TS-SSA: An improved two-stage sparrow search algorithm for large-scale many-objective optimization problems

PLoS ONE, Год журнала: 2025, Номер 20(3), С. e0314584 - e0314584

Опубликована: Март 17, 2025

Large-scale many-objective optimization problems (LSMaOPs) are a current research hotspot. However, since LSMaOPs involves large number of variables and objectives, state-of-the-art methods face huge search space, which is difficult to be explored comprehensively. This paper proposes an improved sparrow algorithm (SSA) that manages convergence diversity separately for solving LSMaOPs, called two-stage (TS-SSA). In the first stage TS-SSA, this (MaOSSA) mainly through adaptive population dividing strategy random bootstrap strategy. second dynamic multi-population manage TS-SSA has been experimentally compared with 10 MOEAs on DTLZ LSMOP benchmark test 3-20 objectives 300-2000 decision variables. The results show significant performance efficiency advantages in LSMaOPs. addition, we apply real case (automatic scenarios generation), result shows outperforms other algorithms diversity.

Язык: Английский

Feature Niching Based Differential Evolution for Feature Selection on High-Dimensional Data

Communications in computer and information science, Год журнала: 2025, Номер unknown, С. 44 - 54

Опубликована: Янв. 1, 2025

Язык: Английский

A sensorless centrifugal pump operation states estimation method based on HSSA-BPNN

Measurement, Год журнала: 2024, Номер 237, С. 115238 - 115238

Опубликована: Июль 9, 2024

Язык: Английский