![]() ![]() In this article, we'll be using it on a discrete search space - on the Traveling Salesman Problem. That being said, Simulated Annealing is a probabilistic meta-heuristic used to find an approximately good solution and is typically used with discrete search spaces. The slow cooling in this algorithm is translated as a lower probability to accept a worse solution than the current solution as the search space is slowly explored. We simulate the annealing process in a search space to find an approximate global optimum. This process serves as a direct inspiration for yet another optimization algorithm. The end result is a piece of metal with increased elasticity and less deformations which makes the material more workable. Successful annealing has the effect of lowering the hardness and thermodynamic free energy of the metal and altering its internal structure such that the crystal structures inside the material become deformation-free. It's a closely controlled process where a metallic material is heated above its recrystallization temperature and slowly cooled. Simulated Annealing is an evolutionary algorithm inspired by annealing from metallurgy. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |