Average of Data

[tariqaziz12]

Hello ,
I have a problem and need your guidance.
I have testing data of four samples (Displacement and Load for each sample). I want to calculate average load and corresponding average displacement. As loading rate is different for all four samples, i am unable to take simple average of four values of load and displacement in every row (Please see the attachment in which i did same).
Please guide me how to take average? or how to take average of Displacement values at 5, 10, 15, 20 . . . . (so on) values of Load?

Thanks in advance

 

[Hui]

Tariqaziz

Firstly, Welcome to the Chandoo.org Forums

I assume you want a weighted average
so using Series s1 as a sample I would use
Displacement x: =SUMPRODUCT($B$3:$B$1681,$I$3:$I$1681)/SUM($I$3:$I$1681)
Load y: =SUMPRODUCT($B$3:$B$1681,$I$3:$I$1681)/SUM($B$3:$B$1681)

I'm not sure if that is correct or not, you should check with a supervisor

If you do that you'll end you with something like :

I'm not familiar with this and so can't comment as to its validity
Please check with a supervisor

I am also worried about the negative displacements
I would normalise the data first, ie: Add the minimum Displacement to each sample ie: -0.00283mm before doing the math above

 

[NARAYANK991]

Hi ,

A few points ; these are just my opinions , and may or may not be applicable in your case.

1. The graph should be drawn so that it is Displacement vs. Load , and not the other way around. This is not so important , but it is essential to understand that load is the independent variable , and displacement is the dependent variable. Correct me if I am wrong.

2. If this is following the generic stress / strain curve , then there are 3 or more sections to the graph ; there is an initial section where some load is required to be applied for any displacement to manifest itself.

Then there is a linear section where the displacement is directly proportional to the load.

Then there is a section where the displacement increases dramatically with a slight increase in the load.

Then there is probably a fracture section where the displacement is irreversible.

Generally , you should first take a look at the individual sample graphs to see whether all the samples are reasonably similar , if not identical.

Averaging should be done only in the linear section , and not in the other sections.

I would suggest that you ignore the first 200 data points , as well as the last 700 ; plot all the 4 sample data in separate charts in the range 200:900 and see whether they are reasonably similar. If possible look at the trendlines and see whether they are also similar.

Only after you confirm all this does it make it OK to average the 4 sets of data.

Narayan