No Brainer: Rebuild the Grid

June 2, 2016


It’s not like there is some kind of choice as to whether we rebuild our electrical grid.  We either do it, and remain a superpower, or we don’t, and our already-obsolete infrastructure degrades further, as we slide gradually into decrepitude and the ash heap of history.

Maybe we should rebuild it to help sustain a liveable planet?


The United States could slash greenhouse gas emissions from power production by up to 78 percent below 1990 levels within 15 years while meeting increased demand, according to a new study by NOAA and University of Colorado Boulder researchers.

The study used a sophisticated mathematical model to evaluate future cost, demand, generation and transmission scenarios. It found that with improvements in transmission infrastructure, weather-driven renewable resources could supply most of the nation’s electricity at costs similar to today’s.

“Our research shows a transition to a reliable, low-carbon, electrical generation and transmission system can be accomplished with commercially available technology and within 15 years,” said Alexander MacDonald, co-lead author and recently retired director of NOAA’s Earth System Research Laboratory (ESRL) in Boulder.

Although improvements in wind and solar generation have continued to ratchet down the cost of producing renewable energy, these energy resources are inherently intermittent. As a result, utilities have invested in surplus generation capacity to back up renewable energy generation with natural gas-fired generators and other reserves.

“In the future, they may not need to,” said co-lead author Christopher Clack, a physicist and mathematician with the Cooperative Institute for Research in Environmental Sciences at the University of Colorado Boulder.

Since the sun is shining or winds are blowing somewhere across the United States all of the time, MacDonald theorized that the key to resolving the dilemma of intermittent renewable generation might be to scale up the renewable energy generation system to match the scale of weather systems.

State of the art transmission from the resource rich plains to population centers east and west would include High Voltage DC power lines – another technology that the US has allowed other countries to take the lead in. Something to rethink?

Ramez Naam:

The continental US is roughly 2,600 miles from east to west. Almost every population center is within 1,000 miles (or far less) of an area with top-notch wind resources. And most are within a few hundred miles of an area with good, if not best-in-class, winds.

HVDC lines are not common in the US, however. Compare the map of HVDC lines in China to that of HVDC lines in the US.



Not just a US problem, notes Bloomberg:

Chile’s solar industry has expanded so quickly that it’s giving electricity away for free.

Spot prices reached zero in parts of the country on 113 days through April, a number that’s on track to beat last year’s total of 192 days, according to Chile’s central grid operator. While that may be good for consumers, it’s bad news for companies that own power plants struggling to generate revenue and developers seeking financing for new facilities.

The main culprit is the northern part of the country, in the Atacama desert. Chile’s increasing energy demand, pushed by booming mine production and economic growth, helped spur the development of 29 solar farms, with another 15 planned, on the country’s central power grid. Now the nation faces slowing demand for energy as copper production slows amid a global glut, and those power plants are oversupplying a region that lacks transmission lines to distribute the electricity elsewhere.

A key issue is that Chile has two main power networks, the central grid and the northern grid, which aren’t connected to each other. There are also areas within the grids that lack adequate transmission capacity.

That means one region can have too much power, driving down prices because the surplus can’t be delivered to other parts of the country, according to Carlos Barria, former chief of the government’s renewable-energy division and a professor at Pontifical Catholic University of Chile, in Santiago.

Inadequate Infrastructure

The government is working to address this issue, with plans to build a 3,000-kilometer (1,865-mile) transmission line to link the the two grids by 2017. It’s also developing a 753-kilometer line to address congestion on the northern parts of the central grid, the region where power surpluses are driving prices to zero.

“Chile has at least seven or eight points in the transmission lines that are collapsed and blocked, and we have an enormous challenge to bypass the choke points,” Energy Minister Maximo Pacheco said in an interview in Santiago. “When you embark on a path of growth and development like the one we’ve had, you obviously can see issues arising.”




9 Responses to “No Brainer: Rebuild the Grid”

  1. Gingerbaker Says:

    Jeezum Crow – what an excellent post, Peter!

    Chock full of great information, excellent links, very salient facts, and brimming with important implications. There is a lot to talk about here.

  2. mboli Says:

    Much as I like this idea, I do have to quibble with the notion that the sun is always shining somewhere in the US.
    Are the authors proposing a transmission line between Guam and North America?
    More seriously, I’ve long been irritated by the argument: a) many renewables are intermittent and have to be backed up by investment in conventional generating stations, b) so there is no point in investing in the renewables.
    Irritated because a lot of people believe this line of reasoning. When I hear it I point out the fallacy: every kg reduction in CO2 emissions is a step toward stabilizing GHGs. No matter if every single watt of renewable generating capacity has to be duplicated by a watt of conventional generating capacity. You still get to use the renewable generation to improve the Earth’s carbon budget.
    For some reason people have trouble seeing the obvious sometimes.

  3. The problem here is that the cited study (MacDonald) understates the actual cost of HVDC transmission by many times.

    The most recently completed overland HVDC lines in North America are the East Alberta line and the West Alberta line. Each is a 2-pole 1GW line, of 485 and 350 km respectively, which cost $1.71 billion and $1.57 billion respectively (2013 USD at then-current exchange rates.) Powerlines have a fixed cost for terminals (regardless of line length) and a variable cost for poles and conductors (which varies with length). From these data, one can easily compute that the fixed cost for these lines was $1.2 million per MW, and the variable cost was $1701 per MW-mile.

    That compares with the numbers used by MacDonald, which were $183,000 per MW fixed (understated by a factor of 6.5) and $701 per MW-mile (understated by a factor of 2.4). And that would be an underestimate for US lines, because Alberta is lightly populated compared to where US lines would run, and land is therefore more expensive here.

    With the correct fixed and variable costs, it is possible to determine how much HVDC adds to the cost of wind power on an LCOE basis, and the results are not encouraging. Using reasonable assumptions, it’s cheaper to build a nuclear plant next door than it is to transmit wind power for ANY distance at all over HVDC.

    Assuming CF of wind=35%, nuclear=90%, wind lifetime=25yr, nuclear lifetime=50yr, wind overnight CAPEX=$2/W, nuclear overnight CAPEX=$7/W, and a 4% discount rate for both, the levelized capital component comes to $53/MWh for wind vs. $57 for nuclear. Adding just the overnight CAPEX for HVDC terminals, and a zero-distance line, pushes wind up to $84/MWh.

    The fixed costs alone put the CAPEX factor of wind+HVDC well above that of nuclear.

    • Gingerbaker Says:

      Your Canadian example does not appear to be representative. Modern HVDC lines can be 7 GW, have 3.5% loss over 1000 km, and cost ~ $1.05 million per mile. Clean Line is claiming they will build inclusive for 2 cents per KWh.

      Every source I read breaks down costs of HVDC to be less than HVAC at 600 km or longer. And HVDC is getting cheaper.

      If your figures were representative, would anybody be constructing HVDC? But it is being constructed all over the bloody place:

    • andrewfez Says:

      America just spent $100B reconstructing Afghanistan, after they spent probably even more tearing it down to begin with.

  4. John Oneill Says:

    From the looks of it, most of the Chinese HVDC is carrying hydro power from the western mountains to the population centres in the east. Similarly, the few HVDC lines in the US are carrying hydro down from the Columbia and Colorado systems to southern California, and from Canada to New England. Hydro usually has a capacity figure ( actual output over nameplate ) of around 45%, not too much better than the best wind sites, and newer 7+ MW designs in the belt of high wind areas from the Dakotas down to Texas are projected to get up to 50%. However wind averages lowest in summer, when demand is highest, and unlike hydro is not dispatchable, so to cover all the permutations of weather and demand, your powerline planners would have a task like that of the travelling salesman conundrum, rather than a simple one way connection from a supply to demand.
    In China, the priority is to reduce coal pollution in the cities, but pollution is worst when the wind is not doing much. That, unfortunately, is in summer and in winter, when coal use peaks for AC and for heating.

  5. andrewfez Says:

    Usually when you see a graph of wind farm aggregate capacities, the production generally doesn’t fall under 30%, supporting U of Delaware’s 3x load overbuild to get us to 90-99% clean energy usage. Here’s one I randomly found just now; it’s the first I’ve seen where there is only 20% ‘base power’ (Ireland), but I supposed you could add those two curves together and divide by 2 to get a higher ‘baseline’, simulating Denmark and Ireland being fully connected to each other for the turbines examined.

    And here’s a measure of the production’s ‘internal’ correlation. Note it takes several hours before correlation peters out, meaning regional production is fairly slow to change, again making the case for a ‘baseline’ of steady power, or at least making the case for ease of forecasting. Note the humps at the 24, 48, and 72 hour marks showing the daily cycle of production:

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: