Boosting quantum rotation gate embedded slime mould algorithm
| Autor: | Ali Asghar Heidari, Lejun Zhang, Weibin Chen, Xue Xiao, Caiyang Yu, Huiling Chen |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
0209 industrial biotechnology
education.field_of_study Boosting (machine learning) Computer science Population General Engineering 02 engineering and technology Computer Science Applications 020901 industrial engineering & automation Rate of convergence Artificial Intelligence Robustness (computer science) Test set 0202 electrical engineering electronic engineering information engineering Trajectory 020201 artificial intelligence & image processing education Rotation (mathematics) Algorithm Randomness |
| Zdroj: | Expert Systems with Applications. 181:115082 |
| ISSN: | 0957-4174 |
| DOI: | 10.1016/j.eswa.2021.115082 |
| Popis: | The slime mould algorithm is an interesting swarm-based algorithm proposed in 2020 based on this entity's trajectory finding abilities in nature. It simulates slime mould movement, foraging, and other behaviors to find the problem's optimal solution. Because of the complexity of the slime mould's trajectory, the SMA has strong randomness and makes the generated population diverse. However, in the late iteration of the algorithm, as the complexity of the problem to be dealt with increases, it tends to drop into the local best, and the convergence rate slows down. Therefore, in this study, an improved SMA, named WQSMA, is proposed to remedy the above imperfections. Specifically, the two strategies of quantum rotation gate and an operation from water cycle are used for the first time to improve the robustness of the original SMA. The purpose of adding both mechanisms is to keep the algorithm in equilibrium among exploration and exploitation inclinations. While expanding the search space of individual population, it also makes a more detailed exploration of the local area. The quantum rotation gate, which rotates by its small angle, can adequately exploit the algorithm and search in the local scope enough. Simultaneously, the water cycle mechanism can help the algorithm search thoroughly in the space to find the optimal solution. The improved algorithm was compared with 14 classical meta-heuristics and 14 advanced algorithms on the test set IEEE CEC 2014, and the results were obtained, with WQSMA ranking first in both comparisons. Also, to further illustrate the role of WQSMA in practical application, three engineering problems are used for verification. Experimental results show that WQSMA also performs well in solving such practical problems. A website at https://aliasgharheidari.com will support this research. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect