A new evolutionary programming approach based on simulatedannealing with local cooling schedule

摘要

The NPOSA (New Population-Oriented Simulated Annealing) techniqueis introduced as an efficient global search tool to solve optimizationproblems. Unlike the conventional simulated annealing or its hybridalgorithms, each individual in the population can intelligently plan itsown annealing schedule in an adaptive fashion to the given problem athand. This not only enhances the search speed but furthermore yields asolution near the global optimum. This technique has been applied tosolve the traveling salesman problem (TSP) for combinatorialoptimization, as well as a continuous function optimization problem, todemonstrate its validity and effectiveness