Try our solar cost and savings calculator
Most Popular Topics
Collapse
First over 40 kWh day
Collapse
This topic is closed.
X
X
-
-
Well, my opinion, and my opinion only, FWIW, is that using PVWatts to predict NOT daily, but monthly performance is an appropriate use of that tool and can ONLY lead to incorrect conclusions if the user doesn't understand the tool for what it is.
If the user understands that PVWatts only compiles historical data and present you the averages based on past weather data and that all it is, then the user will not be led to incorrect conclusion.
For example, based on past weather pattern, we can PREDICT that Minnesota has very cold weather in the winter time. If there's an Indian summer day in MN in the middle of its winter, we will NOT be led to incorrect conclusion that MN winter is not very cold. We will recognize that Indian summer day for what it is, an anomaly. But it doesn't change our oveall prediction that MN winter is very cold. In terms of correct or incorrect outcome, we still expect that MN winter will cost us X amount of heating oil for the season so we still budget Y amount of dollars for X amount of oil based on our PREDICTION that MN winter will cost us that much in heating oil.
The same with PVWatts. If I know that based on historical weather pattern that PVWatts collects over the years, it estimates that my February solar production based on my system size at my location with a 0.87 derating factor should deliver 1,000 kwh of solar energy, then I can use that information combined with the other 11 months to calculate and PREDICT what my payoff period will be.
Furthermore, if my actual solar production for February is only 500 kwh while PVWatts says I should expect 1000 kwh on the average, then I know that something funny is going on and it will help trigger an alarm for me to investigate further. The first thing I'll investigate is not whether something is wrong with my solar system. The first thing I'll investigate is whether the 2014 month of February weather in my location is abnormally more cloudy/raining than normal or not. If not, and the weather is the same as a typical average weather for February, then I'll investigate next to see maybe something is wrong with my system.
I'm being long-winded with an obvious example here to tell you that people are not as dumb as you think they are. So give them some credit and don't start giving out long-winded condescending lectures about how people can use the information to make incorrect conclusion and incorrect outcomes just because they use the word "predict" and to you "predict" is wrong and it has to be "estimate". Who cares about the choice of words if the intent is clear as day already?
For what it's worth, I'm not a retired engineer. But I'm a working electrical engineer myself for the last 30 years.Comment
-
I was going to let this dog sleep too. However:Well, my opinion, and my opinion only, FWIW, is that using PVWatts to predict NOT daily, but monthly performance is an appropriate use of that tool and can ONLY lead to incorrect conclusions if the user doesn't understand the tool for what it is.
If the user understands that PVWatts only compiles historical data and present you the averages based on past weather data and that all it is, then the user will not be led to incorrect conclusion.
For example, based on past weather pattern, we can PREDICT that Minnesota has very cold weather in the winter time. If there's an Indian summer day in MN in the middle of its winter, we will NOT be led to incorrect conclusion that MN winter is not very cold. We will recognize that Indian summer day for what it is, an anomaly. But it doesn't change our oveall prediction that MN winter is very cold. In terms of correct or incorrect outcome, we still expect that MN winter will cost us X amount of heating oil for the season so we still budget Y amount of dollars for X amount of oil based on our PREDICTION that MN winter will cost us that much in heating oil.
The same with PVWatts. If I know that based on historical weather pattern that PVWatts collects over the years, it estimates that my February solar production based on my system size at my location with a 0.87 derating factor should deliver 1,000 kwh of solar energy, then I can use that information combined with the other 11 months to calculate and PREDICT what my payoff period will be.
Furthermore, if my actual solar production for February is only 500 kwh while PVWatts says I should expect 1000 kwh on the average, then I know that something funny is going on and it will help trigger an alarm for me to investigate further. The first thing I'll investigate is not whether something is wrong with my solar system. The first thing I'll investigate is whether the 2014 month of February weather in my location is abnormally more cloudy/raining than normal or not. If not, and the weather is the same as a typical average weather for February, then I'll investigate next to see maybe something is wrong with my system.
I'm being long-winded with an obvious example here to tell you that people are not as dumb as you think they are. So give them some credit and don't start giving out long-winded condescending lectures about how people can use the information to make incorrect conclusion and incorrect outcomes just because they use the word "predict" and to you "predict" is wrong and it has to be "estimate". Who cares about the choice of words if the intent is clear as day already?
For what it's worth, I'm not a retired engineer. But I'm a working electrical engineer myself for the last 30 years.
1.) This whole recent business started for me as an attempt to point out to some posters who seemed to be using PVWatts output data as an indication of single, clear day production, that doing so was not a valid use of the calculator and would perhaps lead to incorrect conclusions. That little factoid may have been overlooked. Given that, if the word "predict" in your 1st para. was "estimate", we'd be in complete agreement on that statement. The two words do not have the same meaning, especially in a statistical sense, where the difference is as clear as day.
2.) If the PVWatts user "understands that PVWatts only compiles historical data...", then, IMO, the user has it wrong. My understanding of PVWatts is that it does not compile historic data at all. Rather, it uses TMY data and other information to estimate long term future system performance and presents it by month.
3.) Do you understand how a "TMY" is determined ? It's a long way from a simple historical average.
4.) I don't think you are long winded at all.
5.) I respectfully ask that you do not try to imply that I think people are "dumb". I never said or wrote that, nor did I imply as such, nor do I think that (separate subject: I'm of the opinion that most folks are about equally intelligent, FWIW, but that's off topic.). If you are going to accuse me of something, get it right. I believe for you to say or imply that I think people are dumb is incorrect and not true.
6.) Sometimes things can't be explained in 25 words or less and be accurate. Sometimes people say :"Gee, I didn't know that" and learn something, rather than feel insulted. Your opinion may be different. Most are, and this is the place to express them - which is what I did.Comment
Copyright © 2014 SolarReviews All rights reserved.
Powered by vBulletin® Version 6.1.3
Copyright © 2026 MH Sub I, LLC dba vBulletin. All rights reserved.
Copyright © 2026 MH Sub I, LLC dba vBulletin. All rights reserved.
All times are GMT-5. This page was generated at 02:48 AM.
Comment