.png){kind=icon}
Return
A complex fractional-order chaotic evolutionary algorithm with memory modulation for continuous optimization
.png){kind=icon}
PRE
AI
DOI：10.1016/j.chaos.2026.119154.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Complex fractional-order derivatives, which combine nonlocal memory properties with additional dynamic modulation capability, have shown great potential for the development of intelligent optimization approaches. To address the insufficient utilization of long-range historical information in existing chaos-driven optimization methods, this paper proposes a Complex fractional-order Chaotic Evolutionary Optimization (CoCEO) algorithm. Within a framework integrating hyperchaotic perturbation and differential crossover, the proposed algorithm introduces a Gr & uuml;nwald-Letnikov complex fractional-order historical memory mechanism and constructs a memory-modulated update rule by coupling the current position with multiple historical states, thereby enhancing the utilization of historical information, the regulation of search trajectories, and the coordination between global exploration and local exploitation. Furthermore, the update relation of CoCEO is rewritten in a matrix-recursive form, and a sufficient condition for the asymptotic decay of the homogeneous historical-memory subsystem under fixed parameters is derived using the characteristic polynomial and the transient historical-mode spectral radius. Experimental evaluations based on CEC 2022 show that CoCEO remains highly competitive with respect to optimization accuracy, convergence rate, and robustness, thereby confirming its effectiveness for complex continuous optimization problems.
Keywords:
Complex fractional-order derivatives
Hyperchaotic perturbation
Differential crossover
Multiple historical states
Convergence
Journal
C
IF:
5.6
Papers:
561
Citations:
0
Organization
View more organizations
Cited Papers
No cited papers available