On a Novel Constraint Handling Methodology for Stochastic Optimization

BLANCO ANÍBAL; SANCHEZ MABEL; ALBERTO BANDONI

Buenos Aires, Argentina

Congreso; XII Interamerican Congress of Chemical Engineering; 2006

Interamerican Confederation of Chemical Engineering

The MaxMin methodology to solve nonlinear systems of equations with genetic algorithms is proposed. The developed methodology consists of an adequate reformulation of the system of equations into a nonconstrained optimization model, which can be effectively addressed with standard genetic algorithms. The proposed methodology is also applied to mixed integer nonlinear constrained optimization problems which present challenging features for deterministic and stochastic algorithms. equations with genetic algorithms is proposed. The developed methodology consists of an adequate reformulation of the system of equations into a nonconstrained optimization model, which can be effectively addressed with standard genetic algorithms. The proposed methodology is also applied to mixed integer nonlinear constrained optimization problems which present challenging features for deterministic and stochastic algorithms. equations with genetic algorithms is proposed. The developed methodology consists of an adequate reformulation of the system of equations into a nonconstrained optimization model, which can be effectively addressed with standard genetic algorithms. The proposed methodology is also applied to mixed integer nonlinear constrained optimization problems which present challenging features for deterministic and stochastic algorithms. equations with genetic algorithms is proposed. The developed methodology consists of an adequate reformulation of the system of equations into a nonconstrained optimization model, which can be effectively addressed with standard genetic algorithms. The proposed methodology is also applied to mixed integer nonlinear constrained optimization problems which present challenging features for deterministic and stochastic algorithms. The MaxMin methodology to solve nonlinear systems of equations with genetic algorithms is proposed. The developed methodology consists of an adequate reformulation of the system of equations into a nonconstrained optimization model, which can be effectively addressed with standard genetic algorithms. The proposed methodology is also applied to mixed integer nonlinear constrained optimization problems which present challenging features for deterministic and stochastic algorithms.