.png){kind=icon}
Return
BTCMOEA: A bidirectional knowledge transfer driven algorithm for constrained multiobjective optimization
.png){kind=icon}
PRE
AI
DOI：10.1016/j.eswa.2026.133649.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Constrained multiobjective optimization evolutionary algorithms (CMOEAs) have been widely applied in practical engineering. However, when dealing with complex feasible region structures, existing methods often weaken the self-exploration ability of the population due to premature knowledge transfer, and the robustness of the algorithm decreases due to parameter sensitivity. Therefore, this article proposes a bidirectional transfer constrained multiobjective optimization evolutionary algorithm (BTCMOEA). This algorithm strengthens the autonomous exploration ability of the main population in the early stage through a segmented evolution strategy, avoiding ineffective migration; In the later stage, this algorithm establishes a bidirectional co-evolutionary mechanism to enhance the information exchange ability between populations, thereby reducing the sensitivity of algorithm parameters while ensuring the convergence and robustness of the algorithm in complex constrained scenarios. The experimental results show that BTCMOEA exhibits superior performance on 43 benchmark test functions and 9 real-world mechanical design optimization problems.
Keywords:
Multiobjective optimization
Evolutionary algorithm
Bidirectional transfer
Knowledge transfer
Mechanical design optimization
Journal
IF:
7.5
Papers:
3.0W
Citations:
10.2W
Organization
Cited Papers
No cited papers available