An uninterruptible power supply (UPS) is defined as, “A device that provides battery backup when the electrical power fails or drops to an unacceptable voltage level. Small UPS systems provide power for a few minutes, enough to power down the computer in an orderly manner, while larger systems have enough battery for several hours.” (Source: PC Magazine Encyclopedia.)
This is the classic definition as applied to the protection of individual computers and other relatively small load demands. Uninterruptible power supplies can range in size from the surge strip that your computer is plugged into to a large commercial or residential emergency backup system. But this is not to confuse a UPS with microturbines, emergency generators, or gensets. Unlike standard backup power systems, a UPS will provide essentially instantaneous protection from power failures.
Also, while gensets are designed for continuous, long-term operation (provided they are properly fueled), a UPS may only operate for a very short period of time. This is especially true of battery UPS systems which can sometimes be designed to operate for only a few minutes. This may seem insufficient, but the buffer or time allotted by a UPS is sufficient for the system to perform a shutdown properly to prevent damage to the powered equipment.
As such, standard UPS are typically used to protect critical systems (telecommunications, emergency systems, data storage, air traffic control systems, commuter rail controls, and operating computers). The surge strip that your computer is probably plugged into is a small-scale example of a UPS. A large example would be the behemoth Battery Electric Storage System (BESS), in Fairbanks, AK. This 46-megawatt battery UPS can power the entire city and surrounding communities in the event of a power outage.
But what about distributed energy and renewable energy sources? How do uninterruptible power supplies work within their operations? How are they used to protect the local grid against power failure from a photovoltaic cell (PVC) array or wind turbine? And what exactly is a failure that would trigger a UPS?
How Energy Supplies Can Fail
A UPS is installed to protect the electrical system from short-term power supply system failure or when the energy supply becomes erratic. In addition to their primary function, uninterruptible power supplies can also smooth out or otherwise compensate for other power failure modes:
- Voltage spikes and surges. These are sudden increases in electrical supply voltage. Increases that last for more than 3 nanoseconds are referred to as surges. Those that last less than 3 nanoseconds are spikes. Voltage spikes can be caused by lightning discharge during thunderstorms, a spark from static electricity, applied magnetic fields and electronic devices, or appliances that turn themselves on or off.
- Brownouts. These are momentary or sustained reduction in input voltage. The opposite of a voltage spike or surge, these also have many potential causes. Local causes include increased voltage applied to the output side and triggering voltage surge protection, or incorrect input voltage not compatible with the electric loads operational setting. System-wide brownouts could result from physical damage or disruption to the grid or be deliberately imposed by the system operator to avoid a worse situation—a blackout with complete failure of the power grid.
- Noise vibrations. These are defined as a high-frequency transient or oscillation, usually injected into the line by nearby equipment. This is electronic noise, not acoustical noise, and is simply a random fluctuation in an electrical current.
- Grid Instability. This is defined as variability between electrical service and electrical load. There is an ongoing debate as to the amount of grid instability created by renewable energy sources linked to the grid.
- Harmonic distortion. This is a temporary distortion of the ideal sinusoidal waveform (sine wave) of the electrical power frequency for alternating current (AC). Linear (constant) load draws current in relation to the sine wave voltage. Non-linear (variable) load, on the other hand, only conducts current near the peak of the sine wave. When these loads are turned on and off, they create non-sine wave current pulses that alter the wave form. This further results in reflective (harmonic) currents being propagated back into the electrical supply distribution system.
UPS units are defined by the category of potential failure mode they are designed to address. Certain types of UPS are designed to manage multiple potential failure modes. The majority of installed UPS capacity is intended to manage localized failure modes by protecting specific valuable assets from the effects of voltage irregularity. For such assets, the effects of failure cannot be tolerated.
However, the saying “failure is not an option” does not really apply to renewable energy sources. Renewable energy sources do not fail so much as they are inherently discontinuous and variable. Some of the conditions that would be considered failure in a standard power supply system are inherent to renewable energy. As such, the UPS systems intended for commercial- and community-scale renewable energy sources tend to be large-scale and regional instead of being for a specific application.
Uninterruptible Power Supplies—Standard Battery Technology
The batteries used to provide uninterruptible power vary in size and capacity in accordance with their specific applications. They are rated according to their total capacity which is a function of their rate of discharge, maximum available charge, and the efficiency of their inverter. This last item is critical to the entire function of the UPS. Batteries generate direct current (DC) electricity. However, electronic equipment, computers, and appliances use alternating current (AC). The inverter converts DC produced by the battery into useable AC. This ability to convert DC into AC combined with the ability to detect failure in electrical power supply is required for those systems that utilize batteries as either emergency backup power or uninterruptible power sources.
Individual batteries and combinations of batteries come in many sizes and capacities. These can be used to provide uninterruptible power to loads as small as an individual laptop computer to large-scale buildings and small communities requiring power in the megawatt range. Batteries are considered to be stationary forms of UPS in that they have no moving parts like a rotary flywheel. In addition to its size and capacity, a battery UPS can be classified according to functionality, operational layout, and complexity. There are four such categories: off-line, line-interactive, on-line, and true.
Off-line UPS is the simplest configuration and is used for small applications such as individual computer and home appliances in the event of a power loss. Being off-line, it is not directly connected to the inverter that converts its DC power into AC power. It needs a power transfer switch to make this connection. In contrast, an on-line UPS is always connected to the inverter and therefore does not need a power transfer switch. If there is a loss of power, the rectifier drops out of the circuit, allowing for unsteady and unchanged power transmission. When the power is restored, the rectifier resumes carrying most of the load and simultaneously charges the batteries.
A line-interactive UPS can function as both an off-line and an on-line UPS. Utilizing a bidirectional AC-DC converter, it can either recharge the battery or convert the current from the battery to power the electrical load. During normal operations, the line-interactive UPS goes into on-line mode and can act to adjust voltage and keep it level. If regulating the voltage is not required, the line-interactive UPS goes into off-line mode and electricity flows freely to the load from the power grid. Should complete power failure occur, a static switch disconnects that grid and the system being protected by the line-interactive UPS goes into islanding mode. This refers to a situation where the local load will continue to receive electrical power from the battery.
A true UPS (a.k.a. “delta conversion”) utilizes two bidirectional converters—one in series and one in parallel with the main power line. Both converters can be used to charge the battery while the parallel converter can transfer power to the local grid when general grid power fails and the system being serviced by the UPS goes into islanding mode. While this is similar to line-interactive UPS operations during a power failure, a true UPS is more suited to loads that use a lot of energy (such as large data centers).
The batteries themselves can be as varied as their configurations: lead-acid, lithium-ion, sodium-sulfur, nickel-cadmium, nickel-metal hydride, and sodium-nickel hydride. Battery UPS systems can be constructed in less than 12 months with flexible locations both inside and outside the building being serviced. Lead-acid is the oldest battery technology and the most widely used, with good efficiencies (from 63% to 89%) and low costs ($50 to $600 per kW). However, they need thermal management during operation. Lead-acid battery systems tend to be used for smaller applications with relatively few utility-scale facilities. Lithium-ion batteries are usually considered for systems that require fast response times, small dimensions, and limited weight, due to their relatively high energy density. They have high operating efficiencies (97%) but often require an on-board computer to manage its operations.
Uninterruptible Power Supply and Distributed Energy
The basic question concerning UPS applications to distributed energy systems is the size of the system. Distributed/renewable energy systems are classified as residential, commercial, or utility in size. Residential-scale applies to individual homes and buildings receiving electrical power from a relatively small dedicated distributed energy system physically attached to the building in question (rooftop solar, e.g.). Commercial-scale encompasses larger business, industrial, and apartment complexes with multiple users but within a well-defined area. This would include factories, stores, shopping malls, campuses, and office buildings that receive power from a local but structurally independent renewable energy source. Utility-scale renewable energy serves as an adjunct to or a replacement for standard utility power sources. These systems are large enough to service entire communities and even municipalities. Entire nations have pledged to achieve 100% renewable energy within the next few decades as they replace older power systems with utility-scale renewable energy sources.
However, unlike standard electrical systems supplied by continuously operating power plants (such as hydroelectric, nuclear, combined-cycle natural gas turbines, and old-fashioned coal fired boiler generators), distributed energy systems produce electricity when available—which is not always when it is needed. By its very nature, renewable energy is variable to a degree that would not be tolerated in a standard steady, continuously operating power source. The sun does not always shine and the wind does not always blow. Not only does the sun go down at night, but its intensity varies throughout the day as it traces an arc across the sky, generating less power at dawn and dusk, and peak power at noon. Furthermore, sunlight varies with latitude, cloud cover, and season. Wind speeds are highly variable, as is the resultant power output. The resultant power supply curve resembles a parabola matching the sun’s location in the sky that is zero at sunrise and sunset and peaks at noon. Wind power is proportional to the wind speed cubed. So, doubling the wind velocity results in an eightfold increase in output power. Such highly variable power cannot be used directly by a supply grid with consistent load demands. However, unlike failure in a standard power grid, these variations can often be anticipated and planned for as part of daily operations.
Demand also determines the nature of the required energy storage and its use as a UPS for a renewable energy system. Industrial and commercial demand can be relatively steady over a work shift when stores are open, offices are occupied, and factories are running production shifts. Commercial and industrial use peaks at about 2:00 in the afternoon to match business activity and the use of environmental controls (heating and cooling). Once they close or shut down, demand drops to almost nothing. Residential use is more variable, at its lowest from midnight to about 6:00 a.m.; it increases at the start of the work day and peaks at about 7:00 in the evening. Combining the two daily power demand curves results in an overall wave that resembles a sine wave, with its peak in the late afternoon and a trough in the early morning hours. Individual use can vary wildly but aggregate demand from a large number of customers remains fairly consistent (with allowances for season changes such as increased heat demand in winter or air conditioner use in summer).
Comparing and overlapping the power supply’s parabolic curve and the energy demand’s sine-like wave shows that power supply typically exceeds aggregate demand from about mid-morning to late afternoon. It is this excess power supply that needs to be stored for later use when demand exceeds power availability, or the renewable power is not available at all. While battery UPS can be integrated into distributed energy operations (certainly for individual home use at the residential scale), most large-scale commercial and utility renewable energy sources often rely on large-scale methods of storing excess solar or wind energy for later use. Given the diffused nature of renewable energy, solar arrays and wind farms can cover a very large area. The electrical energy storage (EES) systems utilized as uninterruptible power supplies at the commercial- and utility-scale levels are often proportionally large as well.
Pumped Hydroelectric Storage (PHS) is the simplest, oldest, most widely used, most efficient, and cheapest means of providing uninterruptible power supply to commercial- and utility-scale electrical power systems. This is a mature, proven technology with PHS plants installed worldwide with power rating ranging from 1 MW to over 3,000 MW. During peak supply hours, the excess renewable energy is used to pump water from a lower reservoir to a pond located at a higher elevation. The available power from the upper stored water is a function of the elevation difference between the two pond surface elevations and the amount of water in the storage pond. The rated power of a PHS facility is a function of the water pressure and the flow rate through the turbines. When the renewable energy supply drops below the demand curve, water is released from the upper pond and used to drive an electrical turbine to generate enough electricity to compensate for the shortfall.
PHS systems can achieve 70 to 85% efficiency, making them the most efficient EES system currently available. However, these systems only make economic sense at very large scales. Studies have been performed with individual buildings utilizing buried reservoirs and storage tanks located on their roofs. Unfortunately, there does not appear to be the necessary economies of scale to make small PHS systems economically viable. Since they are dependent on topography, their available locations are limited. They also have high capital costs and long construction times.
Compressed Air Energy Storage (CAES) is also a mature, economically viable EES technology. Excess energy from the renewable energy source is used to drive an air compressor engine to pump high-pressure air into a storage tank. It later uses the compressed air to spin a turbine and drive a generator to create electricity during times when demand exceeds production. Connection to the grid is via an AC-DC-AC converter to control power quality. It can operate in power ranges in excess of 100 MW. Advanced systems include liquid air energy storage (LAES) and advanced adiabatic CAES which is integrated with a thermal energy storage system.
Flywheels (a.k.a. “rotary UPS”) are a mechanical means of storing renewable energy. As such, there is no need for an electrical motor or generator to act as an intermediary and the actual source of emergency power. They utilize the inertia (mass x momentum) of a high-density spinning flywheel to provide the power to turn a generator and create AC power should a power loss or interruption occur. They can compensate for short-term power spikes or losses without significant reduction in its rotational speed. Given its high inertia, short-term applications don’t have a significant effect on its rate of spin. Since it is continuously spinning, it is considered to be already on-line and can engage near instantly should it be called upon to do so. However, should its full capacity be required, a flywheel UPS typically operates for only a few minutes before the flywheel is slowed to a stop. Yet this is usually sufficient for a backup generator or engine to begin operating and restore electrical service. In short, flywheel technology is mainly used in applications that require a great deal of power but for only a short period of time (minutes or seconds).
Flywheels tend to be simple in design with operational lifetimes of 30 years or more with regular maintenance (primarily ball bearing replacement and recharging of lubricant). This simple design consists of five components: the flywheel itself, a pack of ball bearings, a reversible electric motor/generator, a vacuum chamber with pump, and the axle on which the flywheel rotates. This last item gets the flywheel starting with the application of electrical power. Once engaged, the spinning flywheel drives the reversible motor/generator to create electrical power. Working in the opposite direction, the reversible generator/motor accelerates the flywheel by imparting electrical energy to its rotation. Flywheels are classified into two categories: low speed flywheels which use steel as the flywheel material and rotate below 100 rpm, and high-speed flywheels that use advanced composite materials (typically carbon fiber) for the flywheel and can run up to about 150 rpm.
Biofuels fill a very important niche. Not only does solar energy need to be storable, it often needs to be transportable as well. Biofuel, being liquid or gas, is easily transportable via an existing network of pipelines and fleets of tanker trucks—just like gasoline, diesel, and natural gas. Utilizing microbial and plant processes for production, biofuels include: biodiesel, ethanol, biogas, and algae oil. They have the same energy density that makes traditional fossil fuels so useful. Gasoline and diesel, for example, have an energy density of approximately 46 megajoules per kilogram (MJ/kg). By comparison, a rechargeable lithium-ion battery has an energy density of only 0.4 to 0.9 MJ/kg.
Hydrogen can be considered a fuel when generated by renewable energy. Utilizing sunlight, artificial photosynthesis, via catalysts, splits water into oxygen and hydrogen. The hydrogen is then utilized to power fuel cells that generate DC power like a battery. These catalysts, however, tend to be rare expensive materials. Furthermore, the catalyst chamber requires intensely focused sunlight from multiple heliostats to generate the reaction.
Latent heat sinks function as a thermal energy storage (TES) units. Typically, heat from concentrated solar power (CSP) systems is stored in insulated repositories. However, electrical from wind and solar power sources can be run through resistors that radiate heat into the storage media. These are, in effect, “heat batteries.” Economically attractive with low capital costs, TES systems can store large amounts of energy. Various media can be used to store heat, including water, rock, ceramics, and concrete. The stored heat can be extracted via a heat exchanger to vaporize steam to drive a turbine and generator.
What is the difference between true UPS and backup power systems for renewable energy sources? This is something of a gray area. In many respects, even the largest backup power system for a renewable energy source (think very large pumped hydro storage lake) acts like a UPS supply. But the difference is in response time. A true UPS is designed to respond quickly, often in mere milliseconds. Backup power sources don’t quite work that way, even though they are designed to compensate for loss of power when the sky gets cloudy or the wind stops blowing or the sun sets. There is also a size limitation, with true UPS often relegated to commercial-sized and smaller applications (a large data bank would fall into this category). And there is rarely a UPS acting as a bridge between the main renewable energy source and its backup energy storage system. Yet even though a backup energy storage system is not technically a UPS, it can be treated as such for planning purposes, provided that its limitations and characteristics are taken into account.
Trojan Battery Company is a leading manufacturer of deep-cycle batteries of many types (flooded, AGM, and gel) for a variety of applications. Trojan’s factory sales locations support various battery-powered applications including automotive, floor machine, golf carts, industrial, marine/RV, motorcycle, mobility, renewable energy, and utility vehicles. Early applications for their batteries were in golf carts. Their current product line includes a series of flooded, AGM, and gel batteries for solar energy applications. Their Reliant AGM line of US-made absorbed glass mat (AGM) batteries are the only true deep-cycle AGM batteries on the market today. Reliant is engineered with an advanced technology feature set that provides outstanding sustained performance and total energy output, delivering exceptional quality and reliability. Trojan also offers both Solar Flooded and Solar AGM battery models for the renewable energy market that are designed specifically for solar and other renewable energy applications. Trojan’s Flooded solar batteries are tested up to a 17-year design life under IEC 61427 standard for solar batteries. Trojan’s Solar AGM are true deep-cycle AGM batteries tested to an eight-year design life under the same IEC 61427 standard.
Trojan’s Smart Carbon formula addresses the impact of partial state of charge (PSOC) on deep-cycle batteries in renewable energy, inverter backup, and telecom applications, and is a standard feature in Trojan’s Industrial and Premium flooded battery lines. Smart Carbon provides improved performance when the batteries operate in PSOC, enhancing overall battery life in applications where the batteries are under-charged on a regular basis.
“With batteries being one of the most expensive components of a battery-based solar system, it is critical to maximize the life of the battery bank to reduce the total cost of ownership of a system,” says Elke Hirschman, senior vice president of North American sales and corporate marketing for Trojan Battery. “Trojan’s new Solar AGM line illustrates the company’s commitment to offering reliable energy storage solutions for a wide range of renewable energy market segments. Trojan continues its focus on being an innovative leader in the energy storage space.” Applications include the City of Joy Solar Community center, solar powered tiny houses, and solar powered remote microgrids serving indigenous people in Choco, Columbia. This last project will provide clean energy to 431 homes for the next 20 years.
Schneider Electric is an international leader in renewable energy management. Their advanced microgrid solutions seamlessly integrate with their portfolio of software, hardware, controls, and engineering services. Their integrated, scaled approach across all disciplines allows Schneider Electric to deliver innovative and economically feasible microgrid solutions with greater speed and precision. In doing so, they provide flexible microgrid designs featuring a scalable set of grid components designed to efficiently manage renewable energy infrastructure, including distributed energy resources, storage, and load demand. In addition to this holistic approach, they also provide UPS integration for large-scale applications such as data centers for critical healthcare facilities.
A leading innovator in battery technology, Saft is on the cutting edge of research, advanced manufacturing techniques, and battery design. The company specializes in advanced technology battery solutions for industry, in space, at sea, in the air, and on land in remote and harsh environments from the Arctic Circle to the Sahara Desert. Saft batteries are widely used throughout the health sector in buildings, equipment, and small medical devices. Saft offers an unrivaled portfolio for medical applications and hospital buildings, including a full range of nickel and lithium technology batteries. Saft SBLE, SBM, and SBH ranges of block batteries provide optimum solutions where power is required for an uninterruptible power supply, switching and transmission, emergency and security systems, industrial fire monitors and alarms, process control installations, substation switchgears, and signaling systems. The SBLE blocks are designed to be a reliable source of energy over relatively long discharge periods with current discharges that are relatively low compared to total stored energy or when discharges are infrequent. The SBM blocks sustain electrical loads for between 30 minutes to three hours while providing mixed loads involving a mixture of high and low discharge rates. Saft’s SBH blocks discharge relatively high current over short periods—usually less than 30 minutes of duration.
Not every renewable energy source is natural in origin. Philadelphia’s transit authority (SEPTA) benefits from Saft’s ability to tap the renewable energy source provided by the braking wheels of its commuter trains. Using a technology similar to that used by hybrid vehicles to recapture the energy of braking, their Intensium Max 20P containerized lithium-ion battery recovers up to 10% of SEPTA’s energy costs. It is the world’s first system to capture the kinetic energy of braking trains and store it in a battery system for later use. The storage system, in conjunction with Viridity Energy’s optimized energy recapture system, reduces the grid energy SEPTA consumes by 1,500 MWh, reduces annual costs by $170,000, and provides power insurance against emergency outages. The five train stations that use this system manage 400 stopping trains each workday. This rate is enough to generate over 1,200 MWh annually. Indirect benefits include reduction in greenhouse gases, NOx emissions, and mercury emissions.
As the US military makes a deeper commitment to the use of alternate fuels and renewable energy, Saft has developed a UPS battery system specifically for these needs. At Fort Hunter Liggett in Monterey County, CA, Saft has installed a Supervisory Controller for PV and storage microgrids. This project demonstrates the feasibility and practicality of a system that employs day-ahead optimization and real-time controls to coordinate charging and discharging for a 1-MWh electric storage system servicing a 2-MW photovoltaic generation facility. The system performs energy shifting, storing surplus PV energy for later use. Release energy from the storage system can perform load leveling and reduce the costs of tariff demand charges. It also performs PV integration and mitigates the harmful effects of islanding on the grid. Once a proven success, this system will be expanded to other bases in California.
Falcon Electric Inc. is a leading manufacturer of online UPS systems and backup power for critical applications, including the recent addition of their Lithium Iron Phosphate (LiFePO4) battery option to its award-winning SSG Industrial UPS family. Falcon’s SSG UPS with a hot-swap LiFePO4battery represents a major technology advance with many user benefits. The online rackmount UPS models (1.5 to 6kVA) provide a longer service life, longer backup times, lower weight, and higher safety than UPS with lead-acid batteries or other LiFePO4 batteries. In addition, the SSG UPS lowers total cost of ownership by dramatically reducing costly battery replacements and downtime. Their primary advantages include: long service life (10 years compared to only 4 years for lead-acid batteries), long run times between recharges, a smaller physical footprint with almost half the weight of lead-acid batteries, a wide range of operating temperatures, improved safety, and overall lower lifetime costs.