Register to reply

Determining RPM

by cordines
Tags: determining
Share this thread:
cordines
#1
Feb10-13, 06:17 AM
P: 15
Hello,

I am trying to build a SIMULINK model in Matlab in order to determine the RPM of a wind turbine. Can someone please tell me how I can obtain the RPM of a wind turbine from the voltage fluctuations of a generator?

Any good sites in the subject are also welcome. Thanks in advance.
Phys.Org News Partner Engineering news on Phys.org
Philips introduces BlueTouch, PulseRelief control for pain relief
3-D printing leads to another advance in make-it-yourself lab equipment
Nanoscience makes your wine better
NascentOxygen
#2
Feb10-13, 08:31 AM
HW Helper
Thanks
P: 5,490
Is it a DC generator, or AC?
cordines
#3
Feb10-13, 09:23 AM
P: 15
It's a permanent magnet electrical generator, having a power output of 1.8 kW.

cordines
#4
Feb10-13, 09:32 AM
P: 15
Determining RPM

Generator is AC.
SirAskalot
#5
Feb10-13, 12:29 PM
P: 141
Are you familiar with electric generators/motors? Equivalent circuit? d-q axis? Electromagnetic's?

And do you have the specs. on the generator? Impedances? Is it converter feed, or direct driven? Where do you measure voltage?
NascentOxygen
#6
Feb10-13, 07:07 PM
HW Helper
Thanks
P: 5,490
Quote Quote by cordines View Post
Generator is AC.
So the faster it spins, the higher the output frequency?
cordines
#7
Feb11-13, 01:36 AM
P: 15
No, I'm not that familiar with electric generators. However we did do the equivalent circuits of the generators/motors in my first year in mechanical engineering albeit it being a tiny bit. With regards to the others, I'm not that familiar.
cordines
#8
Feb11-13, 01:37 AM
P: 15
Yes, since the frequency is directly proportional to the rotational speed.
NascentOxygen
#9
Feb11-13, 02:13 AM
HW Helper
Thanks
P: 5,490
Quote Quote by cordines View Post
Yes, since the frequency is directly proportional to the rotational speed.
So that answers your question? You measure the frequency, and that tells you the machine's rotational speed. You do first need to establish the proportionality constant, k, because

RPM = k frequency

It may be possible to determine the number of electrical cycles per revolution by connecting a voltmeter to the generator and rotating the shaft by hand, counting how many times the voltage peaks (in positive polarity). Or else by examining the manufacturer's info on the generator.
NascentOxygen
#10
Feb11-13, 06:18 AM
HW Helper
Thanks
P: 5,490
It might be something as simple as saying if it's a 2 pole generator then you'll get 1 cycle of AC for a full rotation of the shaft; if it's a 4 pole generator then you'll get 2 cycles of AC for one rotation of the shaft, etc.

Do you have a link to your particular model on its manufacturer's web site?
cordines
#11
Feb11-13, 06:41 AM
P: 15
The problem is, how can I do that by Simulink?
skeptic2
#12
Feb11-13, 08:55 AM
P: 1,822
DC generators will produce pulsating DC equal to the rotational speed times the number of poles. If the generator were connected as a DC generator but instead of using a constant field current, the field were connected to the AC line, the output would be the pulsating DC modulated by the line frequency. The output will be an irregular sine wave which, regardless of how fast the generator was turning would have an output at the line frequency. This seems like a much simpler method of regulating and synchronizing the output frequency of myriads of wind turbines than somehow regulating their rotational speed and phase. Assuming this is what is happening, determining the RPM could be done with an FFT of the output and looking at the frequency of the second highest peak frequency and dividing by the number of poles.
milesyoung
#13
Feb11-13, 09:02 AM
P: 533
Are you asking how to calculate the fundamental frequency of one of the phase voltage waveforms in Simulink?

If so, a common way to go about it is to determine the time between zero crossings of the phase voltage in order to calculate its period. You'd probably need to add a little hysteresis so you don't get a bunch of hits at the zero crossing due to "noise" and whatnot. Feeding the waveform through a lowpass filter beforehand will help you decrease the hysteresis band.

I guess there might be a "FFT block" in your Simulink toolkit aswell, so that's probably also an option.

You'd need to know the number of pole pairs though, as NascentOxygen pointed out, to link mechanical to electrical frequency.
cordines
#14
Feb11-13, 03:44 PM
P: 15
Thanks milesyoung. I will try to implement that in Simulink, don't know if I will succeed since I'm new to this. I will get back to you. Thanks again!


Register to reply

Related Discussions
Determining std dev Math & Science Software 2
Determining IFS Calculus & Beyond Homework 3
Determining f Calculus & Beyond Homework 7
Need help in determining the lim Calculus & Beyond Homework 3
Help! determining Keq using NMR Chemistry 0