"The system’s structure determines its properties and behaviors"
February 7, 2020 2:06 PM   Subscribe

David Roberts[twitter] is a journalist at Vox who writes about energy and climate change. He tackled the question of how renewable energy sources, like wind, solar and hydroelectric, can integrate with existing electrical grid architecture and provide power to people in the United States: Clean energy technologies threaten to overwhelm the grid. Here’s how it can adapt. -" Now, I grant you, “grid architecture” is not a term designed to set the heart aflame. But it is extremely important, and the stakes are high. The danger is that policymakers will back into the future, reacting to one electricity crisis at a time, until the growing complexity of the grid tips it over into some kind of breakdown. But if they think and act proactively, they can get ahead of the burgeoning changes and design a system that harnesses and accelerates them. Now is the time to rethink the system from the ground up."

For more on grid architecture and future growth, try the podcast The Energy Transition Show: Grid Architecture of the Future
What kind of grid architecture and markets will we need in order to actually operate the distributed, decentralized grid of the future? What sorts of regulatory models will be needed? And what does it all mean, from a philosophical point of view, about how human society is organized? How can mere mortals begin to understand these subjects? Never fear: We’ve got you covered, in this ultra-geeky yet accessible episode.
some further reading:
The Smart Grid and Renewable Energy
The global transition to clean energy, explained in 12 charts
It’s Time to Tie the U.S. Electric Grid Together, Says NREL Study
We’ve been talking about a national grid for years. It might be time to do it.
100% Renewable Energy Needs Lots of Storage. This Polar Vortex Test Showed How Much.
Getting to 100% renewables requires cheap energy storage. But how cheap?
Renewable Energy Boom Is Pushing The Grid To Its Limits, Prompting Operators To Reinvent Themselves
Energy transitions are usually slow. Here’s why the clean energy transition might be faster.
HOW THE U.S. POWER GRID IS EVOLVING TO HANDLE SOLAR AND WIND
A beginner’s guide to the debate over nuclear power and climate change
How the U.S. Could Have an All-Renewable Energy Grid

Wildfires and blackouts mean Californians need solar panels and microgrids

The Department of Energy's Interactive Grid page has links to applets letting you play with grid costs, carbon dioxide emission, load balancing, and more from UIUC's Cyber Resilient Energy Delivery Consortium (CREDC)
posted by the man of twists and turns (14 comments total) 34 users marked this as a favorite
 
David Roberts consistently writes extremely intelligent and accessible analysis of climate change science and the energy sector .... the headliner VOX piece just another one of his very good pieces. (I work in the financial services industry and quite a few of us were talking about his piece about climate change and the financial sector today)

Thanks for the link roundup! lots of good stuff there.
posted by larthegreat at 2:30 PM on February 7, 2020 [2 favorites]


NREL in the US has done a lot of research and talks on increasing variable renewable energy (VRE; e.g., wind and solar) on the US grid. Short version (for the continental US; Alaska and Hawaii have dedicated papers):
* we're currently running ~7ish% VRE in the continental US
* we could go to 25% VRE with very little expenditure or trouble
* we already have the technology and field-tested systems to manage all this up to about 50% VRE
* going to 90% total renewables and 50% VRE would require energy storage of about 10% of the total installed capacity, or about 140 GW of storage
* 140 GW of storage is about 4.5x what we currently have installed, and we do not have much storage installed right now
* adding that 110 GW of storage would cost us about right around $100B; not cheap, but cheap in the world of grid-scale things

To put it into perspective, for about half of the federal FY2020 "emergency fund" (mostly used to pay for overseas wars) and using technology and systems that already exist today, we could have a power grid that could handle 90% renewables, of which 50% is variable renewables like wind and solar.
posted by introp at 4:22 PM on February 7, 2020 [14 favorites]


What would the 40% non-variable RE be? That's a lot of biomass

How would the breakdown be impacted by decarbonising heat and transport? Does the US look likely to go down the route of electrification of both? Would this make more VRE more tolerable on the system?

Also worth considering what the US could do on energy efficiency. Quite a lot I expect.
posted by biffa at 11:48 PM on February 7, 2020


What would the 40% non-variable RE be? That's a lot of biomass

Hydro is ideal. That's how Denmark routinely has 100% renewable energy days, they use Norwegian hydro as a battery. UK has very little hydro but 12% biomass.

(This is basically my specialist subject but I mostly know about the UK and Euro grids so looking forward to learning about US).

7% seems surprisingly low. In the UK, 2019Q3 was 40% renewable, 39% fossil fuels, 19% nuclear. First quarter we ever did more renewables than fossil fuels which was a big milestone for the industry.
posted by atrazine at 2:29 AM on February 8, 2020


How would the breakdown be impacted by decarbonising heat and transport?

The important thing in understanding what effect this has on managing the impact of intermittency on the grid is how easily those loads can be time-shifted.

Transport is no problem, you just need your chargers to be smart about when they draw power.

Heat makes it much harder, when you need heat you typically need it across a wide geographical area simultaneously and you need it when it is cold (demand cannot be easily time-shifted). With storage heaters you can do a little shifting but cold temperatures are time-series correlated (it's never just cold for an hour and then back to being warm) so at most you get a little load spreading at the beginning.
posted by atrazine at 2:40 AM on February 8, 2020 [1 favorite]


The important thing in understanding what effect this has on managing the impact of intermittency on the grid is how easily those loads can be time-shifted.

I was thinking in terms of what happens if you say double electricity demand to meet heat and transport electrification. The US scope for increasing the firm RE side of things must be pretty limited (ie little potential for new hydro and filling the gap with biomass would come with its own challenges) so the need for VRE will have to increase above the proportions based on current demand. (This is also my speciality area, also more UK/EU than US.)
posted by biffa at 7:53 AM on February 8, 2020


Yes, adding more transportation that can be shifted is good for the grid. We've got a number of pilots and programs doing that now. We need a lit more EVs on the road and grid to match all the renewables being added
posted by CostcoCultist at 8:35 AM on February 8, 2020


"until the growing complexity of the grid tips it over into some kind of breakdown. But if they think and act proactively"

How long has this guy been alive? Or at least living in the US? Is he a technocrat? I'm glad he's asking these questions. But asking questions is easy, getting changes made... in our current system? I hope he's willing to put his money where his mouth is because this is gonna be a long hard fight. Unless I suppose you scream "terrorists" then you can get all the funding you need.
posted by symbioid at 10:50 AM on February 8, 2020 [1 favorite]


From working inside a large electric grid equipment/systems vendor, the sense is that the problem is very solvable on a technical level, it's just that customers (outside of China) aren't buying. National governments have too much debt and they can't collect taxes like they could in the 60s/70s.

I'm curious how the technology paradigm will develop between microgrids (small scale, regional, small solar and batteries) and HVDC interconnections (large scale, '000s of km, sending solar power three time zones forward or back).

At first I was really excited about microgrids, "small is beautiful" and all that. But now I feel they're a technology that allows the rich to pay to avoid power outages and essentially say "fuck you got mine". The nature of today's interconnected AC grids is, despite their technocratic and regulatory nature, inherently egalitarian. Every area is literally synchronized and to some extent dependent on their neighbors, and it's hard to cut off power at too fine-grained a level. In a microgrid land, the rich don't have to care about the reliability of the grid for everyone else.

HVDC, on the other hand, is a large-scale, complicated and finicky technology that allows sending power huge distances and as easily underwater or underground. It's a technology that you need big capital and nation-states to invest in to get working. But the possibilities it enables are to get power from where-ever the wind blows to where-ever it doesn't or to span time zones and people's energy-demanding dinnertime habits. And in a way it's inherently a technology of international cooperation. Power lines are easy targets in war, embargoes, or guerilla conflict. And if we build a grid that really depends on our neighbors to keep the lights on, the rich need to be sure that society is stable enough that nobody decides to drive a truck into a high voltage pylon in the middle of a cold snap.

Anyhow. It is probably possible to run the grid on 100% wind water and solar if we build a crapton of generators, transmission, and electric<>rock heat storage. But step one is getting the (inter)national will and capital to do it.
posted by anthill at 12:29 PM on February 8, 2020 [3 favorites]


Further exploring larthegreat's tangent into climate change vs the financial sector (feel free to skip this if you don't have a keen desire to read about corporate greenhouse gas accounting standards)!

One indirect way to apply pressure to large companies with businesses associated with greenhouse gas pollution is to get those companies to measure and disclose the greenhouse gas pollution associated with their business activities. This doesn't necessarily cause the companies in question to change anything but makes it easier for the public to become aware of the negative impacts associated with the business activities of specific companies, and may make it easier to build support to regulate those activities. (Personally I'd love to see a carbon tax rolled out to force companies to fairly internalise the external costs of greenhouse gas pollution).

Various public financial companies (banks etc) that lend money to finance new fossil-fuel projects or lend to existing fossil-fuel related companies report "exposure at default" risk metrics that estimate how much money the lender might lose if something terrible were to happen to climate-exposed customers (an example of "something terrible" that might happen is "a carbon tax is introduced that forces companies to pay tax for the externalised costs of greenhouse gas pollution" - this might cause a coal mining company to go out of business, which might cause a bank to lose money, which might cause investors or customers of the bank to lose money (alas!)). These "exposure at default" metrics are essentially credit risk metrics but not very useful to estimate other things you might want to know, such as: to what extent does the financial company's business contribute to causing the problem of global warming in the first place?

One greenhouse gas pollution accounting standard used by various companies to report their greenhouse gas emissions are defined by ghgprotocol.org . These standards categorise the greenhouse gas emissions a company contributes towards into three "scopes". Here's a diagram illustrating the three scopes: Corporate-Value-Chain-Accounting-Reporing-Standard -- see page 5.

Scope 1 emissions are emissions caused by activities directly under the company's control. E.g. if your company owns a petrol-powered vehicle and runs it, that's scope 1. If your company burns coal to generate electricity, that's scope 1.

Scope 2 is emissions produced "upstream" in order to generate electricity or heat that the company consumes. E.g. if your company consumes a bunch of electricity that is generated by another organisation that burns coal or gas to produce electricity, the corresponding emissions are attributed to the consuming company are scope 2. Note that the same greenhouse gas emissions are categorised as scope 1 for the producer of the electricity, and scope 2 for the consumer.

Finally, there's scope 3 emissions. These are emissions from the "value chain" that the company is involved with -- they could be emissions related to inputs that the company consumes, or emissions relating to outputs that the company produces.

Scope 3 emissions are very important to measure, in terms of estimating how much pollution a company's business activities contribute towards, but scope 3 emissions don't seem to be estimated or reported very well. Companies that do not disclose scope 3 emissions may massively under-report the greenhouse gas emissions their businesses contribute to. Note also that the same greenhouse gas emissions might be included in the scope 3 emissions of multiple companies/entities involved in a value chain.

For an example of scope 1, scope 2 and scope 3 emissions estimated and reported by a company, have a look at the graph titled: "Microsoft's pathway to be carbon negative by 2030". Note also that the greenhouse gas emissions for "microsoft and it's supply chain emissions" (scope 3) is a lot larger than the "operational emissions". https://blogs.microsoft.com/blog/2020/01/16/microsoft-will-be-carbon-negative-by-2030/

It becomes trickier if we start wanting to do things like figure out which share of the emissions each entity should be associated with. E.g. for generating electrical power from burning thermal coal, if I use it to power my lights, some of the greenhouse gas pollution should be attributed to me, as well as the electricity retailer, electricity distributor, electricity generator, coal distributor, coal transporter, and anyone involved in providing services or inputs, to that entire value chain, and also anyone else profiting from that entire value chain (including the owners and investors in the companies involved, lenders, employees, suppliers, governments extracting tax revenue, etc) -- who all extract some of the value in the polluting economic activity but are likely neglecting to pay for the corresponding externalised costs of pollution that are pushed onto everyone in the world now and in the future.

A couple of examples of the huge difference in greenhouse gas pollution reporting or failing to report scope 3 emissions makes:

1. Let's say I own a thermal coal mine. The mine has some direct greenhouse gas emissions from the machines we run to dig for coal, transport it, sort it, load it. We also use a bunch of electricity which may be generated by fossil fuel electricity generation. These emissions are all measured under scope 1 and scope 2. However, the scope 3 emissions of the coal I sell actually getting burned are going to be at least 100x larger than my reported scope 1 and scope 2 emissions. Scope 1 and scope 2 emissions don't capture the pollution associated with how my business actually makes money: my customers only give me money in exchange for coal because they want to burn the coal, but if I don't measure and report scope 3 emissions arguably I can gloss over the fact that 100% of my company's revenue is value extracted from highly polluting activity of burning thermal coal.

2. Let's say I own a bank. The bank has some direct greenhouse gas emissions from e.g. a fleet of cars we own that employees use to drive around and try to close deals lending businesses money. Scope 1 emissions. We also use a bunch of electricity - Scope 2 emissions. However, like the coal mining company, these emissions don't capture the emissions associated with where my business' revenue comes from. Let's say roughly 100% of my business' revenue comes from interest payments on loans my company makes to companies or individuals to help them fund projects or pay for operational costs. The interest my customers pay me is my share of the value extracted from my customer's business activities. Now, how do my customers generate the revenue to pay me interest? If my bank lends to a lot of businesses associated with large greenhouse gas emissions, arguably my bank should be forced to disclose this. Again, this only shows up in ghg protocol reporting standards as "Scope 3 - category 15 - investments". If a bank actually estimates and discloses this, it is plausible that the greenhouse gas emissions generated by the value streams of the bank's customers are around 100x-1000x larger than the bank's scope 1 and scope 2 greenhouse gas emissions. By measuring the scope 3 emissions of multiple banks we can estimate how much each bank is capturing the value of lending to highly polluting sectors of the economy.

For an example of a bank that appears to be doing a good job of attempting to estimate and disclose it's scope 3 emissions associated with its lending activity (Let's ignore the question of if the bank is doing a good job in refraining from lending to highly polluting sectors of the economy), check out ABN AMRO's 2018 sustainability report .

In particular, pages 8 -- 10 of ABN AMRO's sustainability report estimates the the "obvious" greenhouse gas emissions caused by the bank's consumption of gas, electricity, heat, as well as some other things like scope 3 emissions from business travel of the bank's employees by planes. Then compare those numbers of k tons CO_2-e greenhouse gas pollution with page 12 -- the bank's estimate of scope 3 greenhouse gas pollution associated with its portfolio of lending to different kinds of real estate (total of 3,926 k tons CO_2-e) and on page 14 where the "emissions lending portfolio" breaks down how much the bank lends to each sector of the economy, and what the corresponding scope 3 CO_2-e greenhouse gas pollution estimate is. The total greenhouse gas pollution estimate is 14,474 k tons GHGe (i assume the unit GHGe is equivalent to CO_2-e ). A footnote says this is an under estimate as the bank cannot figure out which economic sector some of its lending is associated with.

So the scope 3 emissions disclosed by the ABN AMRO in 2018 are about ~18,000 k tons CO_2-e --- in contrast all other categories greenhouse gas pollution reported by the bank amount to about 21 -- 58 k tons CO_2-e (depending on if you believe in green certificates or not), so the scope 3 emissions from this "follow the money" analysis of where the bank's revenue comes from discovers emissions that are 300x - 800x (!!) the other emissions reported by the bank.

If you've read this far and want to deep dive, it appears that ABN AMRO is following the additional carbon accounting standards defined by Partnership for Carbon Accounting Financials

> "Follow the money" is a key principle of the global standard, i.e. the money should be followed as far as possible to understand and account for the carbon impact in the real economy. It includes scope 1, 2 and relevant categories of scope 3 of the investee.

> with the publication of the report we could also witness the birth of the next excuse from those that cannot or do not want to change. They said the presented methodology is fine, but we now need an international standard. My answer to them, at the launch of this new PCAF Report, is that the PCAF methodology can be used irrespective of a global standard. Because it is a tool that is fit for its task.
posted by are-coral-made at 6:58 PM on February 8, 2020 [4 favorites]


One of the bigger aspects of this change that I think is fundamental is how dependent the grid should be on communications technology to operate.

The classical 1960s grid design with big spinning steam turbines, motors, and resistance heaters was designed to be inherently stable. It could more or less run itself for hours at a time with the feedback loops based on local measurements and communication happening inherently through the electric AC system dynamics. Operators ran the grid more or less by telephone until the '70s and it's even possible today to some extent.

But now with transistor-switched loads that freak out and switch off if there's a disruption in the grid voltage or AC frequency, the grid is getting less and less inherently stable. So you need to connect more measurements / ICT and for a lot of the new grid operating schemes, low latency real time communication becomes necessary for the grid to operate. So now you have two critical networks to protect - the power network and the communications network. And bugs / maintenance / hacking / etc. of the communications becomes maybe the most critical failure mode.

Do we trust cellphone / internet networks enough to run the whole country off them?
posted by anthill at 6:27 AM on February 9, 2020


(I wish I'd had time to post this yesterday as it looks like this post has faded out a bit.)

To quote anthill above
But now I feel they're a technology that allows the rich to pay to avoid power outages and essentially say "fuck you got mine". The nature of today's interconnected AC grids is, despite their technocratic and regulatory nature, inherently egalitarian.
I am in total agreement and I cannot overstate how much the following can be true at the same time:
  • Microgrids are sometimes amazing community tools and can help bring electricity to a lot of people.
  • A lot of people pushing bottom-up grids like microgrids are either trying to increase inefficiency and skim some of that waste into their pocket or make it easier to dodge regulation
Local optimization and markets (not local generation!) tend to look a lot like deregulated systems and, surprise surprise, that's where they seem to be getting attention. We have a TON of experience around the world that this sort of market-will-solve-it magic is bad for everyone but the investors.
posted by introp at 3:32 PM on February 9, 2020 [1 favorite]


1. Let's say I own a thermal coal mine. The mine has some direct greenhouse gas emissions from the machines we run to dig for coal, transport it, sort it, load it. We also use a bunch of electricity which may be generated by fossil fuel electricity generation. These emissions are all measured under scope 1 and scope 2. However, the scope 3 emissions of the coal I sell actually getting burned are going to be at least 100x larger than my reported scope 1 and scope 2 emissions. Scope 1 and scope 2 emissions don't capture the pollution associated with how my business actually makes money: my customers only give me money in exchange for coal because they want to burn the coal, but if I don't measure and report scope 3 emissions arguably I can gloss over the fact that 100% of my company's revenue is value extracted from highly polluting activity of burning thermal coal.

Yes, but on the other hand they will be counted as scope 1 emissions for the generating company burning the coal and scope 2 for the end-user of the electricity (transmission losses are scope 2 for the transmission operator).

This is why voluntary disclosures are irrelevant. I do not care what ABN Amro's GHG Protocol scope 1/2/3 emissions are. I want all emissions taxed. It does not matter whether that is done by imputation at the point the fuel is mined/pumped, whether it is done at the point energy is converted to electricity, or whether it is done at the point consumers use it. If you do this, you no longer need to count the 1/2/3 emissions because the money does it for you.

For a living, I build models that evaluate whether to make investments in projects in the energy sector. When carbon pricing and/or taxation goes up, coal mines and power plants become unprofitable very quickly. (Especially if we use trade barriers to punish countries not implementing domestic emissions control - otherwise coal could just be exported).

I would prefer banks to drop the emissions disclosure on the grounds that legal reasons have required all their borrowers to incorporate carbon pricing into their operations and therefore GHG disclosures no longer give any additional information.

Anything a company has to declare separately from their bottom-line results might as well be saying, "in addition to our financials which you will read thoroughly, here's some other shit you don't care about".
posted by atrazine at 7:32 AM on February 10, 2020 [3 favorites]


> I want all emissions taxed. It does not matter whether that is done by imputation at the point the fuel is mined/pumped, whether it is done at the point energy is converted to electricity, or whether it is done at the point consumers use it. If you do this, you no longer need to count the 1/2/3 emissions because the money does it for you.

> For a living, I build models that evaluate whether to make investments in projects in the energy sector. When carbon pricing and/or taxation goes up, coal mines and power plants become unprofitable very quickly.

Thanks atrazine, i agree, i also want a carbon tax. Taxing emissions at some point and then letting the price impacts ripple throughout the economy sounds like a relatively simple way of incentivizing more environmentally friendly decisions. Also arguably less likely to be gamed than some kind of emissions trading scheme

> (Especially if we use trade barriers to punish countries not implementing domestic emissions control - otherwise coal could just be exported).

I agree. On the one hand, global warming is a problem with running out of capacity to store pollution in the planet's atmosphere, which is a shared resource between all countries / all people, so ideally the global resource of the atmosphere as a pollution sink needs to be managed at the global level.

On the other hand, pragmatically, different governments will install carbon taxes at different times (or never do), so there needs to be mechanisms to roll it out incrementally, and apply pressure on countries that don't have a carbon tax in place to comply. I've heard about this thing in the media with the EU arguing that future trade agreements are conditional on some level of climate agreement. (I'm an Australian citizen, a chunk of our economy is built around exporting thermal coal, please install trade barriers / regulate the crap out of us until we get our shit together climate wise...)

There's arguably a similar (simpler?) problem with tax avoidance and tax havens, where there's a lot of benefit to avoiding regulation that other people have to comply with. E.g. some countries such as Luxembourg have cottage industries in helping companies or individuals avoid paying tax. Localised profits, socialised losses...
posted by are-coral-made at 1:26 AM on February 11, 2020 [1 favorite]


« Older “outsiders, misunderstood by the haughty...   |   When You Put It On, Something Happens. Newer »


This thread has been archived and is closed to new comments