IDEINFO: An Improved Vector-Weighted Optimization Algorithm

This study proposes an improved vector-weighted averaging algorithm (IDEINFO) for the optimization of different problems. The original (INFO) uses weighted entity structures and three core procedures to update positions vectors. First, rule phase is based on law convergence acceleration generate new Second, vector combination combines obtained vectors with rules achieve a promising solution. Third, local search helps eliminate low-precision solutions improve exploitability convergence. However, this approach pseudo-randomly initializes candidate solutions, therefore risks falling into optima. We, therefore, optimize initial distribution uniformity potential by using two-stage backward learning strategy initialize difference evolution perturb these in stage produce solutions. In phase, range expanded according probability values combined t-distribution strategy, global results. IDEINFO is, tool optimal design considerable efficiency case constraints.