Ruben E. Perez

Bio: Ruben E. Perez is an academic researcher from Royal Military College of Canada. The author has contributed to research in topics: Leading edge & Conceptual design. The author has an hindex of 13, co-authored 67 publications receiving 1823 citations. Previous affiliations of Ruben E. Perez include University of Toronto.

pyOpt: a Python-based object-oriented framework for nonlinear constrained optimization

Abstract: We present pyOpt, an object-oriented framework for formulating and solving nonlinear constrained optimization problems in an efficient, reusable and portable manner. The framework uses object-oriented concepts, such as class inheritance and operator overloading, to maintain a distinct separation between the problem formulation and the optimization approach used to solve the problem. This creates a common interface in a flexible environment where both practitioners and developers alike can solve their optimization problems or develop and benchmark their own optimization algorithms. The framework is developed in the Python programming language, which allows for easy integration of optimization software programmed in Fortran, C, C+?+, and other languages. A variety of optimization algorithms are integrated in pyOpt and are accessible through the common interface. We solve a number of problems of increasing complexity to demonstrate how a given problem is formulated using this framework, and how the framework can be used to benchmark the various optimization algorithms.

Evaluation of Multidisciplinary Optimization Approaches for Aircraft Conceptual D esign

Abstract: This paper presents the evaluation of different MDO architectures using an extended set of metrics, which take into consideration optimization and formulation structure characteristics. Demonstrative comparisons are made for analytic and supersonic business jet conceptual design examples. Results show the promising features of the proposed evaluation metrics to define a standardized guideline when dealing with multidisciplinary optimization formulations which can be applied to aircraft conceptual design problems.

Aerostructural Optimization of Nonplanar Lifting Surfaces

Abstract: to findoptimalnonplanarliftingsurfacesandtoexplainthevariousfactorsandtradeoffsatplayApanelmethodand anequivalentbeam finite-elementmodelareusedtoexplorenonplanarliftingsurfaces,whiletakingintoaccountthe coupling and design tradeoffs between aerodynamics and structures Both single-discipline aerodynamic optimization and multidisciplinary aerostructural optimization problems are investigated The design variables are chosen to give the lifting-surface arrangement as much freedom as possible This is accomplished by allowing a number of wing segments to vary their area, taper, twist, sweep, span, and dihedral, with the constraint that they must not intersect each other Because of the complexity of the resulting design space and the presence of multiple localminima, anaugmentedLagrangianparticle swarmoptimizer isusedto solvethe optimizationproblemsWhen only aerodynamics are considered, closed lifting-surface configurations, such as the box wing and joined wing, are found to be optimal When aerostructural optimization is performed, a winglet configuration is found to be optimal when the overall span is constrained, and a wing with a raked wingtip is optimal when there is no such constraint

Multidisciplinary design optimization: A survey of architectures

Abstract: Multidisciplinary design optimization is a field of research that studies the application of numerical optimization techniques to the design of engineering systems involving multiple disciplines or components. Since the inception of multidisciplinary design optimization, various methods (architectures) have been developed and applied to solve multidisciplinary design-optimization problems. This paper provides a survey of all the architectures that have been presented in the literature so far. All architectures are explained in detail using a unified description that includes optimization problem statements, diagrams, and detailed algorithms. The diagrams show both data and process flow through the multidisciplinary system and computational elements, which facilitate the understanding of the various architectures, and how they relate to each other. A classification of the multidisciplinary design-optimization architectures based on their problem formulations and decomposition strategies is also provided, a...

pyOpt: a Python-based object-oriented framework for nonlinear constrained optimization

Abstract: We present pyOpt, an object-oriented framework for formulating and solving nonlinear constrained optimization problems in an efficient, reusable and portable manner. The framework uses object-oriented concepts, such as class inheritance and operator overloading, to maintain a distinct separation between the problem formulation and the optimization approach used to solve the problem. This creates a common interface in a flexible environment where both practitioners and developers alike can solve their optimization problems or develop and benchmark their own optimization algorithms. The framework is developed in the Python programming language, which allows for easy integration of optimization software programmed in Fortran, C, C+?+, and other languages. A variety of optimization algorithms are integrated in pyOpt and are accessible through the common interface. We solve a number of problems of increasing complexity to demonstrate how a given problem is formulated using this framework, and how the framework can be used to benchmark the various optimization algorithms.

The best-so-far selection in Artificial Bee Colony algorithm

Abstract: The Artificial Bee Colony (ABC) algorithm is inspired by the behavior of honey bees. The algorithm is one of the Swarm Intelligence algorithms explored in recent literature. ABC is an optimization technique, which is used in finding the best solution from all feasible solutions. However, ABC can sometimes be slow to converge. In order to improve the algorithm performance, we present a modified method for solution update of the onlooker bees in this paper. In our method, the best feasible solutions found so far are shared globally among the entire population. Thus, the new candidate solutions are more likely to be close to the current best solution. In other words, we bias the solution direction toward the best-so-far position. Moreover, in each iteration, we adjust the radius of the search for new candidates using a larger radius earlier in the search process and then reduce the radius as the process comes closer to converging. Finally, we use a more robust calculation to determine and compare the quality of alternative solutions. We empirically assess the performance of our proposed method on two sets of problems: numerical benchmark functions and image registration applications. The results demonstrate that the proposed method is able to produce higher quality solutions with faster convergence than either the original ABC or the current state-of-the-art ABC-based algorithm.

Extensions to the design structure matrix for the description of multidisciplinary design, analysis, and optimization processes

Abstract: While numerous architectures exist for solving multidisciplinary design optimization (MDO) problems, there is currently no standard way of describing these architectures. In particular, a standard visual representation of the solution process would be particularly useful as a communication medium among practitioners and those new to the field. This paper presents the extended design structure matrix (XDSM), a new diagram for visualizing MDO processes. The diagram is based on extending the standard design structure matrix (DSM) to simultaneously show data dependency and process flow on a single diagram. Modifications include adding special components to define iterative processes, defining different line styles to show data and process connections independently, and adding a numbering scheme to define the order in which the components are executed. This paper describes the rules for constructing XDSMs along with many examples, including diagrams of several MDO architectures. Finally, this paper discusses potential applications of the XDSM in other areas of MDO and the future development of the diagrams.