Printable View
Hallo All...
I have a system I am trying to calibrate and I want to do this in Software...
I have a table of Inputs and the required output.
Normally I would just use the y = mx + c formula, but the plots I have are not linear.
So my question is: is there anyway I can build a formula from my plot points? One formula that can be used for all the different plots..
Any help would be appreciated.
If you have MSExcel then you can try to see if you can find an appropriate equation. Generate a scatter type graph with a line display for your data set, place your pointer on the displayed graph line, use the right mouse button to bring up an options list, select the "Add Trendline" option to see a window that allows you to test for several types of possible polynominal and log equation fits, select an equation and select the "Show equation on graph" option to see the equation for each selected trendline equation you test.
[QUOTE=JAlberts;7622]If you have MSExcel then you can try to see if you can find an appropriate equation. Generate a scatter type graph with a line display for your data set, place your pointer on the displayed graph line, use the right mouse button to bring up an options list, select the "Add Trendline" option to see a window that allows you to test for several types of possible polynominal and log equation fits, select an equation and select the "Show equation on graph" option to see the equation for each selected trendline equation you test.[/QUOTE]
Yes.... I also think the method of least squares fit (regression) graph can be created with sufficient number of data points..
Check out.. [url]http://terpconnect.umd.edu/~toh/models/CalibrationCurve.html[/url]