الفهرس 
AbstractOnly 14 pages are availabe for public view
 |  | from | 122 |  |  |
| | | from | 122 | | |
Abstract
Remarkable progress in optimization techniques leads to a rapid increase in their usage in industrial planning, scheduling, decision-making and variant complex problems. These techniques, which are classified into classical and soft computing, are seeking near the optimum solution. Thanks to fewer numbers of control parameters and easy implementation of the code, elephant-herding optimization (EHO) has been gained more interest. In spite of all the advantages EHO has, EHO needs some enhancements to deal with complex non-linear problems. Changing to better candidates using EHO equations is not guaranteed, besides, making the parameter α adaptive instead during the evolution of the EHO could be a better approach. As a result, a parametric study of EHO is carried out to come with new three derived algorithms from EHO. To be specific cultural-based EHO, alpha-tuning EHO, and biased initialization EHO are the new derived algorithms. A comparative study has been held between EHO variants and the most well-known state-of-the-art soft computing techniques. Case studies ranging from recent test bench problems CEC 2016, most well-known engineering problems gear train, welded beam, Three-bar truss design problem, in addition to continuous stirred tank reactor and fed-batch fermentor are used to validate and test the performance of the recommended EHO-based algorithms against state-of-the-art algorithms. Furthermore, EHO and the three enhanced algorithms used in the estimation of the best values of unknown parameters of single, double and three diode models of solar cells. These PV models represent 5-, 7- and ten unknown parameters of PV cells. Applications are employed on two types of PV solar cells: The 57mm diameter R.T.C. Company of France commercial silicon for single and double models and multi-crystalline PV solar module CS6P-240P for the three-diode model.