• Structural Engineering and Mechanics
• /
• v.84 no.3
• /
• pp.323-335
• /
• 2022

The main idea of the framework is to seamlessly combine a reasonably accurate and fast surrogate model with the importance sampling strategy. Developing a surrogate model for predicting structures' dynamic responses is challenging because it involves high-dimensional inputs and outputs. For this purpose, a novel surrogate model based on cutting-edge deep learning architectures specialized for capturing temporal relationships within time-series data, namely Long-Short term memory layer and Transformer layer, is designed. After being properly trained, the surrogate model could be utilized in place of the finite element method to evaluate structures' responses without requiring any specialized software. On the other hand, the importance sampling is adopted to reduce the number of calculations required when computing the failure probability by drawing more relevant samples near critical areas. Thanks to the portability of the trained surrogate model, one can integrate the latter with the Importance sampling in a straightforward fashion, forming an efficient framework called TTIS, which represents double advantages: less number of calculations is needed, and the computational time of each calculation is significantly reduced. The proposed approach's applicability and efficiency are demonstrated through three examples with increasing complexity, involving a 1D beam, a 2D frame, and a 3D building structure. The results show that compared to the conventional Monte Carlo simulation, the proposed method can provide highly similar reliability results with a reduction of up to four orders of magnitudes in time complexity.

• International Journal of Fluid Machinery and Systems
• /
• v.10 no.3
• /
• pp.240-253
• /
• 2017

Latin hypercube sampling is widely used design-of-experiment technique to select design points for simulation which are then used to construct a surrogate model. The exploration/exploitation properties of surrogate models depend on the size and distribution of design points in the chosen design space. The present study aimed at evaluating the performance characteristics of various surrogate models depending on the Latin hypercube sampling (LHS) procedure (sample size and spatial distribution) for a diverse set of optimization problems. The analysis was carried out for two types of problems: (1) thermal-fluid design problems (optimizations of convergent-divergent micromixer coupled with pulsatile flow and boot-shaped ribs), and (2) analytical test functions (six-hump camel back, Branin-Hoo, Hartman 3, and Hartman 6 functions). The three surrogate models, namely, response surface approximation, Kriging, and radial basis neural networks were tested. The important findings are illustrated using Box-plots. The surrogate models were analyzed in terms of global exploration (accuracy over the domain space) and local exploitation (ease of finding the global optimum point). Radial basis neural networks showed the best overall performance in global exploration characteristics as well as tendency to find the approximate optimal solution for the majority of tested problems. To build a surrogate model, it is recommended to use an initial sample size equal to 15 times the number of design variables. The study will provide useful guidelines on the effect of initial sample size and distribution on surrogate construction and subsequent optimization using LHS sampling plan.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2018.05a
• /
• pp.41-42
• /
• 2018

This study is about the prediction of ship's roll motion characteristic which has been used for evaluating ship's seakeeping performance. In order to obtain the ship's roll RAO during voyage, this paper utilized machine learning-based surrogate model. By comparing the prediction result data of surrogate model with test data, we suggest the best approximation technique and data sampling interval of the surrogate model appropriate for predicting the ships' roll motion characteristic.

• Nuclear Engineering and Technology
• /
• v.55 no.8
• /
• pp.3071-3079
• /
• 2023

Helium production in the nuclear fuel matrix during irradiation plays a critical role in the design and performance of Gen-IV reactor fuel, as it represents a life-limiting factor for the operation of fuel pins. In this work, a surrogate model for the helium production rate in fast reactor MOX fuels is developed, targeting its inclusion in engineering tools such as fuel performance codes. This surrogate model is based on synthetic datasets obtained via the SCIANTIX burnup module. Such datasets are generated using Latin hypercube sampling to cover the range of input parameters (e.g., fuel initial composition, fission rate density, and irradiation time) and exploiting the low computation requirement of the burnup module itself. The surrogate model is verified against the SCIANTIX burnup module results for helium production with satisfactory performance.

• Journal of Korean Institute of Industrial Engineers
• /
• v.39 no.5
• /
• pp.422-428
• /
• 2013

In many cases, an $\bar{X}$ control chart which is based on the performance variable is used in industrial fields. However, if the performance variable is too costly or impossible to measure and a less expensive surrogate variable is available, the process may be more efficiently controlled using surrogate variables. In this paper, we propose a model for the economic design of a VSI (Variable Sampling Interval) $\bar{X}$ control chart using a surrogate variable that is linearly correlated with the performance variable. The total average profit model is constructed, which involves the profit per cycle time, the cost of sampling and testing, the cost of detecting and eliminating an assignable cause, and the cost associated with production during out-of-control state. The VSI $\bar{X}$ control charts using surrogate variables are expected to be superior to the Shewhart FSI (Fixed Sampling Interval) $\bar{X}$ control charts using surrogate variables with respect to the expected profit per unit cycle time from economic viewpoint.

• Korean Journal of Computational Design and Engineering
• /
• v.17 no.5
• /
• pp.375-386
• /
• 2012

Most engineering design problems require analyses or simulations to evaluate objective functions. However, a single simulation can take many hours or even days to finish for many real world problems. As a result, design optimization becomes impossible since they require hundreds or thousands of simulation evaluations. The surrogate-based optimization (SBO) strategy became a remedy for such computationally expensive analyses and simulations. A surrogate-based optimization strategy has been developed in this study in order to improve global optimization performance. The strategy is a heuristic algorithm and it exploits not only multiple surrogates, but also multiple optimizers. Multiple optimizations of multiple surrogate models yield multiple candidate design points of optima. During the sequential sampling process, the algorithm ranks candidate design points, selects the points as many as specified, and builds the improved surrogate model. Various mathematical functions with different numbers of design variables are chosen to compare the proposed method with the other most recent algorithm, MSEGO. The proposed method shows superior performance to the other method.

• 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.

• 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.

• YAKHAK HOEJI
• /
• v.52 no.5
• /
• pp.331-344
• /
• 2008

Recently, the FDA (Food and Drug Administration) of the United States and many advanced countries remark biomarkers and surrogate endpoints as a critical path tool on model based drug development. Economic, technical and social profit on model based drug development like a reduction of the length of research and development have been achieved. Therefore we summarize previous studies about biomarkers and surrogate endpoints and suggest a development direction of therapeutic agents. In diabetes mellitus (DM) and osteoporosis, there are remarkable increases in number of patients and most of patients take medicine during their whole lifetime. For this reason, many patients with DM and osteoporosis have a tolerance on their medicine. We expect that research and development on biomarkers and surrogate endpoints will contribute to new drug development on DM and osteoporosis. Biomarkers for DM are blood levels of glucose, insulin, ${HbA}_{1c}$, CRP, alpha-glucosidase, adiponectin and DPP-4. Among these, validated surrogate endpoints for DM are blood levels of glucose, insulin and ${HbA}_{1c}$ Biomarkers for osteoporosis are BMD, BMC, trabecular volume, ICTP, DPD, osteocalcin, the activity of osteoclast and production of osteoblast. The validated surrogate endpoints for osteoporosis are BMD only. This review summarizes all suggested biomarkers and surrogate endpoints in DM and osteoporosis. The biomarkers are classified by drugs, and the method of validation for surrogate endpoints is suggested. This information would contribute to suggest a direction of DM and osteoporosis therapeutic agent development.

• International Journal of Aeronautical and Space Sciences
• /
• v.17 no.4
• /
• pp.491-500
• /
• 2016

This paper presents a highly efficient aeroelastic optimization method based on a surrogate model; the model is verified by considering the case of a high-aspect-ratio composite wing. Optimization frameworks using the Kriging model and genetic algorithm (GA), the Kriging model and improved particle swarm optimization (IPSO), and the back propagation neural network model (BP) and IPSO are presented. The feasibility of the method is verified, as the model can improve the optimization efficiency while also satisfying the engineering requirements. Moreover, the effects of the number of design variables and number of constraints on the optimization efficiency and objective function are analysed in detail. The accuracy of two surrogate models in aeroelastic optimization is also compared. The Kriging model is constructed more conveniently, and its predictive accuracy of the aeroelastic responses also satisfies the engineering requirements. According to the case of a high-aspect-ratio composite wing, the GA is better at global optimization.