Critics of renewable energy often cite the fact that technologies like wind and solar only produce energy when the wind is blowing or the sun is shining. They argue that we can’t effectively utilize renewable energy until appropriate energy storage technology is developed. While the fact that wind and solar don’t produce energy around the clock is certainly a major disadvantage, I find that the problems associated with the intermittent nature of many renewables are often exaggerated, and rarely discussed from a practical perspective. With this post, I’ll introduce a few of the main challenges posed by intermittent energy sources, and then discuss three possible solutions.

The Challenges of Intermittent Renewable Energy

The difficulty associated with integrating variable sources of electricity stems from the fact that the power grid was designed around the concept of large, controllable electric generators. Today, the grid operator uses a three-phase planning process to ensure power plants produce the right amount of electricity at the right time to consistently and reliably meet electric demand. Because the grid has very little storage capacity, the balance between electricity supply and demand must be maintained at all times to avoid a blackout or other cascading problem.

Intermittent renewables are challenging because they disrupt the conventional methods for planning the daily operation of the electric grid. Their power fluctuates over multiple time horizons, forcing the grid operator to adjust its day-ahead, hour-ahead, and real-time operating procedures.

Take the example of solar panels. Solar energy is inherently only available during daylight hours, so the grid operator must adjust the day-ahead plan to include generators that can quickly adjust their power output to compensate for the rise and fall in solar generation. Furthermore, power plants that typically produce electricity all day every day might instead be asked to turn off during the middle of the day so that the energy produced from solar can be used in lieu of fossil electricity.

In addition to daily fluctuations caused by sunrise and sunset, the output from solar panels can also change suddenly due to clouds. Variability caused by clouds can make it more difficult for the grid operator to predict how much additional electric generation will be required during the next hour of the day, so it becomes difficult to calculate exactly what the output of each generator should be to accomplish the load-following phase identified in the first graphic above.

Fast fluctuations in output from wind or solar energy don’t only disrupt the hourly load-following phase of grid planning, but also the second-to-second balance between total electric supply and demand. Today, the grid operator sends a signal to power plants approximately every four seconds to ensure the total amount of power injected into the grid consistently equals the total power withdrawn. Because wind and solar increase the magnitude of sudden power generation shortfalls or excesses, the grid operator requires more reserve power ready to respond at a moment’s notice to ensure the grid remains balanced.

While renewables disrupt the grid’s operation in a number of ways, it is not impossible to compensate for the additional intermittency and uncertainty. In fact, many of the strategies to overcome renewable variability are simpler than you might realize. The following sections review strategies that can be used to integrate renewable energy without the need for costly energy storage.

The Law of Large Numbers

While at first glance it might sound like adding too much renewable energy could destabilize the delicate balance of the electric grid, it turns out that renewable energy actually becomes more predictable as the number of renewable generators connected to the grid increases thanks to the effect of geographic diversity and the Law of Large Numbers.

The Law of Large Numbers is a probability theorem, which states that the aggregate result of a large number of uncertain processes becomes more predictable as the total number of processes increases. Applied to renewable energy, the Law of Large Numbers dictates that the combined output of every wind turbine and solar panel connected to the grid is far less volatile than the output of an individual generator.

Because the grid operator is only concerned with balancing the total amount of renewable generation with the rest of the grid, the Law of Large Numbers causes the amount of reserve capacity required to balance renewables with the grid on a second-by-second basis to be a lot less than intuition suggests. In a study commissioned by the Electric Reliability Council of Texas, General Electric calculated how much new reserve capacity will be required as Texas increases the amount of wind energy installed. The report found that an additional 15,000 megawatts of installed wind energy only requires an additional 18 megawatts of new flexible reserve capacity to maintain the stability of the grid. In other words, the spare capacity of one fast-ramping natural gas power plant can compensate for the variability introduced by 5,000 new average-sized wind turbines.

The Power of Prediction

While the law of large numbers and the effect of geographic diversity causes renewable energy to smooth out its own fluctuations on a second-by-second basis, it can still be difficult to predict the expected level of renewable generation during the next hour or two of the day.

Fortunately, experience has shown that it is possible to effectively model and predict the aggregate renewable power available to the grid. Both wind and solar depend on natural systems that can be modeled and forecasted with reasonable accuracy. Today, wind energy makes up over 10 percent of Texas’s annual electricity supply, thanks in part to effective wind generation forecasts. This is especially significant because Texas has a unique isolated grid, with no way to access extra conventional electricity generation from outside the state.

In recent years, an entire industry has emerged around the practice of interpreting data from the web and other sources for the purposes of targeted advertising, political advocacy, and a variety of other practices. Many of the algorithms and strategies developed to predict whether or not you’ll click on a web ad could also be used to predict the expected output of a wind or solar panel.

In the future, I believe there will be a greater need for effective renewable energy prediction, and that increasingly advanced models and algorithms to predict renewable output either an hour or day ahead of time will meet this need.

Incentivizing Energy Production at the Right Time and Place

While it’s possible to manage second-to-second and hour-to-hour fluctuations in renewable energy output through aggregation and prediction, predicting how much renewable energy will be available a day ahead of time is significantly more difficult.

Integrating a large share of intermittent renewable energy into our daily electricity operations will require a mix of sources that complement each other to roughly equal our total energy demand over the day. This is technically possible because continental wind energy tends to peak at night, coastal wind energy tends to peak during the day, and solar can peak at various times over the day, depending on which way it is oriented.

Accomplishing this mix will require an efficient and effective electricity market that incentivizes electricity generation at the right time and place. Existing competitive electricity markets already have prices that vary over the day and over a region depending on the local level of electricity supply and demand. Exposing renewable energy to these prices can help encourage a mix of renewable sources that produces just the right amount of energy when we need it, and reduces the need for costly energy storage.

A Sustainable Electric Grid of the Future

While the challenges posed by the intermittent nature of many renewable energy sources certainly increase the complexity of effectively operating the grid, they are far from insurmountable. In many ways, they pale in comparison to the enormous challenges that were overcome to initially string all the wires, build all the power plants, and implement all of the controls that make up the present grid. Minimizing the costs associated with renewable variability will be a major challenge of the coming years and decades. I don’t doubt that the solutions that emerge will surprise us in more ways than one.


Image Credits: DOE/EERE, CAISO, Perez et al., 2009, NCAR, Flickr user jedavillabali.