At periods of peak electricity demand, the electricity grid comes under strain. Some of the most expensive power plants turn on to meet this demand. What if instead of turning these expensive power plants on we reduce peak load?
This is the idea behind the National Grid’s Demand Flexibility Service. When Loop’s users use less electricity during Turn Down and Save events, the National Grid pays Loop for reducing the cost of running the grid. Loop passes this on to our customers based on how much they reduced their use.
How does Loop calculate your reduction in usage?
Loop’s calculations follow the industry standard “BSC P376 methodology”. This is prescribed by the National Grid.
The idea is to use your historic smart meter data to estimate how you typically use electricity at the time of the Turn Down and Save event. This typical usage is called your “baseline”. For example, if a Turn Down and Save event is from 5pm to 6pm on a Monday, the calculation looks at how you’ve used electricity between 5pm and 6pm on the previous 10 weekdays (excluding bank holidays and days with Turn Down and Save events) to calculate your baseline.
This sounds simple enough. Yet, the methodology then adds in two extra complications.
The "on-the-day adjustment"
The motivation for an “on-the-day adjustment” is to take into account external conditions that mean your usage would be higher (or lower) than your baseline. For example, on cold days homes with electric heating use much more electricity compared to warmer days. The on the day adjustment aims to compensate for this weather effect.
However, there's no exact way to do this - it will only ever be a guess.
The on-the-day adjustment looks at the period between 1 and 4 hours before the Turn Down and Save event. For an event starting at 5pm this is the hours between 1pm and 4pm. The algorithm compares on-the-day usage against typical use during these hours.
If more electricity than normal is used then the baseline is adjusted up. If less electricity than normal is used then the baseline is adjusted down.
There are two big problems with this approach.
The first is that if a user reduces their use throughout the day rather than just during the event they get heavily penalised. An example of this is a home that typically runs at 500Watts throughout the afternoon and evening. They decide to reduce this to 100Watts to take part in Turn Down and Save but they make this reduction throughout the afternoon and evening. As the house has reduced its use by 400Watts during the hours leading up to the event, their on-the-day adjustment reduces their baseline from 500Watts to just 100Watts. They then use 100Watts through the event. Relative to their baseline this comes out as no reduction and they receive no reward.
This often isn't a fair adjustment. The algorithm tells customers who've made large reductions that in fact they've made no reduction. Understandably this has lead to customers complaining to Loop.
The other disadvantages is that the on-the-day adjustment can't be calculated before the event. So Loop cannot set a target usage ahead of the event. In our opinion, this is a major flaw.
Half-hourly timescales
The second complexity is that the electricity system runs on half-hourly timescales. A one-hour Turn Down and Save event covers two half-hour blocks. The National Grid has decided to calculate savings per half-hour, then add up to get to total savings. So rather than calculating a single baseline for an hour-long Turn Down and Save event, we calculate one for each half-hour.
For example, suppose a user’s baseline is 1kWh for both the first and second half-hours of an event. If they use 0kWh in the first half-hour and 2kWh in the second half-hour then overall they’ve not reduced their usage. But because the savings are calculated half-hourly, they are rewarded for the 1kWh saved in the first half-hour.
This decision makes communicating savings much more difficult. It breaks the direct link between energy saved over an event and the reward a customer receives.