4 ms·

The article's about two months old - but I thought it was a fascinating idea. I know that one of the challenges of harnessing renewable energy is battery techn

The article's about two months old - but I thought it was a fascinating idea. I know that one of the challenges of harnessing renewable energy is battery technology not being up to he task (yet), but I'd never thought about creating a grid large enough that demand is smoothed out over a large region.

reacweb 10y ago
Could it be feasible at the scale of the planet ?
- binalpatel 10y ago
  I'm not an expert - so I can't say for sure. The article does mention that Japan apparently runs two different "grids" that are so different, that it's close to impossible to connect the two. I'd imagine that issues like that would compound as more and more countries are added?
  - msl 10y ago
    Not every two countries use mutually incompatible mains systems; see for example [0] to see a real time map showing an electricity market with multiple nations as participants. Note also, that your AC grids do not need to be compatible to transfer power between them, and even if they are, you might want to convert to and from DC in between - see [1] for examples of this. I have no idea why it is impossible for the Japanese connect their grids, but I am guessing the problem is not one of technology. Edit: Reading the Wikipedia article on the subject [2] it seems that the incompatible Japanese grids are indeed interconnected to a degree. It also seems that the electricity market in Japan being currently liberalized, which I take to mean that it has traditionally not been particularily open for competition. That might explain why nobody has been in a hurry to improve the system as a whole. [0] http://www.statnett.no/Kraftsystemet/ http://www.statnett.no/Kraftsystemet/ [1] https://en.wikipedia.org/wiki/High-voltage_direct_current https://en.wikipedia.org/wiki/High-voltage_direct_current [2] https://en.wikipedia.org/wiki/Electricity_sector_in_Japan https://en.wikipedia.org/wiki/Electricity_sector_in_Japan
  - reacweb 10y ago
    It seems 50Hz and 60 Hz are the only relevant standards: https://en.wikipedia.org/wiki/Utility_frequency#Standardization https://en.wikipedia.org/wiki/Utility_frequency#Standardizat...
  - rurounijones 10y ago
    Japan has two grids that run at different frequencies. (50hz and 60hz) They are actually connected but can only transfer about 2000MW from one to the other via the one connection point. They have plans to add more interconnects starting in 2020.
- peller 10y ago
  I suspect the transmission losses would be too great for it to make too much sense.[1] But even if that weren't a problem, well, politics. [1] https://en.wikipedia.org/wiki/Electric_power_transmission#Losses https://en.wikipedia.org/wiki/Electric_power_transmission#Lo...
  - treehau5 10y ago
    Not just politics, security as well. One bad actor could wipe out the power grid for most of the world's superpowers.
    - binalpatel 10y ago
      How would that hypothetical work out? I know that the grids - as they stand right now - must balance generation and consumption. Would a large outage (or large spike) in any given part of the interconnected grids be able to take out all of it?
      - msl 10y ago
        The Wikipedia article on Cascading failure [0] lists a few examples of this kind of a large scale grid failure. You start by taking one (or some) power plants down, thus overloading others trying to reblance the system, and possibly failing too. Such a chain reaction can, over multiple generations, take a great many power plants down. [0] https://en.wikipedia.org/wiki/Cascading_failure https://en.wikipedia.org/wiki/Cascading_failure
  - perfunctory 10y ago
    from the article: Nine regional interconnected grids are already functioning within China, so the technical feasibility of transmitting electricity across long distances at high voltages has already been proven.
    - peller 10y ago
      Eh yea, but there's an economic feasibility limit too. From the Wiki article I linked: "As of 1980, the longest cost-effective distance for direct-current transmission was determined to be 7,000 km (4,300 mi). For alternating current it was 4,000 km (2,500 mi), though all transmission lines in use today are substantially shorter than this." [It's unclear when "today" means, or at what voltage these lines were running. But the 1000KV range this renewable grid is running is pretty typical, and, is in the ballpark of these quoted distances.] For reference, the circumference of earth is about 40,000km (as parent was pondering). The thing is, transformers and wire haven't exactly changed much since 1980. So until we can figure out room-temperature super conductivity, or a more efficient way of massively scaling up/down voltages than transformers, I'm inclined to believe the situation hasn't changed much since then. Then again, I'd love to be proved wrong.
- Retric 10y ago
  You can link US / Russia with a 50 mile under sea electric cable. Which would join, Afro-Eurasia with the Americas fairly easily. However, local storage (https://en.wikipedia.org/wiki/Pumped-storage_hydroelectricity https://en.wikipedia.org/wiki/Pumped-storage_hydroelectricit...) is going to be cheaper than shifting solar power around the globe. Further, while you can connect any two grids (worst case generator / alternator pair) that does not make it cost effective.
  - maxerickson 10y ago
    The Bering Strait? Easily? There's all sorts of things that are easy if you focus on an easy part that doesn't accomplish anything.
    - Retric 10y ago
      You may have missed my point. Under sea cables that distance are proven tech. EX: (https://en.wikipedia.org/wiki/Isle_of_Man_to_England_Interconnector https://en.wikipedia.org/wiki/Isle_of_Man_to_England_Interco... a 40 MW 65 mile under sea cable.) However, my point was geography was not the problem.
      - maxerickson 10y ago
        I was thinking more about the thousands of miles separating the straight from any large groups of people (on both sides). Yes, crossing the straight is nicely feasible. Putting power to that crossing is not so nicely feasible.
        Retric 10y ago
        Sure, sending 100+GW of power though there would be ridiculously expensive and a huge waste of resources. However, there are some people in the area so sending "nominal" amounts of power would not be an issue. I do wonder how connected these areas are to a wider power grid. It's probably cheaper to just have local power generation than run power cables to these remote areas I mean https://en.wikipedia.org/wiki/Wales,_Alaska https://en.wikipedia.org/wiki/Wales,_Alaska has 150 people. ED: Anyway, it's an economic not a technical problem.
- carapace 10y ago
  Bucky Fuller promoted the idea of a World Grid. One of his foundational principals is that as the system you design for gets larger you can engineer greater efficiencies. I think a "mesh network" for power would be better than a global "client/server" format, but economics, politics, etc...
  - kefka 10y ago
    Solar proves the opposite, as it is hyper-local. Why bother with remote power generation and such, when I can hang stuff on my roof and get free energy? (I am for large scale clean power generation, along with solar panels and other local tech. Just the ones with big money aren't - you know, the ones that make the laws and pass them on to Congress.)
    - hx87 10y ago
      > Why bother with remote power generation and such, when I can hang stuff on my roof and get free energy? Because the sun is down for about half a day each day. However, halfway across the world, the sun is shining. Wouldn't it be nice if you could get power from there?
    - _ph_ 10y ago
      Sun has two large problems: a) the nights and b) bad weather. A) is quite predictable and usually electricity consumption is lower in the night but b) needs a fix. So large scale grids can not only extend the time per day solar energy is available but also compensate for bad weather. Over large enough areas, the total solar output is more constant. Add in a good amount of wind energy a pretty constant load should be bearable with renewal energies with very little of storage.
  - carapace 10y ago
    *Principles D'oh!

reacweb 10y ago

Could it be feasible at the scale of the planet ?

binalpatel 10y ago
I'm not an expert - so I can't say for sure. The article does mention that Japan apparently runs two different "grids" that are so different, that it's close to impossible to connect the two. I'd imagine that issues like that would compound as more and more countries are added?
- msl 10y ago
  Not every two countries use mutually incompatible mains systems; see for example [0] to see a real time map showing an electricity market with multiple nations as participants. Note also, that your AC grids do not need to be compatible to transfer power between them, and even if they are, you might want to convert to and from DC in between - see [1] for examples of this. I have no idea why it is impossible for the Japanese connect their grids, but I am guessing the problem is not one of technology. Edit: Reading the Wikipedia article on the subject [2] it seems that the incompatible Japanese grids are indeed interconnected to a degree. It also seems that the electricity market in Japan being currently liberalized, which I take to mean that it has traditionally not been particularily open for competition. That might explain why nobody has been in a hurry to improve the system as a whole. [0] http://www.statnett.no/Kraftsystemet/ http://www.statnett.no/Kraftsystemet/ [1] https://en.wikipedia.org/wiki/High-voltage_direct_current https://en.wikipedia.org/wiki/High-voltage_direct_current [2] https://en.wikipedia.org/wiki/Electricity_sector_in_Japan https://en.wikipedia.org/wiki/Electricity_sector_in_Japan
- reacweb 10y ago
  It seems 50Hz and 60 Hz are the only relevant standards: https://en.wikipedia.org/wiki/Utility_frequency#Standardization https://en.wikipedia.org/wiki/Utility_frequency#Standardizat...
- rurounijones 10y ago
  Japan has two grids that run at different frequencies. (50hz and 60hz) They are actually connected but can only transfer about 2000MW from one to the other via the one connection point. They have plans to add more interconnects starting in 2020.
peller 10y ago
I suspect the transmission losses would be too great for it to make too much sense.[1] But even if that weren't a problem, well, politics. [1] https://en.wikipedia.org/wiki/Electric_power_transmission#Losses https://en.wikipedia.org/wiki/Electric_power_transmission#Lo...
- treehau5 10y ago
  Not just politics, security as well. One bad actor could wipe out the power grid for most of the world's superpowers.
  - binalpatel 10y ago
    How would that hypothetical work out? I know that the grids - as they stand right now - must balance generation and consumption. Would a large outage (or large spike) in any given part of the interconnected grids be able to take out all of it?
    - msl 10y ago
      The Wikipedia article on Cascading failure [0] lists a few examples of this kind of a large scale grid failure. You start by taking one (or some) power plants down, thus overloading others trying to reblance the system, and possibly failing too. Such a chain reaction can, over multiple generations, take a great many power plants down. [0] https://en.wikipedia.org/wiki/Cascading_failure https://en.wikipedia.org/wiki/Cascading_failure
- perfunctory 10y ago
  from the article: Nine regional interconnected grids are already functioning within China, so the technical feasibility of transmitting electricity across long distances at high voltages has already been proven.
  - peller 10y ago
    Eh yea, but there's an economic feasibility limit too. From the Wiki article I linked: "As of 1980, the longest cost-effective distance for direct-current transmission was determined to be 7,000 km (4,300 mi). For alternating current it was 4,000 km (2,500 mi), though all transmission lines in use today are substantially shorter than this." [It's unclear when "today" means, or at what voltage these lines were running. But the 1000KV range this renewable grid is running is pretty typical, and, is in the ballpark of these quoted distances.] For reference, the circumference of earth is about 40,000km (as parent was pondering). The thing is, transformers and wire haven't exactly changed much since 1980. So until we can figure out room-temperature super conductivity, or a more efficient way of massively scaling up/down voltages than transformers, I'm inclined to believe the situation hasn't changed much since then. Then again, I'd love to be proved wrong.
Retric 10y ago
You can link US / Russia with a 50 mile under sea electric cable. Which would join, Afro-Eurasia with the Americas fairly easily. However, local storage (https://en.wikipedia.org/wiki/Pumped-storage_hydroelectricity https://en.wikipedia.org/wiki/Pumped-storage_hydroelectricit...) is going to be cheaper than shifting solar power around the globe. Further, while you can connect any two grids (worst case generator / alternator pair) that does not make it cost effective.
- maxerickson 10y ago
  The Bering Strait? Easily? There's all sorts of things that are easy if you focus on an easy part that doesn't accomplish anything.
  - Retric 10y ago
    You may have missed my point. Under sea cables that distance are proven tech. EX: (https://en.wikipedia.org/wiki/Isle_of_Man_to_England_Interconnector https://en.wikipedia.org/wiki/Isle_of_Man_to_England_Interco... a 40 MW 65 mile under sea cable.) However, my point was geography was not the problem.
    - maxerickson 10y ago
      I was thinking more about the thousands of miles separating the straight from any large groups of people (on both sides). Yes, crossing the straight is nicely feasible. Putting power to that crossing is not so nicely feasible.
      - Retric 10y ago
        Sure, sending 100+GW of power though there would be ridiculously expensive and a huge waste of resources. However, there are some people in the area so sending "nominal" amounts of power would not be an issue. I do wonder how connected these areas are to a wider power grid. It's probably cheaper to just have local power generation than run power cables to these remote areas I mean https://en.wikipedia.org/wiki/Wales,_Alaska https://en.wikipedia.org/wiki/Wales,_Alaska has 150 people. ED: Anyway, it's an economic not a technical problem.
carapace 10y ago
Bucky Fuller promoted the idea of a World Grid. One of his foundational principals is that as the system you design for gets larger you can engineer greater efficiencies. I think a "mesh network" for power would be better than a global "client/server" format, but economics, politics, etc...
- kefka 10y ago
  Solar proves the opposite, as it is hyper-local. Why bother with remote power generation and such, when I can hang stuff on my roof and get free energy? (I am for large scale clean power generation, along with solar panels and other local tech. Just the ones with big money aren't - you know, the ones that make the laws and pass them on to Congress.)
  - hx87 10y ago
    > Why bother with remote power generation and such, when I can hang stuff on my roof and get free energy? Because the sun is down for about half a day each day. However, halfway across the world, the sun is shining. Wouldn't it be nice if you could get power from there?
  - _ph_ 10y ago
    Sun has two large problems: a) the nights and b) bad weather. A) is quite predictable and usually electricity consumption is lower in the night but b) needs a fix. So large scale grids can not only extend the time per day solar energy is available but also compensate for bad weather. Over large enough areas, the total solar output is more constant. Add in a good amount of wind energy a pretty constant load should be bearable with renewal energies with very little of storage.
- carapace 10y ago
  *Principles D'oh!