Premium
Shuffled Complex Evolution model calibrating algorithm: enhancing its robustness and efficiency
Author(s) -
Muttil Nitin,
Jayawardena A. W.
Publication year - 2008
Publication title -
hydrological processes
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.222
H-Index - 161
eISSN - 1099-1085
pISSN - 0885-6087
DOI - 10.1002/hyp.7082
Subject(s) - robustness (evolution) , simplex algorithm , algorithm , simplex , population , computer science , mathematical optimization , test suite , calibration , global optimization , suite , mathematics , machine learning , linear programming , test case , statistics , chemistry , biochemistry , geometry , demography , regression analysis , sociology , gene , history , archaeology
Shuffled Complex Evolution—University of Arizona (SCE‐UA) has been used extensively and proved to be a robust and efficient global optimization method for the calibration of conceptual models. In this paper, two enhancements to the SCE‐UA algorithm are proposed, one to improve its exploration and another to improve its exploitation of the search space. A strategically located initial population is used to improve the exploration capability and a modification to the downhill simplex search method enhances its exploitation capability. This enhanced version of SCE‐UA is tested, first on a suite of test functions and then on a conceptual rainfall‐runoff model using synthetically generated runoff values. It is observed that the strategically located initial population drastically reduces the number of failures and the modified simplex search also leads to a significant reduction in the number of function evaluations to reach the global optimum, when compared with the original SCE‐UA. Thus, the two enhancements significantly improve the robustness and efficiency of the SCE‐UA model calibrating algorithm. Copyright © 2008 John Wiley & Sons, Ltd.