One of the most commonly offered-up reasons for not incorporating higher levels of renewable energy (RE) into our power system is this: Renewable energy is intermittent and unreliable, subject to seasonality—sunshine, wind speed, rainfall—and therefore can’t replace fossil fuels completely. Even relatively more predictable RE sources like hydropower, some argue, are susceptible to drought.
Well, Costa Rica has done it: over 99% of its electricity is generated from RE (check out this awesome TED Talk). And with the rapidly declining costs of RE, especially for technologies like wind and solar, integrating more RE into the energy mix is becoming increasingly economically attractive.
I recently attended the Asian Clean Energy Forum (ACEF), an annual conference hosted by the Asian Development Bank which brings together top players in the clean energy field, including policymakers, donors, academics, and private sector developers. This year, discussions centered around how to integrate higher levels of RE without compromising stability of the power grid, and how to procure RE more efficiently and competitively.
Balancing the Grid
The cost of utility-scale solar photovoltaics (PV)—the technology that converts sunlight into electricity—has fallen by 73% between 2010 and 2017, while for wind power, the cost reduction has been almost 25%. Speaking at ACEF, the Vice President of JinkoSolar, a Chinese solar developer, claimed that the latest solar prices in China now match those of local coal. These cost declines make the economic case for switching more of our electricity generation away from fossil fuels and towards wind and solar.
However, unlike biomass and hydropower (or fossil fuels like natural gas and coal) which can generate electricity on demand—a characteristic known as ‘dispatchable’—solar power can only be produced during the daytime (on sunny days) and wind power can only be produced on windy days. This intermittency threatens the stability of the power grid because in order to keep running, the grid must maintain a certain frequency: around 60 Hz. If the frequency drops too low or goes too high, it can cause power outages.
To illustrate frequency, this article uses a clever analogy of a person riding a bike, where frequency is the rate of pedaling, and the demand and supply of electricity determine the slope of the hill the bike is on. If demand exceeds supply, the slope gets steeper and more power is required to pedal uphill at the same rate; conversely, if supply exceeds demand, the slope becomes a sharp downhill and the biker pedaling at the same rate can go crashing to the bottom. Large amounts of intermittent power, e.g. from solar, connected to the grid can cause a great degree of uncertainty and threaten the stability of the grid’s frequency. Soaring solar production during the day would flood the grid (sending the biker speeding downhill) and drop off by sunset (steep uphill).
How can we take advantage of free and clean renewable resources while making sure the lights don’t go out?
Taking on more variable renewable energy (VRE) without threatening stability requires increasing grid flexibility. Since production of wind and solar depend not only on the time of day but also uncertain weather conditions, the power system needs to be able to respond quickly to changes in output from those sources.
One solution is to improve existing power plants’ ramping abilities; that is, the ability to be turned up or down, as well as the speed at which it can be done. This requires upgrading power plants, like upgrading a stove from simply having an on/off switch to one that has a dial for adjusting the heat level. When solar or wind power output is high, existing fossil fuel-based power plants can respond by ramping down production.
Let’s look at a case study of India: according to a representative from India’s largest power utility, coal-based generation—which accounts for 58% of India’s energy mix—can be made more flexible to accommodate higher levels of solar coming online. Below is a crude illustration I put together with my rudimentary Power Point skills.
Under the conventional model (left), coal-fired plants would run at a constant output level throughout the day. In the absence of storage (which I discuss later), solar generation in excess of demand is curtailed, resulting in significant wasted capacity. However, if coal power plants are cycled such that they turn down generation during peak solar production hours (right), less solar capacity is wasted and green house gas emissions are reduced.
Another solution for reducing intermittency of VRE is hybridization. Last month, India finalized its wind-solar hybrid policy to promote not only the construction of new combined wind and solar plants, but also additional installations to existing wind or solar farms.
Thailand’s recent attempt to circumvent the intermittency issue was last year’s Small Power Producers (SPP) Hybrid-Firm auction which—though the name misleads—allowed but did not require participating projects to be hybrids. It did, however, require projects to supply firm power, defined as guaranteed capacity during certain periods. In Thailand this meant that, once in operation, each plant would deliver 100% of the contracted capacity during peak hours and 65% during off-peak hours. But not only did the policy do little to encourage the development of hybrid projects (most winning bids went to single-technology biomass plants), it would also potentially only increase stability and reliability at the individual plant level, rather than strengthen the power system as a whole.
The biggest buzz in the RE field recently has been around storage. Combining RE with storage is a form of hybridization—one that could potentially solve the problem of intermittency in wind and solar technologies. One of the winning projects in the Thai auction, and the only one to utilize storage, combines a 100kW solar farm with 130kWh batteries, allowing it to procure firm power. In Chile’s latest auction, a Spanish developer won a bid to supply solar PV power at night, presumably by combining PV with batteries. Chile’s power auction is divided into three distinct time-blocks in a 24-hour period, so by bidding for the night shift, this developer was able to circumvent competition from other solar developers who bid for the daytime block.
The biggest barrier to utilizing storage, however, is cost. Batteries remain expensive, but one would expect that over time the cost will become increasingly competitive, as has been the case with solar and wind technologies. Another obstacle is storage capacity. A recent episode on NPR’s Planet Money podcast explains the problem faced by California, which is generating a lot of solar power with no means of storing it all for night-time consumption. Lithium batteries alone can’t provide enough storage capacity. The solution, the podcast concludes, may lie in pumped storage: solar power is used to pump water from a lower reservoir to an elevated reservoir. When electricity is needed, water from the upper reservoir is released through a turbine, generating electricity, and back into the lower reservoir.
This large-scale pumped storage could help balance output not only of the individual plant, but potentially at the system level. Of course, as with the construction of any dam or reservoir, the impact on aquatic biodiversity and the surrounding ecosystems is significant. The question is whether or not the benefit from displacing fossil fuel-generated electricity with clean renewable energy will outweigh the ecological cost, or whether there is a way to employ existing hydro facilities to store solar and/or wind power. These are questions that no doubt will continue to be debated at length.