.png){kind=icon}
.png){kind=icon}
Return
Neuro-PSO algorithm for large-scale dynamic optimization
.png){kind=icon}
OA
AI
DOI：10.1016/j.swevo.2025.101865.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Over the last few decades, dynamic optimization and large-scale optimization have been two challenging research topics. In this context, dynamic optimization with high dimensionality is undoubtedly another important research topic. For such a combined problem, this paper develops: (1) an algorithm that incorporates problem decomposition to deal with high dimensionality, (2) a search algorithm for optimization, and (3) a prediction strategy to deal with dynamic changes. Firstly, a decomposition method is introduced to divide the problem into multiple subproblems based on the level of interactions among the decision variables. For optimization, a multi-population search algorithm is proposed, where each subpopulation evolves individually. Finally, a machine learning-based prediction strategy is developed to learn information from historical solutions and predict some solutions that maybe useful for the new environment. The proposed algorithm is tested using the generalized moving peaks benchmark problems. The results show that the proposed algorithm can find better solutions than existing approaches.
Keywords:
Evolutionary dynamic optimization
Large-scale dynamic optimization problems
Tracking moving optimum
Generalized moving peaks benchmark
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
8.5
Papers:
2.1K
Citations:
1.0W