Optimization of Saw Process Parameters Using Particle Swarm Optimization

Welding process is used in most of the manufacturing Industrie which requires metal joining in a large scale. In any welding process it is very essential to optimize process parameters in order to achieve desired weld bead characteristics. In this work, combined objective function of maximizing the bead penetration, minimizing the dilution, reinforcement and width w/as considered. Four SAW process parameters (voltage, wire feed rate, welding speed, and nozzle to plate distance) were identified for optimization subjected to realistic process constraints. Several conventional techniques had been suggested in the literature for solving this problem. But these techniques are not robust and take iot of time to find the global optimum and are difficult to understand and implement. In order to over come the difficulties with conventional techniques a new technique called particle swarm optimization is implemented In this work. PSO is a simple and powerful technique based on the concept of social interaction to problem solving. In PSO a swarm search of n individuals communicate either directly or indirectly with one another for getting the search direction. This algorithm starts with 20 particles (solutions) and searches for the new ones by updating the velocities. Maximum of 500 iterations were performed and the solution was obtained. Software has been written using VC++ language. The solution obtained by this procedure is found to be superior. The computational effort is very less and easy to implement.