It seems everyone wants a clean and green energy grid these days, and the reasons are all good ones; be it cutting emissions or reducing demand for expensive and scarce foreign resources. Unfortunately there is a problem with this dream of powering our world solely on wind and solar electricity; we use power 24 hours a day, 7 days a week. The sun isn’t always shining and the wind isn’t always blowing, it’s called intermittency, and today I am going to provide a quick tutorial on it.

So what are we supposed to do on a calm summer night it August, when its 90 degrees out, and we all have our air conditioners cranked up?

The common sense solution to this problem is to just use batteries or other methods of storage to keep wind and solar energy on standby for when we need it. However, this just isn’t feasible because our present energy storage technology isn’t cost effective enough to implement a system like this. Just think about how short the battery life is on your brand new Smartphone or Tablet PC, and how expensive electric cars like the Nissan Leaf or Chevy Volt are, and how limited their range. If we haven’t cleared these battery hurdles yet, how can we deploy a vast battery network to backup solar and wind power for the entire country? (There are some promising developments in battery and energy storage technologies out there, but that’s a topic for another blog post).

If we don’t yet have the technology to store it, is it really that big of a deal that it’s not generating 100% of the time, can’t we get around this somehow?

Well, unfortunately it is a big deal. The local energy grid (operated by PJM) requires a regulation of electricity, which pretty much means the supply and demand must always be the same. It’s an incredibly complicated process to maintain these grids. I have had the opportunity to sit at a trading desk of an energy company and watch as they ramp up and down generation from different power plants at the request of PJM operators who are dispatching the needed load around the grid. It’s quite something to see in person.

Then if supply and demand must always be the same, what about solar powered buildings?

Solar systems use a somewhat novel way to get a building to a $0 electric bill; it’s called net metering. It works like this; at any given point the solar system is not producing what the building needs, but has been designed to generate enough over the entire year to meet its annual demand. Sometimes the system will need to buy power from the grid, other times it is selling excess power back into the grid, but it is netting out to zero over the year.This can work for an individual home or business, as they are a blip in the radar for the overall grid, but dispatching gigawatts (home electric use is measured in kilowatts, a gigawatt is a million kilowatts) of solar to power the entire grid, you’d have nothing to net against when the sun isn’t available. Net metering typically is not used as much for wind power, since large wind farms often dispatch power directly to the energy grid like any other power plant.

Even if wind and solar aren't generating power all the time, don't we know when they usually do, and isn't it around the same time when we need it most?

Not only are wind and solar not producing power 100% of the time, when they are producing electricity they just do not match the pattern of demand we have. You’ll notice the image in Figure 1 shows the average demand each hour (broken down by season) in the PJM Mid-Atlantic Region Grid (in red), as well as average production for both solar (yellow) and wind (blue) from those periods (based on the wind and solar systems at the ACUA). Common sense would say solar production should be a close match to our demand, since we use lots of power to run factories or cool our homes during the day when the sun is out, but while solar power peaks around mid day, our use is at its highest in late afternoon, around 5 and 6 pm.

If solar, which is fairly predictable, doesn’t match our use, what about wind power?

On Figure 1 wind appears relatively constant, even if it doesn’t completely match our demand curve. However, Figure 1 is the only average production curve for the month; Figure 2 demonstrates what wind generation really looks like on a daily basis. The first chart in Figure 2 shows wind generation every day in January 2010, each colored line represents a different day in the month. As you can see there is a lot of variability in every hour of every day. Even if the average (heavy black line) is fairly stable, each day has significant peaks and valleys, making wind very difficult to count on for reliable demand over the course of a day. The second chart in Figure 2, which looks a lot like some kind of modern art, shows every day for an entire quarter (Jan – Mar, 2010), and as you can see, the production is so varied there is hardly any blank space on the chart at all.

If that didn’t get too technical, what should be apparent is that wind and solar power just don’t supply power that can reliably meet our demand presently. This is not to say the sources don’t have their uses; as mentioned, on an individual building scale solar power with net metering can be a great solution. And while wind resources can be incredibly intermittent, offshore wind resources are far more constant, thus the push for wind farms in the Atlantic Ocean. However, when someone is trying to tell you we could and should be powered only by clean energy resources today, they need their head examined.

As the old saying goes “If we can put a man on the moon…” then I think we can certainly develop batteries or other storage technologies to solve this problem. In the mean time these obstacles should not stop us from continuing to install wind and solar, but they have to be complementary to “baseline” load for now. So no matter how much we want to be 100% clean and green, there will, and should, be a push for sources such as nuclear and natural gas to provide power until have reached the point when we can rely solely on the clean and green wind and solar power in the future.