In the past few months, electric utilities’ reliability and power quality have come under scrutiny. The U.S. Department of Energy (DOE) set up a task force, the Power Outage Study Team, to investigate last summer’s power outages and disturbances, and Bill Richardson, Secretary of DOE, has gone as far as encouraging lawmakers to enact strict reliability legislation. In addition, reliability and power quality issues have prompted many commercial and industrial customers to explore avenues, such as on-site power generation and switching electricity suppliers, to make them less dependent on their current electric utility.
These actions should not surprise utilities, given the fact that next to high energy rates, customers invariably rank power interruptions high on their lists of problems with local utilities. Outages, whether momentary or sustained, usually cause some service disruption to the customer, and for commercial and industrial customers, the economic consequences can be significant.
Although reliability and power quality are somewhat related, they are really two separate issues. The simplest definition for reliability is power that’s there when it’s needed. Power quality can be defined as the degree to which power supplied by the utility conforms to “pure” sinusoidal waveforms of exactly 60 cycles per second-60 Hz.
It is fairly easy to understand how poor reliability can affect customers. The effects of poor power quality on the other hand are harder to measure. Variations from the pure waveform can create anything from a minor annoyance to a major disruption of customers’ processes. As computer-based operations become more common, high power quality becomes more important.
EPRI-owned PEAC Corp. of Knoxville, Tenn., is at the forefront of power quality solutions, with a portfolio of services that includes research, testing, on-site power quality assessments, training, diagnostic software tools and a power quality hotline. “Our business over the past years is rapidly increasing,” said Tom Geist, EPRI PEAC’s power quality engineer. “People have been calling from all types of industries, looking for solutions to power quality problems. As much as 90 percent of the power quality problems we see are characterized by the inability of computer-based process controls to tolerate voltage sags and very short power interruptions, often only a few cycles or less,” Geist said.
There is almost always some debate over whether utilities are obligated to provide “clean” power, but these days a utility needs to balance the cost of addressing customer problems with the prospect that a customer may seek alternative suppliers. Currently there are no universally accepted definitions or industry standards for power quality, although numerous organizations are developing and promoting such standards.
Any variation from the pure waveform is considered a degradation of power quality. Such variations can include:
“- Voltage. Utilities have standards that define the acceptable voltage range and methods for correcting voltage if it gets out of range. Voltages that are too high may overheat or damage customer equipment; voltages that are too low may cause stalling or dropout of motors, flickering or dimming of lights, or high currents in constant-power devices such as compressors.
“- Fundamental frequency other than 60 Hz. Frequency is tightly controlled by the utility. Mismatches between generation and load are the main cause of abnormal frequency, and automated control systems react quickly to restore balance in most cases.
“- Unequal phase currents and/or voltages. These can be caused by imbalances in the utility supply voltage, unequal phase impedance of transmission and distribution elements (lines, transformers, etc.), or unbalanced customer loads. Non-linear loads such as rectifiers or power-electronics’ controls are also sources. In the United States, residential loads are usually single-phase, therefore, distribution feeders almost always have some degree of imbalance.
“- Harmonics. These are frequencies other than 60 Hz and are the result of the imbalances described above, propagated through the system.
“- Noise. Noise is defined as small, random fluctuations in voltage or current, caused by things such as corona (surface sparking) on transmission lines, partial discharge (internal sparking) in transformers, scintillation on dirty insulators, arcing in the contacts of tap changers, faulty or loose hardware, radio frequency interference (RFI), lightning or solar disturbances.
“- Transients (“spikes” or “ringing”) in voltage or current. Transients are fluctuations caused by switching, relaying or other short-term disturbances. Their source can be in the utility’s system or in customer equipment. Transients originating in one customer’s equipment can affect another customer on the same feeder, and may not be apparent to the utility at all.
The anomalies mentioned above can cause problems for customers in varying degrees. Since there are many causes, it is not surprising that the solutions to power quality problems are many and varied. The following technologies are often used to solve power quality problems:
“- Uninterruptible power supplies (UPSs)-electronically controlled battery systems that charge from the grid and provide auxiliary power.
“- Motor-generator sets-an AC motor drives a generator at synchronous frequency; the rotating inertia of the system filters out power quality-related problems and provides ride-through of momentary interruptions.
“- Storage systems-includes large-scale battery installations, capacitors, flywheels and superconducting magnetic storage (SMES).
“- Power conditioning systems-power electronics that utilize conversion/inversion techniques to clean up the power to protect sensitive loads.
“- Solid state transfer switches-power electronic devices that can switch from one source to another in as little as half a cycle, greatly reducing the impact of momentary interruptions.
“- Surge arrestors-nonlinear resistors that “clamp” the voltage to suppress voltage surges due to lightning, switching or other system disturbances.
“- Isolation transformers-provide a 1:1 voltage transformation and smooth out voltage variations through the magnetic buffering properties of the transformer.
“- Dynamic voltage restorers-power electronics-based systems to support loads through voltage sags and outages.
“- Proper grounding procedures-improper wiring or grounding is a very common cause of poor power quality, in addition to being a safety concern.
Any of these devices can act as a buffer between the utility and the customer, ensuring clean, steady power (within limits). Geist explained that most problems PEAC investigates are resolved with an external solution, such as installing a UPS or a constant voltage transformer, or improving grounding.
“PEAC recommends economical solutions, that use capacitors and power electronics, to enable customers’ controls to ride through short-duration power quality events. The big-dollar fixes like motor-generator sets are only needed if a customer is worried about keeping processes going during sustained outages, which is more of a reliability problem than a power quality problem,” Geist said. The ultimate solution is when vendors and manufacturers offer power quality solutions embedded in their products, he added.
Even though power system reliability has been getting a lot of bad publicity lately, reliability in the United States is still quite high. Ideally, electricity would be available to every customer all the time, no exceptions. But the real world isn’t perfect. Various factors contribute to lower reliability, but fortunately measures can be taken to improve reliability. These measures differ between transmission and distribution systems.
A transmission system’s reliability index is normally expressed in percent of system average availability and is typically more than 99 percent. Even though this percentage is a testament to the North American transmission grid’s robustness, it is rarely translated into average customer outage minutes per year. Even if it were, the number would be much less than the number of outage minutes attributable to distribution reliability.
Transmission systems’ design plays the biggest role in providing such high availability levels. They are meshed networks (grids) that deliver large amounts of electric power at high voltages. Because transmission systems are networks, the loss of any one segment, such as a transmission line, transformer or generator, usually causes only a minor disturbance to the system. The network allows the power to take different paths from the generation source to the load, and it usually can take a second or even third contingency before disastrous results occur. Typical causes of transmission outages include lightning strikes, transformer failures, line splice failures, switching surges, wind toppled towers, lines in contact with trees or vegetation, and insulator flashovers due to animals or contamination buildup.
While distribution systems are pretty reliable, they do not enjoy the high availability rate that transmission systems do. Just like with transmission systems, distribution system design plays a significant role in system performance. Distribution systems are usually, but not always, “radial” systems, which means power flows from the supply point, usually a substation connected to the transmission system, downstream to customers distributed along the line. Unlike the transmission system, little if any redundancy exists on a distribution feeder: It is a series of segments and components, resembling a chain, and like a chain, it is only as strong as its weakest link.
If any link fails, all customers downstream of that link are out of service until that link is restored. Equipment failure, trees, lightning, wind, birds and car-pole accidents are some of the most likely causes of distribution outages. Most U.S. distribution systems are designed with strategically placed sectionalizing switches, so that loads can be switched between feeders to restore service to customers while repairs are being made to faulted line sections. Software tools allow engineers to model the feeder as a system, and to identify the most cost-effective fixes to maintain acceptable reliability.
A momentary outage is defined as an outage that lasts less than five minutes, corresponding to the time allowed for automatic reclosing schemes to try to restore the circuit if the fault was temporary (a flashover, for example). A sustained outage lasts longer than five minutes, is due to non-recoverable failure of some component, and requires utility personnel to perform repairs to restore service.
Utilities normally express distribution system reliability in terms of either percent availability, or sometimes as average yearly minutes of outage per customer.
As this article mentioned earlier, some utilities are feeling regulatory pressure to improve reliability. One way this is occurring, especially in areas where deregulation is introducing market-based incentives, is through performance-based rates (PBRs). PBRs typically set reliability targets and build reward/penalty systems into the utility rate structure. Reliability targets are based on industry-accepted outage indices; the most important of which are described in the table.
These indices (and others) are included in IEEE Standard 1366: Electric Power Distribution Reliability Indices. These indices capture the effects of the number of outages, however short, as well as outage duration, and are usually computed from the past several years’ worth of utility data. Under PBR rate structures, reliability standards can be an effective tool for ensuring that the utility stays focused on customer service.
At least one recent PBR filing also included a proposed electric maintenance performance clause. This clause calls for a penalty or reward based on a performance index called maintenance repair and replacement outages (MR&RO). Its objective is to provide an incentive for the utility to perform effective line inspections so that faulty equipment is detected and repaired before failures occur.
Reliability is first and foremost a function of design, operation and maintenance practices. T&D systems designed with bigger conductors, higher insulation levels, and newer equipment, with redundancy built in where possible, will be more reliable and less likely to fail. In some cases, typically in downtown or commercial areas, distribution systems can be connected in a networked configuration to achieve supply-path redundancy, similar to transmission systems.
In the maintenance area, reliability-centered maintenance (RCM) is coming into vogue. An information-based analysis tool, RCM depends on an expert system database of maintenance data to allow distribution engineers to predict which components are likely to fail and how soon. Real-time sensors can be employed on transformers, circuit breakers and other expensive equipment to telemeter data directly into the RCM database, saving considerable technician time and labor.
One approach utilities can take is to offer a Premium Power service to customers. This may include equipment and methods to counteract power quality problems, or measures to assure greater supply reliability. In the latter case, a utility or electric service provider can offer a solution that includes distributed generation located at or near a customer. Another option might be to offer a customer a slightly higher rate in return for moving the customer’s load farther down in the load-shedding pecking order during contingencies.
Distributed generation has been touted as a possible reliability enhancer for distribution systems. Highly sophisticated automated controls for distributed generation systems, allow them to be remotely dispatched for local voltage support and supply. If a distributed generation application is supplying local load, it may be possible to have the generator ramp up its output to supply other customers or sell to the grid, like a standby generator, but probably faster. The voltage profile on the feeder would be improved, line losses reduced, and ride-through of short interruptions would be possible, benefiting customers.
Distribution automation (DA) systems can also help utilities improve reliability by providing real-time monitoring of voltages, feeder status and other critical parameters. System operators can remotely control switches, capacitors and other components to restore power more quickly after sustained outages. Some researchers are investigating artificial intelligence technologies, such as expert systems, that will provide DA systems with a higher degree of automated response.
Reliability and power quality have historically been high priorities with utility customers. As the electric utility industry transitions into a market-driven system, customers will be even more insistent on getting good value for their money, or they will seek alternate suppliers. Utilities and service providers will have to compete on the basis of service as well as commodity cost to retain customers, and the cost-effective solutions being developed by EPRI and numerous others will help enable this competition.
Lloyd Cibulka is a registered electrical engineer with 25 years of electrical engineering and management experience in the electric power industry. He currently works as an electrical engineering consultant and freelance writer.