GUI for Grey Relational Analysis

The software for the Grey Relational Analysis to optimize the inputs for minimizing the difference between target and outputs

The software, using the Grey Relational Analysis to optimize the inputs for minimizing the difference between target and outputs and the code for the MATLAB GUIDE (GUI) is shown here.

The Grey Relational Analysis is a method to study the design and analysis of experiments for improving product quality if the parameters optimized are more than two. There are three types of loss functions: 1) “the nominal the best; “the smaller the better;” and “the larger the better.” These loss functions also are used in the Grey Relational Analysis. The quality characteristic “the nominal the best,” occurs whenever the output ‘y’ has a finite target value, usually nonzero, and the quality loss is symmetric on either side of the target. The smallest of the best of the loss function is used in this study because the targets, such as the nameplate parameters, are given. So, this particular loss function is used to optimize the input parameters to make the minimum difference between outputs and targets and to improve the shape of the time domain of output parameters of the dynamic simulation model of a squirrel-cage induction motor that has been developed by Ariunbolor Purvee.

The following four steps are used to calculate the evaluated S/N ratios:

Step 1. Distribution of 1s, 2s, and 3s into orthogonal arrays: The Rs, Rb, Re, Ls, Lb, Le columns of orthogonal arrays kept in orth18.txt have been divided into three columns each, matching the three experimental levels. For example, Column Rs is divided into columns labeled Rs1, Rs2, and Rs3. All 1s are in Rs1, all 2s are in Rs2, and all 3s are in Rs3.

Step 2. Replacement by S/N ratio: All 1s in Rs1, Rb1, Re1, Ls1, Lb1, Le1, and all 2s in Rs2, Rb2, Re2, Ls2, Lb2, Le2 and all threes in Rs3, Rb3, Re3, Ls3, Lb3, Le3 were replaced by the S/N ratios of the Grey relational grade corresponding to the experimental distribution of the row inTable X. The evaluated S/N ratio is calculated by (1).

Step 3. Reconstruct a table: The evaluated S/N ratios of the Grey relational grade are reconstructed into a table with six rows and three columns.

Step 4. Plot a Grey relational graph: The graphs are plotted