Electricity generated from wind power can be highly variable at several different timescales: from hour to hour, daily, and seasonally. Annual variation also exists, but is not as significant. Because instantaneous electrical generation and consumption must remain in balance to maintain grid stability, this variability can present substantial challenges to incorporating large amounts of wind power into a grid system. Intermittency and the non-dispatchable nature of wind energy production can raise costs for regulation, incremental operating reserve, and (at high penetration levels) could require an increase in the already existing energy demand management, load shedding, or storage solutions or system interconnection with HVDC cables. At low levels of wind penetration, fluctuations in load and allowance for failure of large generating units requires reserve capacity that can also regulate for variability of wind generation.

A series of detailed modelling studies which looked at the Europe wide adoption of renewable energy and interlinking power grids using HVDC cables, indicates that the entire power usage could come from renewables, with 70% total energy from wind at the same sort of costs or lower than at present. Intermittency would be dealt with, according to this model, by a combination of geographic dispersion to de-link weather system effects, and the ability of HVDC to shift power from windy areas to non-windy areas.

Pumped-storage hydroelectricity or other forms of grid energy storage can store energy developed by high-wind periods and release it when needed. Stored energy increases the economic value of wind energy since it can be shifted to displace higher cost generation during peak demand periods. The potential revenue from this arbitrage can offset the cost and losses of storage; the cost of storage may add 25% to the cost of any wind energy stored, but it is not envisaged that this would apply to a large proportion of wind energy generated. Thus the 2 GW Dinorwig pumped storage plant(figure 2)adds costs to nuclear energy in the UK for which it was built, but not to all the power produced from the 30 or so GW of nuclear plants in the UK.

In particular geographic regions, peak wind speeds may not coincide with peak demand for electrical power. In California and Texas, for example, hot days in summer may have low wind speed and high electrical demand due to air conditioning. Some utilities subsidize the purchase of geothermal heat pumps by their customers, to reduce electricity demand during the summer months by making air conditioning up to 70% more efficient; widespread adoption of this technology would better match electricity demand to wind availability in areas with hot summers and low summer winds. Geothermal heat pumps also allow renewable electricity from wind to displace natural gas and heating oil for central heating during winter, when winds tend to be stronger in many areas. Another option is to interconnect widely dispersed geographic areas with a relatively cheap and efficient HVDC "SuperGrid". In the USA it is estimated that to upgrade the transmission system to take in planned or potential renewables would cost at least $60 billion. Total annual US power consumption in 2006 was 4 thousand billion kilowatt hours. Over an asset life of 40 years and low cost utility investment grade funding, the cost of $60 billion investment would be about 5% p.a. ie $3 billion p.a. Dividing by total power used gives an increased unit cost of around $3,000,000,000 x 100 / 4,000 x 1 exp9 = 0.075 cent / kWh.

According to a 2007 Stanford University study published in the Journal of Applied Meteorology and Climatology, interconnecting ten or more wind farms allows 33 to 47% of the total energy produced to be used as reliable, baseload electric power, as long as minimum criteria are met for wind speed and turbine height.

In the UK, demand for electricity is higher in winter than in summer, and so are wind speeds. Solar power tends to be complementary to wind. On daily to weekly timescales, high pressure areas tend to bring clear skies and low surface winds, whereas low pressure areas tend to be windier and cloudier. On seasonal timescales, solar energy typically peaks in summer, whereas in many areas wind energy is lower in summer and higher in winter. Thus the intermittencies of wind and solar power tend to cancel each other somewhat. A demonstration project at the Massachusetts Maritime Academy shows the effect. The Institute for Solar Energy Supply Technology of the University of Kassel pilot-tested a combined power plant linking solar, wind, biogas and hydrostorage to provide load-following power around the clock, entirely from renewable sources.

A report from Denmark noted that their wind power network was without power for 54 days during 2002. Wind power advocates argue that these periods of low wind can be dealt with by simply restarting existing power stations that have been held in readiness or interlinking with HVDC. The cost of keeping a power station idle is in fact quite low, since the main cost of running a power station is the fuel (see spark spread and dark spread).

Penetration

Wind energy "penetration" refers to the fraction of energy produced by wind compared with the total available generation capacity. There is no generally accepted "maximum" level of wind penetration. The limit for a particular grid will depend on the existing generating plants, pricing mechanisms, capacity for storage or demand management, and other factors. An interconnected electricity grid will already include reserve generating and transmission capacity to allow for equipment failures; this reserve capacity can also serve to regulate for the varying power generation by wind plants. Studies have indicated that 20% of the total electrical energy consumption may be incorporated with minimal difficulty. These studies have been for locations with geographically dispersed wind farms, some degree of dispatchable energy, or hydropower with storage capacity, demand management, and interconnection to a large grid area export of electricity when needed. Beyond this level, there are few technical limits, but the economic implications become more significant.

However In evidence to the House of Lords Economic Affairs Select Committee, the UK System Operator, National Grid have quoted estimates of balancing costs for 40% wind and these lie in the range £500-1000M per annum. "These balancing costs represent an additional £6 to £12 per annum on average consumer electricity bill of around £390."

At present, few grid systems have penetration of wind energy above 5%: Denmark (values over 18%), Spain and Portugal (values over 9%), Germany and the Republic of Ireland (values over 6%). The Danish grid is heavily interconnected to the European electrical grid, and it has solved grid management problems by exporting almost half of its wind power to Norway. The correlation between electricity export and wind power production is very strong.

Denmark has active plans to increase the percentage of power generated to over 50%.

A study commissioned by the state of Minnesota considered penetration of up to 25%, and concluded that integration issues would be manageable and have incremental costs of less than one-half cent ($0.0045) per kWh.

ESB National Grid, Ireland's electric utility, in a 2004 study that, concluded that to meet the renewable energy targets set by the EU in 2001 would "increase electricity generation costs by a modest 15%"

A recent report by Sinclair Merz saw no difficulty in accommodating 50% of total power delivered in the UK at modest cost increases.

Predictability

Related to variability is the short-term (hourly or daily) predictability of wind plant output. Like other electricity sources, wind energy must be "scheduled". The nature of this energy source makes it inherently variable. Wind power forecasting methods are used, but predictability of wind plant output remains low for short-term operation.