• Journal of the Korea Institute of Military Science and Technology
• /
• v.25 no.4
• /
• pp.412-424
• /
• 2022

Since traditional derivative-based optimization for noisy simulation shows bad performance, evolutionary algorithms are considered as substitutes. Especially in case when outputs are binary, more simulation trials are needed to get near-optimal solution since the outputs are discrete and have high and heterogeneous variance. In this paper, we propose a genetic algorithm called SARAGA which adopts dynamic resampling and fitness approximation using surrogate. SARAGA reduces unnecessary numbers of expensive simulations to estimate success probabilities estimated from binary simulation outputs. SARAGA allocates number of samples to each solution dynamically and sometimes approximates the fitness without additional expensive experiments. Experimental results show that this novel approach is effective and proper hyper parameter choice of surrogate and resampling can improve the performance of algorithm.

• Smart Structures and Systems
• /
• v.32 no.1
• /
• pp.61-81
• /
• 2023

This study presents an ensemble learning based Bayesian model updating approach for structural damage diagnosis. In the developed framework, the structure is initially decomposed into a set of substructures. The autoregressive moving average (ARMAX) model is established first for structural damage localization based structural motion equation. The wavelet packet decomposition is utilized to extract the damage-sensitive node energy in different frequency bands for constructing structural surrogate models. Four methods, including Kriging predictor (KRG), radial basis function neural network (RBFNN), support vector regression (SVR), and multivariate adaptive regression splines (MARS), are selected as candidate structural surrogate models. These models are then resampled by bootstrapping and combined to obtain an ensemble model by probabilistic ensemble. Meanwhile, the maximum entropy principal is adopted to search for new design points for sample space updating, yielding a more robust ensemble model. Through the iterations, a framework of surrogate ensemble learning based model updating with high model construction efficiency and accuracy is proposed. The specificities of the method are discussed and investigated in a case study.

• Computers and Concrete
• /
• v.32 no.4
• /
• pp.373-381
• /
• 2023

Fly ash, granulated blast furnace slag, marble waste powder, etc. are just some of the by-products of other sectors that the construction industry is looking to include into the many types of concrete they produce. This research seeks to use surrogate machine learning methods to forecast the compressive strength of self-compacting concrete. The surrogate models were developed using Gradient Boosting Machine (GBM), Support Vector Machine (SVM), Random Forest (RF), and Gaussian Process Regression (GPR) techniques. Compressive strength is used as the output variable, with nano silica content, cement content, coarse aggregate content, fine aggregate content, superplasticizer, curing duration, and water-binder ratio as input variables. Of the four models, GBM had the highest accuracy in determining the compressive strength of SCC. The concrete's compressive strength is worst predicted by GPR. Compressive strength of SCC with nano silica is found to be most affected by curing time and least by fine aggregate.

• Steel and Composite Structures
• /
• v.24 no.2
• /
• pp.227-237
• /
• 2017

This paper presents optimization of a long-span portal steel frame under dynamic wind loads using a surrogate-assisted evolutionary algorithm. Long-span portal steel frames are often used in low-rise industrial and commercial buildings. The structure needs be able to resist the wind loads, and at the same time it should be as light as possible in order to be cost-effective. In this work, numerical model of a portal steel frame is constructed using structural analysis program (SAP2000), with the web-heights at five locations of I-sections of the columns and rafters as the decision variables. In order to evaluate the performance of a given design under dynamic wind loading, the equivalent static wind load (ESWL) is obtained from a database of wind pressures measured in wind tunnel tests. A modified formulation of the problem compared to the one available in the literature is also presented, considering additional design constraints for practicality. Evolutionary algorithms (EA) are often used to solve such non-linear, black-box problems, but when each design evaluation is computationally expensive (e.g., in this case a SAP2000 simulation), the time taken for optimization using EAs becomes untenable. To overcome this challenge, we employ a surrogate-assisted evolutionary algorithm (SAEA) to expedite the convergence towards the optimum design. The presented SAEA uses multiple spatially distributed surrogate models to approximate the simulations more accurately in lieu of commonly used single global surrogate models. Through rigorous numerical experiments, improvements in results and time savings obtained using SAEA over EA are demonstrated.

• Smart Structures and Systems
• /
• v.22 no.5
• /
• pp.561-574
• /
• 2018

In this paper, for efficiently reducing the computational cost of the model updating during the optimization process of damage detection, the structural response is evaluated using properly trained surrogate model. Furthermore, in practice uncertainties in the FE model parameters and modelling errors are inevitable. Hence, an efficient approach based on Monte Carlo simulation is proposed to take into account the effect of uncertainties in developing a surrogate model. The probability of damage existence (PDE) is calculated based on the probability density function of the existence of undamaged and damaged states. The current work builds a framework for Probability Based Damage Detection (PBDD) of structures based on the best combination of metaheuristic optimization algorithm and surrogate models. To reach this goal, three popular metamodeling techniques including Cascade Feed Forward Neural Network (CFNN), Least Square Support Vector Machines (LS-SVMs) and Kriging are constructed, trained and tested in order to inspect features and faults of each algorithm. Furthermore, three wellknown optimization algorithms including Ideal Gas Molecular Movement (IGMM), Particle Swarm Optimization (PSO) and Bat Algorithm (BA) are utilized and the comparative results are presented accordingly. Furthermore, efficient schemes are implemented on these algorithms to improve their performance in handling problems with a large number of variables. By considering various indices for measuring the accuracy and computational time of PBDD process, the results indicate that combination of LS-SVM surrogate model by IGMM optimization algorithm have better performance in predicting the of damage compared with other methods.

• Structural Engineering and Mechanics
• /
• v.65 no.2
• /
• pp.173-181
• /
• 2018

In this paper, finite element (FE) model updating based on multi-objective optimization with the surrogate model for a steel plate girder bridge is investigated. Conventionally, FE model updating for bridge structures uses single-objective optimization with finite element analysis (FEA). In the case of the conventional method, computational burden occurs considerably because a lot of iteration are performed during the updating process. This issue can be addressed by replacing FEA with the surrogate model. The other problem is that the updating result from single-objective optimization depends on the condition of the weighting factors. Previous studies have used the trial-and-error strategy, genetic algorithm, or user's preference to obtain the most preferred model; but it needs considerable computation cost. In this study, the FE model updating method consisting of the surrogate model and multi-objective optimization, which can construct the Pareto-optimal front through a single run without considering the weighting factors, is proposed to overcome the limitations of the single-objective optimization. To verify the proposed method, the results of the proposed method are compared with those of the single-objective optimization. The comparison shows that the updated model from the multi-objective optimization is superior to the result of single-objective optimization in calculation time as well as the relative errors between the updated model and measurement.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.6B
• /
• pp.559-567
• /
• 2009

Distributed Content Delivering Network (DCDN) will make use of the existing resources of the common Internet users in terms of storage space, bandwidth and Internet connectivity to create it. However DCDN has some limitations that are inefficient using of storage space, reliability and having special load balancing (LB) algorithm. So, this paper proposes Re-distribution based CDN (RDCDN) that overcomes the limitations of DCDN. RDCDN has the content re-distribution algorithm and separates surrogates to main surrogate and sub surrogates. Main surrogate can help service reliability be improved by storing all contents as back-up system. And content re-distribution algorithm also can help storage space be saved because all contents are not stored in every surrogate. Especially, when RDCwDN uses content re-distribution algorithm, it can work active load balancing function without extra LB algorithm like as DCDN. Results of simulation show that the proposed architecture can improve reliability and efficiency of storage space, and it also can offer the same performance as that of commercial CDN and DCDN.

• Journal of Convergence for Information Technology
• /
• v.10 no.7
• /
• pp.138-146
• /
• 2020

The paper deals with comparative study of characteristics of surrogate models and sensitivity evaluation using design of experiments in order to enhance and analysis the structure design of an automatic salt collector under various design load conditions. Orthogonal array design based on numerical analysis was used for the design of experiments. The thickness sizing variables of main structure member were considered the design factors, and the output responses were selected from the strength performances as well as the weight. The quantitative effects on responses for each design factor were evaluated from the orthogonal array experiment. Optimum design case was also identified to improve the strength performances with weight minimization. Using the orthogonal array experiment. various surrogate models such as response surface model, Kriging model, and Chebyshev orthogonal polynomial were generated. The orthogonal array experiment results were validated by the surrogate modeling results. The most suitable surrogate model was the response surface model for the exploration of design space of the automatic salt collector.

• Steel and Composite Structures
• /
• v.27 no.4
• /
• pp.427-438
• /
• 2018

In this paper, the single and multi-objective optimization of thin-walled conical tubes with different types of indentations under axial impact has been investigated using surrogate models called metamodels. The geometry of tapered thin-walled tubes has been studied in order to achieve maximum specific energy absorption (SEA) and minimum peak crushing force (PCF). The height, radius, thickness, tapered angle of the tube, and the radius of indentation have been considered as design variables. Based on the design of experiments (DOE) method, the generated sample points are computed using the explicit finite element code. Different surrogate models including Kriging, Feed Forward Neural Network (FNN), Radial Basis Neural Network (RNN), and Response Surface Modelling (RSM) comprised to evaluate the appropriation of such models. The comparison study between surrogate models and the exploration of indentation shapes have been provided. The obtained results show that the RNN method has the minimum mean squared error (MSE) in training points compared to the other methods. Meanwhile, optimization based on surrogate models with lower values of MSE does not provide optimum results. The RNN method demonstrates a lower crashworthiness performance (with a lower value of 125.7% for SEA and a higher value of 56.8% for PCF) in comparison to RSM with an error order of $10^{-3}$. The SEA values can be increased by 17.6% and PCF values can be decreased by 24.63% by different types of indentation. In a specific geometry, higher SEA and lower PCF require triangular and circular shapes of indentation, respectively.

• YAKHAK HOEJI
• /
• v.54 no.2
• /
• pp.75-90
• /
• 2010

Biomarkers are likely to be important in the study of various pulmonary diseases for many reasons. Research efforts in developing biomarkers and surrogate endpoints of lung diseases have resulted in the identification of new risk factors and novel drug targets, as well as the establishment of treatment guidelines. Government agencies, academic research institutions, diagnostic industries, and pharmaceutical companies all recognize the importance of biomarkers in new drug development and advancing therapies to improve public health. In drug development, biomarkers are used to evaluate early signals of efficacy and safety, to select dose, and to identify the target population. Identification of suitable end points not only would help investigators design appropriate clinical trials but would assist clinicians in caring for this patient population. Though the area of pulmonology has received much attention in the past decades, it still lags behind with regard to the development of biomarkers, particularly those of health effects and susceptibility. This review critically summarized several biomarker researches such as Evaluation of COPD Longitudinally to Identify Predictive Surrogate End-points (ECLIPSE) study with objectives of identifying the parameters that predict disease progression of COPD, as well as biomarkers that may serve as surrogate end-points.