.png){kind=icon}
.png){kind=icon}
Return
Data driven strategy on structural optimization using beetle-genetic hybrid algorithm
.png){kind=icon}
PRE
AI
DOI：10.1016/j.istruc.2025.108314.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
The local searching ability of adaptive genetic algorithm (AGA) cannot match the global searching ability under high pressure selection, leading to the decline of its comprehensive searching ability. In order to tackle this problem, a beetle antennae hybrid genetic algorithm (BAGA) is proposed based on data driven strategy. Firstly, the beetle antennae operator (BA) which is refined from the beetle antennae search (BAS) is inserted into the genetic algorithm to partly reform new individuals generated from the genetic algorithm. According to the cask effect, the comprehensive searching ability of the algorithm can then be hugely improved with the employ of adaptive elite reservation strategy by balancing the local search ability and global search ability. Secondly, the data-driven strategy is adopted to ameliorate the complexity caused. Finally, 10-bar spatial truss and 72-bar spatial truss illustrative examples are carried out to test the performance of the algorithm, the optimization results obtained indicate that the BAGA has higher feasibility and stability compared to other meta-heuristic algorithms. Moreover, the heuristic rule adopted in the algorithm is highly in coincidence with the transmission path achieved from the topology optimization results, proving the feasibility of data-driven strategy in the application of the BAGA.
Keywords:
Genetic algorithm
Data driven
Beetle antennae
Heuristic rule
Sensitivity analysis