Читать книгу Nature-Inspired Algorithms and Applications - Группа авторов - Страница 33

FA is the swarm-based metaheuristic approach which is introduced by Xin. The behavior such as flashing lights of the fireflies is inspired and utilized in the algorithm. The algorithm utilizes the concept that fireflies are always both sex and implies that any firefly can be engrossed by some firefly and the ability of the desirability of the firefly is directly relational to the ability of its brightness which depends upon the goal work. A firefly will be pulled in to the firefly with more brightness.

The working function FA has the following steps.

1 Objective function is initialized by absorbing the light intensity.

2 Initial population of the firefly is generated.

3 For every firefly, the light intensity is determined.

4 Attractiveness of the firefly is calculated.

5 The firefly which has brightness level of minimum is moved toward the firefly which has brightness level of maximum.

6 Light intensity of the firefly is updated.

7 Fireflies are ranked based on the intensities and best solution is found.

The advantages of the FA are that it has an ability to die with nonlinear more effectively, optimization of multimodal problem can be solved naturally, there is no need of velocity as it is needed in PSO, solution to the global optimization problem can be found as soon as possible, it is flexible to integrate with other technique of optimization, and initial solution is not required. The disadvantage of FA is it consumes more time to reach the optimal solution. FA is used in the field of semantic web composition, classification and clustering problems, neural network, fault detection, digital image compression, feature selection, digital image processing, scheduling problems, and TSP.