ABSTRACT
The goal of optimization is finding the minimum value for the target function, determining the initial values for algorithm parameters is important. In case that the initial values are not chosen rightly, the algorithm may diverge or may converge to a suboptimal solution. Important parameters in optimization algorithm include min and max values of speed, min and max values of position, and learning parameters c1 and c2. The particle that is closer to the target has more competence. First a group of particles are created randomly, and by updating groups, one of them may seek to optimize the solution. The best position obtained so far is called best. The other best value used by the algorithm is called best which is the best position obtained so far by the population. In some literatures the particle only chooses other particles which are its neighbors topologically; in such situation the best local solution is called best. If enough particles are given, PSO guarantees to get the global optimum solution of the objective function. A gas quality of very from 0.90 to 0.94 was obtained by using initial velocity state of particle swarm technique, by using tow factorized flow arrangements occurred as of time for the heterogeneous and homogenous network processing micro channel kept under fixed inlet manifold conditions and the single micro-channels was reassessed with two-phase parallel flows. To prevent Mixing of different parts, Wrong parts or a Parts box of less than unit quantity, set the handling rules of partial parts boxes, having been generated unavoidably, and have operators obey the rule.
