How to calculate activation function value

Dear all,

I am currently trying the ANN to formulate a mathematical model for my system.

I read a paper stated :

I would like to know whether we can calculate the activation function of f1 and f2. And if we can, what is "n"? is it the number of output layer or hidden layer?

Since this guy can simplyfy his equation into :

In his case, T1 and T2 is the output variable while p is the input variables (there are 12 inputs variables).

Could someone share me some lights here?

Thank you.

Best regards,

Wegik

Answer 1

[JESUS DAVID ARIZA ROYETH]

"n" is the output of the second layer before going through the transfer function, therefore, and

[Wegik Prasetyo]

Thank you Jesus.

How to calculate the f1 and f2 itself?

From the screenshot, it is straightforward right, that I need the value of f1 and f2 and input those value on the generic model and then i will get my mathematical model.

However, I dont understand how we can get the f1 and f2.

And also from the screenshot, this paper used logsig transfer function. And you provide me with the equation to calculate "n". However, in your equation, it still contains the "f1".

I actually like want to know how to get the value of f1 and f2, so i can put them on the generic model to get my spesific model.

Thank you.

Best regards,

Wegik

[JESUS DAVID ARIZA ROYETH]

this pdf can help you http://128.174.199.77/matlab_pdf/nnet.pdf page 2-8