Dynamic Optimization of Nonlinear Processes with an Enhanced Scatter Search Method

Abstract

An enhanced scatter search method for the global dynamic optimization of nonlinear processes using the control vector parametrization (CVP) approach is presented. Sharing some features of the scatter search metaheuristic, this new method presents a simpler but more effective design which helps to overcome typical difficulties of nonlinear dynamic systems optimization such as noise, flat areas, nonsmoothness, and/or discontinuities. This new algorithm provides a good balance between robustness and efficiency in the global phase, and couples a local search procedure to accelerate the convergence to optimal solutions. Its application to four multimodal dynamic optimization problems, compared with other state-of-the-art global optimization algorithms, including an advanced scatter search design, proves its efficiency and robustness, showing a very good scalability.

Modeling and Optimization Techniques with Applications in Food Processes, Bio-processes and Bio-systems

“…In order to surmount this difficulty, hybrid methods and metaheuristics have been recently presented for the solution of dynamic optimisation problems [7,16] or parameter estimation problems [30]. They speed up these methodologies while retaining their robustness and, provided a gradient based local method is used, they guarantee convergence to a gradient zero solution.…”

Embedded multiple shooting methodology in a genetic algorithm framework for parameter estimation and state identification of complex systems

“…The MS being a local search method, its performance in parameter estimation and state variable identification very much depends on the initial guesses of the parameters. This multimodal behavior for the CSTR example was studied by the procedure outlined in Egea et al (2009) and using MS with fmincon for 100 diverse initial guesses that are randomly generated in the large search space interval specified earlier. The MS runs with the stopping criteria δ ¼ 10 À 6 (Fig.…”

“…We now present the results of comparing the performance of EMSGA with an established and efficient parameter estimation solver that employs an enhanced scatter search (eSS) method for the dynamic optimization of nonlinear processes by a control vector parameter (CVP) optimization approach (Egea et al, 2009(Egea et al, , 2010 available in the Scatter Search Method based Global Optimization Toolbox (SSmGO). The eSS uses a solution combination method that does not restrict the search to the relative directions defined by solutions in the population, but allows movement to other areas to increase the search diversity (Egea et al, 2009).…”

“…The eSS uses a solution combination method that does not restrict the search to the relative directions defined by solutions in the population, but allows movement to other areas to increase the search diversity (Egea et al, 2009). Our choice of employing eSS for comparison with EMSGA is because of the following reasons.…”

“…In fact, we show how this approach makes EMSGA amenable to high accuracy and efficiency in parameter estimation by using the fundamental Taylor series formulation for derivative calculations and integration purposes. We also aim to assess the suitability of EMSGA by comparative studies with a well-known global-local strategy for optimization employing an enhanced scatter search method (Egea et al, 2009).…”