MCS PV Output Estimation and Shading

[The Solar King]

Some of you may remember that both myself and Gavin A wrote Mr Angry letters to Gemserve/MCS on this subject. (http://www.electriciansforums.co.uk...tallation-photovoltaic-systems-shading-3.html) Without boring you with the background, we got ourselves invited to the next MCS PV working group this coming Thursday to discuss the issue. Between us we have written a paper which has now been circulated to all working group members. I should say, that the large bulk of this document covers a detailed technical critique by Gavin based on his own experiences of using a bespoke estimation method. I use expert system software to the same end. A good job he has done too.

Below are the recommendations we are making in our paper:


Proposed Policy Changes - Overview
1. Alternative performance estimation methods should be permitted to be used instead of the MCS methodology providing they are at least as accurate as the current MCS method, and based on the same PVGIS-SAF climate data as the MCS method.

2. Alternative shading assessment methods should be permitted to be used instead of the MCS methodology providing they can be demonstrated to be at least as accurate as the current MCS method.

Proposed policy changes - details
1. Alternative performance estimation methods should be permitted to be used instead of the MCS methodology providing they are at least as accurate as the current MCS method, and based on the same PVGIS-SAF climate data as the MCS method.
a. The MCS tables are derived from the PVGIS Climate SAF database, so there shouldn’t be a problem with companies being allowed to use this database directly either via the portal or specialist software that draws its data from this source.
b. The company must be able to demonstrate their competence to use the alternative performance estimation method selected.
c. If the PV-GIS portal is used directly, then records should be kept detailing the assumptions used in producing the estimate, which the company must be able to justify at their MCS surveillance visit. These assumptions should include at least the losses associated with inverter efficiency, AC & DC losses, and any panel uprating and derating due to temperature coefficients, low light performance, positive or negative tolerances and similar issues.
d. Alternatively a standard assumption of losses of 10% can be applied directly in PVGIS if the information in point b is not known (as used by MCS to produce the generation tables).

2. Alternative shading assessment methods should be permitted to be used instead of the MCS methodology providing they can be demonstrated to be at least as accurate as the current MCS method and the following criteria are met:-
a. The company can demonstrate their competence to use the alternative method.
b. Precise details and methodology statements should form part of a company’s MCS Quality and Procedures Manual, and should be subject to assessment at the annual MCS surveillance visit.
c. The estimated percentage losses due to shading should be presented to the customer and incorporated within the overall performance estimates used.
d. If 3d modelling is used in preference to the Sun Path diagram, then a sample of images showing the shading at different points in the year should be included in the information pack.

3. Specialist software such as PVSOL or PVSYS should be viewed as an acceptable alternative method to be used instead of the MCS method providing the company can demonstrate competence in the use of this software for this purpose.

4. A Statement along the following lines to accompany all output estimates:
“The performance estimate has been prepared using a method that meets the requirements of MIS3002, the Microgeneration Certification Scheme standard for the installation of PV systems.”

5. Further study and monitoring to improve accuracy
a. Performance estimation methods should be subjected to periodic review and comparison with real world performance data to assess and improve their accuracy.

We would ask that you please register your support with us either in this thread or by PM to either myself or Gavin A. We don't want them thinking it is just us two, but we are talking for everyone.

Rather than just leaving your 'handle', please give your real name and company name.We do have to sign confidentiality agreements to attend the meeting, but we will endevour to let you know how we get on without breaching this.

Thanks in anticipation of your support.

Peter Randall
Solar Kingdom Ltd

 

[Gavin A]

anyone who wants a copy of the full paper please email me.. info at leeds-solar.co.uk

also anyone wanting to support us less publicly, can email me too - name and company name please

we're aiming to get support from within the industry on this to avoid the potential for the steering group being able to sideline us by saying there's no evidence we have any support / this is a widespread issue etc.

...but we want to try to do this without washing the industries dirty linen in public. Also we have to sign some confidentiality stuff before we go to the meeting, so don't want this making properly public, so anyone who gets a copy, please keep it within the industry only.

I could be wrong, but I reckon if we're at the meeting effectively as representatives of those who've signed up to supporting our position, then we ought to be able to feed back to those who we're representing afterwards regardless of the confidentiality clause... besides, one of my objections to the whole situation is that it was all decided behind closed doors, and wasn't consulted on, so best not to repeat that mistake really IMO.

I'd also welcome any additional comments positive or negative about our proposals / a straw poll on whether people have got used to the new MCS method and are all now using it happily, or like us, are pretty much ignoring it completely (obviously we'll not drop you in it with MCS).

 

[jerriais]

Matt Thornington
SunSmart Energy Ltd

Fully support proposal to allow a good PV*Sol or similar design to replace MCS shade simulation.

 

[Gavin A]

cheers Matt...

for others, please also include your position in the company.

 

[Worcester]

Excellent proposal Gavin and Peter, you've got my full support and thanks not only for putting the paper together, also for attending the meeting at your own time and expense.
Rgds
G
p.s. can you email me the document (you've got my email address )

 

[whinmoor]

Good stuff, guys. A bit confused by this clause:

"d. Alternatively a standard assumption of losses of 10% can be applied directly in PVGIS if the information in point b is not known (as used by MCS to produce the generation tables)."

Should "Point B" = "Point C"?

It may be possible to get a list of approved software - although I'm not sure if that would be desirable as a cartel/monopoly/duopoly may develop and hold us to ransom by hiking prices on updates. That's what HMRC have on their website to do VAT returns:

HM Revenue & Customs

I think it's crazy to give the customer a sunpath diagram with "No Shade" written across it so we don't do that. We just say it, get the nod from the customer and then write it on the quotation that that's one of the assumptions behind the predicted output.

Count me in on your side:

Philip Chow
MD
Medoria Solar, Wakefield
[email protected]

 

[Gavin A]

well spotted, yes that should have been point C, we obviously must have moved the points around at some point and missed that.

tis the problem with MCS giving us barely any notice to get the report submitted.

worcester - was just sending you a copy when you wrote that message. Philip, you also have mail. Matt - I seem to have lost your email, let me know if you want a copy.

 

[The Solar King]

If we get anywhere with this, it hopefully would not be about lists of approved software. It should be about businesses evidencing that their chosen method meets the requirements and that they can show their competence to use it.

 

[TedM]

My only question would be around the "at least as accurate as the current MCS method" wording.

How are you going to define that?

e.g. does it mean any alternate method needs to come to the same result as the MCS method within plus or minus x%?

- or -

if the MCS method is only 95% accurate (under predicts compared to reality) then any alternate method also needs to produce at least a 5% under-prediction.

And if using 'compared to reality x%' then is this based on figures for a single property or an average figure over many properties?

Gavin, I would guess that you have a ton of data from customer's that could be used, in a statistical way, to back up any reasoning?

 

[The Solar King]

This was all a little rushed as Gavin has pointed out. We had thought of 'a method equal or better than' as this not only covers the process, but also the outcome. Trying to get the correct form of words and cover all the bases is often difficult. We will have the opportunity in the meeting to cover this sort of point. If we are able to establish the principal of the two summarised points, the rest should follow.

May we have your support please Ted?

 

[TedM]

Not meant as a criticism - just a 'heads-up' as to what might get thrown at you in the meeting.

Happy to give my support, but I don't install PV.

 

[Gavin A]

Good question Ted

for example, MCS claims +/- 10% accuracy for their shading estimates. We guaratee - 5% accuracy for ours, so our method would meet that criteria.

Any method that used 10% losses inputted into PVGIS directly would automatically be at least as accurate as the MCS method for performance estimates - as that's the metric they used (despite the methodology they say they used). It'd actually be significantly more accurate as it would be geographically accurate so would lose the +/-10% for geographic inaccuracies inherent in the MCS method.

If an installer calculated the actual inverter and cable losses etc and used these to input to PVGIS then this would be even more accurate.

It'd be up to the likes of PVSOL to make a statement to the effect that their software met / exceeded this rule and was therefore MCS compliant as an alternative to the MCS method.

That's the way I envisage it anyway, but yes it might be a good idea for me to be able to outline this at the meeting as it's an obvious question to be asked and potentially trip us up.

 

[whinmoor]

Thanks for the document, Gavin. It looks comprehensive and I wish you both a good trip to London. (It might have been a good idea to locate Skegness, Sheffield and Skipton on the map because the suits in Whitehall probably don't travel much north of Watford so wouldn't be so clued up about Butlins at Skeggy).

Peter, my point about approved software was to demonstrate a precedent that HMRC endorses certain commercial packages so MCS theoretically could but obviously the limited scale of the market may be problematic.

 

[t4sltd]

I am fully behind your proposal. Please send a copy of your report: [email protected]
Paul Chandler
Director
T4 Sustainability Ltd

 

[SolarCity]

Fully behind the proposal. I'd love to receive a copy too if you don't mind - [email protected]

As Gordon says, fair play to you both for taking your own time and expense to further this.

Chris Hammond
Director
Solar City