Title :
Derivative-free optimization with use of problem structure: Applications to oil production
Author :
Caio Merlini Giuliani;Eduardo Camponogara
Author_Institution :
Department of Automation and Systems Engineering, Federal University of Santa Catarina, Florianó
Abstract :
Derivative-free methods allow optimization in cases where the objective function is calculated by computer simulation and one does not have access to its derivatives. Such methods are known to require numerous function evaluations in order to achieve final convergence. This is especially true when the number of variables is large. This work studies the use of the Alternating Direction Method of Multipliers to exploit the structure of the optimization problem, allowing the solution of problems with fewer variables to be carried out by a derivative-free trust-region method. An application to the maximization of oil production of gas-lifted wells is made.
Keywords :
"Optimization","Manifolds","Modeling","Interpolation","Linear programming","Polynomials"
Conference_Titel :
Automation Science and Engineering (CASE), 2015 IEEE International Conference on
Electronic_ISBN :
2161-8089
DOI :
10.1109/CoASE.2015.7294173