Adaptive Fish-Reborn Optimized Algorithm

A new strategy of optimization--------Artificial fish algorithm simulates the fish foraging,cluster and collision behavior.The optimized algorithm has characteristic of cluster intelligence, good parallelism, the robustness of the initial value of parameters and strong advantages, and has been used in engineering widely, such as combinatorial optimization, distribution decision and mechanical fault diagnosis,etc.The design idea of artificial fish algorithm is simple, and when dealing with low dimensional optimization function,it is the ability to maintain a higher precision and also obtain the global optimal solution quickly.But we encountered in the actual engineering problems tend to be large, the dimensions of the decision variables is higher,these will led to the search space complexity.Then using traditional artificial fish algorithm is easy to fall into local optimum, precision and speed of convergence will follow off.The well adaptive rebirth fish optimization algorithm is putted forward. First in each process of iteration, by using the method of reverse learning, constantly to the fish into the "new life" makes possible the rebirth of the fish, it enriched the diversity of population and search scope of the artificial fish, also increase the opportunity to jump out of local optimum.Then the possibility of improved algorithm to obtain the global optimal is increasing;Second using normal distribution dynamic to adjust the upper limit of crowded degree factor makes the algorithm more close to the fish search food process.The experimental results show that the improved algorithm not only ensure convergence speed, increase the likelihood of algorithm to obtain the global optimal, but also is suitable for solving large-scale optimization problems.In this paper, two examples of shoals of fish with the improved algorithm is optimized and the optimized result with the actual have good consistency, illustrates the effectiveness and practicability of the improved algorithm.