Smart Meters (Do they go far enough?)

I am convinced that a major revolution is about to occur in the way electrical grids operate. I am worried that the groups running the Smart Meter pilots in Ireland could miss an opportunity to be ahead of the curve.

Electricity generation and supply has, since the time of Edison, been about meeting inflexible but predictable demand with flexible supply. This inflexible demand has been met by huge generation plant which is dispatched in synchronisation with the predicted demand. This description of electricity networks is about to change.

1) A growing percentage of energy will be generated by non dispatchable supply such as wind, wave and solar.

2) Microgeneration at the domestic level is going to make demand less predictable.

3) Electricity consumption will increase dramatically. As we move away from fossil fuels in our homes we will use more electricity for heat pumps and for automotive (EV and PHEV). This will result is a doubling of our electrical consumption, even if we hold our overall energy consumption static, by the time we achieve our 33% renewable target in 2020. These new loads will be shiftable in time because of the use of hot water and battery storage.

These three changes will make the electrical grid partially inflexible on the suppply side. To compensate for this inflexibility on the supply side flexibility on the demand side will be required. So I believe that the Electric Grid 2.0 will look like this.

1) There will be a realtime price for electricity broadcast and available to every consumer in real time. Currentlyy the wholesale markets operate in this way. Soon the retail markets will follow suit.

2) There will be a spread between the purchase and sale price for electricity. You can decide to produce and sell or buy and consume at any time.

3) Standard time of day tariffs (STOD) will completely disappear. It will make more sense to have tariffs based on wind availablility. For example on 30 March and 16 April this year the wholesale price profile was inverted. On those days wholesale electricity was dearer at night than during the day.

4) Realtime response based on frequency will be widely implemented across the grid to stabilise short term supply and demand mismatch while price signals are being propogated.

5) Demand response will be mostly automated and not depend on user behaviour.

An unusual visionary in this change is Google.

“I could imagine a smart garage where I would plug in my car and the computer handles it. I could even make money by cost shifting….  It solves energy security, energy prices and job creation… and by the way, climate change.”  Eric Schmidt CEO of Google.   Google and GE have formed an alliance to exploit smart grid concepts.

It is vital that Ireland install smart meters that have 15minute power measurement intervals and can cope with real time pricing and not STOD tariffs.

Even though these ideas will apply globally, Ireland is aspecial case. We will, by 2020, have 30% of our energy coming from wind and we ahave a tiny grid size. We cannot copy others in thisregard. We must lead.

I am afraid that we will miss an opportunity.

Advertisements

2 Responses to Smart Meters (Do they go far enough?)

  1. Dundevan says:

    Regarding your comments on electric Grid 2.0;

    1) Real-time Pricing
    The biggest challenge in enabling this is not the meter itself, but the capability of Billing engines to deal with this level of variability across millions of customers. This no doubt may be possible in future but currently is way beyond the ability of today’s systems anywhere in the world. Wholesale Markets don’t have to deal with this issue as there are only a small number of buyers in the wholesale market.

    2) Spread between Purchase and Sale Price
    Laws of economics dictate that selling price and buying price will seek to converge. In general I expect that if sustainable generation ( incl. CHP ) is available it will be economically viable to export it to the grid.

    3) Standard Time of Day Tariffs
    Whereas wind availability will play a role in determining price, the SMP (System Marginal Price ) which determines wholesale prices is determined by matching demand with the cost of the highest priced generation dispatched, hence until a significant proportion of generation is wind based the impact on wholesale price will remain minimal.

    4) Real-time Response based on frequency
    This is the system on which the system is currently stabilised using thermal generation plant. Higher usage of Wind generation impacts the ability to use frequency response of thermal plant to stabilise the system.

    5) Demand Response Automated
    Agreed, however in Ireland unlike in California and Canada where Automated Demand Response has been tested, the extent of “discretionary” load is unlikely is to be as effective.

    6) 15 minute intervals
    Initial meters being installed for trials in Ireland are based on 30 minute intervals aligned with the 30 minute wholesale trading interval.

    NOTE – The extent of the Smart Meter programme for Ireland is such that it will position Ireland well “ahead of the curve”.

  2. ciablasar says:

    Hi Dundevan, I reply as follows:-

    1) The entire database size per month of meter readings for 1,000,000 meters (total Irish market) could be calculated as follows:
    1,000,000 meters x 48 billing periods x 31 days x 10 Bytes = 15GBytes
    Performing a billing calculation on this is trivial. I am in contact with a company in Hydrabad that do similar billing systems for mobile phone companies and thay are looking for business. The output of their software can be fed to ERP systems such as Oracle and SAP to do customer billing. Costs of implementation is under €100,000.

    2) Absolutely. That is what I want to encourage. I am all in favour of democratising the grid. Anyone should be able to buy or sell electricity at any time based on the current market spread.

    3) Yes, and that floor is currently around 6 cent per kWh. I would be very happy to drive the price to that floor more often and to get rid of the more than fifty spikes that occured on the SEM this year so far above 25 cent. I don’t see how this point relates to STOD tariffs. My point is that as we move to sometimes having 150% of our electricity needs met by wind (July windy night 2021) STOD tariffs will not make sense. Even at the moment they don’t fully make sense. For a reason I can’t explain the SEM D+4 price was settled on 30 March and 16 April with an inverted price curve. The price of electricity on those two days was dearer at night than during the day. STOD does not take thermal plant failures or renewable availability or interconnect availability into account.

    4) I am talking about doing frequency on the demand side as per companies like RLTec and Dynamic Demand in the UK. I am nottalking about the current systems on the supply side.

    5) I estimate that EV, PHEV and heat pumps (all used in fossil fuel replacement scenarios) must add at least 3GW to our national load if we are to hold our overall energy levels constant but move to 30% renewables by 2020. These loads are very moveable. Remember out automotive energy needs are three times our electricity energy needs. And our non electric heating needs are almost equivalent to our total electricity load.

    6) To paraphrase George Orwell, 30 minutes good, 15 minutes better, I like 15 minutes because it gives you more granular control over weather affected market pricing systems.

    Ahead of what curve? If the project is limited to STOD tariffs an opportunity to develop a real time market will be lost.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s

%d bloggers like this: