Presentation structure

• Introduction:
  • What problem? (imbalance caused by unpredictable consumers and porudercs. consumers becoming producer + intermittent sources vs old fashioned powerplant-household configuration:
  • Why a problem? (introduction in to the grid; there is a physical limit on capacity)
    • Physical grid is already clusters (kind of tree).
    • On the lowest level the imbalances are the smallest, not yet a problem
    • However combined they become large ->overload inter-cluster connections
    • Solve the problem by the root, first try to balance in the cluster, then on 1 level higher, and so on.
• The balancing (easier said than done):
  • Explain notion of backup supplier:
    • Drop-in replacement; can't cut people off, different price production consumption, maybe even cap the production (very likely in the future; 2020).
  • Clients submit their baseline consumption to keep the household running (all of this by a software agent on their behalf). This is guaranteed; first locally else backup suppliers.
  • Next to that also their flexibility: Don't really need to charge my car yet, but if it's cheaper then backup currently might as well do it now. Or might stop charging my car if someone needs it more than me (i.e. willing to pay more for it than it costs me). smart appliances, (fridges, overnight laundry stuff etc.) but also new businesses specialized in storing energy can join and trade energy (cost of wear and tear vs current price etc.)
  • All of this is information enters the 'energy exchange market'. Loosely based on stock market. Buy price is maximum price, sell is minimum and you settle in the middle so it's fair for both (always pay less then you want, or get more than you asked). Bigger overlap is bigger chance of match making. That's why (just as the stock market) we first match highest buyer with lowest seller. (they generate overlap and therefore enable more matching). also because highest buyer is matched with lowest, seller he might pay less than the second highest buyer. (buy 9 + sell 1 = 5, buy 8 + sell 4 = 5,5). Or the second highest buyer might not even match at all.
  • After possible trades are made in a cluster, data is aggregated (minimize data flow) and remaining flexibility is send up, to see if energy can be traded intra-cluster, if nit move up until root.
  • All of this is implemented in a 100% deterministic way so that every unique set of incoming order always generate the exact same set of trades in the exact same order (and no rounding or float errors). Important for consensus later on.
• The blockchain part:
  • Why blockchain:
    • Multiple parties involved in the grid; but not necessarily trust eachother, immutable data; tamper resistant
  • Why tendermint:
    • Proof of stake (proof of work defies the purpose: waste energy to solve energy problems)
    • Permissionsed
    • Agnostic of application but does implement all of the consensus reaching and messageing stuff

https://www.cpb.nl/sites/default/files/omnidownload/CPB-Beantwoording-vragen-verschillen-Regeerakkoord-en-doorrekening-CPB.pdf (source unlinknig productionc onsumption price at 2020)