This group brings together the best thinkers on energy and climate. Join us for smart, insightful posts and conversations about where the energy industry is and where it is going.


Impacts of Variable, Intermittent Power on Grids

Willem Post's picture
President Willem Post Energy Consuling

Willem Post, BSME'63 New Jersey Institute of Technology, MSME'66 Rensselaer Polytechnic Institute, MBA'75, University of Connecticut. P.E. Connecticut. Consulting Engineer and Project Manager....

  • Member since 2018
  • 613 items added with 219,544 views
  • Nov 11, 2010

The increased generation of variable, intermittent wind power will present a challenge to the management of the power on the New England Electric Grid, NEEG. Because wind power is variable and intermittent, spinning reserves (nuclear plants, regulation velocity less than 1%/min, thermal plants, 1%/min; combined cycle gas turbines, 2.5%/min; simple cycle gas turbines, 4%/min, hydro plants with or without pumped storage, 100%/min) need to be kept in operation to quickly supply power as needed and when wind power is absent during no wind and too much wind conditions. Accordingly, the CO2 reductions due to wind power will be less than forecast by wind proponents, etc. Note: other renewables, such as solar power, are not mentioned because at current build-out rates they are likely to remain only a minor fraction of wind power.


The recent Eastern Wind Integration and Transmission Study, EWITS, states that the impacts on grids from wind power short-term variability and intermittency can be reduced by interconnecting all US grids and building wind turbines everywhere in windy areas. The thinking is that wind is likely to blow somewhere and a certain quantity, at most about 20% of the installed wind turbine capacity, of smoother wind power will be available to consumers at all times. That means at least 80% would continue to be mostly from existing sources. 


EWITS also states it is possible for the US to get 20% of its power from wind by 2030, and that wind power from the sparsely populated, windy Northern Plain states can be brought to the densely populated East Coast by high voltage direct current, HVDC, transmission facilities at a cost of about $90 billion (2009$) by 2030. 


A University of Delaware study regarding future East Coast offshore wind power recommends many offshore wind farms be built along the East Coast from Maine to North Carolina, about 1,500 miles, and that they be interconnected. The geographical area needs to be at least 1,500 miles long to cover about 2-3 weather patterns. 


This arrangement will have results similar to EWITS, however it would be closer to the East Coast population and avoid the above $90 billion transmission facilities. Offshore wind farms cost about 50% more per kW installed than land-based wind farms and require more expensive maintenance. See below.


The North Sea Countries Offshore Grid Initiative proposes a similar grid for the nations around the North Sea. The grid may also be connected to the hydro plants in Scandinavia to smooth the wind power and allow more power storage. 


Integrating wind power already is a challenge in Texas, about 4% wind consumption and Germany, about 6.5% wind consumption, because they lack sufficient hydro plants to smooth their wind power. The grid management challenge of wind power has been successfully met by Denmark, Spain and Portugal during the past 25 years. Denmark (about 20% PRODUCTION from wind, 9% is consumed in Denmark and 11% exported) uses hydro plants in Norway and Sweden to smooth its wind power. Spain, its grid weakly connected to nearby grids, about 13.7% from wind, uses its own hydro plants, some with pumped storage, and CCGT plants to smooth its wind power. Portugal, about 15% from wind, largely mimics Spain.




Spinning reserves are necessary for maintaining voltage on the grid. They typically operate at about 50% of rated capacity otherwise they cannot modulate effectively. They contribute only a small percentage of power to the grid. Standby and peaking plants, that are turned on and off, usually contribute larger percentages. Spinning plants respond to quick changes, such as occur from a sudden plant outage or reduction in its output, or a sudden change in variable, intermittent wind power.


With increased penetration of wind power, grid operators found that increased CO2 producing spinning reserve capacity is required. In fact, that capacity needs to be much larger with wind power penetrations of about 5% and up. 


Changes of PV solar power are less sudden than wind power. However, in Germany, with rapidly changing cloud cover, changes in PV solar power is becoming a problem for grid operators. They already have a problem dealing with 6.5% wind power on their grid. The rapid rate of PV solar system installation in 2009 and 2010 to beat FIT reduction deadlines contributed to the problem. 


Changes of CSP with storage is much less sudden than PV solar. CSP, feasible in places with high levels of insolation, such as Southern Spain, the US Southwest and Northern Africa, is likely to remain only a small part of total power production.




Denmark has huge wind power potential. It started to develop it after the oil shock of 1973. In 1996, Denmark, Norway, Sweden, and Finland created Nord Pool, which trades in and manages power flow between these nations. The main sources of power are hydro (56.9%), nuclear (21.9%), coal (6.3%), biofuel (5.1%) and wind (2.6%, mostly Danish); only about 13% is from fossil fuels. As the generating modes differ and are distributed differently in the various nations, the need for power will vary from nation to nation and at different times. Nord Pool helps to optimize the use of available power and reduce local deficits. Electricity prices would be higher if all the Nordic nations had to build enough generating capacity to be individually self-supporting. 


Denmark has about 5,500 wind turbines (about 89% are from VESTA), total capacity about 3,125 MW; this capacity has not changed by more than 1% since 2004. All wind turbines are controlled from a single center. Denmark has two electric grids: West grid (about 4,300 wind turbines, capacity 2,430 MW, output 5.6 TWh/yr) and East grid (about 1,200 wind turbines, capacity 695 MW, output 1.6 TWh/yr). They are not interconnected. The West grid has robust connections to Norway, Sweden and Germany. The East grid has robust connections to Sweden and Germany. 


As a result of Denmark’s early start in wind power, VESTA has become the No. 1 turbine supplier in the world with about 19.8% of the world market; GE is No. 2 with 18.6%. VESTA has about 4,900 wind turbines with a total capacity of 2,434 MW in Denmark. It has about 39,000 wind turbines worldwide with a total capacity of 35,400 MW. It installs one turbine every 3 hours around the clock, as does GE. 


Wind Power, Production, Consumption and Exports 


Denmark’s 5-yr average wind power PRODUCTION is about 19-21% of its total production; wind varies year-to-year. Denmark’s 5-yr average wind power CONSUMPTION is about 9% of its total consumption. After 30 years of rebuilding its two electric grids and using nationwide electric demand/supply management (smart meters, smart appliances, load control switches), Denmark’s grids are capable of accommodating about 10% of variable, intermittent wind power. During windy periods and when electric demand decreases in Denmark, etc., selected wind farms are idled, as part of electric supply/demand management. Any production beyond about 9%, and production due to future increases of Denmark’s wind capacity, currently mostly offshore, are/will be exported. Denmark’s production cannot rise quickly because modifications to the grids of Germany, Sweden and Norway would need to occur in tandem requiring major coordination and “horse trading” to move forward. 


Graphs of the daily power supply profiles and the daily production and exports of wind power for both grids show that more than 50% of all wind power is exported to the grids of Norway (total production 137 TWh/yr of which 27,528 MW of hydro plants provide 98% = 135 TWh/yr) and Sweden (total production = 135 TWh/yr of which nuclear plants provide 47% = 63.5 TWh/yr and 16,209 MW of hydro plants provide 44% = 59.4 TWh/yr) and Germany (total production = 606 TWh/yr of which mostly coal-fired thermal plants provide 62% = 375.7 TWh/yr and nuclear plants provide 28% = 169.7 TWh/yr). 


The only reason Denmark’s high level of wind power production “works” is because robust connections exist to LARGE nearby grids that are willing to cooperate (by modulating the outputs of their hydro plants and pumped storage) and because the exported wind power is mostly sold about 5-10% below spot prices; i.e., a mutually beneficial arrangement. 


However, the spot prices for wind are below Danish production costs, i.e., Danish households are subsidizing wind power exports which has contributed to Denmark having the highest RESIDENTIAL electric rates in Europe (energy $0.15/kWh + fees, taxes, transmission 0.19/kWh = 0.34/kWh, about double the price in the UK and about triple the price in France which gets about 80% of its power from its load-following nuclear plants and most of the rest from hydro. France has one of the lowest residential electric rates in Europe. The Danish COMMERCIAL rate is kept at about 1/3 of the residential rate for international competitive reasons; an illegal trade subsidy? 


Future Wind Power Plans 


Denmark has a population of 5.5 million with about 2.5 million households connected to district heating loops. Denmark has about 550 distributed small (coal, gas, biomass) combined heat power, CHP, plants, a.k.a. cogeneration plants. Instead of exporting all of the excess wind power, it has been proposed to use some of it for heating HTHW loops of the district heating systems, i.e., a form of thermal storage. Denmark’s announced goal of 50% of its electricity PRODUCTION from wind by 2025 means that nearly all of it will be exported and/or used for augmenting hydro power in Sweden and Norway, for heating HTHW loops (proposed), and for charging hybrid/all-electric vehicle batteries (far into the future). 


If it took Denmark, the paragon of energy efficiency in Europe, 30 years to accommodate about 10% of variable power with help from Norway and Sweden, how will all this play out within the NEEG? Where is the mutually beneficial arrangement? Is Hydro-Quebec, HQ, needed for “smoothing” variable power? Will NEEG-wide supply/demand management systems be needed? 


IAEA data for 2004, 2005; Danish Annual Energy Statistics 2007; Danish Energy Authority October 2008.   




Spain and Portugal have huge wind power potential. In November, 2009, a major weather front passed over Spain and Portugal, almost all of their wind turbines were spinning, and Spain’s wind power production was more than 50% of Spain’s normal consumption for about 5 hours. Spain’s electricity consumers never noticed this power surge, because it was smoothed by increased exports, by increased pumping to fill up hydro power reservoirs and by ramping down hydro power and using quick-starting combined-cycle gas turbine, CCGT, plants.


The Spanish grid (capacity 93,000 MW of which 19,149 MW {at end 2009} is wind; production 282.1 TWh in 2008, of which about 31.4 TWh, or about 11.1%, was from wind) has little connected to nearby grids, as do Portugal, Ireland and the UK. Spain has about 18,000 MW of hydro plants of which 3,272 MW are with pumped storage. Hydro power production varies from 30 TWh/yr to 40 TWh/yr, depending on rain fall and pumped storage. At least 2,500 MW of hydro plants with pumped storage are under construction or planned. 


IBERDROLA is a utility company that has a worldwide electric generating capacity of 43,925 MW. It is Spain’s second largest electric utility. Its capacity in Spain is 26,700 MW, including 5,130 MW wind and 8,800 MW hydro of which 2,300 MW is pumped storage. It has the mix of power plants that enables it to play a major role in smoothing variable, intermittent wind and solar power.


In Spain (and Portugal and Denmark), hydro power plants is the preferred option to smooth wind power because:

–  they are the most flexible of the technologies in performing continuous startups and shutdowns without a significant detrimental effect on the equipment’s service life. 

–  their load variation speed is high. For example, it is possible to vary the power by about 100%/min. 

–  their minimum load is low, often less than 10% of rated capacity. 

–  their fuel cost is zero. 

–  they do not produce CO2.




The German grid (capacity 133,000 MW of which about 25,777 MW {at end 2009} is wind; production 611.9 TWh in 2008, of which about 37.2 TWh, or 6.4%, was from wind, about 22 TWh/yr, or 3.4%, was from hydro, and about 4.5 TWh, or 0.7%, was from PV solar) is strongly connected to nearby grids. Germany has insufficient hydro power to smooth its wind power.




The Texas electric grid (capacity 105,000 MW of which about 9,410 MW {at end 2009} is wind; consumption 404.8 TWh in 2008, of which about 14.2 TWh, or 3.5%, was from wind) is not connected to nearby grids. Texas has hydro plants that produces about 1.0 TWh/yr, insufficient to smooth its wind power. For comparison: US 2008 consumption 4,119.4 TWh, New England 130 TWh, Vermont 6.0 TWh. 


At 3 p.m. on February 26, 2008, wind power was supplying a little more than 5% of Texas’ 40,000 MW demand at that time. But over the course of the next 3.5 hours, an unforecast/unexpected wind lull caused wind power to fall from 2,000 MW to 350 MW, just as evening demand was peaking. Grid operators declared an emergency and blacked out 1,100 MW of load in a successful attempt to avoid a system collapse. The Texas power system narrowly avoided a breakdown by rapidly shedding loads and increasing the output of its spinning reserve and standby power plants. 


According to the Electric Reliability Council of Texas, ERCOT, this was not the first or even the worst such incident in ERCOT’s area. Of 82 alerts in 2007, 27 were strongly correlated to a drop in wind. To minimize future emergencies, ERCOT is planning to connect to nearby grids to utilize their spinning reserve and standby power plants.  If Texas has so many alerts with 3.5% wind, what will be the alerts with 20% wind, as envisioned by EWITS and the US Department of Energy by 2030?




The BPA, a non-profit, was created in 1937 to deliver and sell power from hydro power plants on the Columbia River. Over the decades, 31 hydro plants and one nuclear plant, total capacity about 10,500 MW, and a transmission system were built covering the states of Washington, Montana, Oregon and Idaho. The system is interconnected with the Western Interconnection which covers the  Western United States, Alberta, British Columbia, and small portions of Mexico. Starting in 2000, total wind capacity connected to the BPA system was about 150 MW, will be about 2,100 MW (22 wind farms) by end 2009, about 6,500 MW by end 2013, about 10,000 MW by end 2016. 


The BPA reserves parts of its hydro power system to back up wind in case unscheduled wind power up/down ramps occur unexpectedly. Historically, the BPA has used the Federal hydro system to provide reserves for all variability that occurs within its transmission network, but wind has presented unprecedented variability. The PBA is considering adding pumped storage to its hydro power system and CO2-producing gas-fired gas turbine-generators to provide more flexibility for integrating future wind power. This follows the way taken by Denmark and Spain. If a total of 1,000 mW of pumped storage and gas turbines is added, the cost will be about $2 billion. Will this be paid for by the BPA, the utilities it sells to (i.e., ratepayers), or the wind farms?


In March, 2009, there was an instantaneous peak of 1,733 MW from wind; almost all of the wind turbines were spinning at about 83% of rated output. Graphs of wind power and total demand show wind power usually varies from 0 MW to 1,000 MW and usually is about 0% -10% of total demand. Water is quickly diverted from and to hydro plant turbines to deal with the variability of incoming wind power. The impact on fish limits some operations. The diverted water could have produced CO2-free hydro power at about $0.025/kWh, well below the cost of wind power. The BPA is building additional weather stations to better predict and anticipate wind power variability. Wind integration costs vary from about $2/MWh-$9/MWh nationwide, in the Northwest about $5.5/MWh. Currently, the BPA is charging about $5.7/MWh to integrate wind power.’-wind-power-reaches-milestone-on-bonneville-power-administration-system.html




The UK (England, Wales and Scotland) and Ireland have huge wind power potential. Development of even a small percentage of it would overwhelm their electric grids with variable, intermittent wind power. Robust inter-connections with the European mainland are needed to spread this power over much larger grids. The grid may also be connected to the hydroelectric plants in Scandinavia to allow more power storage. The grid is expected to cost some 30 billion Euros. Europe aims to have 20% of its electricity from renewables by 2020.


On December 7, 2009, energy ministers from the UK, Germany, France, Belgium, the Netherlands, Luxembourg, Denmark, Sweden and Ireland signed an agreement to develop the world’s first large-scale offshore wind energy grid in the North and Irish Seas, providing a boost to Europe’s fast-expanding offshore wind industry.


The North Seas Countries’ Offshore Grid Initiative requires member nations to cooperate on the development of a new offshore energy grid that would allow energy generated by offshore wind farms to be transmitted between North Sea nations. The ministers said the initial aim of the initiative is to develop a strategic work plan in early 2010 that would coordinate offshore infrastructure development. This plan would then be formally enshrined in a Memorandum of Understanding to be signed later in 2010. 


Europe has about 28 offshore wind farms in operation, about 1,684 MW (only two of the 28 farms have GE wind turbines, the rest have European wind turbines, mostly from Vestas and Siemens), another 17 farms, about 2,792 MW, mostly less than 200 MW each, will be operating by 2011 and an additional 14 farms, mostly greater than 500 mW each, are proposed; the US has zero offshore wind farms in operation. The experience gained by Europeans will be useful all over the world, including the US. If the US does not quickly get going, the train will have left the station, as it did for land-based wind power.




The NEEG, managed by ISO New England, capacity about 34,020 MW, power supplied about 130,000 GWh/yr, includes over 350 central power plants and 8,000 miles of high-voltage transmission lines to provide power to about 6.5 million customers. The NEEG power is 62% from CO2-producing fossil fuels, 26% from CO2-free nuclear, 6% from CO2-free hydro, 4% from CO2-producing wood waste, 2% from CO2-producing solid waste and 1% other (i.e., CO2 -free wind, solar, etc.). Almost all of this power is STEADY power and the NEEG is designed accordingly. Note: The reason the 5 New England nuclear plants, a total of 4,486 MW, produce so much electricity is because their average CF is about 0.92, much higher than of most plants on the grid. 


The VPIRG “Repowering Vermont” report, proposed renewable power to be 15.4% solar and 27.4% wind, for a total of 42.8%  = 3,595 GWh/yr by 2032. Based on Denmark’s and Spain’s 25-year experience with wind power, that quantity of variable, intermittent power cannot be fed into the Vermont grid, unless an economical way is found to store it, such as heating HTHW loops, charging hybrid/all-electric vehicle batteries, using compressed air energy storage, CAES, and augmenting hydro power with pumped water storage, none of which are feasible in Vermont at present. That means Vermont has to feed the subsidized, expensively produced wind power and subsidized, very expensively produced solar power into the NEEG and likely sell it at about 5-10% below spot prices, just as the Danes do. 


Each New England state has its own renewable power targets. For example: Maine, with huge wind power potential and better winds than Vermont, has 5 operating wind farms, totaling 174 MW, plans to have 2,000 MW by 2015 and 3,000 MW by 2020. Maine will need to feed almost all of this power into the NEEG or the Hydro-Quebec grid. Massachusetts’ Cape Cod, also with huge wind power potential and better winds than Vermont, will need to do the same. Their wind power will be more competitive than of Vermont, New Hampshire and Rhode Island. Many years of major coordination and “horse trading” would be needed to move forward.


To sum up:

–  at present the NEEG is not designed for large variable, intermittent power inputs

–  there are few HTHW loops for thermal storage in the NEEG grid area

–  using CAES may not be feasible in the NEEG area 

–  charging batteries of hybrid/all-electric vehicles and NEEG-wide supply/demand management systems are many years hence 

–  smoothing NEEG wind power using hydro power plants will be limited because hydro power is only 6% of the NEEG power production 




There may be a way forward similar to the one found by Denmark, Portugal and Spain about 15 years ago when their wind power production exceeded the stability limits of their grids. They smoothed their wind power; Denmark by exporting to the hydro power systems of Norway and Sweden; Portugal and Spain by building more hydro power and pumped storage capacity and CCGT plants within their borders. 


The US Northeast and Eastern Canada have significant CO2-free hydro power and could have a lot more. With modifications, including adding pumped storage, the Hydro-Quebec and Maine hydro plants will be of great value for storage and smoothing of wind power in the US Northeast and Eastern Canada, especially at greater percent wind penetration in the future, say more than 5%. Store during periods of high wind power production and during low nighttime demands and distribute during the daytime. The power from regional wind farms needs to be aggregated and part of it (as much as the grid can take) consumed and the rest sent via high voltage direct current, HVDC, lines (DC has less line loss) to hydro plants for storage and smoothing, and then redistributed as required. 


Hydro-Quebec, HQ, in Canada, has 59 hydro power plants and one nuclear plant, total capacity 36,429 MW, serves 3.9 million customers who enjoy among the lowest power rates in North America. HQ’s present and planned hydro capacity may be large enough to smooth the power of the US Northeast and Eastern Canada. Significant expansion of HVDC systems will be required. A 750 mile, 2,000 MW HVDC line to New England was completed in 1992. 


The US Northeast grid is currently arranged mostly for fossil and nuclear plants. 

Wind power is coming. About 10,000 MW was installed in the US in 2009. It is the only CO2-free renewable that is near-competitive with fossil. The grid needs to be rearranged for wind power. 


HQ and Vermont signed a 328 MW power supply agreement in 1990 that expires in 2015; negotiations are being held to renew it by 2010.

HQ exported 21,300 GWh to neighboring Canadian states and the US in 2008.

HQ signed an agreement with New Brunswick on October 29, 2009 enabling HQ to double its power sales to New England by 2011. 

HQ aims to smooth the power of 4,000 MW of Canadian wind farms by 2015.ébec



Paul O's picture
Paul O on Nov 13, 2010

Thanks Willem for providing this highly informative post, I have often Googled for some kind of documentation of the opperational realities for wind power in places where it has been used for a long time.

Some thoughts:

1) Obviously a lot of work needs to  be done to make Wind Power a reasonable alternative, even in those countries where there has been long term acceptance and adoption of Wind Power,

2) It would make sense to me to require that Wind Farm supplier and operators must build and operate storage facilities equivalent to say 10-20% of their generating capacity along side their Wind Farms.


To me, my reluctance for reliance on mother nature for our energy future seems even more reasonable.


Willem Post's picture
Thank Willem for the Post!
Energy Central contributors share their experience and insights for the benefit of other Members (like you). Please show them your appreciation by leaving a comment, 'liking' this post, or following this Member.
More posts from this member

Get Published - Build a Following

The Energy Central Power Industry Network is based on one core idea - power industry professionals helping each other and advancing the industry by sharing and learning from each other.

If you have an experience or insight to share or have learned something from a conference or seminar, your peers and colleagues on Energy Central want to hear about it. It's also easy to share a link to an article you've liked or an industry resource that you think would be helpful.

                 Learn more about posting on Energy Central »