Building a Smart Grid

Roadblocks on the Path to Hawaii’s Energy Future

April, 2010

High on a ridge overlooking Maalaea Bay, a small group of students from the Horizons Academy scramble out of vans into the vast open space at the top of the Kaheawa wind farm. They gape for a moment in the brilliant morning light. It’s an impressive sight: the giant white turbines of Kaheawa – 20 in all – standing majestically along the ridge that slopes to the sea, as astonishing as the heads of Easter Island.

“Cool,” says one of the young students.

Noe Kalipi, a spokeswoman for First Wind, the company that built the wind farm, tells the students that each turbine is 168 feet tall. Taken together, the 20 turbines have a capacity of 30 megawatts, or more than 10 percent of Maui’s peak load. That combination of majesty and power capacity makes the Kaheawa wind farm a symbol of the state’s rush to meet the goals of the Hawaii Clean Energy Initiative: 70 percent clean energy and 40 percent renewables by 2030.

Cool, indeed.

But Kaheawa poses as many questions as it answers. How do you integrate variable electricity generation from wind farms and photovoltaic systems into the electrical grid without compromising reliability? How do we pay the enormous cost of modernizing the grid to accommodate these renewables? And how do we monitor and regulate these changes to make sure the grid is ready for the 21st century? Most experts believe that the answer to these questions lies in a collection of new technologies and practices known collectively as the “smart grid.” In fact, Maui Electric Co. (a division of Hawaiian Electric Industries) has joined the Hawaii Renewable Energy Institute, the U.S. Department of Energy, General Electric and a few other partners in a smart grid pilot project in Wailea. But it’s not clear that MECO (or the rest of the state) is fully ready for the smart grid.

How “Smart” Works

There are many definitions of a smart grid. “No two people can agree on what that means,” says Robbie Alm, senior vice president of HECO. But for the purposes of MECO and what they want out of the Wailea pilot project, the smart grid is all about communication. In some cases, that communication empowers consumers with smart meters and advanced metering infrastructure, which let utility customers monitor their energy use in real time. In combination with time-of-use rates, smart meters may reduce peak load on the grid by encouraging consumers to move some of their consumption to off-peak times, when rates are cheaper.

Probably more important is to have more data about the grid itself. “For example,” says Chris Reynolds, MECO’s superintendent of operations, “if we could see that power was being injected into the system from a PV (photovoltaic) system, then we’d know what was going on. Then we could find ways to mitigate that if it should drop off all of the sudden.”

Speaking to Power

It sounds easy enough. The problem is that even the traditional grid – a model that’s nearly 100 years old – is surprisingly complicated. Sure, its basic elements are familiar: a power plant that generates electricity, high-voltage transmission lines that carry the power long distances, substations and transformers that step the voltage down to useful levels, and a network of distribution lines to deliver electricity to the end user. This generic view of the electrical grid makes it seem almost mechanical: add fuel – oil, coal, bagasse – to the hoppers at one end, and 120-volt electricity comes out the wall sockets on the other. In its particulars, though, the grid is complex; more like an organism than a machine, it’s full of fidgets and sensitivity. The utility is constantly monitoring its vital signs, especially frequency and voltage.

Reynolds points out that the pulse of the modern grid throbs at the remarkably consistent frequency of 60 hertz. Maintaining this frequency depends upon a fairly steady balance of power generation and load. If MECO loses a generator, or the wind dies suddenly at Kaheawa, the frequency falters.

Most utilities handle these problems with what’s called “spinning reserve,” having extra generators up and running and ready to come online. “In place of spinning reserve, MECO uses load-shedding,” Reynolds says, basically cutting power to prearranged customers. “At 59.3 hertz, there are some pump leads at HC&S that will come off-line. Below 58.7 hertz, then we’ll start opening up distribution points on our customers.” In other words, a 2 percent drop in frequency can mean localized blackouts. It can also fry customers’ electronics.

The traditional grid has evolved tools to deal with normal fluctuations in load. MECO’s power plant at Maalaea, for example, isn’t just one generator; it’s 21 generators of various sizes and types. They range from small, “fast-start” generators to deal with sudden outages, to enormous combustion turbines that are much more efficient, but take longer to start. An automatic system controls the generators’ output based on variations in load.

These controls work fine for a grid dominated by consistent power, like diesel generators or hydroelectric, but they’re not responsive enough to handle Maui’s increasing suite of wind and PV power. Instead, MECO has to limit renewables.

For example, through a process known as curtailment, the utility routinely dials back power generation at Kaheawa. Sometimes curtailment at the wind farm is partial; sometimes it is 100 percent. Similarly, MECO restricts the installation of PV systems to less than 3 percent of the system’s peak load, or less than 10 percent of the load on any one circuit. Although these strategies run counter to the utility’s own preferences, probably nothing short of a smart grid will ease the restrictions.

Technical Problems

One of the challenges facing MECO’s smart-grid aspirations is an aging infrastructure. Over the past several years, the utility has modernized its systems, particularly by improving its SCADA, the supervisory control and data acquisition system it uses to control critical elements on the grid.

But the utility still has many substations that haven’t been integrated into its SCADA system, and the system has no means to see beyond the substations to monitor the load of its customers (or the production of most independent PV systems). Also, many of MECO’s generators are aging and inefficient – the oldest, a steam generator in the Kahului plant, was first put online in 1947 – meaning MECO’s high-voltage transmission lines carry 69,000 volts in some areas and 23,000 volts in others. These are all challenges on the journey from existing infrastructure to smart grid.

But the greatest technical challenge is isolation. On the Mainland, most local grids are linked to one another in a super-grid. It’s possible, for example, for a customer on the East Coast to buy electricity from a power provider in Texas or even Canada. That’s important because this interconnectedness makes it easier for utilities to provide some of the ancillary services that are essential to an effective electrical system. As Carl Freedman, one of Hawaii’s most respected experts on utility regulation, likes to point out, an electric company provides customers much more than kilowatt-hours of electricity.

“For example,” Freedman says, “they also have to provide reliability,” a quality that includes things like operational and spinning reserves. Operational reserves ensure the grid has the capacity to supply the maximum load. Freedman explains: “If somebody turns on a 1,000 horsepower motor or turns off a 1,000 horsepower motor, operational reserves mean it isn’t going to shut lights off and destabilize the grid.” Spinning reserves, on the other hand, represent the utility’s ability to handle the loss of a generator (or wind on a wind farm). “On Oahu,” he says, “they have a spinning reserve sufficient for the loss of their largest unit. In other words, they would have enough units up and spinning so that they could lose that unit without dropping load.

“Spinning reserves and operational reserves are both identifiable services,” Freedman says, as are basic utility functions like voltage regulation, transmission and power generation. “On the Mainland, there’s a huge market for all this stuff. If you don’t have something, you can go out and get it.” Freedman points out how this simplifies the way a utility operates. “Each utility, for example, needs to carry sufficient capacity – or contracts for capacity – to meet its loads. But they don’t need to provide the emergency capacity of the largest load like we do here, because they can buy that. In fact, they can buy it for free by having a bilateral agreement with somebody else, saying, ‘You cover me, and I’ll cover you.’ ”

This highlights the challenges facing MECO and HECO as they build their smart grids. Because they’re island grids, they’re completely isolated. Freedman notes: “Each one of these systems has to supply all the ancillary services: all the generators, all the reserve capacity, all the reliability. We have to do it all on each system. So, the job of a smart grid here is a tall order.”


Capital Problems

Not all the challenges facing the smart grid are technical. Rebuilding something as complex as the grid – even a small one like MECO – will be fabulously expensive. “As an example,” says Chris Reynolds, “the meter that’s on a typical home costs about $25. For the smart grid demonstration project in Wailea, we’re looking at a cost of about $400 per meter.” He adds that MECO has about 67,000 meters.

Freedman takes an even broader perspective. “According to DBEDT,” he says, “we’re about to spend $16 billion – that’s billion with a ‘B’ – on capitalization for this energy transition.” He notes that, although the goal is to reduce our $7 billion annual expenditure on fossil fuel, that’s still a fantastic upfront investment. “The question is how are we going to capitalize this. This is a major issue for the state that hasn’t been addressed by anyone, really.”

It’s certainly hard to see how the Hawaiian Electric companies can afford it. “I don’t know what we’re counting in the smart grid,” Freedman says, “but if you include the (undersea, interisland) cable, then you’re talking a billion dollars just to hook up Lanai and Molokai. If you’re talking, like the utilities, about hooking up Maui as well, then you’re talking several billion dollars. Well, the whole capitalization of all the electrical infrastructure right now is something on the order of $3 or $4 billion.” Even if, as now seems likely, the state decides to finance the construction of the cable, Freedman points out, ratepayers will have to repay the debt. It’s still a capital liability on the utilities books.

One of the ironies in this smart grid bagatelle is that many of the policy initiatives intended to promote more renewables further aggravate the capital problems for the utility. For example, the financial arrangements that underpin distributed generation – power-purchase agreements, net-energy metering, feed-in tariffs –all appear on the utility’s books as liabilities. Each, after all, is a commitment to purchase power from customers. The feed-in tariff, at least, also shows up on the income side of the books because the customer still buys the same amount of energy as before. With net metering, the customer’s PV output simply rolls his meter backwards, reducing his bill.

Also, most of the utility’s assets – and collateral – traditionally were in its physical plant: generators, power lines, substations. “Looking forward,” Freedman says, “it looks like they’re not going to be increasing generation anymore. The new generation is going to be in renewables, it’s going to be distributed, and loads are going to met by energy efficiency. And none of those things have the utility’s own capitalization.” Hawaiian Electric Industries is publicly traded; it’s hard to see how these changes in capitalization won’t affect the company’s market valuation. “In the long run,” Freedman says, “the utility’s business model is being challenged a little bit by the whole move to renewables.”

A local smart grid is thus far from inevitable, even with Hawaii’s incomparable resources for renewable energy; even with an ambitious agenda for reform in the Hawaii Clean Energy Initiative; and even with a cadre of utilities and citizens committed to the idea of a clean, distributed power generation.

Up at the Kaheawa wind farm, the students from Horizons Academy gather in the scant shade of a giant turbine to pose for a group photograph. Squinting into the late morning sun, the children smile for the camera. It’s supposed to be a picture of Hawaii’s future – the children and the energy that will power their adult lives – but that future is not yet fully in focus.


Kaheawa Wind Farm

• Minimum: As little as 6 mph of wind will turn the long, elegant blades of the Kaheawa turbines.

• Maximum: When the wind reaches 55 mph, the blades feather and each turbine stops spinning.

• RPMs: Regardless of the wind speed, the turbines top out at 21 rpm – slow enough for nene to fly through in formation.

• Best wind: At 23 knots, the optimum wind speed, each turbine produces 1.5 megawatts of electricity.

source: first wind inc.


P.A.C.E.: Supercharging the Solar-Energy Industry

Many homeowners and businesses want solar energy to lower their electric bills but can’t afford the upfront cost – as much as $25,000 for a standard residential installation. But a new form of funding called PACE – property-assessed clean energy – offers a nearly painless solution.

How PACE Works

People who want to purchase clean-energy technology, such as solar water heating or photovoltaic systems, for their homes or businesses will be able to borrow from a special revolving fund established by the state. In return, they agree to pay the money back (plus interest and administrative costs) through an added assessment on their property taxes. In most scenarios, PACE funding will have no effect on the availability of federal or state tax credits.

How It’s Funded

To establish the PACE revolving fund, the state would issue general-obligation bonds. These would be guaranteed by the incremental increase in property taxes. In theory, PACE shouldn’t add any costs to the state budget. It’s even possible that federal grants would pay for the administrative costs of setting up the program and establishing a certification process.

Who Would Be Eligible?

One of the charms about PACE funding is that it’s tied to the house, not the homeowner’s credit. As long as you can afford to keep up with the property taxes, you would be eligible to borrow money for any qualified clean-energy system. What’s more, when you sell your home or business, the obligation to pay goes with the property. That makes sense, because an investment like a PV system adds value to your home, but is worthless to you when you sell.

Will It Happen Here?

The Sierra Club and Blue Planet Foundation are advocating strongly for PACE. It also enjoys broad support in the Legislature and with Gov. Linda Lingle. Legislation introducing the program, HB 2643, has already passed unanimously in the state House, but it still faces challenges in the Senate and in conference. Advocates such as Sen. Kalani English warn that, given the state’s fiscal troubles, it may take more than one session to pass.

Related Stories

Magazine Promo

On Newsstands Now


HB October 2014 Wahine Issue


Dennis Hollier