Title :
A Hybrid Vector Artificial Physics Optimization with One-Dimensional Search Method
Author :
Yang, Gangjun ; Xie, Liping ; Tan, Ying ; Zeng, Jianchao
Author_Institution :
Key Lab. of Contemporary Design & Integrated Manuf. Technol., Northwestern Polytech. Univ., Xi´´an, China
Abstract :
The direction of each individual velocity vector is very useful information for search in Artificial Physics Optimization (APO) algorithm. APO´s simple way to dealing with the errant individual exceeded the lower or upper bounds constrains changes the direction of its motion vector. A shrinkage coefficient is introduced to ensure that each individual is within the decision space without changing its direction. One-dimensional search method is merged into the vector model of APO (VM-APO) to improve its local search capability. The simulation results confirm that the performance of the hybrid vector APO with one-dimensional search method is effective.
Keywords :
optimisation; physics; search problems; hybrid vector artificial physics optimization; one dimensional search method; shrinkage coefficient; velocity vector; Algorithm design and analysis; Electronic mail; Force; Optimization; Search methods; Upper bound; Artificial physics optimization; Global optimization; Newton´s Second law; Physicomimetics; Virtual force;
Conference_Titel :
Computational Aspects of Social Networks (CASoN), 2010 International Conference on
Conference_Location :
Taiyuan
Print_ISBN :
978-1-4244-8785-1
DOI :
10.1109/CASoN.2010.12
