melonfool wrote:Octopus sound as if they are doing the same to you as Eon did to me a year ago - ignoring my readings, billing on estimates, then when I got them to use the readings dumped them all in at once which made the annual usage estimate go up about 8-fold.
It's taken a year to work that through.
I'm interested in this:
use of annual consumption estimates from the centralised industry balancing and settlement systems for the purposes of estimating consumptions for billing customers. Balancing and settlement values are inappropriate for billing when there is a significant change in consumption patterns such as when properties become or cease to become empty.
For the longest time my electricity estimated annual usage was 2572, and that figure is used to keep the bills level through the year. When they dumped in the historic readings it went to 16182 and they wanted £500pm (I spend about £75pm on combined fuels, on the current rates, which are higher than at that time!).
They've gradually come down over the year and in August hit exactly 2572 again. It's weird how they do it and it lands on the same number!
But then in Sept it jumped again to 5867. No idea why. The only difference is that I did have one whole day of using the oven. And I got a smart meter......so I suspect it's the system only using the smart meter recordings and they are now suffering from the same problem I had with my gas usage figures - too short a reference period. It seems to me they are not referencing back to the readings from the previous meyer. So I expect I have another year of this rubbish again!
They're asking for £400pm.
By my own calculations my annual average should be around 2000. When I try to explain this to them they just say "we don't do it that way". I have no idea what way they do it but I am confident that it doesn't work!
Mel
At a fairly superficial level there is an explanation for electricity of how the Balancing and Settlement system converts meter readings into an estimate of annual consumption that you can find buried in this presentation from Elexon -
https://www.elexon.co.uk/operations-set ... llocation/ Elexon is the electricity industry's central body which undertakes all the Balancing and Settlement calculations. The relevant part is the 10 minutes between 11:30 and 21:30, the rest is how these annual consumption estimates fit into balancing and settlement.
Domestic customers are non-half hourly (NHH). Because the Elexon presentation is aimed at industry insiders, I'll add my own explanations to supplement the rather dry commentary of the presentation. There are two types of annual consumptions involved.
First when a meter reading is obtained (and customer reads count just as much as reads by an actual meter reader), the consumption between this read and the preceding read is determined and then annualised by using the relevant profile. The profile comprises 365 daily consumption proportions which sum to 1. If, for example, the sum of the profile values for the period between the two meter readings is 0.306, and the metered consumption over this period is 670 kWh then the annualised value is 670/0.306 = 2190 kWh. This annualised value is called the annualised advance (AA). The AA is a property of the period between two successive meter readings, so the AA might change substantially at each meter reading dependent on whether or not your consumption pattern is well represented by the profile.
A second estimate called the estimated annual consumption (EAC) is also updated whenever a meter reading is obtained. The updated EAC is a weighted average of the newly calculated AA and the previous EAC. Like AA's, EAC's are also a property of the period between two meter readings, but unlike AA's, an EAC exists for the period after the most recent meter reading. Balancing and Settlement calculations are undertaken for each half-hourly period but repèated a number of times, the first calculation is undertaken shortly after period ends and then repeated a number of times up to a last calculation around (IIRC) around 15 months after the half hour. Each calculation aggregates the annual consumptions of all a supplier's NHH customers in a particular profile class, utilising for each customer the AA from the relevant period if it exists and the EAC if it doesn't. The later part of the Elexon video explains how the aggregated annual consumption of what it calls the "super customer" is then changed into the appropriate half hourly consumption value applicable in aggregate across all the supplier's customers.
With it so far? For electricity (unlike for gas), I do not know the detailed rules that apply to the calculations of AA's and EAC's: for example I don't know if there is a minimum or maximum permissible period between successive meter readings that must be satisfied for a new AA to be calculated. Similarly, I don`t know what weights are used when updating the EAC. However, if short periods are possible or too much weight is given to the AA when updating the EAC then the calculated values are likely to display some volatility. Mathematically, the use of EAC's alongside AA's suggests that volatility might be a problem. In addition, I have no idea at all how smart meters fit into all this - I suspect not very well at the moment - or what happens when a meter is exchanged.
AA's and EAC's exist to support balancing and settlement. However, supply companies have decided mainly because they do not do their own load research (development of understanding of patterns of consumption by different types of customers) to use the balancing and settlement information (profiles and AA/EAC values) to estimate consumptions for billing purposes. I think that's lazy but understandable when the focus has been on the-more-the-merrier competition, minimising costs and an absence of monitoring of the quality of estimates made for billing purposes (thank OFGEM for all of these). Whilst volatility in the AA's and EAC's of individual customers presents the sort of problems you have identified, at the aggregate level (used in balancing and settlement) this volatility diminishes to the point of unimportance. It is this difference in volatility at individual and aggregate level that prompted to my comment that "Balancing and settlement values are inappropriate for billing". For electricity, I'm tempted drop the qualification "when there is a significant change in consumption patterns such as when properties become or cease to become empty."
Gas, where I have more detailed knowledge, is a bit more sophisticated. For starters, there are more profile classes intended to represent different types of customers: those who use gas but not for space heating, those who use it for some space heating, for full space heating, etc. Second, there's only one type of annual consumption. It is the equivalent of electricity's AA and is only updated when a new meter read is taken if there is a preceding meter reading available between 9 and 15 months previously. The 9-15 month consumption is annualised in a similar way to electricity. Overall, the annual consumption estimates on the gas side suffer fewer volatility problems than electricity but the condition about significant changes in consumption patterns still makes them less than ideal for use in customer billing systems.
I'm sure you already know this but I'll write it anyway. I would stick to taking regular meter readings of your own, use these to determine your own estimates of annual consumption, use your estimates to rubbish your supplier's versions (and the suggested monthly payment) and, if they don't listen, raise a complaint and when that gets to failure to agree take it to the energy ombudsman. Do it by email, phoning call centre agents will get you nowhere when pressing simple technical arguments. As I noted upthread, there is a lot of abuse by energy companies in determining DD values. A bit of pushback is no more than they deserve.
modellingman