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

4-Participant

## Maximize function - Multiple output issue.

I am working through a simplified Pipe Stress calculation and having an issue with the Maximize function. I have tried to use the Maximize function to calculate the allowable free span of the pipeline, without success.

I am looking at the stresses in two pipelines and therefore require two results from the function. I have set up a Given/Find solve block without any issues. Can anybody explain if that a limitation of the Minimize function?

Cheers

Mike

1 ACCEPTED SOLUTION

Accepted Solutions
15-Moonstone
(To:MikeArmstrong)

How about the attached? Surely, sigma.s should be vectorised?

Alan

16 REPLIES 16
15-Moonstone
(To:MikeArmstrong)

Mike,

From the Help file: "If you are solving for n variables, the solve block must have n equations"

Alan

4-Participant
(To:AlanStevens)

I appreciate that Alan, but still this issue is driving me mad.

I have two values for sigma.c and sigma.t and I'm trying to find the maximum length that x can be (x is used for calculating the free span), so the combined longitudinal stressed do not exceed 95% yield.

M11 worksheet included.

Mike

15-Moonstone
(To:MikeArmstrong)

How about the attached? Surely, sigma.s should be vectorised?

Alan

4-Participant
(To:AlanStevens)

Cheers for taking the time Alan, but the calculated lengths should be 59.3m & 57.9m

Mike

15-Moonstone
(To:MikeArmstrong)
 Mike Armstrong wrote:Cheers for taking the time Alan, but the calculated lengths should be 59.3m & 57.9mMike

Are you sure? These are the results you obtain without vectorising sigma.s. I've vectorised sigma.s. Without vectorising the sigma.s function just returns a single value. Is that what you intended?

Alan

4-Participant
(To:AlanStevens)

Positive Alan,

Using the calculated spans in your example, the spanning stress would be above the specified yield with the other expected stresses.

Mike

4-Participant
(To:MikeArmstrong)

Sorry for the mulitple replies, the website is giving me grief.

I get the following.

Mike

15-Moonstone
(To:MikeArmstrong)

Ok, in that case, the attached should do it!

(Actually, my previous file was wrong also because I forgot to put subscripts on sigma.c and sigma.t!)

Alan

4-Participant
(To:AlanStevens)

Cheers Alan, nice work.

You have to admit that it looks horrible though. Maybe I'll stick with the Given/Find solve block for this issue.

Cheers

Mike

15-Moonstone
(To:MikeArmstrong)

I think you should compare Wp_sub.Z^-1 with Wp_sub/Z. They're not the same!

Alan

4-Participant
(To:AlanStevens)

Good spot Wayne, because I had brought the Z up, the formula required the vectorize operator.

Mike

4-Participant
(To:AlanStevens)

I have changed this to be correct Alan.

Cheers to both you and Wayne for looking.

Mike

Hi Mike,

Your much too bright, try dimming it down to my level.

4-Participant
(To:wayne)

Hahahahahah, stop being modest Wayne.

I am actually solving for x, I'm looking for the maximum allowable span, which I have found using a Given/Find solve block, but I can't get the Maximize solve block to work.

Mike

MIke,

(Thanks Alan, for how to use the solve block

My original comment was in error because I did not take the sq root to get xmax, but the point is that is easy to solve for directly. Why use a solve block?

Anyway, I get these lengths (Maybe I goofed)?

4-Participant
(To:wayne)
 Wayne Reid wrote:MIke,(Thanks Alan, for how to use the solve blockMy original comment was in error because I did not take the sq root to get xmax, but the point is that is easy to solve for directly. Why use a solve block?Anyway, I get these lengths (Maybe I goofed)?

I goofed, not you Wayne.

When I submit the calculation I will transpose the formula as you have done. I was simply playing around with the Maximize function.

Cheers

Mike

Announcements
Check out the latest