Calculate average daily peak sun-hours from .epw file data?

[bernard]

Hi,

Just to mention I am currently not interested in building an actual PV panel but just informing myself for the start, and learning about this technology.
What confuses me is the sizing of Photovoltaic panels. I assume this is a complicated issue, and depends on a lot of factors.
Still I googled a bit an looks like all explanations of Photovoltaic panels include a so called 'average daily peak sun-hours' (units: kW/m^2/day or solar peak hours). I found a large portion of data for specific US, and Europe locations.

But is there any way 'average daily peak sun-hours' could be calculated through the use of data from EnergyPlus weather files (.epw)? These are solar irradiation measures from .epw file:

- directNormalRadiation (hourly amount of solar radiation in Wh/m2 received directly from the solar disk on a surface perpendicular to the sun's rays)

- diffuseHorizontalRadiation (hourly amount of solar radiation in Wh/m2 received from the sky (excluding the solar disk) on a horizontal surface during the number of minutes preceding the time indicated)

- globalHorizontalRadiation (total hourly amount of direct and diffuse solar radiation in Wh/m2 received on a horizontal surface during the number of minutes preceding the time indicated)

Which one of these three can be used to calculate 'average daily peak sun-hours'?

Thank you.

[Naptown]

Gone gaisma link in my signature has thousands of global irradiance measurements.

[bernard]

Thank you for the Naptown.

I would still like to know if there is a way calculating 'average daily peak sun-hours' through some of these: directNormalRadiation, diffuseHorizontalRadiation, globalHorizontalRadiation from EnergyPlus weather files.

[J.P.M.]

Hi,

Just to mention I am currently not interested in building an actual PV panel but just informing myself for the start, and learning about this technology.
What confuses me is the sizing of Photovoltaic panels. I assume this is a complicated issue, and depends on a lot of factors.
Still I googled a bit an looks like all explanations of Photovoltaic panels include a so called 'average daily peak sun-hours' (units: W/m^2/day or solar peak hours). I found a large portion of data for specific US, and Europe locations.

But is there any way 'average daily peak sun-hours' could be calculated through the use of data from EnergyPlus weather files (.epw)? These are solar irradiation measures from .epw file:

- directNormalRadiation (hourly amount of solar radiation in Wh/m2 received directly from the solar disk on a surface perpendicular to the sun's rays)

- diffuseHorizontalRadiation (hourly amount of solar radiation in Wh/m2 received from the sky (excluding the solar disk) on a horizontal surface during the number of minutes preceding the time indicated)

- globalHorizontalRadiation (total hourly amount of direct and diffuse solar radiation in Wh/m2 received on a horizontal surface during the number of minutes preceding the time indicated)

Which one of these three can be used to calculate 'average daily peak sun-hours'?

Thank you.

1.) All 3 to a greater or lesser degree are used and needed to calculate the solar irradiance incident on a solar device.

2.) What you have referenced is a listing of the "TMY" (Typical Meteorological Year) database. Check NREL and see the definitions for a better description of TMY than I can give here.

3.) There are programs/software that will translate the irradiance data in the raw form as available from your sources to a POA (Plane Of Array) value. That is, how much solar energy is the solar device "seeing" in a fashion that is perpendicular to the surface of the solar device. It can and was done by hand before calculators and computers. Things are better now, but doing by hand was a true learning experience probably now lost.

4.) Depending on what you need, there is also software that will do the same as the TMY data but estimate clear sky values by the hour, or minute if you want, and do the same transform to POA values. I did some stuff of that nature in Fortran IV about 35 years ago or so. Things have improved some since then.

5.) I believe I understand the concept of "Average Daily Peak Sun Hours". Given the progress of software and the state of things, that has become a somewhat anachronistic concept. It mostly came about because it was a commonly available parameter before common availability of computing power when empirical correlations were about the best and usually only game in town.

6.) Check out something called PVWatts from NREL. Spend a few minutes reading all the info screens, and do a run or two. Use the hourly output option and notice the column headings.

[bernard]

Thank you for the reply J.P.M.

I am not looking for a piece of software, but for an explanation on how to calculate the photovoltaic panel size using solar radiation data from EnergyPlus weather files by hand.
So the thing that confused me was 'average daily peak sun-hours' as I did not know how to calculate those.

If there is some other way of sizing PV panels with data taken from EnergyPlus weather files, I would be grateful if somebody could show it to me.


Most of USA EnergyPlus weather files, are in fact TMY 1,2,3 data (TMY 1 being the hourly data for one year based on 1948-1980 data; TMY 2 is based on 1961 and 1990, TMY 3 on 1991-2005 period).

I have already seen the PVWatts thank you. But it is online based, without the step by step explanations of each calculation equations behind the scene.

[J.P.M.]

Thank you for the reply J.P.M.

I am not looking for a piece of software, but for an explanation on how to calculate the photovoltaic panel size using solar radiation data from EnergyPlus weather files by hand.
So the thing that confused me was 'average daily peak sun-hours' as I did not know how to calculate those.

If there is some other way of sizing PV panels with data taken from EnergyPlus weather files, I would be grateful if somebody could show it to me.


Most of USA EnergyPlus weather files, are in fact TMY 1,2,3 data (TMY 1 being the hourly data for one year based on 1948-1980 data; TMY 2 is based on 1961 and 1990, TMY 3 on 1991-2005 period).

I have already seen the PVWatts thank you. But it is online based, without the step by step explanations of each calculation equations behind the scene.

You are welcome.

Can you elaborate a bit on what you mean when you say "how to calculate the photovoltaic panel size" ?

[bernard]

Can you elaborate a bit on what you mean when you say "how to calculate the photovoltaic panel size" ?

Ok. Here is a typical example I have been founding all around the web:

annual energy consumption (kWh/year) / 365 days = average daily consumption (kWh/day)
average daily consumption (kWh/day) / average daily peak sun-hours (kW/m^2/day) = array size (kW)
array size (kW) / efficiency factor = array size (kW)

But for now calculation of average daily peak sun-hours (kW/m^2/day) is bugging me. And I would like to calculate that from EnergyPlus weather file data (using these: directNormalRadiation, diffuseHorizontalRadiation, globalHorizontalRadiation), not use an online chart, website or piece of software to calculate this for me.

[J.P.M.]

Ok. Here is a typical example I have been founding all around the web:

annual energy consumption (kWh/year) / 365 days = average daily consumption (kWh/day)
average daily consumption (kWh/day) / average daily peak sun-hours (kW/m^2/day) = array size (kW)
array size (kW) / efficiency factor = array size (kW)

But for now calculation of average daily peak sun-hours (kW/m^2/day) is bugging me. And I would like to calculate that from EnergyPlus weather file data (using these: directNormalRadiation, diffuseHorizontalRadiation, globalHorizontalRadiation), not use an online chart, website or piece of software to calculate this for me.

Bernard: Take a deep breath and keep in mind I don't have it in for you.

Now:

Equation 1 above is correct, but of limited use.

What I believe you have described with equation 2 is a rule of thumb. To my experience they are good for measuring thumbs, similar to a 1st approx., but not as good. More like a dart throw with a low probability of hitting the board.

Equation 3 above is incorrect unless the efficiency factor is 1.

To give an honest opinion, I believe if you had more knowledge of this subject you would realize that your approach will not get you the desired information (that information being something called average daily peak sun hours) which itself is mostly specious to begin with. I'm not trying to rain on your parade, but IMO, your approach so far will lead to GIGO results and belies an ignorance of solar energy, at least with regard to resource availability and determination.

I encourage your continued curiosity and respectfully suggest you consider that your current approach has the cart before the horse. FWIW, I'm also impressed with your attitude of doing things without the aid of charts, graphs and software. I think that's one of the best ways to acquire more than superficial knowledge, and one of my mantras. To that end, one other approach I'd suggest for your consideration might be to find a serious textbook on the fundamentals of solar energy and read, study, think and learn. If you do that, you may see the futility of your current attempts and, perhaps more importantly, a straighter path to your goal.

Basically - you want better information, get closer to the source. Get the fundamentals right before you succumb to the alleged technical fluff in magazines, utube and most of the ad laden crap on the web and elsewhere.

Good luck.

[bernard]

Equation 1 above is correct, but of limited use.

What I believe you have described with equation 2 is a rule of thumb. To my experience they are good for measuring thumbs, similar to a 1st approx., but not as good. More like a dart throw with a low probability of hitting the board.

Equation 3 above is incorrect unless the efficiency factor is 1.

That is exactly what I am trying to do for the start - an approximate measure of the final array size in kW.

Equation 3 above is incorrect unless the efficiency factor is 1.

I suppose I should have named the final variable: "modified array size" for example:

array size (kW) / efficiency factor = modified array size (kW)

To give an honest opinion, I believe if you had more knowledge of this subject you would realize that your approach will not get you the desired information (that information being something called average daily peak sun hours) which itself is mostly specious to begin with. I'm not trying to rain on your parade, but IMO, your approach so far will lead to GIGO results and belies an ignorance of solar energy, at least with regard to resource availability and determination.

I encourage your continued curiosity and respectfully suggest you consider that your current approach has the cart before the horse. FWIW, I'm also impressed with your attitude of doing things without the aid of charts, graphs and software. I think that's one of the best ways to acquire more than superficial knowledge, and one of my mantras. To that end, one other approach I'd suggest for your consideration might be to find a serious textbook on the fundamentals of solar energy and read, study, think and learn. If you do that, you may see the futility of your current attempts and, perhaps more importantly, a straighter path to your goal.

Basically - you want better information, get closer to the source. Get the fundamentals right before you succumb to the alleged technical fluff in magazines, utube and most of the ad laden crap on the web and elsewhere.

Thank you for the opinion and advice once again J.P.M. Will try checking on some fundamentals.

As much as I appreciate your concern for me getting the precise result and understanding of solar energy, I can still notice that in these couple of replies, you did not answer my initial question:


How to calculate average daily peak sun-hours (kW/m^2/day) by using: directNormalRadiation, diffuseHorizontalRadiation, globalHorizontalRadiation data from EnergyPlus weather file (again without further explanations on how this approach can be inaccurate).

Do you know, or you do not know how to do that?

Bernard: Take a deep breath and keep in mind I don't have it in for you.

Of course this is just a discussion, no hard feelings.

Thank you.

[J.P.M.]

The short answer to your question: I do not.

A bit more: I do not, mostly because I do not know how to give a useful meaning to a term that has at best an imprecise definition and therefore, at least to my limited knowledge, usefulness. That has been most of the reason behind my initial and ongoing responses to your questions.

Perhaps others more knowledgeable on the subject than I am can tell you (with all possible candor on my part) what, IMO, you seem to want to read/hear to make reality fit your view of the situation.

Again, Good Luck in your future efforts.

J.P.M.

[bernard]

Thank you for all the help and replies J.P.M.

[sensij]

I have already seen the PVWatts thank you. But it is online based, without the step by step explanations of each calculation equations behind the scene.

There is in fact detailed documentation describing how PVWatts arrives at its numbers, with additional documentation describing what they've done in version 2. Version 5 documentation is also available for the update just released last month. More factors are accounted for than just the irradiance numbers, and it will be somewhat more difficult to follow their estimation strategy than it would be to perform the calculations you have mentioned. I too am a fan of understanding the math, but often more importantly, understanding what assumptions and exclusions apply.

You may disagree, but in the case of PVWatts, I find that the numbers it provides relating energy output to panel size and installation factors are "good enough", and my time is better spent analyzing my electricity usage for ways in which it might be more efficient, and projecting my long term consumption in a way that can be used to justify (or not justify) the cost of a PV install.

[bernard]

Thank you sensij,
Wonderful reply!

Take care.