cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
Showing results for 
Search instead for 
Did you mean: 

Community Tip - Visit the PTCooler (the community lounge) to get to know your fellow community members and check out some of Dale's Friday Humor posts! X

how to synchronize all array ?(attached mathcad file)

Mathcad111
11-Garnet

how to synchronize all array ?(attached mathcad file)

is there any inbuilt function to synchronize all the array so that i will get something like (given picture"file name: output.png")?

1 ACCEPTED SOLUTION

Accepted Solutions

Comparing the two curves with index 9 (red) and 10 (green): How would an "equalization program" decide that the red increasing at 100 is wrong and should be decreasing like the green curve?

B.PNG

View solution in original post

21 REPLIES 21
-MFra-
21-Topaz II
(To:Mathcad111)

Hi,

Synchronizing seems excessive to me, but it is possible to average:

DES.jpgDES1.jpgDES2.jpg

please ask if you don't understand my question..

thanks for modification can please upload in mathcad file instead of picture ?
-MFra-
21-Topaz II
(To:Mathcad111)

Hi,

I did this analysis to see if something particular came out. But there is not. So it does not seem useful. However I attach the file.
  Greetings FM

DES3.jpg

You will have to explain what exactly you mean by "synchronize" and how the plot you show in your picture should be derived from the data you provide.

If averaging as shown by FM is what you are after, it can be done that way, too:

B.PNG

i have uploaded new files check once. 

LucMeekes
23-Emerald III
(To:Mathcad111)

See if this helps.

 

Success!
Luc

thank you for reply, but have problem with "min" function 

LucMeekes
23-Emerald III
(To:Mathcad111)

Create Nans this way:

 

Nans:=0

Nans[ORIGIN,ORIGIN+cols(Data)-1:=0

Nans:=Nans+NaN

 

 

Success!
Luc

i could not able to insert this symbol "["

in this line "Nans ORIGIN,ORIGIN+cols(Data)-1:=0"

LucMeekes
23-Emerald III
(To:Mathcad111)

Try this.

 

Essentially what the Equalize routine does is take away the (assumed !!) slope of every trace.

From the result it can be seen that that assumption may be wrong.

 

Success!
Luc

thanks a lot but i have to check this...!!!

why we can't able to sync. at center ?

LucMeekes
23-Emerald III
(To:Mathcad111)

That is possible too, but I guess that your graphs will diverge at the endpoints then.

I think the graphs are the result of a repeated measurement of some phenomenon, and that the measurements suffer from a non-constant 'offset'. It is not constant over the horizontal axis, and not constantly in- or decreasing (linear), so it is 'warped' in a more complex way. It may be quadratic or higher order, it may be exponential. I think you have to work that out before you can successfully 'align' these graphs.

In addition I wonder if the starting point for each of these graphs is the same with respect to the phenomenon that you are investigating.

 

Success!
Luc

yes the graph is not a constant offset, but only some channels (or cols of data) is miss matched. i wonder how to do that linearization in middle part of graph ?      

LucMeekes
23-Emerald III
(To:Mathcad111)

Here's a sheet with the EqualizeAt function. It allows you to select a point (on the x-axis) where all graphs (y-values) must be equal (to the average of all graphs at that point).

I've set it to the middle (133) but you can change it to any value from ORIGIN to ORIGIN+rows(Data)-1.

 

Success!
Luc

thank you but instead of selecting only one point in x-axis ('r') and solving this problem, why can't we set a linear line (i'm mean a straight line) and why can't we optimize or sync. according to that ?   (i'm not sure it maybe work or not)

LucMeekes
23-Emerald III
(To:Mathcad111)

Sure! That is what the (first) Equalise function does. But it chose then first and last points of the x-axis variable. And you were able to see how warped the graphs are...

Luc

sorry for late response, so we cannot achieve exact result like i have mention in the figure "output_2" ?

LucMeekes
23-Emerald III
(To:Mathcad111)

Not, unless you have information on how to 'unwarp' the measured traces. As demonstrated, assuming a linear in/de-crease (or less) does not suffice.

 

Success!
Luc

Comparing the two curves with index 9 (red) and 10 (green): How would an "equalization program" decide that the red increasing at 100 is wrong and should be decreasing like the green curve?

B.PNG

Top Tags