Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me.
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127
Posted: 01:16am 02 Jan 2011
Copy link to clipboard
Print this post
Too often I am asked the question, If I build a windmill, How much power will it make?
I usually reply with, You need to know how much wind you get at the site, and what is the wind distribution over the day and the year at the site? This usually results in blank faces.
Windspeed data of some form is available for most of the planet. Windspeed data available for aviation purposes and available from say a Meteorological source is often used.
I have found that there is a high variability in the same data depending on how it is presented.
daily windspeeds from Met. sources often consists of maximum and minimum windspeeds for the day. A lot can happen in a day, and max/mins don't reflect this.
Here are some examples.
this graph is data from my weather station. The data is for a single day 311210, and consists of typical 10min averages [144data points for the day], and the maximum value for each period, plotted together on the same graph. The maximum reading for the day was 39mph, and the minimum was 4mph.
this is the same day period for 311210, from my windspeed logger. This consists of logged 10sec averages [8624data points for the day]. The maximum recorded reading for the day was 31.4mph, and the minimum was 0mph.
The data for the day is quite difficult to view on a graph, so here is a selected 1hour period.
I have crunched some numbers, and a median windspeed for the day based on max/mins, is 21.5mph. The same data gave an average of 11.247mph based on 10min averages, and 20.833mph based on averages of the maximum readings during these 10min time windows. The average windspeed based on 10sec readings is 9.729mph.
My 2 windspeed measuring cup anemometers [Davis type] are within 1m of each other, so essentially measuring the same airspeed.
There is approx a factor of 2 difference in the data from the different measurements. Unfortunately this reflects an 8fold difference in wind energy based on the data.
We often think a location is windy, but without a standard comparison of measurements we can be very dissapointed when the windmill does not seem to make as much power as was estimated from the data.
Gordon.
become more energy aware
VK4AYQ Guru Joined: 02/12/2009 Location: AustraliaPosts: 2539
Posted: 01:58am 02 Jan 2011
Copy link to clipboard
Print this post
Hi Gordon
I can see the collection methods vary in a way to cause confusion when looked at as single unit average, which isn't a reflection of the power available in the wind.
The problem as I see it is that it is a theoretical figure rather than a practical unit designed to mimic the power absorbed by a HAWT.
For data collection the logger is very precise, but unless coupled to the rpm of the alternator it doesn't reflect the power absorbed, as it isn't taking into account the inertia of the blades.
From a practical point the wind data from the weather station gives a visual representation closer to the power absorbed by its low sample rate, as the peaks and troughs in the logger program wouldn't be followed by the blade speed due to inertia in the system and relative loading on the alternator at varying power output.
To my simple mind the recording head needs more weight to add inertia and represent a closer to the average wind speed as would be felt by the windmill also loaded with inertia and power loading. From a theoretical point of view the logger will be more accurate due to high sample rates and averaging on a very short timebase, but to me that doesn't reflect the performance of the mill without other factors added to the equation.
Even my el cheapo weather station exposes the sad state of winds at my place, your readout makes me envious of your location.
All the best
Bob Foolin Around
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127
Posted: 12:27pm 02 Jan 2011
Copy link to clipboard
Print this post
Hi Bob,
I think the anemometer really needs to accurately measure the true windspeed, and not a smudged value to represent how a windmill might be working. You can only make an instrument to do what you outlined after you have made a windmill to calibrate it against. Every time you made a change to your windmill, the sensor would need to be recalibrated as well.
Anyway, re correlation between a windmill rpm and windspeed.
Here is a graph showing my windmills rotor performance relative to windspeed. This data was logged at 1 reading per second, so this is the limit of timing accuracy.
The whole point of a windmill is for the rotor to track the windspeed quickly. There is no point in manipulating the windspeed measurement to suit how a poorly tracking windmill works.
In the graph above the top graph is the cup anemometer windspeed measured data in purply-pink. The light blue is the windmill rotor rpm. The red graph is power into the battery loading.
I think you can agree that there is good correlation of the rotor rpm to the changing measured windspeed. This is the intended outcome. If there is lack of response of the windmill, this will show up with a time difference between the graphs, and reduced maximum rotor rpm's obtained compared to calculated rpm v windspeeds. If the windmill rotor has high inertia, then this will result in sluggish rotor response to the changing windspeed. My windmill does not show this problem. Rotor Inertia is an undesirable feature of a windmill and compounds problems with electrical capacitance and the ability of a control system to make changes to get optimum power from the wind.
I would prefer, and have a windmill that captures the bulk of the wind energy that passes my home. The windmill I have, rapidly changes speed, as seen by the graphs, to follow the windspeed. During storm events with high peak wind energy, my windmill ignores the peak energy now. Soft furling and rpm limited by loading has resulted in wind energy shedding during storms and reduced the need to process power electronically during these times. The weather data recording has shown that storm event wind energy contributes little to maintaining the battery, and only places significant demand on electronic power diversion systems. The few percent of the time that storms play a part does not justify the effort and componentry in capturing every last watt, with the potential to test all the special seals that keep the magic smoke in.
I feel that averaging highly dynamic data [like from a windmill] too much, reduces the information it contains. Averaging the windspeed compounds the meaningful calculation of energy available from it as well. The accurate measurement of windspeed in close proximity to the windmill and a reasonably fast measurement of key windmill parameters can show if a windmill is working optimally or not.
Gordon.become more energy aware
VK4AYQ Guru Joined: 02/12/2009 Location: AustraliaPosts: 2539
Posted: 01:24pm 02 Jan 2011
Copy link to clipboard
Print this post
Hi gordon
I like the graphs and analysis of the wind mill and its ability to track the wind speed very well representing the maximum performance of the mill.
But I thought that we where talking about the initial site selection and assessment, as that is how you started the post.
Downwind Guru Joined: 09/09/2009 Location: AustraliaPosts: 2333
Posted: 02:57pm 02 Jan 2011
Copy link to clipboard
Print this post
Bob,
When you prove you can have a computer system that can work with a picaxe, i will give you a logger that can log your data so it can be seen by you and all.
To have your windmill data in a format that is standard to the rest of us, would be a great in sight to many here.
Pete.Sometimes it just works
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127
Posted: 12:46am 03 Jan 2011
Copy link to clipboard
Print this post
Hi Pete,
Logging the windmill data does help, but the performance and tracking of the windmill rotor with windspeed is dependent heavily on the windmill mechanics and loading. To modify a windspeed recording unit to somehow estimate wind energy at a site is what Bob is hinting to.
My power recording and windspeed logging do not correlate that well across the wide daily ranging of the wind energy passing. If they did, I would have just given the mathematical relationship. I have measured windspeed to a high standard, and it correlates between independent sensing and recording systems.
I have found that smaller windmills wind energy harvest correlates better with windspeed sensing. A larger windmill, or a windmill with slower rotor dynamics or a sluggish electronic power control system has a lower energy harvest in variable wind conditions.
I am finding it really difficult to answer posts systematically now and am tending to cram possibly seemingly unrelated info into my posts that could be in other threads. This is a work around with a 3post limit imposed on me with the forum administration. I was thinking maybe that a new year might be different. There are still many unanswered questions to me on other threads. I suppose eventually I will stop posting as discussions are difficult with days between replies. better stop ranting!
Hi Bob,
I suppose I really intended to highlight that predicting windmill output from normally available weather data is not an easy task. I have a lot of tools at my disposal, and many years recorded data and I am still finding holes in computer model theories, to predict windmill output from the windspeed. It helps when the windmill output follows the wind energy.
Gordon.
become more energy aware
VK4AYQ Guru Joined: 02/12/2009 Location: AustraliaPosts: 2539
Posted: 01:29am 03 Jan 2011
Copy link to clipboard
Print this post
Hi Gordon
Sorry I didn't fully understand the issues as I now recognize you where trying to kill three birds with one stone, and late at night after a glass or two of rich red I took a simplistic view that you where talking about sight suitability, rather than the overall cause and effect of wind energy collection, on the positive side you made me decide to reduce the weight of my blades to get a better energy track on gusts.
All the work gone into modeling under ideal conditions doesn't tell a true story as you so ably demonstrated with your post.
All the best
Bob Foolin Around
VK4AYQ Guru Joined: 02/12/2009 Location: AustraliaPosts: 2539
Posted: 01:41am 03 Jan 2011
Copy link to clipboard
Print this post
Hi Pete
New second hand Puter ex government lease dual core thingo and its got USB 2 that works
Not to bad for $300 with everything.
I salvaged the HD from the old one onto a backup drive so didn't loose anything important, now loading programs and drivers to get it all together.
I lost the program for the weather station unfortunately but that is the biggest issue so far. I really appreciate your offer and will talk to you on the land line.
All the best
BobFoolin Around
RossW Guru Joined: 25/02/2006 Location: AustraliaPosts: 495
Posted: 09:01pm 03 Jan 2011
Copy link to clipboard
Print this post
I took a rather different aproach to visualising the wind....
Basically, I sample the wind every 5 seconds. I then display its speed and direction by plotting a dot at it's position on the "rose". In order to see the third important element - time - I colour code the time over a 24 hour period. There's a little line graph at the bottom that shows time/colour, and wind run for each minute to give an idea of windspeed trend.
It conveys a lot of information at a glance, once you get used to it.
GWatPE Senior Member Joined: 01/09/2006 Location: AustraliaPosts: 2127
Posted: 10:41pm 03 Jan 2011
Copy link to clipboard
Print this post
Hi Ross,
This looks like a pic of the south pole.
This is similar chart info to a Met. wind rose. The coloured timebase adds a nice touch.
I have wind direction available as well on my wind sensor, but I don't normally log it, except on my weather station. Are you able to change the timebase from a coded 24hr period to say an update of the rolling last 24hr period?
The dot chart I use for plotting windmill power v rpm or windspeed could be made to output similar information as a part of my windmill analyzer. I would consider adding this if you did not mind.
The addition of (c) info embedded into the chart is a useful touch as well, for publishing, that I don't currently do.
Real time graphical data update is so useful. I think that a reduced timebase may be good as well, as In my own plotting I have found that the most recent data is hard to see in amongst a sea of dots in a 2D and 3D representation. I also notice that Glenn has added a real time graph to the PicLog program. I found that when I added real time graphing to my windmill analyzer that I had so much more info available without needing to manually delve into historical data. Clients using my analyzer found this really good on the public viewing screen.
I think that having some sort of standardized visual data format would be very useful. Comparing aples with apples would certainly clear up a lot of confusion with weather and certainly windspeed data.
The wind direction does not change the amount of power my windmill makes. Logging wind direction has been a low priority for me for this reason. Displaying the full wind data on a single chart would be most useful in looking at seasonal changes of wind patterns. The raw logged data from my Davis weatherstation is binary encoded, so extraction of the basic data will need some work. I have records going back to 2005 so a wind rose viewing might offer fresh light compared to overlaying strip charts. I should add wind direction to the suite of parameters I now log. It is certainly easier to view day to day, or month to month trends with a dot graph like you have done.
Gordon.
become more energy aware
VK4AYQ Guru Joined: 02/12/2009 Location: AustraliaPosts: 2539
Posted: 10:48pm 03 Jan 2011
Copy link to clipboard
Print this post
Hi Ross
Thats a nice touch and it sure adds dynamics to a graph for a presentation.
All the best
BobFoolin Around
RossW Guru Joined: 25/02/2006 Location: AustraliaPosts: 495
Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me.
Print this post
