An Adaptive Flamingo Algorithm and Application for Compression Spring Design

The Flamingo Search Algorithm (FSA) is a new swarm intelligence optimization algorithm proposed in 2021. It has a unique search mechanism and mathematical model. However, when handling complex optimization problems, it has defects such as low convergence accuracy, low stability, and easy fall into local optimization. Given these shortcomings, a Flamingo algorithm with the adaptive and lens learning strategy (LILAFSA) is proposed. First of all, the population is initialized with the theory of good point set. Secondly, a nonlinear adaptive factor is added to balance the exploration and production capabilities of the algorithm. Then, the reverse learning strategy of convex lens imaging is introduced to improve the diversity of the population in the later stage. At the same time, the water wave dynamic adaptive factor is added to the attacker's location update to enhance the ability of jumping out of the local optimum. Finally, six benchmark test functions are used to test the ability of the improved algorithm and apply it to mechanical design optimization problems. The experimental results show that FSA has good performance in engineering design optimization problems.