Too much renewable energy.

It’s not a problem you would expect many people to complain about today — especially in Hawaii, the only U.S. state with a 100% renewables mandate.

But for some utility-scale renewable developers in the Aloha State, too much wind and solar — especially on the distribution system — is a growing issue. Faced with fast-growing rooftop solar and increased economy-wide efficiency, many large renewable facilities are being forced to turn their generation down to meet demand — a practice known as curtailment.

For the Kaheawa Wind Project, a 51 MW wind installation on island of Maui, curtailment is a fact of daily life. The facility is actually two separate projects — one installed in 2006 and the other in 2012. For KWP II, the latter 21 MW installation, the amount of power not accepted by Maui’s utility can equal amore than a quarter of its annual potential output.

“Last year’s curtailment was on the order of approaching 30% for KWP II,” Steven Rymsha, the plant manager, told Utility Dive during a visit to Maui in March. “That statistic is probably slightly high. It's in a 25-30% range on a yearly basis.”

While a 30% curtailment rate is the highest of the three wind facilities in Maui — more on why later — the issue is one that’s familiar to both developer and utility alike.

Last year, Maui Electric (MECO) curtailed 28.5 GWh of wind energy, accounting for more than 9% of the potential wind generation from three utility-scale facilities — KWP II, the 30 MW Kaheawa I wind farm, and the 21 MW Auwahi Wind project.

Curtailment was higher in 2015 than in 2014, MECO said in an April filing with state regulators, when the utility lost 20.6 GWh of power to the wind. But it was significantly less than the 43 GWh curtailed in 2013, amounting to more than 13% of the island’s potential wind generation.

The issue isn’t unique to MECO. As a whole, the Hawaiian Electric Company curtailed more than 2% of its potential curtailable generation in the last quarter of 2015. Each organized market on the mainland has also deployed the practice to some extent, especially for utility-scale wind:

Curtailment is often viewed as an unfortunate waste of usable energy — especially on the mainland, where renewable penetration remains significantly lower than in Hawaii. The situation on Maui, however, shows the issue isn’t black and white.

While a practice utilities and developers seek to minimize, curtailment is essential to the functioning of Maui's grid today and its hopes of reaching 100% renewables by 2045. The island's experience with curtailment offers some important lessons for mainland utilities as their renewable energy portfolios expand.

Rooftop solar prompts curtailment

When First Wind first decided to expand the Kaheawa project, the plan wasn’t for a quarter of its potential generation to go unused.

The facility sits on a ridge in the West Maui Mountains overlooking the sea — a site chosen for its abundant wind resource. Blessed with steady trade winds, the wind farm’s capacity factor hovers between 40% and 45%. A 10 MW (20 MWh) battery provides ramp rate control for the facility and other grid services to MECO. The plant is operated by SunEdison which assumed control when it bought First Wind in 2012.

While the development team always knew the KWP II would face some level of curtailment, it never suspected it would be this high. What they didn’t account for was the explosion of rooftop solar, Rymsha said.

Virtually negligible 10 years ago, high power prices and decreasing solar costs have spurred a wave of rooftop installations. Today, Hawaiian Electric has approved rooftop applications for more than 17% of its customers, a figure that easily overshadows the much-lower penetrations of rooftop solar experienced by mainland utilities.

“What's occurred since the PPA was signed is the rapid growth of photovoltaics due to low cost rooftop PPAs [and] tax credits in place,” Rymsha said. “So, it's really had an impact that wasn't envisioned when this contract was originally negotiated, but we continue to provide services to support the integration of all renewables on the island.”

Historically, most of the excess energy challenges on Maui happened during the early morning hours, when power use was low, he said. As the island woke up, the wind energy would be uncurtailed to serve rising demand.

The rooftop solar boom changed that pattern, creating more excess energy challenges. While the utility can manage generation from large scale plants like KWP, it has no control over the rooftop solar on its system. Rather than acting as a new source of generation, the distributed solar power simply shows up as decreased load on the utility grid.

“As solar starts increasing, the load levels off and then starts decaying,” Rymsha said. "So you're going to see curtailment of wind energy increasing until such time that the solar resources start decreasing."

As a result, load on Maui is actually lower at noon than it is at 4 p.m. The peak in Hawaii comes in the evening, creating a load curve that many in the state power sector say resembles the Loch Ness Monster:

Even when the KWP II project is curtailed, its battery continues offering frequency control and other services to MECO to help reduce the need for fossil generation. But despite persistent curtailment, it doesn’t store unused wind generation for later use.

Plant staff members “always try to keep the batteries at a certain percentage of charge so we can inject and absorb power, so there's no real load shifting in the battery,” Rymsha said.

“[The main concern is] how do we optimize its state of charge to provide grid services to Maui Electric when there are high winds and all projects are operating,” he said. “If we're within normal frequency range, we'll be managing ramp rate [of the wind farm], and outside normal frequency range, Maui Electric has the opportunity to pulse the generator and ask for power to be delivered to the grid.”

Seniority is everything

While the two Kaheawa wind facilities are situated right next to each other, the difference in their curtailment rates is immense. In 2015, KWP I had just 350.78 MWh of generation curtailed, compared with 25,685.809 MWh of curtailment for KWP II.

The difference all depends on when they were built, said Mat McNeff, power supply manager at MECO.

While generation in organized markets is typically dispatched according to price, that’s not the case with Maui Electric’s large-scale renewables. Instead, the facilities are dispatched in order of seniority, with the oldest plants providing generation first. Newer facilities, by contrast, are the first ones to be curtailed. Since KWP I was deployed in 2006, six years before its sister facility, it is curtailed less often.

“Economic curtailment is not allowed for our PPAs, so the curtailment order is reverse chronological order by the PUC approval date, typically,” McNeff told Utility Dive during a visit to the MECO control room. “It's kind of like, whoever got here first.”

The curtailment order aims to ensure that older, more expensive facilities can continue to sell power as renewable energy prices decline. If economic curtailment was in place, “you would continually rob those earlier facilities,” McNeff said, “and what that eventually will do is cause no one to invest.”

While the scheme often means the lowest-price generation facilities end up selling the least power, curtailment has resulted in some cost savings for island residents as of late.

In 2015, as oil prices dipped, curtailment of wind generation actually saved MECO money.

While monthly prices for fossil fuel generation—mostly from fuel oil and diesel—never dropped below $0.20/kWh in 2014, they averaged just over $0.13/kWh last year. By contrast, the cost of curtailed wind energy averaged nearly $0.16/kWh, meaning that curtailment actually saved MECO $769,000 in 2015, according to the utility’s filing. That trend has continued into 2016, with MECO estimating it saved $629,000 by curtailing wind in the first three months of the year alone.

But those savings are likely incidental and temporary. In their recent Power Supply Improvement Plan, Hawaiian Electric officials estimate that fuel oil costs will rebound to prices well above where they were in 2014, making even Maui’s most expensive wind generation cheaper than the fossil plants. Even if Hawaiian Electric succeeds in its push to shift from fuel oil to liquefied natural gas for its fossil plants, curtailment could start costing the utility money again soon, as it did before 2015.

Curtailment for the grid’s sake

Even as curtailment began to save MECO money over the past year, its grid operators have been working to reduce the amount of wind curtailed on the system.

There have been a lot of improvements on Maui Electric's side to reduce operating reserves," Rymsha said.

Beyond simply looking at reserves and unit minimums, McNeff said MECO has been investigating more innovative ways to reduce curtailment.

Manually managing the fluctuations in electricity supply and demand from Hawaii’s variable renewables and its glut of uncontrollable rooftop solar would be “beyond the capabilities of a typical operator — pretty much beyond any operator,” McNeff said. So the utility teamed up with its generation control provider to devise a program to “automate the curtailment based on the different rules that we’ve programmed in.”

“With Alstom, we built a system where it takes inputs on what the potential output is from all these facilities and incorporates that as it changes the curtailment signal for other facilities as the resources at other facilities move, and it does that every four seconds,” McNeff said. “So that's an innovative thing that we came up with.”

But while MECO works to minimize curtailment, McNeff said it will remain part of utility operations in the foreseeable future. And that’s not a bad thing.

Without any curtailment or a massive deployment of energy storage, the maximum amount of renewables capacity that a grid could conceivably handle would be the utility's minimum load, he explained. Otherwise, a particularly sunny or windy day could push renewables generation above electricity demand, necessitating either storage or curtailment.

The same logic applies to utility systems across the mainland, though they have the option of exporting renewable generation to surrounding areas.

Non-curtailment, McNeff said, would ignore a huge opportunity for more renewables on the system. Peak demand on electricity grids across the U.S. is regularly over 50% higher than the minimum demand, and in places like Maui and California the discrepancy is approaching 100%. Without storage or curtailment, fossil resources would have to make up that demand, which won’t work for a state with a 100% renewables mandate.

While renewable energy developers know they won’t be able to sell power every hour of every day, they are willing to take that risk in return for the opportunity to sell more electricity over the minimum load, McNeff said. The result is more clean energy not being utilized during times of low demand, but also more of it being ready for when load increases.

“With that being said, after a facility is installed, the general public and some developers point to this as a sign of waste,” McNeff said.

Alternatives to curtailment

On the mainland, curtailment of renewable resources is not a pressing issue. Virtually only the Southwest—especially southern California—faces the prospect of widespread renewables curtailment in the next few years due to a proliferation of distributed energy.

For the California grid operator, the answer is simple: Expand the market for renewables. The need to export excess solar and wind has been a driving force behind the push for the Energy Imbalance Market in the West, which seeks to ease power flows between the region’s numerous balancing authorities.  

But Hawaii does not have that option. The state lacks power interconnections between its many islands, let alone the mainland U.S.

Besides curtailment, that leaves one option: energy storage. But McNeff said that while developers have all the reason in the world to invest in ways to store their renewable generation, they have yet to do so on Maui.

“Every one of the developers has had the opportunity to install their own battery and capture that energy that we couldn't take and discharge it later when we could,” he said. “But that's never proven economical.”

Even as storage prices continue to decline, the MECO grid manager said it remains cheaper for many developers to simply not sell a portion of their power than to invest in a battery and store it.

“To a certain extent, depending on the percentage, curtailment is the cheapest option,” he said. “It's kind of like if you have a mango tree, you can get all the bottom mangoes sitting under it, and buy a ladder for the middle ones, but you need a bucket truck for the top 10%. It's probably not worth it.”

The need to find strategies and technologies to minimize curtailment is not lost on the state’s energy leaders. At the Maui Energy Conference in March, Commissioner Mike Champley spoke to the need to add more flexible resources as the island approaches higher levels of renewables.

“Once you've hit 100% from a capacity perspective, the laws of physics take over,” he said. “You have to real-time match supply and demand, and if your capacity is 100% and you add capacity beyond that, it's going to lead to curtailment immediately or some consideration of storage or load shedding.”

As the state pushes toward 2045, it should shift from “dumb renewables” to “smart renewables,” Champley said, such as distributed solar-plus-storage facilities with advanced controls that allow them to be aggregated and managed by the utility.

“The technology matters,” Champley added. “The more we rely on technologies with lower capacity factors, the more installed capacity we will need to get us to 100% ... And once you get above 100% of installed renewable capacity, a lot of other considerations come into play.”

Back at the control room, McNeff echoed Champley’s sentiment.

“One of the attributes that will be highly desired in future generation resources will be flexibility, perhaps maybe at the expense of efficiency or other things that used to be king,” he said. “If anything it's shown that you're never quite sure what the future holds and it's best to invest in something that's adaptable.”