China
Africa
Explore chapters and articles related to this topic
Big Data Analysis on Smart Tools and Techniques
Published in Gautam Kumar, Dinesh Kumar Saini, Nguyen Ha Huy Cuong, Cyber Defense Mechanisms, 2020
Jabar H. Yousif, Dinesh Kumar Saini
Deterministic global optimization (DGO) is numerical optimization methods that seek to find a solution for the global optimization problem. DGO is implemented in some cases that seek to get a global solution for mathematical models with a minimum optimization problem. It aims to use the analytical properties of the variables for generating a series of features that help to concentrate on the optimal global solution. The deterministic global optimization techniques fail to give a rigorous result of implicit black-box code to return function bounds [22]. Therefore, the computational graph techniques are used to represent the system with a deterministic global optimization problem. The typical methods for DGO are Inner and outer approximationCutting-plane methodsBranch and bound methodsInterval methods.
Parameter Estimation
Published in Alex Martynenko, Andreas Bück, Intelligent Control in Drying, 2018
Deterministic optimization techniques rely in most cases on variational calculus. Well-known methods aiming at an iterative approximate solution are gradient methods like the Levenberg-Marquardt and Gauss-Newton algorithms which use (analytical or numerical) derivatives of the cost function, and direct search techniques like the Nelder-Mead simplex algorithm which depends on evaluation of the cost function at a variety of parameter values. The success rate of this method depends significantly on the choice of the initial estimate. In general, it cannot be guaranteed that a found minimum is a global one and thus one cannot be certain whether the estimated parameter set is the optimal one. For this reason, deterministic optimization is often repeated with different initial parameter estimates. Furthermore, advanced deterministic global optimization techniques like branch and bound algorithms can be applied, which yield the global optimal parameter sets. Further details can be found in Bonnans et al. (2006).
Multiobjective optimization and nonlinear model predictive control of the continuous fermentation process involving Saccharomyces Cerevisiae
Published in Biofuels, 2022
ANTIGONE is a deterministic general mixed-integer nonlinear global optimization framework. Deterministic global optimization of mixed-integer nonlinear programs is broadly applicable in diverse domains ranging from molecular biology to refinery operations to computational chemistry to synthesizing sustainable processes.