Surrogatebased modeling and optimization applications in. Surrogatebased particle swarm optimization for largescale. S, where s is the set of all feasible designs, is said to be nondominated with respect to a set a. It should be noted that this surrogatebased optimization algorithm does not give global optimality guarantees, same as ga. A systematic framework for applying the surrogate based optimization methods in transportation research is then developed. Technically, this might be best illustrated by the generic sequence of steps performed during surrogatebased optimization sbo. Surrogatebased optimization methods differ in two aspects. Surrogatebased particle swarm optimization for largescale wind farm layout design. To enhance the model robustness, proper sampling techniques are required to cover the entire domain of the. A surrogatebased optimization method with rbf neural network enhanced by linear interpolation and hybrid infill strategy.
The user can choose beween different options for the surrogate model the sampling strategy the initial experimental design. Most engineering design problems require experiments andor simulations to evaluate design objective and constraint functions as a function of design variables. To address this limitation, we propose a framework leading to substantially. Heuvelink, optimization of sample patterns for universal kriging of environmental variables, geoderma, 8 2007 8695. Accurate, highfidelity simulations are used not only for design verification but, even more importantly, to adjust parameters of the system to have it meet given performance requirements. This can be a blackbox model constructed from realworld data. The application of surrogate modelling is subject to the variation of the input parameters of the model. An introduction dimitri solomatine introduction this paper should be seen as an introduction and a brief tutorial in surrogate modelling. Surrogate models generated using this methodology can be used in an algebraic optimization framework with. Multiscale modeling, surrogate based analysis, and optimization of lithiumion batteries for vehicle applications by wenbo du a dissertation submitted in partial fulfillment of the requirements for the degree of doctor of philosophy aerospace engineering in the university of michigan 20 doctoral committee. Applications in engineering koziel, slawomir, leifsson, leifur on. It can be also used by students who would like to choose this area as a topic for their msc studies. A surrogate model is an engineering method used when an outcome of interest cannot be easily directly measured, disputed discuss so a model of the outcome is used instead.
Global optimization of general constrained greybox models. The term refers to models of a system that is fast and simple enough that you can tune their inputs to optimize the output. A generic surrogatebased optimization code for aerodynamic and multidisciplinary design central composite design ccd, boxbehnken and doptimal design dod, were developed for laboratory experiments. A generic surrogatebased optimization code for aerodynamic and multidisciplinary design central composite design ccd, boxbehnken and d. Bayesian optimization is a type of surrogatebased optimization, where the surrogate is a probabilistic model to. Henao, maravelias, 2010, henao, maravelias, 2011 introduce a surrogatebased superstructure framework based on the stnotoe approach. Mixing is considered as a critical process parameter cpp during process development due to its significant influence on reaction selectivity and process safety. Surrogatebased superstructure optimization framework. Energies free fulltext impact of sampling technique on.
In optimization context, the goal is to find optimal solution and not to predict responses away from optimality. A superstructure optimization approach based on mixedinteger nonlinear. For the purpose of economic design of the combined process of raw natural gas treating and co2 compression, a new superstructure combining co2 removing and compression processes is presented. Methodology for surrogate based superstructure optimization. Surrogatebased optimization, realworld applications of genetic algorithms, olympia roeva, intechopen, doi. Surrogate model optimization toolbox file exchange matlab. This means, in particular, that surrogates must be accurate only in promising regions of the design space. An overview of surrogatebased analysis and optimization was presented in this journal by queipo et al. Rigorous design optimisation for combined process of raw. Surrogatebased global optimization of composite material parts under dynamic loading. Surrogate based optimization has been explored in the flowsheet optimization caballero and grossmann, 2008.
Surrogatebasedanalysis and optimizationsbao has been shown to be an effective. Datadriven models are essential tools for the development of surrogate models that can be used for the design, operation, and optimization of industrial processes. Superstructure optimization of oleochemical processes with. Optimization ego algorithm 3, which is a surrogate based optimization sbo approach. Mathematical programming for piecewise linear regression. Section 3 describes the proposed surrogate based optimization methodology in detail, and presents a pseudocode for the optimization algorithm. Surrogate model toolbox for unconstrained continuous constrained integer constrained mixedinteger global optimization problems that are computationally expensive. Hannsjorg freund, andreas peschel, kai sundmacher, kai sundmacher. The optimization methodologies in the framework consist of simplex, subplex, genetic algorithm. Energies free fulltext impact of sampling technique.
Surrogatebased optimization has been explored in the flowsheet optimization caballero and grossmann, 2008. To accomplish a reliable optimum solution, the surrogate based optimization. In supero, piecewise linear representations of different process segments can be used as the building blocks of a superstructure optimization bertran et al. Oct 01, 2017 the optimal process structure and operating conditions are determined by minimizing the total annual cost tac. The general optimization problem we address is min f. Surrogatebased modeling and optimization applications. Methodology for surrogatebased superstructure optimization. The surrogatebased optimization sbo approach has been shown to be an e. We havedevelopedthe packagealamothatimplements the. Goel t, vaidyanathan r 2005 surrogatebased analysis and optimization. Grossmann, an algorithm for the use of surrogate models in modular flowsheet optimization, aiche journal, 54 2008 26332650. They covered some of the most popular methods in design space sampling, surrogate model construction, model selection and validation, sensitivity analysis, and surrogatebased optimization.
Optimization of reaction selectivity using cfdbased. Uncertaintyintegrated surrogate modeling for complex system. A systematic framework for applying the surrogatebased optimization methods in transportation research is then developed. This kind of optimization tries to minimize an objective.
Parego is a surrogatebased multiobjective optimization algorithm based on the krigingdace model, designed speci. Surrogate model optimization toolbox file exchange. Superstructure optimization based process synthesis is generally regarded as theoretically powerful. The manyobjective optimization performance of the krigingsurrogatebased evolutionary algorithm ea, which maximizes expected hypervolume improvement ehvi for updating the kriging model, is investigated and compared with those using expected improvement ei and estimation est updating criteria in this paper.
Help us write another book on this subject and reach those readers. Approximation models such as surrogate models provide a tractable substitute to expensive physical simulations and an effective solution to the potential lack of quantitative models of system behavior. The goals should be to find good or near optimal solutions when enough resources are provided. The co2 can be further compressed for storage or utilization to promote environmental protection. Variable density fluid reactor network synthesisconstruction of the attainable region through the ideas approach.
To accomplish a reliable optimum solution, the surrogatebased optimization sbo is performed by. This paper introduces a novel methodology for the global optimization of general constrained greybox problems. The performance of different forms of surrogate models is compared through a numerical example, and regressing kriging is identified as the best model in approximating the unknown response surface when no information. In the present surrogatebased framework, the optimization framework works as integrated module. A surrogate based optimization method with rbf neural network enhanced by linear interpolation and hybrid infill strategy. Surrogatebased analysis and optimization ntrs nasa. A framework for surrogate based aerodynamic optimization. The main challenges are how to overcome metamodeling and optimization difficulties in highdimensional design space. Boukouvala and ierapetritou, 20 and extended to the superstructure optimization. In contrast, the modern doe methods such as latin hypercube sampling lhs, orthogonal array. Multiscale modeling, surrogatebased analysis, and optimization of lithiumion batteries for vehicle applications by wenbo du a dissertation submitted in partial fulfillment of the requirements for the degree of doctor of philosophy aerospace engineering in the university of michigan 20 doctoral committee. Maravelias, a superstructurebased framework for bio separation. A surrogatebased optimization method with rbf neural. In the following sections i will briefly describe the first two steps and then will focus on the model choice and model fitting steps of the framework.
Performance of surrogate models in reliabilitybased. Nevertheless, mixing issues are difficult to identify and solve owing to their complexity and dependence on knowledge of kinetics and hydrodynamics. Henao, maravelias, 2010, henao, maravelias, 2011 introduce a surrogatebased superstructure framework based on. Uncertaintyintegrated surrogate modeling for complex. The steps of the methodology are presented as follows. Optimization strategies this framework is based on an optimization framework 23 programmed in python 24. Preferencebased surrogate modeling in engineering design. Framework of a typical sbotype optimization surroopt is a stateoftheart, surrogatebased optimization code, which can be used to efficiently solve. These capabilities not only enable the efficient design of complex systems, but is also essential for the effective analysis of physical phenomenacharacteristics in the different domains of.
To obtain a good tradeoff between calculation accuracy and efficiency, a surrogate based optimization framework is presented to address the nlp problem. This section offers some guidance on choosing from among the available surrogate model types. To solve the above difficulties, a developed surrogatebased optimization framework combining highdimensional model representation hdmr. The design optimization of laminated composite structures is of relevance in automobile, naval, aerospace, construction and energy industry. Surrogatebased particle swarm optimization for large. Surrogatebased optimization of simulated energy systems. In the present surrogate based framework, the optimization framework works as integrated module. Learning surrogate models for simulationbased optimization. Second, ga is replaced with surrogatebased optimization based on rbf, which has been proven to be more efficient than ga. In section 4 we describe the mixedinteger surrogate optimization miso framework and. There are several options for each of these steps, as well as several advantages and disadvantages of each option.
In the following sections i will briefly describe the first two steps and then will focus on the model choice and model fitting steps of. Henao, maravelias, 2010, henao, maravelias, 2011 introduce a surrogate based superstructure framework based on the stnotoe approach. Superstructure approaches are the solution to the difficult problem which involves the. On the other hand, in superstructure optimizationbased methods, a process. Numerical experiments are conducted in 3 to 15objective dtlz17 problems. Miller1 1national energy technology laboratory, pittsburgh, pa 15236 2department of chemical engineering, carnegie mellon university, pittsburgh, pa 152 october 5, 20 abstract we address a central problem in modeling, namely that of learning.
Superstructure optimizationbased process synthesis is generally regarded as theoretically powerful. Superstructure optimizationbased process synthesis is generally regarded as. Surrogate based optimization using kriging based approximation. A surrogatebased optimization method with rbf neural network. A study on manyobjective optimization using the kriging. A greybox problem may contain a combination of blackbox constraints and constraints. Learning surrogate models for simulationbased optimization alison cozad1,2, nikolaos v. Process superstructure optimization using surrogate models. Pdf surrogatebased optimization approach to membrane. It begins by presenting the basic concepts and formulations of the surrogatebased modeling and optimization paradigm and then discusses relevant modeling techniques, optimization algorithms and design procedures, as well as stateoftheart developments.
In order to assess the algorithm we couple weap21 water resources management model 4 with matlab and thus extend its capabilities by using it within pso framework named weap21pso. Data driven surrogatebased optimization in the problem. Optimization of chemical processes using surrogate models based. Concurrent surrogate model selection cosmos framework is applied to identify the best surrogate model to represent the wind farm energy production as a function of the reduced variable vector. The traditional surrogatebased optimization techniques are facing severe challenges for highdimensional engineering optimization problems. In this paper, we proposed an optimization methodology using computational fluid.
88 336 1200 170 414 983 529 756 935 1075 538 1662 994 1265 1333 1521 1255 767 164 525 344 830 833 382 704 1586 1607 732 658 420 636 11 138 12 519 686 270 1400 589