PSO-Based Neural Network Prediction and its Utilization in GMAW Process.

This paper presents a Particle Swarm Optimization (PSO) technique in training an Artificial Neural Network (ANN) which is used for predicting Gas Metal Arc Welding (GMAW) process parameters for a given input set of welding parameters. Experiments were conducted according to central composite rotatable design with full replication technique and results are used to develop a multiple regression model. Multiple set of data from multiple regression are utilised to train the intelligent network. The trained network is used to predict the weld bead geometry. The welding parameters welding current, welding speed, contact tip to distance, welding gun angle and pinch are predicted with consideration of performance of bead width, penetration, reinforcement and dilution. Instead of training with conventional back propagation algorithm a new concept of training with PSO algorithm is used in this paper. The proposed ANN-PSO model developed using MATLAB function is found to be flexible, speedy and accurate than conventional ANN system. [ABSTRACT FROM AUTHOR]

Copyright of Jordan Journal of Mechanical & Industrial Engineering is the property of Hashemite University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)