.png){kind=icon}
.png){kind=icon}
Return
MNEARO: A meta swarm intelligence optimization algorithm for engineering applications
.png){kind=icon}
PRE
AI
DOI：10.1016/j.cma.2023.116664.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Artificial rabbits (AR) optimization is a newly proposed swarm intelligence algorithm. The idea of AR optimization (ARO) is inspired by the nature behaviors of the AR including the detour foraging, the random hiding and the energy shrink. Although ARO has better performance than many classical natural heuristic algorithms, it still has some disadvantages such as lacking the diversity of the population and easy to trap in the local optimal. To mitigate the influence of the above defects, this paper respectively introduces the mutation strategy, the prey identification strategy and the elite opposition-based learning strategy into the original algorithm, and proposes a new meta swarm intelligence optimization algorithm termed MNEARO. Firstly, the mutation strategy is introduced into the two natural behaviors of ARO to expand its exploration range in the solution space and increase the probability of finding the global optimal. Secondly, replacing the captured individuals with the predators that take better action trajectories to achieve the population evolution. Another candidate population is generated by the prey identification strategy, thus improving the population diversity and dynamics of ARO. Finally, the elite opposition-based learning strategy improves the exploitation ability of ARO and the overall quality of the population by selecting some individuals with the better fitness values as the elites and using the greedy selection before the end of each iteration. A series of performance tests are conducted on the CEC2020 test set in four dimensions and the CEC2022 test set in the highest dimension respectively to verify that the proposed MNEARO has better convergence and robustness. In addition, the practicality of MNEARO is further highlighted through five mechanical optimization designs, two truss topology optimization designs and the vehicle cruise control system. The experimental results show that the proposed MNEARO has a strong competitiveness in solving complex optimization problems with different dimensions.
Keywords:
AR optimization
The mutation strategy
The prey identification strategy
The elite opposition-based learning strategy
Optimization problem
Journal
IF:
7.3
Papers:
1.3W
Citations:
5.6W