You can use anything you want as a surrogate, as long as you have some way
of proving it works. In this case, if you could somehow get real time PV
data for a set of sites that also had actual insolation and then looked at
your results, I think you could derive a typical relationship that wouldn't
lead you too far astray. The problem will come when there are passing
clouds and the PV system is "hunting" all the time trying to catch up with
the changing insolation. I bet you could find those and apply a bias
correction that could then be used in multiple sites with real time PV
data.
This is an intriguing idea (using PV output as a surrogate for
irradiance), but there are a number of complications that need to be
investigated:
1. PV efficiency degrades with temperature; therefore, you have to have
some way to adjust the output by the temperature of the PV (not the air
temperature)
2. you also need to adjust for the tilt and azimuth of the PV panels to
derive the equivalent global horizontal radiation.
3. you also need to split the global horizontal into direct and diffuse
radiation, which could be done using a number of direct/diffuse split
models, which is what's done for 99% of the weather files anyway because
I've yet to find any measured direct normal radiation outside of
strictly research applications.
When I look at this list, all three could be done using a building
energy simulation program, provided that you know the absorptivity of
the PV and its conversion efficiency as a function of temperature, but
at this point it's getting more hairy than calculating the solar
radiation from standard weather station data or, beginning in Sept 2015
for US locations, downloading the solar radiation from NREL.
I'm not being critical, and actually I am quite intrigued by this
technique, and would be happy to help evaluate how well it works.
Joe
Joe Huang
White Box Technologies, Inc.
346 Rheem Blvd., Suite 205A
Moraga CA 94556
yjhuang at whiteboxtechnologies.com
http://weather.whiteboxtechnologies.com for simulation-ready weather data
http://www.whiteboxtechnologies.com
(o) (925)388-0265
(c) (510)928-2683
"building energy simulations at your fingertips"
Another note I forgot to mention: how easy is it to get the hourly
output from the PV panels? If the need for the solar radiation data is
to evaluate actual building DSM, then the PV output data would have to
have been collected at the same time as the utility billing data.
Joe
Joe Huang
White Box Technologies, Inc.
346 Rheem Blvd., Suite 205A
Moraga CA 94556
yjhuang at whiteboxtechnologies.com
http://weather.whiteboxtechnologies.com for simulation-ready weather data
http://www.whiteboxtechnologies.com
(o) (925)388-0265
(c) (510)928-2683
"building energy simulations at your fingertips"
Yes, interesting idea and glad to see it discussed here.
This idea also occurred to us when evaluating PV systems on a number of
buildings in Hawaii. In that project, we were fortunate enough to have
a solar monitoring site right down the
street, and he had 15min and 1hour PV energy generation for every panel
(micro-invertors) on 15 different buildings, each oriented in the same
direction. However, some other issues, besides the one's Joe described,
came to mind in trying to develop a model that linked the two:
* In addition to temperature, PV panels energy production also
degrades as the panels get dirty. That seemed like the variable
that is toughest to compensate for.
* Data connection between the inverters to the cloud was pretty good,
but we did find gaps a number of gaps; sometimes for up to a month
or two.
Still yes, an interesting idea, and sounds like a good research project...
David
In addition to dirt, there is snow cover. ?I live in NH and have a 5 kW
PV system with a real-time display of power output. ?Even minor snow
cover drops the output to essentially 0. Chip Barnaby
--
Chip Barnaby
chipbarnaby at gmail.com
> Yes, interesting idea and glad to see it discussed here.?
This idea also occurred to us when evaluating PV systems on a number
of buildings in Hawaii.? In that project, we were fortunate enough
to have a solar
monitoring site[1] right down the street, and he had 15min and
1hour PV? energy generation for every panel (micro-invertors) on 15
different buildings, each oriented in the same direction.? However,
some other issues, besides the one's Joe described, came to mind in
trying to develop a model that linked the two:
degrades as the panels get dirty.? That seemed like the variable
that is toughest to compensate for.?
good, but we did find gaps a number of gaps; sometimes for up to
a month or two.??
project...
David
irradiance), but there are a number of complications that need to
be investigated:
1. PV efficiency degrades with temperature; therefore, you have to
have some way to adjust the output by the temperature of the PV
(not the air temperature)
2. you also need to adjust for the tilt and azimuth of the PV
panels to derive the equivalent global horizontal radiation.
3. you also need to split the global horizontal into direct and
diffuse radiation, which could be done using a number of
direct/diffuse split models, which is what's done for 99% of the
weather files anyway because I've yet to find any measured direct
normal radiation outside of strictly research applications.
When I look at this list, all three could be done using a building
energy simulation program,? provided that you know the
absorptivity of the PV and its conversion efficiency as a function
of temperature,? but at this point it's getting more hairy than
calculating the solar radiation from standard weather station data
or, beginning in Sept 2015 for US locations, downloading the solar
radiation from NREL.
I'm not being critical, and actually I am quite intrigued by this
technique, and would be happy to help evaluate how well it works.
Joe
White Box Technologies, Inc.
346 Rheem Blvd., Suite 205A
Moraga CA 94556
(o) (925)388-0265
(c) (510)928-2683
"building energy simulations at your fingertips"
long as you have some way of proving it works. In this case,
if you could somehow get real time PV data for a set of sites
that also had actual insolation and then looked at your
results, I think you could derive a typical relationship that
wouldn't lead you too far astray. The problem will come when
there are passing clouds and the PV system is "hunting" all
the time trying to catch up with the changing insolation. I
bet you could find those and apply a bias correction that
could then be used in multiple sites with real time PV data.?
Chris Yates >>>> Ok, so there's no substitute for doing
weather data right.
real-time pv yield be used as a surrogate?
typos, etc.
_______________________________________________
Bldg-sim mailing list
To unsubscribe from this mailing list send? a blank
message to BLDG-SIM-UNSUBSCRIBE at ONEBUILDING.ORG
Bldg-sim mailing list
To unsubscribe from this mailing list send a blank message to BLDG-SIM-UNSUBSCRIBE at ONEBUILDING.ORG
Bldg-sim mailing list
To unsubscribe from this mailing list send a blank message to BLDG-SIM-UNSUBSCRIBE at ONEBUILDING.ORG
You can use anything you want as a surrogate, as long as you have some way
of proving it works. In this case, if you could somehow get real time PV
data for a set of sites that also had actual insolation and then looked at
your results, I think you could derive a typical relationship that wouldn't
lead you too far astray. The problem will come when there are passing
clouds and the PV system is "hunting" all the time trying to catch up with
the changing insolation. I bet you could find those and apply a bias
correction that could then be used in multiple sites with real time PV
data.
This is an intriguing idea (using PV output as a surrogate for
irradiance), but there are a number of complications that need to be
investigated:
1. PV efficiency degrades with temperature; therefore, you have to have
some way to adjust the output by the temperature of the PV (not the air
temperature)
2. you also need to adjust for the tilt and azimuth of the PV panels to
derive the equivalent global horizontal radiation.
3. you also need to split the global horizontal into direct and diffuse
radiation, which could be done using a number of direct/diffuse split
models, which is what's done for 99% of the weather files anyway because
I've yet to find any measured direct normal radiation outside of
strictly research applications.
When I look at this list, all three could be done using a building
energy simulation program, provided that you know the absorptivity of
the PV and its conversion efficiency as a function of temperature, but
at this point it's getting more hairy than calculating the solar
radiation from standard weather station data or, beginning in Sept 2015
for US locations, downloading the solar radiation from NREL.
I'm not being critical, and actually I am quite intrigued by this
technique, and would be happy to help evaluate how well it works.
Joe
Joe Huang
White Box Technologies, Inc.
346 Rheem Blvd., Suite 205A
Moraga CA 94556
yjhuang at whiteboxtechnologies.com
http://weather.whiteboxtechnologies.com for simulation-ready weather data
http://www.whiteboxtechnologies.com
(o) (925)388-0265
(c) (510)928-2683
"building energy simulations at your fingertips"
Another note I forgot to mention: how easy is it to get the hourly
output from the PV panels? If the need for the solar radiation data is
to evaluate actual building DSM, then the PV output data would have to
have been collected at the same time as the utility billing data.
Joe
Joe Huang
White Box Technologies, Inc.
346 Rheem Blvd., Suite 205A
Moraga CA 94556
yjhuang at whiteboxtechnologies.com
http://weather.whiteboxtechnologies.com for simulation-ready weather data
http://www.whiteboxtechnologies.com
(o) (925)388-0265
(c) (510)928-2683
"building energy simulations at your fingertips"
Yes, interesting idea and glad to see it discussed here.
This idea also occurred to us when evaluating PV systems on a number of
buildings in Hawaii. In that project, we were fortunate enough to have
a solar monitoring site
street, and he had 15min and 1hour PV energy generation for every panel
(micro-invertors) on 15 different buildings, each oriented in the same
direction. However, some other issues, besides the one's Joe described,
came to mind in trying to develop a model that linked the two:
* In addition to temperature, PV panels energy production also
degrades as the panels get dirty. That seemed like the variable
that is toughest to compensate for.
* Data connection between the inverters to the cloud was pretty good,
but we did find gaps a number of gaps; sometimes for up to a month
or two.
Still yes, an interesting idea, and sounds like a good research project...
David
In addition to dirt, there is snow cover. ?I live in NH and have a 5 kW
PV system with a real-time display of power output. ?Even minor snow
cover drops the output to essentially 0. Chip Barnaby
--
Chip Barnaby
chipbarnaby at gmail.com
> Yes, interesting idea and glad to see it discussed here.?
This idea also occurred to us when evaluating PV systems on a number
of buildings in Hawaii.? In that project, we were fortunate enough
to have a solar
monitoring site[1] right down the street, and he had 15min and
1hour PV? energy generation for every panel (micro-invertors) on 15
different buildings, each oriented in the same direction.? However,
some other issues, besides the one's Joe described, came to mind in
trying to develop a model that linked the two:
degrades as the panels get dirty.? That seemed like the variable
that is toughest to compensate for.?
good, but we did find gaps a number of gaps; sometimes for up to
a month or two.??
project...
David
irradiance), but there are a number of complications that need to>>>> Ok, so there's no substitute for doing
be investigated:
1. PV efficiency degrades with temperature; therefore, you have to
have some way to adjust the output by the temperature of the PV
(not the air temperature)
2. you also need to adjust for the tilt and azimuth of the PV
panels to derive the equivalent global horizontal radiation.
3. you also need to split the global horizontal into direct and
diffuse radiation, which could be done using a number of
direct/diffuse split models, which is what's done for 99% of the
weather files anyway because I've yet to find any measured direct
normal radiation outside of strictly research applications.
When I look at this list, all three could be done using a building
energy simulation program,? provided that you know the
absorptivity of the PV and its conversion efficiency as a function
of temperature,? but at this point it's getting more hairy than
calculating the solar radiation from standard weather station data
or, beginning in Sept 2015 for US locations, downloading the solar
radiation from NREL.
I'm not being critical, and actually I am quite intrigued by this
technique, and would be happy to help evaluate how well it works.
Joe
White Box Technologies, Inc.
346 Rheem Blvd., Suite 205A
Moraga CA 94556
(o) (925)388-0265
(c) (510)928-2683
"building energy simulations at your fingertips"
long as you have some way of proving it works. In this case,
if you could somehow get real time PV data for a set of sites
that also had actual insolation and then looked at your
results, I think you could derive a typical relationship that
wouldn't lead you too far astray. The problem will come when
there are passing clouds and the PV system is "hunting" all
the time trying to catch up with the changing insolation. I
bet you could find those and apply a bias correction that
could then be used in multiple sites with real time PV data.?
Chris Yates
weather data right.
real-time pv yield be used as a surrogate?
typos, etc.
_______________________________________________
Bldg-sim mailing list
To unsubscribe from this mailing list send? a blank
message to BLDG-SIM-UNSUBSCRIBE at ONEBUILDING.ORG
Bldg-sim mailing list
To unsubscribe from this mailing list send a blank message to BLDG-SIM-UNSUBSCRIBE at ONEBUILDING.ORG
Bldg-sim mailing list
To unsubscribe from this mailing list send a blank message to BLDG-SIM-UNSUBSCRIBE at ONEBUILDING.ORG
Links:
1. http://www.nrel.gov/midc/nelha/