A Nature-inspired Algorithm based on Classical-conditioning Theory

Nature-inspired algorithms are the imitation of nature opened a new era in calculations for solving optimization problems. In this thesis, we will provide an optimization algorithm inspired by nature using the instinctive behavior of birds. In this thesis, particles learn to have a conditional normal behavior towards an unconditioned stimulus using the classical conditioning learning behavior of birds. Particles will be divided into multiple categories in the problem space. If any particle had a low-level category, it will try to move towards its best personal experience. If any particle had a high-level category, it will learn to move towards the global optimum in its category. Using the idea of birds’ sensitivity towards the environment, in which birds are flying, we tried to move particles in incompetent spaces more quickly so that the particle goes far away from that space, and vice versa, we will bring down the particles’ speed in valuable spaces to search for more. We selected a population based on the particles’ merit in the initial population selection using the instinctive behavior of birds. The proposed method was implemented in MATLAB software, and the results have been compared in several different ways. The results showed that the proposed method is a reliable algorithm to solve the static problems.