EVs charge mostly at night. In fact they can be set on a timer to charge after midnight when rates are cheaper. In many areas this provides an excellent match to wind power. During the high demand summer months, California offers a great example of this with wind picking up as the sun goes down.
Consider Texas, the most wind rich state in the country. They are building out wind rapidly. However perhaps not as much as they would like. The problem is that wind in Texas is night peaking but the load from air conditioning peaks in the daytime. Wind is not a good match to the Texas load as it stands. But if EVs were ubiquitous they could charge at night enabling Texas to better take advantage of their wind resource while reducing carbon emissions from autos and from the coal and natural gas power plants that would otherwise be required to charge them.
The graphs below are typical of summer load and generation profiles in Texas and illustrates the point well.
While the graphs above show the resource for a day, the graph below shows the behavior for the entire year.
The subscription based educational portal Solar Learning Lab provides access to a variety of solar data sites from around the country. The screenshot below shows data from an Archer City Texas solar array. The match with wind above is quite good. If Texas pursues solar to augment wind they will have an excellent complementary resource.
Solar resource in Archer City Texas.
It is important to note that the fit is not perfect. Around 7:00pm, solar has dropped off considerably while wind has not yet ramped. This can be addressed by use of tracking arrays. In the Southwest they tend to be cost effective on an energy basis and in this case would have the added benefit of making the solar / wind pairing eligible for a much higher capacity credit. In other words, the tracking arrays would reduce or possibly eliminate the reliance on peakers for the 1/2 hour transition from solar to wind. Also, due to the very short transition period, this could also be addressed with demand side management. The technology for utility control of load such as hot water heaters and air conditioners is mature. Businesses could save by reducing AC from 6:30pm to 8:00pm and homes could similarly reduce air conditioning in a tighter band to address this as well.
In the graphs below, a fixed and tracking array are compared. These two solar arrays are at Rensselaer Polytechnic Institute. They are the same size and use the same inverters. One is on a fixed mount and the other on a tracking mount. Late in the day there is harvest of electricity during the transition period which otherwise would be underserved by renewable resources. This is high value and further justifies the expenditure for electricity.
Graph showing improved ability of the PV array to match load as it ramps to peak in the late afternoon.
Similar profiles are observed in California. While it is true that the wind resource in late fall and winter is low in California, so to is their load. In the summer when there is the highest demand on the grid, wind power is very reliable and available through the night. In fact, if you look at the data, it turns out that wind power is very predictable in California.
The term intermittent can mean many things. The general understanding is that intermittent means unpredictable or random. However, the term is often applied even in the case of predictable renewable resources such as wind in TX or solar in the Southwest. In each case, suggesting that the resource is random is naive at best. See graph below of solar production from a CA based solar array. Even on the rainy day the array produces considerable energy. Of course on a rainy day, temps and therefore AC demand would be lower. Generally , clouds and rain are predictable in CA. The combination of advanced notice and load reduction minimize and possibly eliminate the grid stability ramifications that may be existent for high penetration of solar in other areas.
Energy production from a 600kW solar array in Hemet CA. Energy production is consistent and predictable.