A modified quasisecant method for global optimization
© 2017 Elsevier Inc. This paper presents an algorithm for global optimization problem whose objective functions is Lipschitz continuous but not necessarily differentiable. The proposed algorithm consists of local and global search procedures which are based on and inspired by quasisecant method, res...
| Main Authors: | , , , |
|---|---|
| Format: | Journal Article |
| Published: |
Elsevier
2017
|
| Online Access: | http://hdl.handle.net/20.500.11937/57291 |
| _version_ | 1848760044576833536 |
|---|---|
| author | Long, Q. Wu, Changzhi Wang, Xiangyu Wu, Z. |
| author_facet | Long, Q. Wu, Changzhi Wang, Xiangyu Wu, Z. |
| author_sort | Long, Q. |
| building | Curtin Institutional Repository |
| collection | Online Access |
| description | © 2017 Elsevier Inc. This paper presents an algorithm for global optimization problem whose objective functions is Lipschitz continuous but not necessarily differentiable. The proposed algorithm consists of local and global search procedures which are based on and inspired by quasisecant method, respectively. The aim of the global search procedure is to identify “promising” basins in the search space. Once a promising basin is identified, the search procedure skips from an exhausted area to the obtained basin, and the local search procedure is then applied at this basin. It proves that the proposed algorithm converges to the global minimum solution if the local ones are finite and isolated. The proposed method is tested by academic benchmarks, numerical performance and comparison show that it is efficient and robust. Finally, The method is applied to solve the sensor localization problem. |
| first_indexed | 2025-11-14T10:09:31Z |
| format | Journal Article |
| id | curtin-20.500.11937-57291 |
| institution | Curtin University Malaysia |
| institution_category | Local University |
| last_indexed | 2025-11-14T10:09:31Z |
| publishDate | 2017 |
| publisher | Elsevier |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | curtin-20.500.11937-572912017-10-30T08:35:19Z A modified quasisecant method for global optimization Long, Q. Wu, Changzhi Wang, Xiangyu Wu, Z. © 2017 Elsevier Inc. This paper presents an algorithm for global optimization problem whose objective functions is Lipschitz continuous but not necessarily differentiable. The proposed algorithm consists of local and global search procedures which are based on and inspired by quasisecant method, respectively. The aim of the global search procedure is to identify “promising” basins in the search space. Once a promising basin is identified, the search procedure skips from an exhausted area to the obtained basin, and the local search procedure is then applied at this basin. It proves that the proposed algorithm converges to the global minimum solution if the local ones are finite and isolated. The proposed method is tested by academic benchmarks, numerical performance and comparison show that it is efficient and robust. Finally, The method is applied to solve the sensor localization problem. 2017 Journal Article http://hdl.handle.net/20.500.11937/57291 10.1016/j.apm.2017.06.033 Elsevier restricted |
| spellingShingle | Long, Q. Wu, Changzhi Wang, Xiangyu Wu, Z. A modified quasisecant method for global optimization |
| title | A modified quasisecant method for global optimization |
| title_full | A modified quasisecant method for global optimization |
| title_fullStr | A modified quasisecant method for global optimization |
| title_full_unstemmed | A modified quasisecant method for global optimization |
| title_short | A modified quasisecant method for global optimization |
| title_sort | modified quasisecant method for global optimization |
| url | http://hdl.handle.net/20.500.11937/57291 |