The Sailfish Optimizer: A novel nature-inspired metaheuristic algorithm for solving constrained engineering optimization problems
| Autor: | Hamid Reza Naji, Vahid Khatibi Bardsiri, S. Shadravan |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
0209 industrial biotechnology
education.field_of_study Optimization problem Computer science Population Swarm behaviour 02 engineering and technology Engineering optimization 020901 industrial engineering & automation Local optimum Artificial Intelligence Control and Systems Engineering Metaheuristic algorithms 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Electrical and Electronic Engineering education Metaheuristic Algorithm Global optimization |
| Zdroj: | Engineering Applications of Artificial Intelligence. 80:20-34 |
| ISSN: | 0952-1976 |
| DOI: | 10.1016/j.engappai.2019.01.001 |
| Popis: | Nature-inspired optimization algorithms, especially swarm based algorithms (SAs), solve many scientific and engineering problems due to their flexibility and simplicity. These algorithms are applicable for optimization problems without structural modifications. This work presents a novel nature-inspired metaheuristic optimization algorithm, called SailFish Optimizer (SFO), which is inspired by a group of hunting sailfish. This method consists of two tips of populations, sailfish population for intensification of the search around the best so far and sardines population for diversification of the search space. The SFO algorithm is evaluated on 20 well-known unimodal and multimodal mathematical functions to test different characteristics of the algorithm. In addition, SFO is compared with the six state-of-art metaheuristic algorithms in low and high dimensions. It also indicates competitive results for improvement of exploration and exploitation phases, avoidance of local optima, and high speed convergence especially on large-scale global optimization. The SFO algorithm outperforms the best algorithms in the literature on the majority of the test functions and it shows the statistically significant difference among other algorithms. Moreover, the SFO algorithm shows significantly great results for non-convex, non-separable and scalable test functions. Eventually, the promising results on five real world optimization problems indicate that the SFO is applicable for problem solving with constrained and unknown search spaces. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect