GA-optimized ANN model - multiple response - multi-objective optimization

Published: 24 October 2023| Version 1 | DOI: 10.17632/4rtjmc4bss.1
Contributor:
BISWANATH MAHANTY

Description

The data set is two predictor and two response models - ANN developed using GA-optimized architecture. Both the responses are optimized using multi-objective optimization. LIME and SHAP value analysis plots

Files

Steps to reproduce

IFBR.mat is the data file with two predictor and two responses. a1_ANN_GA_opt.m performs ANN architecture optimization using GA. The optimal network is stored in a2_ANN_opt_para.mat. a3_Network_and_plots.m used the information stored in a2_ANN_opt_para.mat to formulate the network, generate plots, and store the final network as a4_network.m. a5_multiobj performs multi-objective optimization. Other file names are self-explanatory

Institutions

Karunya University

Categories

Artificial Neural Network, Multi-Objective Optimization, Fluidized Bed Reactor

Licence