Improved Opposition-Based Particle Swarm Optimization Algorithm for Global Optimization

Particle Swarm Optimization (PSO) has been widely used to solve various types of optimization problems. An efficient algorithm must have symmetry information between participating entities. Enhancing efficiency relative the symmetric concept is a critical challenge in field security. PSO also becomes trapped into local optima similarly other nature-inspired algorithms. The literature depicts that order pre-mature convergence for algorithms, researchers adopted parameters such as population initialization and inertia weight can provide excellent results with respect real world This study proposed two newly improved variants termed Threefry opposition-based ranked (ORIW-PSO-TF) Philox (ORIW-PSO-P) (ORIW-PSO-P). In variants, we incorporated three novel modifications: (1) pseudo-random sequence utilization population; (2) increased diversity learning used; (3) introduction rank-based amplify execution standard acceleration speed. are examined on sixteen bench mark test functions compared conventional approaches. Similarly, statistical tests applied simulation obtain an accurate level significance. Both show highest performance stated benchmark over addition this, ORIW-PSO-P training artificial neural network (ANN). We performed experiments using fifteen datasets obtained from repository UCI. Simulation shown ANN algorithms provides best than traditional methodologies. All observations our simulations conclude ASOA superior optimizers. addition, predict how method profoundly impacts convergence.