Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Structural optimization

Status
Not open for further replies.

Maverick2021

Mechanical
Mar 17, 2021
1
SE
I am trying to create a response surface of a simple L-bracket. I am running 25 simulations, where in each run I change the bracket's dimensions. These points are distributed randomly between the minimum and maximum value. For each simulation I record the mass and the max Von-Misses stress.

After the simulation when I attempt to fit a curve between the dimensions and the stress, my machine learning alogrithm returns empty. Basically the algorithm could not find the relationship between the stress and the dimensions. I tried Linear, Polynomial, SVR, Neural networn and decision tress. All of them could not fit the data.

This is so strange because I find many scientifc researches and also commerical software that are doing this. Am I simplifying something or am I missing something? Any and all response are greatly appreciated!
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor