Correlation factors between optimization paramaters, objectives and constraints (plot_OP_correlations)

Commandplot_OP_correlations
Module OP.pp
Description plot_OP_correlations plots the correlations between variables, objectives and constraints
Parameters
Restrictions To be run after optimization, is_save must be set to 1 in the Optimization Inputs file
See also plot_MS_correlations How to use multi-objective optimization with MANATEE

The aim of this post-processing is to show the correlations between parameters, objectives and constraints by using the correlation factor. It computes the correlation factors for all the designs evaluated during the optimization process.

The correlation factor between two variables X and Y Corr(X,Y) is given by dividing their Covariance Cov(X,Y) by their standard deviations \sigma_X and \sigma_Y.

Corr(X,Y)= {Cov(X,Y)\over\sigma_X . \sigma_Y}

The Correlation Factor varies between -1 and 1. A Correlation Factor of 1 means a perfect positive linear correlation between the two variables, a correlation of -1 means a negative linear correlation.

The results showed here come from the project demo_optim.m used in the How to use multi-objective optimization with MANATEE? article.

Results

Full Correlation Table
Full Correlation Table
Correlation factor between design variables and objectives
Correlation factor between design variables and objectives
Correlation Factor between objectives
Correlation Factor between objectives
Correlation Factors between constraints and objectives
Correlation Factors between constraints and objectives

Read more