Anders Björk

QUOTE (torekp @ Jul 11 2008, 05:51 PM) Sorry for not answering this so fast, have been on vacation. Since the scores from a PCA models are a few new coordinates they are eaiser to apply some classification rule than the orignal variables. The scores are the new coordinates of your data (the first, second,.. largest variation in your measured variables). You could see a PCA model as model capturing the most of variation in your data and where you can select the number of underlying phenomenas you want to discribe by determining the number of principal components, PC (new coordinates). The principal components are preprendicular to each other, orthagonal. The PCA model X=TP'+E T scores matrix (Objects row, columns the score value for PC 1, 2 to n) P Loading matrix E Error matrix (residuals) X Data matrix (Objects rows, Variables columns)

QUOTE (torekp @ Jul 3 2008, 08:23 PM) Using the SVD-vi or coding up the NIPALS-algorithm. There are also examples on PCA here on Lava with some vis. The SVD approach would work well if you dont have extremely many scrap piecies, say less than some thousands. This since I guess you dont have more than 10 meaures per scrap piece.

Why not use principal component analysis to cluster different scrap classess togheter instead, from my perspektive that would be eaiser (probally more robust). Ones you found the classes you could build a classifier based the on PCA scores. Another method to use would be discrimant analysis using partial least squares.

Run your matlab code through the MATLAB-node or modify your code so it is runable in Mathscript-node. Whats possible depend on your labview version.

Dear all We got a strange problem when moving from a Fieldpoint with a FP-AI-110 to CompactDaq rack with a 9203 (and also a AO, DO, DI module to in additon to the 9203, we used a 6009 before), there is one pressure sensor 4-20mA and one temperature sensor 4-20mA. Before the change the readings were normal now they have very high noise overlaid, resulting in a pressure from -2 to 3 bar when unpressurized. We have tried about 10 different ideas to eliminate the problem but with no success yet. Any ideasor suggestions are highly appreciated ! Best regards Anders

Only the old t-shirt is availble in the store :/ On other things the new figures is there...

I wonder if do not need to something similar to that discribed in this http://forums.lavag.org/COM-component-that...l&pid=29816 ? Otherwise import your dll and supply a header file, using Tools, Import and shared library.

QUOTE (Siconi @ Jun 10 2008, 11:54 AM) Look at this link it maybe useful http://www.morechemistry.com/labview/drive..._adam_4000.html

Does not ADAM-modules use Advantecs own device manger, check under Advantec on the program menu? Atleast the usb-module I use their own device manager, where set you the device number.

In Sweden they sometimes call to after one have been looking at some webinar or checked the preliminary price for some equipment. I try to be proactive to look at new thing before I really need them. Another thing is that since I am work for research institute we need approximate prices for grant application that may take 6-12 month before we get the decision. I feel it like this is the curse of the Information age. I think they should treat the web the same as their paper catalog.

QUOTE (jbrohan @ Jun 9 2008, 11:40 AM) Some windows programs are runable under linux using Wine. But I guess it would require a lot of tweaking with labview applications, do use a lot of IOs?

QUOTE (zhpipo16 @ Jun 4 2008, 08:54 PM) First you really dont need a flat sequence spanning over the complete diagram. Second you really want a 50 second delay, from the final write to the first read? Third if you would start reading immidately after the last write, what whould get then?

QUOTE (spectric @ May 29 2008, 03:03 PM) Does the code inside the while loop run fast? If yes, you might need to put a small delay into the while, say 10-100 ms. Otherwise you can get a nonresponsive user interface.

I am happy with my HP4400 on the road, last time it was three hours of wiring on a train and that was not problem. Although I only get 1024x768 in resolution, in the office I use 1280x1024. However I am not a full-time developer.

The 0.5 you are talking about is probally not 0.5Hz! It is most likely 0.5*SampleRate. This is standard signal processing.

Solidus Integration has a freeware library for read/write of matlab 6 files http://www.solidusintegration.com/si_projects.htm

QUOTE (Aristos Queue @ May 5 2008, 11:30 PM) Do all VIs need to been in memory at the same time? Couldnt this be made more robust?

farnoosh See your previous post, and do not re-post the same question.

Here you go. This can most possibly be done better. But I think it would work.

Seems like you need to do some unbundling, on the blue-with daq-data-line or re configure the DAQ-assistant so that it gives a double vector.

If only labview would have had a vector for axis tick values and axis tick labels similar to Matlab. Possible the mathscript node has this builtin, simple plot is built in.

Your problem is not related to some like this http://forums.lavag.org/COM-component-that...l&pid=29816

Sorting the amplitudes and taking the value at 50% or if no value exist at 50% do an interpolation between value below and above 50%. Now you would be able to the frequency indexes and interpolate.

What you wire to shift-register is what is hold by the register,unless you have broken wires (this happen if it not the same datatype on left and right side). It is an array that is save in register in this case.

How about storing your natural freq. array using shiftregister placed on the while loop? Then will be stored there until a new natural frequency test is done.