|Home||Energy Physics||Nuclear Power||Electricity||Climate Change||Lighting Control||Contacts||Links|
ENERGY STORAGE SYSTEMS:
An energy storage system absorbs energy from the electricity grid at times when the uncontrolled demand for electricity is low (and hence energy is relatively inexpensive) and/or releases energy to the electricity grid at times when the uncontrolled demand for electricity is high (and hence energy is relatively expensive). Usually an energy storage system is only financially viable when there is an electricity valuing mechanism such as a combination of peak power and energy billing that adequately recognizes the benefits to the electricity system of high capacity factor generators and high load factor customers.
Energy storage systems are characterized in the file titled:
Energy Storage Primer.
In the absence of sufficient energy storage, a combination of generation constraint and load constraint is required to match generation to load. Both forms of constraint increase the average cost of electricity for most consumers.
Problems common to all renewable energy forms are the intermittency and seasonality, and insufficient capacity for seasonal storage of energy when it is plentiful for later use when it is scarce. Energy storage for a few hours is economically possible but in most jurisdictions energy storage for weeks or months is both inefficient and prohibitively expensive.
Energy storage using lithium batteries has been highly developed for automotive propulsion. However, this form of energy storage is generally too expensive for seasonal energy storage applications.
A mountainous region blessed with consistent rainfall and a low average population density, such as British Columbia, Quebec or Norway, can rely on hydraulic energy storage if the population is willing to build and maintain the required large hydraulic reservoirs. Adjacent regions with intermittent solar and wind generation may be able to access this hydraulic energy storage capacity by integrating their electricity grids with the region containing the hydraulic energy storage. However,there are significant constraints on hydraulic balancing of intermittent wind and solar generation imposed by dam storage capacity, high and low limits on downstream river flow and impacts on fisheries and indigenous populations.
A related major issue is thatriver water that is used for hydraulic electricity production is river water that is not available for agricultural irrigation or for recharging depleted aquifers. The amount of river water required per kWh for hydraulic electricity production is far greater than the amount of river water required per kWh for nuclear electricity production. As fresh water aquifers are depleted the resulting increased requirement for river water for agricultural irrigation will reduce the amount of river water that is available for hydraulic electricity generation and will reduce the hydraulic power available for balancing wind and solar electricity generation.
In theory the river water shortage could be mitigated by use of pumped storage, but the practical political problems attendant to large scale pumped storage systems are immense. From a geographical perspective the obvious location in North America for implementation of large scale pumped storage is between Lake Ontario and Lake Erie. However, there are so many conflicting governmental and water front property interests around these two great lakes and on the Niagara River that realizing a cost effective pumped storage implementation agreement is thought to be impossible. Pursuit of this project would likely require a change to the US constitution.
An alternative means of energy storage is electrolysis of water to make hydrogen. The hydrogen can be stored as a compressed gas in deep underground salt caves or can be compounded with toluene in a reversible chemical reaction for storage as a liquid. The hydrogen can subsequently be used for meeting the peak heating and/or electricity load. However, this process is only about 60% efficient on heat recovery and 25% to 35% efficient on electricity recovery. Hence the stored energy is relatively expensive. However, in the absence of sufficient seasonal hydraulic energy storage hydrogen storage is better than nothing. Stored hydrogen is suitable as a supplementary peak winter heating fuel and as a vehicle fuel.
In urban areas with high winter peak heating loads it is usually more cost efficient to deliver the stored energy to the load via a hydrogen gas pipeline than to deliver the stored energy to the load via the electricity distribution system. However, in rural areas gas pipelines are often uneconomic.
A relatively new form of energy storage is crushed rock thermal energy storage with hot oil or nitrate salt as a heat transfer fluid, as described in a paper titled 100-Gigawatt-Hour Crushed-Rock Heat Storage for CSP and Nuclear and the related slide set. This thermal energy storage system type can operate at up to 400 degrees C using oil as the heat transfer fluid and up to 600 degrees C using nitrate salt as the heat transfer fluid and can store thermally stratified heat directly discharged from a nuclear reactor for subsequent use for electricity generation. The initial cost of such an energy storage system is anticipated to be comparable to the initial cost of a large crushed rock hydroelectric dam. The sites of such energy storage systems will be practically constrained by nuclear reactor siting criteria, suitable geography and local availability of suitable igneous rock. There are non-trivial issues related to reliable containment of the heat transfer fluid. The high temperature version can only be used with rocks that are chemically nitrate salt compatible.
ENERGY STORAGE FINANCIAL CONSTRAINT:
An example of an unappreciated problem with energy storage is a plan to use stored high temperature heat and to oversize the turbine generator portion of a nuclear power plant in order to arbitrage electricity price spreads between day and night in the wholesale electricity market. The plan proponents have failed to understand that the electricity market price is not static. Both the day and night price are affected by supply/demand. If a large nuclear power plant (NPP) stores power at night and supplies power during the day that NPP will raise the nighttime price and lower the daytime price. The day/night price spread will totally disappear if the NPP size is large enough compared to the market. The expected additional cash flow to help pay for the energy storage portion of the NPP will disappear and the project will lose money. That market price dynamic has already been tested by a 2,000 MW merchant hydroelectric storage facility on the US east coast that went bankrupt trying to arbitrage the price of electricity.
Day/night energy storage price arbitrage only makes financial sense for generators or loads that are so small compared to the electricity market that they do not significantly affect the market price.
ENERGY STORAGE SYSTEM TYPES:
Each type of energy storage system has a characteristic cycle time for which it is most cost effective for storing energy. The viable energy storage system types for multi-day or longer energy storage are gravitational hydraulic and synthetic liquid fuel. The viable energy storage system types for daily energy storage are gravitational hydraulic, thermal, electrochemical and compressed gas. The viable energy storage system types for short term energy storage for control stabilization and power factor correction are flywheel kinetic, electrostatic and electromagnetic.
Note that synthetic liquid fuel and thermal energy storage systems are most efficient when their inputs and/or outputs are respectively hydrogen and/or heat rather than electricity. These two energy storage system types can be used to shift energy from times of high supply to times of high demand.
MATCHING GENERATION TO LOAD:
In any electricity system, in order to regulate the system voltage, it is necessary to continuously match total instantaneous power supplied to the grid to the total instantaneous power thermally dissipated or drawn from the grid. Available nuclear and renewable energy is more efficiently used if the electricity system contains energy storage subsystems that can absorb and/or release energy at controllable rates.
DISTRIBUTED ENERGY STORAGE:
In terms of minimizing the requirement for transmission/distribution the optimum location for energy storage is behind uncontrolled generator and uncontrolled load customer meters. By reducing the metered individual generator output power variations and the metered individual load input power variations voltage regulation is improved, transmission/distribution costs are minimized and the difference between total uncontrolled generation and total uncontrolled load that must be met with dispatched generation and dispatched load is minimized.
APPLICATIONS OF ENERGY STORAGE:
Wind, solar and tidal energy generators at remote locations, if motivated by suitable compensation rates, will use behind the meter energy storage to reduce the amplitude of variations in their net power outputs. Reducing the amplitude of these output power variations reduces transmission capacity requirements, reduces voltage regulation problems and reduces the need for both external energy storage and balancing generation to meet temporary shortfalls in total primary generation.
Transmission connected dispatched energy storage can be used to match variations in total generator output power to variations in total load customer electricity demand. Due to statistical averaging of the outputs of many geographically distributed renewable generators the total generator output power variation is less than the sum of the individual generator random output power variations. Similarly the total load variation is less than the sum of the individual random load variations. This statistical averaging mechanism is known as diversity. If the energy storage is centrally located diversity reduces the amount of energy storage required to match the total generation to the total load. However, use of central energy storage in place of behind the meter energy storage increases the cost of transmission/distribution.
A major problem with wind and solar generation is lack of diversity. At night all the solar generation is off. At low wind times all the wind generation is off. Provision of sufficient energy storage to convert intermittent wind and solar generation into dependable base load generation is extremely expensive.
Behind the load customer meter energy storage can be used to reduce the amplitude of variations in the customer's net power input. Reducing the amplitude of these power variations minimizes transmission/distribution requirements. Another benefit of behind-the-meter energy storage is potential protection of the load customer from short term failures in the external electricity system. A secondary benefit of behind-the-meter energy storage is that it conceals from the electricity utility the actual moment by moment power usage profile, which some customers consider to be private information.
Nuclear reactors are usually operated at a constant thermal power to minimize component thermal stress. Hence, nuclear generation is most economic when operated constantly at full rated power. Electricity systems usually try to minimize the cost of electricity to retail customers by using other generation types and/or energy storage to achieve load following.
In the past use of fossil fuels inherently provided the electricity system with energy storage for load following. As fossil fueled generation is taken out of service due to CO2 induced climate change and increasing cost of fossil fuels, more non-fossil energy storage will be required to efficiently match the total load customer power demand to the available total generation. In the absence of sufficient energy storage, a combination of generation constraint and load constraint is required to match generation to load. Use of generation constraint increases the average cost of electricity. Use of load constraint decreases both the marginal value of electrical energy and the return on investment for the affected industrial installations.
An important technical issue is that energy storage systems should be designed to ride through transient grid faults and not to rely on other generators for black start or to provide voltage regulation.
The metering system and electricity rates should adequately reward uncontrolled generators and uncontrolled load customers for providing energy storage that reduces local power fluctuations and that enhances overall grid voltage stability.
FINANCIALLY ENABLING ENERGY STORAGE:
At this time electricity rates in Ontario do not financially enable energy storage. The present generator compensation rates do not sufficiently encourage storage of non-fossil energy when it is surplus for later use when non-fossil energy is in short supply. The load customer electricity rates do not sufficiently encourage off-peak energy storage at the load to relieve on-peak grid congestion. The Green Energy Act Feed-In Tariffs do not adequately reflect the value to transmission/distribution of high generator capacity factor.
Most of the aforementioned rate failures are caused at least in part by the present use of fossil fuel electricity generation for load following. As long as generators bear no cost related to their CO2 emissions and natural gas remains relatively cheap, natural gas will contine to be used for fueling load following and peaking electricity generation. However, sooner or later an increase in the cost of natural gas will force a significant change in the electricity rate structure.
Energy storage systems will not be built or operated unless there is sufficient on-going benefit to the energy storage system owner to justify the required capital investment, the ongoing operating and maintenance costs and the loss of revenue that the energy storage system owner would otherwise receive for alternative use of the same land or building space. There must also be long term electricity rate certainty.
A load customer's electricity bill is primarily made up of peak demand and energy charges. Use of marginal electricity at peak demand times affects both the peak demand and energy charges. Use of marginal electricity at non-peak demand times affects only the energy charges. Hence the cost of on-peak electricity includes both peak demand and energy charges whereas use of off-peak electricity causes only energy charges.
Any energy storage system relies on the Rate Ratio:
Rate Ratio = (Cost of on-peak electricity / kWh) / (Cost of off-peak electricity / kWh)
being sufficient to fund the storage system energy losses as well as the storage system's capital amortization, operating, maintenance and overhead costs. To financially enable a practical energy storage system this Rate Ratio should be in the range:
2.5 < (Rate Ratio) < 3.5
and the energy storage system efficiency should be greater than 50%.
However, the importance of this Rate Ratio is not presently recognized in the Ontario electricity rate structure. The financial requirements of energy storage systems should be taken into account in any future determination of electricity rates, global adjustments, carbon taxes or generation incentives. Ideally, behind the meter energy storage systems should be used in conjunction with a daily peak demand dependent electricity rate. The energy charges should also be time dependent. The global adjustment should vary with time in direct proportion to the energy charge.
A major issue is that the fossil fuel industry does all that it can, including political corruption, to prevent electricity utiliites offering to consumers low cost interruptible clean electricity for charging energy storage, because such low cost electricity reduces fossil fuel consumption.
Some understanding of the funding issues relating to behind the load customer meter energy storage systems can be gained by considering two simple time dependent electricity rate examples:
Unfavorable Case (Rate Ratio = 2.5, Efficiency = 50%):
Assume the on-peak electricity rate is $.24 / kWh, the off-peak electricity rate is:
($.24 / kWh) / 2.5 = $.096 / kWh
and the energy storage system efficiency is 50%. If the energy storage system charges off-peak and discharges on-peak, the gross income per discharged kWh is:
$.24 / kWh - (($.096 / kWh) / .5) = $.048 / kWh
Favorable Case (Rate Ratio = 3.5, Efficiency = 75%):
Assume the on-peak electricity rate is $.24 / kWh, the off-peak electricity rate is:
($.24 / kWh) / 3.5 = $.06857 / kWh
and the energy storage system efficiency is 75%. If the energy storage system charges off-peak and discharges on-peak, the gross income per discharged kWh is:
$.24 / kWh - (($.06857 / kWh) / .75) = $.14857 / kWh
Note that seemingly minor changes in energy storage efficiency and off-peak electricity rate tripled the amount of money available for funding the energy storage system capital, operating and maintenance costs. Thus,the economic viability of energy storage is very sensitive to both the energy storage system efficiency and the ratio of on-peak to off-peak electricity costs.
The net result of the present electricity rate failure is that there is less energy storage connected to the Ontario electricity system than in the past. During the 1960s Ontario Hydro offered electricity rates with 10 year guarantees that encouraged many developers of large buildings to incorporate energy storage systems into their buildings. By 1983 there were over 20,000,000 square feet of commercial space in Metro Toronto with behind the load customer meter energy storage for electrical load control. These energy storage systems were almost all taken out of service during the period 1983 to 1997 when changes in the electricity rates induced by availability of load following fossil fuel generation diminished the financial benefits to end users of energy storage systems. This author, then acting on behalf of the Urban Development Institute, advised the OEB in 1981 that the then trend in Ontario Hydro electricity rates would lead to these energy storage systems being taken out of service and that is exactly what transpired.
REQUIRED RATE CHANGES:
In Ontario, in order to financially enable energy storage, the Ontario Energy Board must increase the Rate Ratio by increasing the peak demand charge and decreasing the energy charge.
On this web site it is proposed that the blended electricity rate for all uncontrolled load customers should decrease with increasing load factor. Similarly the uncontrolled generator compensation rates should decrease with decreasing generator capacity factor. This proposal financially enables behind-the-meter energy storage while avoiding system instability problems inherent in simple block Time-Of-Use (TOU) electricity rates.
It is of fundamental importance that costs associated with load following be applied to the demand rather than the energy portion of end users electricity bills. An example of these charges is constraint payments such as amounts paid to owners of natural gas fuelled generators when these generators are not actually producing electricity.
Renewable generation in Ontario has a seasonal component. The monthly average wind generation during the winter is about two times the monthly average wind generation during the summer. One potential means of meeting the peak summer electricity load is to use a very large reservoir (Lake Erie-Lake Ontario) hydraulic energy storage system to store surplus energy during the winter for use during the following summer. The alternative to this seasonal energy storage is to build sufficient additional generation with daily energy storage to meet the summer peak load and to also build additional load such as synthetic liquid fuel plants that can be dispatched off during the summer. Neither of these means of meeting the peak summer load is inexpensive. When the total costs of meeting the peak summer load with renewable generation are taken into account, nuclear generation is a bargain.
RIVER FED RESERVOIRS:
River fed hydro-electric reservoirs are a highly efficient and readily controllable means of energy storage. If a river fed reservoir feeds a hydro-electric hydraulic turbine, and if the reservoir has sufficient area and depth to store half of the river flow for 12 hours, the flow through the turbine for 12 hours at night can be set at 0.5 X the daily average river flow and the flow through the turbine for 12 hours during the day can be set at 1.5 X the daily average river flow. This arrangement results in a 3:1 change in the downstream water flow rate and results in an electricity generation rate during the day that is three times the electricity generation rate during the night. This arrangement is a practical way of realizing daily load following generation provided that a daily 3:1 change in the downstream water flow rate is acceptable.
In principle, river fed reservoirs can also be used to balance minute by minute variations in wind generation. However, there are occasional weather conditions that cause the total wind generation to drop very quickly. Under those circumstances the downstream water flow rate from hydro-electric facilities used to balance wind generation increases very quickly, causing dangerous conditions for anyone in, on or close to the downstream river. To mitigate these dangerous conditions it is necessary to constrain the rate of increase of downstream river flow and to partially balance a rapid drop in wind generation by shedding dispatched load or by using another type of quick responding energy storage such as distributed electro-chemical energy storage.
An advantage of river fed reservoirs is that the energy storage efficiency is generally over 90%. There is a small decrease in energy storage efficiency as the storage level decreases.
Construction of river fed reservoirs generally requires low cost land and favourable river valley geography and topography. It is usually economically unfeasible to build such reservoirs in urban areas, so there are usually substantial related transmission costs.
A significant problem with river fed reservoir hydro-electric generation is that it uses fresh water for on-peak electricity generation that might otherwise be used for irrigation of high elevation farm crops. As global warming increases irrigation requirements, the cost of fresh water for on-peak hydro-electric generation will increase.
An example of a large river fed reservoir is the Grand Coulee Dam on the Columbia River in Washington State, USA and its up-river seasonal storage dams in British Columbia. In past years Columbia River hydro-electric power was used to meet US electricity demand peaks as far east as Chicago.
PUMPED HYDRAULIC ENERGY STORAGE:
If the available river fed reservoirs are all fully utilized, then the alternative is pumped hydraulic energy storage. With pumped hydraulic energy storage when there is surplus electricity water is pumped up hill from a low level reservoir to a high level reservoir, storing gravitational potential energy. When there is a deficiency of electricity water runs back downhill from the high level reservoir to the low level reservoir via a hydraulic turbine to generate electricity. A pumped storage system between two nearby water bodies with different elevations can easily store energy for about one day. Pumped hydraulic storage is a good way of efficiently converting nuclear base load power into daily load following power.
Centrally located pumped hydraulic energy storage may be used for absorbing energy surpluses and for meeting energy deficits.
Pumped hydraulic storage between two very large lakes such as Lake Erie and Lake Ontario can potentially store sufficient energy for seasonal balancing of wind generation in all of Ontario and New York State.
Pumped hydraulic energy storage systems are characterized by rapid changes in the water flow rate between the high level reservoir and the low level reservoir. These rapid flow rate changes are inherently dangerous. To ensure safety the public must be completely excluded from the river or canal system interconnecting the two reservoirs.
A disadvantage of pumped hydraulic storage is that, due to its rural location, there is usually an associated major transmission cost. There are also changes in water level that have minor ecological and marine consequences. The daily water level change is similar to the tide change at a sea port.
An example of a pumped hydraulic energy storage system is:
Racoon Mountain in Tennessee. View Racoon Mountain Cross Section
Another example of a pumped hydraulic energy storage system is:
Dinorwig in the UK.
Another example of a pumped hydraulic energy storage system is:
Robert Moses Dam at Niagara Falls in New York.
Another example of a pumped hydraulic energy storage system is:
Bath County Pumped hydraulic power station
Most hydrogen for industrial purposes is presently obtained by reforming natural gas. However, production of hydrogen from natural gas releases large amounts of CO2 to the atmosphere. Also hydrogen obtained by reforming natural gas contains carbon contaminants that shorten the operating life of fuel cells.
Purer hydrogen can be obtained via electrolysis of water. Electrolysis of water requires large amounts of electrical energy. The produced gaseous hydrogen is difficult to store, so an energy storage system based on electrolysis of water to produce hydrogen generally has to include either a methanol (CH3OH) or an ammonia (NH3) production facility or a toluene storage facility to efficiently absorb the hydrogen as it is produced. Large variations in available electric generation and customer load can be balanced by varying the hydrogen storage rates under dispatch control.
The hydrogen produced by electrolysis of water can be combined with biomass to form methanol (CH3OH). Methanol, also known as methyl hydrate, is a room temperature liquid that is easy to store and transport and that can conveniently provide energy when and where required with little environmental impact. Methanol has long been used as a relatively safe fuel for both racing cars and model airplanes. In a refinery methanol can be dehydrated to form liquid hydrocarbons like methylcyclopentane (CH3C5H9), a high energy density liquid fuel similar to gasoline. At times of exceptionally high electricity demand methanol can also be directly used as a fuel for electricity generation.
The hydrogen produced by electrolysis of water can also be combined with nitrogen to form ammonia (NH3), which as a liquid can readily be stored and transported. Ammonia is an essential input for making a wide range of chemical products including nitrogen fertilizers.
The hydrogen produced by electrolysis can be stored by chemically combining the hydrogen with toluene to form tri-methyl cyclane (also known as spera hydrogen) which is a manageable liquid. With the aid of a suitable catalyst this chemical reaction can be reversed to yield hydrogen gas.
ELECTRO-CHEMICAL ENERGY STORAGE (BATTERIES):
In an electro-chemical energy storage system surplus AC electricity is converted into DC and then is stored as chemical energy. When there is a deficiency of electricity the chemical energy is converted to DC and then is converted again into three phase AC using a static inverter.
Electro-chemical energy storage may be used at generation and load sites to provide voltage regulation, reactive power and black start when and where required.
Electro-chemical energy storage has quick response allowing it to be used for balancing rapid changes in wind generation.
An advantage of electro-chemical energy storage is that it is transportable and can be located at both distributed generation sites and at urban load sites to minimize transmission/distribution costs.
Lithium ion batteries are simply too expensive for utility scale energy storage.
The chemical systems that are most suitable for stationary electricity energy storage are the calcium-calcium chloride-antimony liquid metal battery supplied by Ambri, sodium-sulfur-nickel chloride (Na-S-NiCl2) and vanadium. These chemicals have a high DC electrical efficiency (84% - 91%) allowing 65% to 79% of stored electricity to be recovered as AC. The liquid metal batteries are good for many thousands of full charge-deep discharge cycles. The Na-S-NiCl2 energy storage systems are suitable for daily capacity factor or load factor improvement. Vanadium energy storage systems are suitable for daily or weekly generator capacity factor or load factor improvement.
The materials used in construction of liquid metal batteries, sodium-sulfur-nickel chloride and vanadium energy storage systems are common, relatively inexpensive and readily available in Canada. Toronto puts hundreds of tons of sodium chloride on its streets after every snowfall. Canada exports mountains of sulfur obtained from processing metal-sulfide ores and from refining of petroleum. Canada has large reserves of the other required materials: calcium, antimony, iron, nickel, and vanadium.
Liquid metal batteries are in use by an electricity utility in the USA. The supplier is Ambri which is headed by MIT professor Don Sadoway
Sodium-sulfur-nickel chloride electro-chemical energy storage is presently used on a medium scale in Japan, and on a smaller scale in trial installations in the USA, Canada, Ireland, Uk, Switzerland and Italy. The principal suppliers of sodium-sulfur-nickel chloride energy storage systems are NGK Insulators Ltd. in Japan and MES-DEA in Switzerland. The high energy density of sodium-sulfur-nickel chloride energy storage modules allows them to be used for fleet vehicle propulsion. These modules can also be used for daily load factor improvement in applications as small as single family homes and small commercial establishments. Disadvantages of sodium-sulfur-nickel chloride technology are high temperature operation (~300 degrees C), complex charge control and a limited number of charge-discharge cycles (typically 2000) before the energy storage cells must be recycled. A feature of Na-S-NiCl2 energy storage systems is that when installed adjacent to major buildings the waste heat can be recovered for use in domestic hot water and space heating.
A vanadium based electro-chemical energy storage system is manufactured by Prudent Energy in Richmond, British Columbia. This vanadium based system has the advantages that: it operates at room temperature allowing efficient energy storage over long time periods, its charge control is inherently simple and it offers more charge-discharge cycles than are presently available from sodium-sulfur-nickel chloride energy storage systems. The disadvantages of a vanadiaum based energy storage system are that the energy density is low, making this technology unsuitable for vehicle propulsion, and that the energy storage chemicals are liquid (sulfuric acid) rather than solid at room temperature. In this author's practical experience, widespread use of this energy storage technology in residential or small commercial applications would almost certainly result in significant amounts of dilute sulfuric acid being dumped into city sewers.
Electro-chemical energy storage systems may involve sliding tap transformers and 12 pulse inverters to match the three phase line voltage to the charge dependant electro-chemical DC bus voltage while controlling the direction, rate and phase angle of energy exchange. A dedicated controller monitors the amount of chemical energy contained in each cell string and compensates for single cell shorts. A control loop with maximum and minimum power limits sets the direction and rate of energy exchange in proportion to the difference between the RMS line voltage and the software controlled RMS line voltage setpoint. Phase control of the static inverter permits reactive power compensation.
THERMAL ENERGY STORAGE:
In thermal energy storage an electrically powered heating or cooling system is run during electricity off-peak periods to generate and store large tanks full of molten salt, hot water or ice pellets. The stored thermal energy is used for heating or cooling during subsequent electricity on-peak periods. Thus the heating or cooling electricity load is effectively shifted from the electricity on-peak period to the electricity off-peak period. Thermal energy storage has the advantage of relatively high overall efficiency as compared to other forms of daily energy storage.
Thermal energy storage systems are both large and heavy and hence either must be built into a building during its original construction or must be retrofit built on an adjacent property. However, between the retrofit energy storage system location and the existing building there is often a public road that requires a difficult to obtain easement to cross underground with fluid pipes and electrical cables. Municipal public utilities frequently oppose such easements as infractions on their monopoly rights. Such easements are usually only granted to municipal district heating/cooling systems. Mortgage funders also oppose any arrangement whereby the essential services of one property are dependent upon privately owned equipment on another property. Such opposition makes many retrofit thermal energy storage opportunities uneconomic.
A novel way of achieving large scale thermal energy storage is to use a mixture of hot rocks and oil where the oil dribbles through a porus pile of hot rocks. The oil permits heat transfer and storage at a low pressure but at a temperature much higher than can be achieved with water. This technique has the additional safety advantage that in the event of a heat exchanger failure a sodium-oil chemical reaction is much less violent than a sodium-water chemical reaction. However, the oil used must be suitable for reactor to oil heat transfer above 320 degrees C to prevent deposition of NaOH on the hot side of the heat exchange bundle. A variation on this idea is to use a low melting point nitrate salt such as NaNO3 instead of oil as described in 100-GWH CRUSHED ROCK ULTRA-LARGE STORED HEAT (CRUSH) SYSTEM . A leader in hot rock thermal storage is professor Charles Forsberg of MIT.
Another thermal energy storage medium is solar salt, which is a mixture of KNO3 and NaNO3. Solar salt has been successfully used for thermal energy storage and transport in focused solar heating systems. Its melting point is sufficiently low that it is thermally compatible with steam generators and steam turbines. However, there are serious potential chemical compatibility problens between solar salt and liquid sodium.
The simplistic reaction is:
5 Na + NaNO3 --> 3 Na2O + (1/2) N2
The exotherm is:
-111.5 = 3(-99.4) + Hr ;
Hr = -186.7 kcal/mol NaNO3 or -781 kJ
It is slightly higher for the KNO3 which has a smaller heat of formation.
The exotherm is real. How does it propagate and what damage can occur are logical questions. The impulse from the N2 release could cause damage that could contribute to larger leaks and worse outcomes.
5 K + KNO3 = 3 K2O + (1 / 2) N2
KINETIC ENERGY STORAGE:
In kinetic energy storage surplus electricity is stored as kinetic energy in a flywheel. When there is a deficiency of electricity on the grid the kinetic energy in the flywheel is converted back into electricity. Flywheels typically store energy from seconds to minutes. Flywheel energy storage systems are primarily used to dampen voltage oscillations on electricity transmission/distribution systems. Utilities generally include the costs of kinetic energy storage in their transmission/distribution costs. A representative supplier of flywheel energy storage systems is Beacon Power.
ELECTROSTATIC ENERGY STORAGE:
In electrostatic energy storage the surplus energy is stored as an electic field in a device known as a ultra capacitor. Ultra capacitors can be realized using carefully controlled alternating thin layers of metal and barium titanate. Ultra capacitor technology is relatively new. Ultra capacitor use in utility applications is presently aimed primarily at the wind turbine market. This author has safety concerns relating to utility size ultra capacitors. These devices are potentially explosive.
ELECTROMAGNETIC ENERGY STORAGE:
In electromagnetic storage electrical energy is stored in magnetic fields and is released in a controlled manner a few milliseconds later. The primary uses of electromagnetic energy storage are for AC/AC voltage change, for AC/DC/AC static power inversion and for power factor correction. Energy can be efficiently stored for a long time in a magnetic field realized using superconductoring electomagnets, but the capital cost of such energy storage is generally prohibitive for electric utility applications.
ENERGY STORAGE APPLICATIONS:
Run-of-river generators provide electricity generation that tends to be much higher in the spring than in the late summer. Run-of-river generators can easily be constrained for load following. However, the constrained portion of the generator's potential energy output is usually discarded.
Wind generators need nearby energy storage to average their daily outputs. Failure to purchase energy storage along with wind generation has the practical effect of committing Ontario to extra transmission costs and long term large scale use of other generation for balancing of wind generation. Due to the seasonality of wind generation in Ontario wind generation will be most economic when the total grid load is consistently higher in the winter than in the summer. Seasonal shifting of the provincial load profile requires large scale production of synthetic liquid fuels or pumped hydraulic energy storage between Lake Erie and Lake Ontario.
All non-fossil fuel generation types except solar need to shift energy from the daily off-peak period to the daily on-peak period, where the energy has greater value.
Energy storage at wind generator sites serves several important functions:
1. Shifting excess energy from times of energy surplus to times of energy deficiency;
2. Filtering out random short term variations in generator output;
3. Providing generator black start capability and reactive power;
4. Reducing the cost of switchgear required for network protection;
5. Reducing associated transmission capital costs;
6. Proving local voltage regulation.
Energy storage at load sites serves several other additional important functions:
1. Filtering out random short term variations in load power input;
2. Providing emergency backup power;
3. Reducing the high I^2 R thermal losses that occur at times of peak load.
The overall electricity system performance can be further enhanced by use of a "smart grid" in which each energy storage unit has sufficient intelligence and communication network connectivity to recognize times when there is a local energy supply surplus and times when there is a local energy supply deficiency. In response to a local energy supply surplus the energy storage units should absorb electrical energy. In response to a local energy supply deficiency the energy storage units should release electrical energy. This result is achieved by programming the energy storage units to have a negative slope output power versus voltage characteristic. Above nominal voltage the unit charges (absorbs energy). Below nominal voltage the unit discharges (liberates energy). The nominal voltage setting can be adjusted to meet system control requirements and energy storage constraints.
If there is a loss of network communications the energy storage control should default to regulating the net power flow out of each generator and the net power flow into each load customer so as to regulate local voltage and optimize use of local distribution and local energy storage.
This power control methodology is extremely robust and is highly resistant to computer hacking because the distributed generator and distributed load power setpoints are independently controlled by the equipment owners, not the public utility.
ENERGY RATE FOR UNCONSTRAINED WIND GENERATION:
Via a number of 20 year wind power purchases the Ontario Power Authority (OPA) has determined that in 2009 the value of unconstrained on-shore wind generation, with no energy storage, is about:
$.135 / kWh.
At a favorable site wind generation typically has a winter capacity factor of 40%, a summer capacity factor of 20% and an annual capacity factor of 30%.
One kW of such generator peak plate capacity produces its owner a gross revenue of about:
(8766 h / year) X 0.3 X ($.135 / kWh) = $355.02 / kW-year
The OPA has further found that the capital cost of land based wind generation is about $2800 / kW wind generator peak plate capacity, indicating that the maximum simple payback period for wind generator financing is:
($2800 / kW wind generator peak plate capacity) / ($355.02 / kW wind generator peak plate capacity-year) = 7.89 years
The actual payback period for investors is longer due to the costs of site leasing and wind generator maintenance. However, an investment in unconstrained wind generation may make sense to the investor if this investment shelters other income from tax.
ENERGY RATE FOR WIND GENERATION WITH DAILY ENERGY STORAGE:
Assume that about 25% of the electrical energy generated is lost in the daily energy storage system, so on an annual basis 1 kW of wind generator peak plate capacity with daily energy storage potentially exports to the grid:
1 kW X .30 X 8766 h / year X .75 = 1972.35 kWh / year
The maximum allowable cost of financing the daily energy storage system is reasonably estimated to be about $.10 / kWh of filtered net output energy. Hence the maximum allowable capital cost of the daily energy storage system per kW of wind generator peak plate capacity is:
1972.35 kWh / KW-year X $.10 / kWh X 7.89 years= $1556.18 / kW wind generator peak plate capacity
Since the output of the daily energy storage unit inverter is only about 0.5 kW per kW of wind generator peak plate capacity, the capital budget for the energy storage unit can be as much as:
$1556.18 / kW X (1 kW / .5 kW) = $3112.37 / kW of inverter output capacity.
The cost of wind generated energy delivered to the grid from daily energy storage becomes:
($.135 / kWh) / .75 = $.18 / kWh.
After the cost of financing the storage system is included, filtered energy costs:
$.18 / kWh + $.10 / kWh = $.28 / kWh
If due to seasonal generation variation 33% of this generation (half of the average winter generation) has to be constrained off the cost of the remaining generation is:
($.28 / kWh / .67 = $.4179 / kWh
and the average output in the summer is:
(.2 kW X .75) = 0.15 kW / kW wind generator peak plate capacity.
Thus, subject to the aforementioned assumptions relating to the efficiency and cost of daily energy storage, with no seasonal energy storage the average cost of wind generated electricity for feeding a constant load is $.4179 / kWh.
In principle the average cost of wind generated electricity can be slightly reduced if the surplus energy available in the winter can be stored for use during the following summer.
SEASONAL WIND ENERGY STORAGE:
If instead of being constrained off half of the average winter surplus energy is stored at an efficiency of 50%, the total net extra output energy per annum per kW of wind generator peak plate capacity is:
(1972.35 kWh / year-kw wind generator peak plate capacity) X (1 / 6)
= 328.725 kWh / year-kW wind generator peak plate capacity
VALUE OF SEASONALLY STORED WIND ENERGY:
The additional value of this extra seasonally stored filtered energy is:
$.4179 / kWh X 328.725 kWh / year-kW wind generator peak plate capacity
= $137.374 / year-kW wind generator peak plate capacity
The extra capital that can reasonably be committed to store this energy is:
($137.374 / year-kW wind generator peak plate capacity) X 7.89 years
= $1083.88 / kW wind generator peak plate capacity
CAPITAL COST OF SEASONAL STORAGE:
The seasonal storage system must be capable of meeting energy storage shortfalls in the local energy storage systems, suggesting that the seasonal storage output capacity must be 0.2 kW / kW wind generator peak plate capacity, implying a seasonal storage system cost not exceeding:
($1083.88 / kw wind generator peak plate capacity) / (0.2 kW /(kW wind generator peak plate capacity))
=$5419.4 / kW
It has been shown that based on assumptions of a storage efficiency of 50% and a storage cost of $0.10 / recovered kWh the combined inefficiency and cost of energy storage for smoothing wind generator output increases the average cost of wind generated energy from $.135 / kWh to about $.4179 / kWh. Addition of seasonal energy storage enormously improves the certainty of availability of energy in the summer but does not significantly reduce the net average cost of wind energy. Hence the true net cost of base load on-shore wind energy, exclusive of transmission/distribution, is close to $.42 / kWh.
ENERGY STORAGE AND TRANSMISSION:
In the absence of electrical energy storage located adjacent to a generator, the associated transmission must be sized to match the peak generator output. For unconstrained wind generation without daily energy storage (annual capacity factor = 30%) the cost of transmission / kWh-km is about three times the cost of transmission / kWh-km for base load nuclear generation (annual capacity factor = 90%). Hence, there is an economic incentive for locating energy storage close to wind generation to reduce transmission costs.
IMPACT OF ENERGY STORAGE ON TRANSMISSION COST:
It is generally accepted that the capital cost of rural dual circuit 500 kVAC transmission is about:
$3,000,000 / km-3000 MW peak = $1.00 / km-kW peak
The annual blended cost of interest, capital amortization and maintenance for the transmission of 1 kW of peak generation without energy storage may be as much as:
0.2 / annum X $1.00 / km-kW = $.20 / kW-km-year
When the average transmission distance becomes of the order of 1000 km, the annual cost of peak transmission becomes:
1000 km X $.20 / kW-km-year = $200.00 / kW-year
If the generator capacity factor is only 30%, this cost must be amortized over:
0.3 X 8766 kWh / kW-year = 2629.8 kWh / kW-year
leading to a transmission cost per kWh of:
($200.00 / kW-year) / (2629.8 kWh / kW-year) = $.076 / kWh
If behind the meter daily energy storage improves the net generator capacity factor from 30% to 60% the transmission cost savings are:
0.5 X $.076 / kWh = $.038 / kWh
This transmission cost saving is significant, but alone is not enough to justify the cost of an energy storage system. Power supply reliability is the dominant issue.
UNCONSTRAINED WIND ENERGY COST SUMMARY:
|STORAGE TYPE||CAPACITY FACTOR||ENERGY COST||TRANSMISSION COST|
|No Storage||30%||$.135 / kWh||$.076 / kWh-1000 km|
|Daily Storage||60%||$.28 / kWh||$038 / kWh-1000 km|
|Seasonal Storage||90%||$.42 / kWh||$.0253 / kWh-1000 km|
The average per kWh wholesale electricity rate must more than double from $.065 / kWh to about $.135 / kWh to financially enable unconstrained wind energy. The average per kWh wholesale electricity rate must about double again to financially enable daily wind energy storage. The average per kWh wholesale electricity rate must increase by a further 50% to also financially enable seasonal wind energy storage. Hence reliance on wind as a source of prime energy in Ontario would cause the cost of electrical energy to increase about six fold. When the cost of transmission is considered the overall electricity price increase might be four fold. However, this price increase is still very dramatic.
It would be much less expensive for Ontario to build nuclear generation instead of wind generation for meeting the electricity base load.
This web page last updated April 10, 2021.
|Home||Energy Physics||Nuclear Power||Electricity||Climate Change||Lighting Control||Contacts||Links|