1. I'm computing this integral

Int(exp(At)*B*BT*exp(AT*t))

Where A is a nxn matrix
B is a n x 1 vector
BT = transpose(B)
AT = transpose(A)

If you could factor out the B*BT as a constant then you might be able to write it as

B*BT* integral(exp(A + AT)*t))

I can integrate that rather easily.

I don't think you can factor B*BT out however. (since the order of matrix multiplication usually matters.)

There are no integral tables for things like this.

Anyways just looking for any thoughts or help.

2.

3. Well this is the first time i've ever heard mention of integrating matrices, so i won't be much help, but doing some matrix multiplication got me this.

If the entries of 'A' and 'B' contain no dependence on 't' then the integral is made really simple, but i'm guessing this isn't the case in general?

4. I found a way to integrate numerically in matlab finally. I'll plug this in later and see if I get the same results. It looks like you used the definition of exponential function as a series to simplify it a bit.

5. Originally Posted by GenerationE
I found a way to integrate numerically in matlab finally. I'll plug this in later and see if I get the same results. It looks like you used the definition of exponential function as a series to simplify it a bit.
That is what i did, although if you're using matlab then there's no point simplifying it... unless your matrix is very large, then the fact that is symmetric will help to (almost) halve the number of operations.

 Bookmarks
##### Bookmarks
 Posting Permissions
 You may not post new threads You may not post replies You may not post attachments You may not edit your posts   BB code is On Smilies are On [IMG] code is On [VIDEO] code is On HTML code is Off Trackbacks are Off Pingbacks are Off Refbacks are On Terms of Use Agreement