
The document recommends that NY-ISO staff work with the owners of synchronous condensers to identify the unique issues associated with operation and cost-recovery needs, and develop a proposal that allows fair compensation for service provided to the New York market.
The concerns expressed in this draft report certainly are valid. The New York Power Authority (NYPA) recently took a 40-MVAR synchronous condenser out of service because it was not economical to operate. Dynamic reactive power reserves are shrinking. As loading has increased and transmission systems have been “propped up” with more and more capacitor banks, and as less transmission capacity is available, the voltage collapse curve has become steeper. The power voltage curve “nose” has become very sharp, and voltages will fall immediately when there is no reactive power available to support more load or transfer. This is shown in Figure 1.1
The United States Military Academy has a 1,500-kW diesel generator and two steam turbine-driven 1,250-kW generators. The turbines are older non-condensing turbines used only in the winter, and the exhaust steam is used to heat campus buildings. The generators are vintage 1973 and operate at 1,800 rpm.
These generators would be ideal for use as synchronous condensers because they are connected to the turbines with removable couplings. In addition, there is a newer 1,500-kW steam turbine. These generators are installed with switchgear connecting them to the West Point distribution system, and they could be controlled either to regulate voltage or control the power factor on the West Point distribution system.
A critical point, however, is that reactive power does not travel well, and there are some locations where it is needed and some locations where it is not.
The local transmission operator at West Point is Orange and Rockland. Orange and Rockland has two capacitor banks on the 34.5-kV line that feeds West Point, and they usually operate only one. The second one is only connected during a heat wave when load is unusually high, and has been used only once in the last two years. Orange and Rockland has no need for reactive power from West Point.
For the purposes of discussion, let us assume that West Point is in a load pocket, an area of potential voltage susceptibility, that Orange and Rockland has grown weary of replacing capacitor switches and load tap changers, and that Orange and Rockland is looking for an alternate method of supplying dynamic reactive power and regulating local voltage. In this case, it may be possible to supply dynamic reactive power from West Point as a service. As mentioned above, we have a total of 5.5 MW of real power available from the four generators.
To supply reactive power when the generator is running, all we need to do is to control the generator excitation to provide the appropriate phase angle, either for voltage regulation or power factor correction. The problem is that the turbine generators are running only during the winter, and the diesel generator is only running during emergency conditions. Removable couplings are needed to allow the generators to “motor.”
If we assume the power factor of these generators is 0.8, we have 4.1 MVAR of dynamic reactive available. In theory, we may expect compensation equivalent to that paid by the ISO for generators located on the transmission system. This compensation is $4,000/MVAR per year, or $16,500 per year for the 4.1 MVAR.
The benefit from providing reactive power from the synchronous generators in the West Point distribution system may be approximated as follows. The generators are connected to 4,160-V buses. The peak West Point load is 17 MW in the summer and 12 MW in the fall, winter and spring. The West Point power factor is close to 0.9 for all months. If we assume an average load of 7 MW at 0.9 power factor, this results in reactive power absorption of 3.39 MVAR. This is smaller than the 4.1 MVAR of reactive power available from the onsite generators, so the reactive output of the generators could be used to correct the campus net power factor to 1.
For a rough estimate, let’s assume that the 4,160-V bus feeder cables are two 750 kCM in parallel, a common size seen in bus-feeder cables. These two cables in parallel would have a resistance of 0.073 ohms per 1,000 feet. The extra current associated with the 0.9 power factor is 107 amps at 4,160 V. The additional I squared R loss associated with this extra current is only about 16 kW. In addition, if we assume that the friction, windage and I squared R losses of the generators is 3 percent—a reasonable estimate—then the total losses of the 5.5 MW of generation is 165 kW. Thus the losses in the machines easily negate the possible savings in distribution system losses if the machines were to be operated solely for the purpose of providing reactive power to reduce West Point losses. Some transmission operators will compensate a synchronous condenser for the losses in the machine when the machine is supplying a reactive-power service. This would be essential for synchronous generators to economically supply reactive power.
The West Point generators were not purchased with removable couplings between the engine and generator; these would be required at perhaps $5,000 each as a minimum. A local generation excitation controller would be required at perhaps $5,000 each. The total cost to enable the generators to supply reactive power and control the net West Point power factor may be at least $40,000 to $50,000. The ISO pays $4,000 per year for reactive power to generators that are officially connected to their transmission system, where the ISO can control the reactive power being supplied.
If we were able to get some similar payment, it would probably only be for the amount that West Point could actually go leading. This would be the 4.1 MVAR available minus 3.39 MVAR absorbed, or 0.71 MVAR. If it could be paid, this would only yield roughly $2,800 per year. If friction, windage, and I squared R losses are not compensated for, the payback time for reactive power supply would be a very long time.
There are alternate possibilities, however. West Point may have some large motors used for pumps or fans that could be used to supply reactive power if they were equipped with adjustable speed drives.
Adjustable speed drives (ASDs) change the voltage magnitude and frequency at the motor terminals. Adjustable speed drives are tremendous energy savers because motors that drive pumps or fans can be easily controlled to supply just the amount of water or air that is needed, with no wasted energy. When a pump or fan is used in an application where the flow requirement varies, as they often are, controlling the pump flow with an adjustable speed drive instead of a throttle valve can save energy equal to one half the horsepower rating of the motor. Adjustable speed drives often have payback periods of less than one year.
Today’s adjustable speed drives can be used to change power factor. The option to control power factor is called an active front end. With the use of adjustable speed drives at West Point, it is easy to imagine that the net power factor could be corrected to near 1.0 for the entire year. We can roughly approximate the savings as follows.
Motor-driven equipment accounts for 64 percent of the electricity consumed in the U.S. industrial sector.2 Let’s assume that just 30 percent is large motor load. To simplify, let’s assume that 10 motors draw 200 kW at 0.8 power factor, slightly less than 30 percent of the 7-MW average load discussed above. If we equipped these motors with adjustable speed drives, we could supply about 1.6 MVAR of reactive power from these adjustable speed drives. Importantly, we could supply this reactive power at the motor terminals, where it does the most good in reducing losses.
How much energy goes into heating in the cable and transformer system that feeds the 480-V motors? Average distribution system losses account for 2 percent of plant annual energy use.3 If we use this 2 percent to calculate a system resistance for the circuits feeding the motors, the extra current flow associated with the 0.8 power factor accounts for roughly 15 kW. Installing adjustable speed drives to correct the power factor to 1.0 would save roughly this amount of power. The 15 kW at an average energy cost of 0.062 $/kWh gives a total dollar savings of about $8,000 per year.
The cost of installing adjustable speed drives usually is amortized by the energy savings realized by the reduction of losses in the air or water flow. Drives often are paid back in six months or less. In addition, some utilities offer rebates for the installation of adjustable speed drives. For this reason, we will not consider the cost of installing the drives. If they are warranted by the conventional energy savings analysis, their cost will be quickly amortized. The savings of $8,000 per year simply will be an additional incentive.
The site kilowatt savings in cable and transformer heating from using adjustable speed drives to correct power factor was estimated above to be 15 kW. Over a one-year period this is 131 MWh. Using the conversion factor of 11,850 BTU/ kWh, this converts to 1,557,000 source BTU savings.
This is a meaningful level of source BTU savings, especially if it can be repeated at other commercial and industrial installations. Supplying reactive power from ASDs will provide an important service to distribution and transmission operators. The power factor of the load could be kept at 1.0, or even leading when needed. This will result in greater distribution and transmission system efficiency and reliability.
The results may be common to a wide range of industrial, commercial, and even residential distribution systems. The ASDs could be controlled easily to provide leading power factor when needed. A meter is available that measures power factor at the load and reports the power factor to the utility using a built-in cell phone. The meter also can transmit a local radio signal to the ASDs. The utility could thus request that power factor go leading when they are dealing with a system under stress and have inadequate reactive reserves.
There is yet another possibility that we should consider. Alliant Energy provides a power-factor credit to customers that can correct power factor to 0.95 or higher. Although Alliant is in Wisconsin and West Point is in New York, consider the hypothetical case of using the Alliant credit at West Point. This is a reasonable situation to consider as this sort of credit hopefully will become common across the nation.
We will assume an average West Point peak load of 14.5 MW. Alliant’s demand charge is about $10 per on-peak kilowatt per month. If West Point could correct their power factor from .9 to 1.0, Alliant would credit West Point ($10 x 14,500 x .1), or $14,500 per month. The annual credit would be about $174,000.
Inverters are used in both microturbines and adjustable speed drives. Inverters with active front ends can be used to control power factor. Next generation microturbines will have this capability. If microturbines could be used at West Point in a combined heat and power application, they could have an especially fast payback because of this power factor added credit. West Point already has a demonstrated need for the waste heat from the turbines. They use boilers to heat water for campus use. Microturbines or small turbines would be ideal to provide heat and real power from the turbine and reactive power from the microturbine’s inverter.
The savings to the system owners as well as to distribution system operators, who could control the inverters to provide leading power factor when needed, would be truly significant. Presenting a net power factor of near 1.0 is becoming increasingly important as the nation’s bulk power system is under more and more stress, and inverter-based distributed energy resources are a readily available and highly efficient method to accomplish this.
1. Reactive Power for Planning and Operation, Harrison K. Clark.
2. U.S. DOE, Office of Energy Efficiency and Renewable Energy, Industrial Technologies Program, Motors, Pumps and Fans Fact Sheet.
3. Energy Tips – Motor Systems, Motor Systems Tip Sheet #8, September, 2005, U.S. DOE, EERE, Industrial Technologies Program.