Some integration challenges to contemplate before we commit to a blockchain electricity trading future

Hardly a week goes by these days without an article announcing that blockchain is going to totally transform the way electricity markets operate.  The technology shows huge potential, and there is wide range of processes where it might be applied (or is even already in operation at a small scale). The greatest hype is around its use as the platform for all electricity trading on the connected grid, often presented with a premise along the lines of “kiss goodbye to your electricity retailer”. Given that electricity trading sits on top of a physical power system, and is just one of the processes that make the system and market function, how easy is it going to be to integrate blockchain trading, and can it be as transformational as hoped?

This broader systems view has been missing from the articles I’ve seen, so I thought it might be useful to start a discussion by identifying some of the potential challenges in the context of the NZ market: 

1. Are you sure those are your electrons?

The power system is a ‘common pool’ – everyone puts their electricity into, and takes it from the same big bucket and you have no means of determining that your customer has consumed ‘your’ electricity. Everyone participating in the market relies on everyone else for ensuring there is an accurate record of what went in and what came out, and that these balance (‘reconciliation’). 

In NZ we currently manage this by making the retailer (‘trader’) responsible for accurate metering (through contracting a suitable metering equipment provider) and providing accurate volumes into the reconciliation process. If we want this same level of veracity to continue (and veracity and accuracy seems to be one of the key selling points for blockchain), either the blockchain provider has to become a certified trader, or the consumer still ends up having to use a retailer or some other agent(s) to meet these requirements. (That said, rationalising the essentially identical reconciliation processes currently undertaken by individual traders into a central system would seem to make sense from an overall cost and efficiency perspective). 

We also need to remember that power systems suffer losses as electricity is moved between generators and consumers. In an electricity market you can throw in some additional losses from factors such as input errors and theft (though these are reducing as the smart meter rollout progresses). While each consumer’s share of these losses will be tiny, they become consequential as more and more consumers are supplied via the blockchain. These losses will need to be accommodated in the blockchain, and can presumably be readily embedded in smart contracts, but the market (and regulator) will need to have confidence this is being done correctly.

2. We care about power quality, and not just energy

While consumers buy electricity in units of energy (kWh), they also expect that electricity to be delivered at the right frequency and voltage, and to be resilient to supply disturbances. The price they pay for their electricity needs to incorporate this quality component, including the reserves, frequency keeping, voltage support, etc. (‘ancillary services’) that are currently purchased at an aggregate system level through the system operator. While these services will need to change over time, the basic requirement for power quality is not going to go away.

In the future residential consumers will likely provide some of these services, and will expect this to be managed alongside their energy trading within the trading platform. This throws up challenges around how the information can be provided from the trading platform up to the system operator at an aggregate level, how they optimise the purchases to achieve the lowest total cost, and how they operationalise their decisions, all within the timeframes necessary to ensure the safe and reliable operation of the power system. While computing power continues to increase rapidly, is it going to be fast enough, and secure enough for operational purposes? (Having made the decisions, the payment process looks like it would probably be relatively straightforward).

3. A transactional grid still needs the network

Getting power between one consumer and another is going to require the use of the distribution network(s), with some associated costs (and we are now starting on a path towards these costs reflecting the specific services provided). At the moment, retailers are the parties that contract for these services on behalf of consumers, and who deal with the commercial nitty gritty. Making the payment transactions work within blockchain seems simple enough, but who deals with all of the commercial stuff? The blockchain provider? Or each customer through standard terms and conditions? The latter seems more likely, but will require a change in mindset and operating model for the networks.

4. Who you gonna call?

It is inevitable that somewhere down the line you are going to lose your power supply as a result of a planned or unplanned outage somewhere on the system. In extreme cases (for example the Penrose substation fire in 2015) this might even be for a few days. At the moment, the party that you deal with as a consumer in this sort of situation is your retailer. How is that going to work in the world of blockchain when your energy provider is another consumer somewhere on the system (and potentially a different one for each half hour trading period)? Again, in the absence of a retailer the network company seems the most likely option, but will require a reasonably significant change to processes and operations.

5. It is possible to consume electricity without paying for it

The blockchain establishes a running ledger of who needs to pay whom, but would seem to rely on every consumer maintaining positive credit, i.e., there is a direct debit from the consumer’s account to the generator’s account and the transaction does not proceed if there are insufficient funds available. The problem is that it is difficult to prevent the consumer from using the electricity if they don’t have the funds to pay for it. In theory, the remote disconnection capability of smart meters provides a means of turning off the supply to those consumers who aren’t making their payments, but past experience tells us that this is not an approach you’d want to take indiscriminately.

If this is not how it works in practice, then either our market clearing and settlement systems and processes need major changes to manage the payments (and absence of payments) from and to all consumers, or we are back to having agents filling the credit management role that retailers play today.

 What have I missed?

These are five broad areas that raise questions for me about the practical application of blockchain energy trading in NZ. This will not be a complete list – there will be other things to consider such as how to accommodate consumers without smart meters – but it seems sensible to start a discussion on practical considerations before we get too carried away with the hype. This is not to say that it won’t work, and the discussion also relates to only one of the many potential uses of blockchain in the sector. An interest group already seems to have been established to look at this potential more closely, and it will be interesting to see what emerges from their discussions. If anyone from that group, or elsewhere around the sector wants to discuss this post in more detail, please get in touch, otherwise I look forward to any feedback in the comments below.