Solar panels and battery - any real world reccomendations?

94JDH said:
Nearly, 10 day average for the two hour session but only on the day of the session for the 3 hours window. Apparently there is an algorithm that varies the benefit based on 1, 2, 3 hours etc session - in this case it works on around 2/3 of the usage difference.
Click to expand...

I don't think it is.
before we had the technical definition rather than the power point which was in some weird semi coded language and equations.
However I think below bolded is key, it infers like we thought before that same days baselines would be used as the days used to calculate the actual usage.

"To correct for all day specific effects such as weather conditions the baseline profile will be in day adjusted by using actual Metered Data over the three hour period up until one hour before the relevant Settlement Period where the unit is delivering the service. This will be compared to the calculated baseline values and an additive adjustment applied to ensure that the profile created by the baseline best matches real data for the run up to the Settlement Period."

The net effect of 10 very low "normal" days or one "normal" day vs a highly increased single 3 hour period prior to the actual session is going to be very similar. Otherwise what are they comparing to to calculate the adjustment.
As ever they get a little wonky with language since you could take the very first statement "To create a Baseline Profile, at unit level, the last 60 days of actual metered data is used" as that when they are doing the in day check vs baseline they are in fact referring to this 60 day actual metered data.
However reading this (below) and per the original semi understandable version I am pretty convinced its the same benchmark being used (last 10 days of weekday non saving session usage for a weekday, or last 4 weekend days for a weekend one)
 
Mercenary Keyboard Warrior said:
I don't think it is.
before we had the technical definition rather than the power point which was in some weird semi coded language and equations.
However I think below bolded is key, it infers like we thought before that same days baselines would be used as the days used to calculate the actual usage.

"To correct for all day specific effects such as weather conditions the baseline profile will be in day adjusted by using actual Metered Data over the three hour period up until one hour before the relevant Settlement Period where the unit is delivering the service. This will be compared to the calculated baseline values and an additive adjustment applied to ensure that the profile created by the baseline best matches real data for the run up to the Settlement Period."

The net effect of 10 very low "normal" days or one "normal" day vs a highly increased single 3 hour period prior to the actual session is going to be very similar. Otherwise what are they comparing to to calculate the adjustment.
As ever they get a little wonky with language since you could take the very first statement "To create a Baseline Profile, at unit level, the last 60 days of actual metered data is used" as that when they are doing the in day check vs baseline they are in fact referring to this 60 day actual metered data.
However reading this (below) and per the original semi understandable version I am pretty convinced its the same benchmark being used (last 10 days of weekday non saving session usage for a weekday, or last 4 weekend days for a weekend one)
Click to expand...
I found this elsewhere, not say it's correct but certainly plausible and pretty much confirms my values :

The IDA is added 100% to the baseline average. It's just that the 3 hour (6 x 30 minutes) average for the IDA is applied to whatever length the saving session is. In the case of the 12 December session that was 2 hours (4 x 30 minutes) so the total usage in the IDA reference window was counted at 2/3 in this particular session whereas it was only 1/3 for each of the previous 1 hour saving sessions.

Should there ever be a longer session then the effect would be amplified further. A 4 hour session would count 4/3 of the IDA usage.
 
94JDH said:
I found this elsewhere, not say it's correct but certainly plausible and pretty much confirms my values :

The IDA is added 100% to the baseline average. It's just that the 3 hour (6 x 30 minutes) average for the IDA is applied to whatever length the saving session is. In the case of the 12 December session that was 2 hours (4 x 30 minutes) so the total usage in the IDA reference window was counted at 2/3 in this particular session whereas it was only 1/3 for each of the previous 1 hour saving sessions.

Should there ever be a longer session then the effect would be amplified further. A 4 hour session would count 4/3 of the IDA usage.
Click to expand...

I don't think thats right at all, certainly isnt my take from either of the versions I have seen

Remember the provider are doing this for each individual but also for the total for all those signed up. They cant just add ALL the usage in that period it would be wrong.

I think if you look at the graph it clearly shows this is incorrect since the in day (red) line is below the (mustard?) baseline line. You cant simple add usage from 3 hours to a baseline and get a lower number unless the usage is negative.

Honestly I think its the same data set. So tak ethe 3 hour window, compare actual to baseline to create in day adjustment (ie does it look like usage is high or low today) and use that to flex the actual target.
It was unclear before and remains so now, but my suspicion is its a simple calc.
Say you baseline for the 3 hours was 1kwh, and you make that 11kwh on the session day, they do something (again was unclear before) and add(or subtract) this from the baseline.
My suspicion is its a % calc, so going back to 11 and 1, they would say 11/1 = 1100%, apply 1100% to each baseline saving number in the 30 minute slots. Eg say 17:00 was slot 1 and you used 1.5kwh average in that 30 minutes, they would baseline that to be 1.5x1100% = 16.5kwh for addition to baseline for comparison to actuals.

Oh P376 baselining methodology is the technique if you want to look it up but its a bit nasty unless your used to the terminology and maths equations.
 
Last edited:
math-hangover.gif
 
All I know is my batteries were low so charged in the hours before and went from earning 6p to £6 profit…. I definitely know what to do next time though
 
Last edited:
Mercenary Keyboard Warrior said:
I don't think thats right at all, certainly isnt my take from either of the versions I have seen

Remember the provider are doing this for each individual but also for the total for all those signed up. They cant just add ALL the usage in that period it would be wrong.

I think if you look at the graph it clearly shows this is incorrect since the in day (red) line is below the (mustard?) baseline line. You cant simple add usage from 3 hours to a baseline and get a lower number unless the usage is negative.

Honestly I think its the same data set. So tak ethe 3 hour window, compare actual to baseline to create in day adjustment (ie does it look like usage is high or low today) and use that to flex the actual target.
It was unclear before and remains so now, but my suspicion is its a simple calc.
Say you baseline for the 3 hours was 1kwh, and you make that 11kwh on the session day, they do something (again was unclear before) and add(or subtract) this from the baseline.
My suspicion is its a % calc, so going back to 11 and 1, they would say 11/1 = 1100%, apply 1100% to each baseline saving number in the 30 minute slots. Eg say 17:00 was slot 1 and you used 1.5kwh average in that 30 minutes, they would baseline that to be 1.5x1100% = 16.5kwh for addition to baseline for comparison to actuals.

Oh P376 baselining methodology is the technique if you want to look it up but its a bit nasty unless your used to the terminology and maths equations.
Click to expand...
I've gone back through my previous 10 weekdays and have no usage from the Grid in that 2 hour period ( only from 4 hour cheap Go period)
I agree something is making a difference, it's definitely not the whole 36kWh or so that I used as a payment as the value only reflects around 25kWh worth of the total used.
 
94JDH said:
I've gone back through my previous 10 weekdays and have no usage from the Grid in that 2 hour period ( only from 4 hour cheap Go period)
I agree something is making a difference, it's definitely not the whole 36kWh or so that I used as a payment as the value only reflects around 25kWh worth of the total used.
Click to expand...

Yeah its doing something but unclear what exactly. Increasing usage in that window will increase your adjusted baseline for the actual savings windows.

Is it over compensating? looks likely, but overall its probably a fairly small impact, and its within the rules so I doubt they would really care that much
 
gpuerrilla said:
Thanks for the feedback. What is this bit?

Is there any way of knowing what export they are able to accept or is this just made up on the spot when their local engineer ganders at the local site?

Asking as we are rural but paid for our connection when building. We have 100amp service which splits off from a pole transformer assuming 11kVa. If you could reverse calculate what their max export would be allowed you could then just spec the inverter to that or cap a bigger one.
Click to expand...
Someone on my street within the past few years agreed to pay towards a transformer upgrade in our sub station. Some large detached houses have been rebuilt as flats so I’m guessing it may have been needed for that. Simply because that was paid, as a “second comer” who would benefit (nonsense really as my 100amp supply here hasn’t changed in 40 years) I was told I’d have to pay towards it.
The workaround was to cap export to 3.68kw.
 
94JDH said:
Nearly, 10 day average for the two hour session but only on the day of the session for the 3 hours window. Apparently there is an algorithm that varies the benefit based on 1, 2, 3 hours etc session - in this case it works on around 2/3 of the usage difference.
Click to expand...
This is very interesting. So for those of us with batteries who use a negligible amount from grid during the day, we should be importing/charging the max we can in the 3 hours before the window, and then zero during the two hour period, and based on the algo will get a weighted average “regular use” figure 2/3 weighted towards that 3 hour window 1/3 to the ten day average (close to 0). Is that correct?
 
SBo said:
Someone on my street within the past few years agreed to pay towards a transformer upgrade in our sub station. Some large detached houses have been rebuilt as flats so I’m guessing it may have been needed for that. Simply because that was paid, as a “second comer” who would benefit (nonsense really as my 100amp supply here hasn’t changed in 40 years) I was told I’d have to pay towards it.
The workaround was to cap export to 3.68kw.
Click to expand...

I am nearly there then, when we paid for the connection (and transformer) as we were the only user we were told if any new people add to that we would be reimbursed some of the cost. What I am interested in is as I am the only one on the line for that run wondering if there is no cap so to speak and it would allow greater than 3.68 inverter?
 
SBo said:
This is very interesting. So for those of us with batteries who use a negligible amount from grid during the day, we should be importing/charging the max we can in the 3 hours before the window, and then zero during the two hour period, and based on the algo will get a weighted average “regular use” figure 2/3 weighted towards that 3 hour window 1/3 to the ten day average (close to 0). Is that correct?
Click to expand...
Yes definitely weighted to the 3 hour window (don't use anything from the Grid or 1 hour before the start or during) - from the comments and replies above, we still aren't 100% sure on the algorithm that defines the return.
It obviously was a risk trying it as it may have been unsuccessful and ultimately charged the high rate for the amount used but it definitely worked.
However, there is no guarantee it may not be tweaked going forward to reduce the benefit.
 
I don't think they would tweak it quickly, its based on an accepted methodology they use for other reasons, just adopted to be used for this as well.

Personally my thought was and remains, I wont charge energy I wouldnt use that day. BUT if I would be using say 4kwh from grid from say 7pm-0:30am then I may as well charge that into my battery, and go zero from grid later on.
Its kind of gaming but I am not using any energy I wouldn't be using anyway. Oh plus the minor charge loss of course.
 
You must log in or register to reply here.
Back
Top Bottom