Artificial neural network-based sequential approximate optimization of metal sheet architecture and forming process

Han Seong-Sik,김흥규 한국CDE학회 2024 Journal of computational design and engineering Vol.11 No.3

This paper introduces a sequential approximate optimization method that combines the finite element method (FEM), dynamic differential evolution (DDE), and artificial neural network (ANN) surrogate models. The developed method is applied to address two optimization problems. The first involves metamaterial design optimization for metal sheet architecture with binary design variables. The second pertains to optimizing process parameters in multi-stage metal forming, where the discrete nature arises owing to changing tool geometries across stages. This process is highly non-linear, accumulating contact, geometric, and material non-linear effects discretely through forming stages. The efficacy of the proposed optimization method, utilizing ANN surrogate models, is compared with traditionally used polynomial response surface (PRS) surrogate models, primarily based on low-order polynomials. Efficient learning of ANN surrogate models is facilitated through the FEM and Python integration framework. Initial data for surrogate model training are collected via Latin hypercube sampling and FEM simulations. DDE is employed for sequential approximate optimization, optimizing ANN or PRS surrogate models to determine optimal design variables. PRS surrogate models encounter challenges in dealing with non-linear changes in sequential approximate optimization concerning discrete characteristics such as binary design variables and discrete non-linear behavior found in multi-stage metal forming processes. Owing to the discrete nature, PRS surrogate models require more data and iterations for optimal design variables. In contrast, ANN surrogate models adeptly predict non-linear behavior through the activation function’s characteristics. In the optimization problem of metal sheet architecture for design Target C, the ANN surrogate model required an average of 4.6 times fewer iterations to satisfy stopping criteria compared with the PRS surrogate model. Furthermore, in the optimization of multi-stage deep drawing processes, the ANN surrogate model required an average of 6.1 times fewer iterations to satisfy stopping criteria compared with the PRS surrogate model. As a result, the sequential global optimization method utilizing ANN surrogate models achieves optimal design variables with fewer iterations than PRS surrogate models. Further confirmation of the method’s efficiency is provided by comparing Pearson correlation coefficients and locus plots.

Experimental validation of FE model updating based on multi-objective optimization using the surrogate model

황용문,진승섭,정호연,김세훈,이종재,정형조 국제구조공학회 2018 Structural Engineering and Mechanics, An Int'l Jou Vol.65 No.2

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.

Approximation Models for Multi-Objective Optimization

Yeun, Y.S.,Yang, Y.S.,Jang, B.S.,Ruy, W.S. 대진대학교 생산기술연구소 2000 생산기술연구소 논문집 Vol.3 No.-

In engineering problems, computationally intensive high-fidelity models or expensive computer simulations hinder the use of standard optimization techniques because they should be invoked repeatedly during optimization, despite the alarming growth of computer capability. Therefore, these expensive analyses are often replaced with approximation models that can be evaluated nearly free. However, due to their limited accuracy, it is practically impossible to exactly find an actual optimum(or a set of actual noninferior solutions) of the original single(or multi-objective) optimization problem. Significant efforts have been made to overcome this problem, The model management framework is one of such endeavours. The approximation models are sequentially updated during the iterative optimization process in such a way that their capability to accurately model original functions especially in the region of our interests can be improved. The models are modified using one or several sample points generated by making good use of the predictive ability of the approximation models. However, theses approaches have been restricted to a single objective optimization problem. It seems that there is no reported management framework that can handle a multi-objective optimization problem. This paper will suggest strategies that can successfully treat not only a single objective but also multiple objectives by extending the concept of sequentially managing approximation models and combining this extended concept with the Genetic Algorithm which can treat multiple objective s(MOGA). Consequently, the number of exact analysis required to converge an actual optimum or to generate a sufficiently accurate Pareto set can be reduced considerably. Especially, the approach for multiple objectives will lead to the surprising reduction in the number. We will confirm these effects through several illustrating examples. Key words : optimization, approximation model, model management framework, multi-objective

Experimental validation of FE model updating based on multi-objective optimization using the surrogate model

Hwang, Yongmoon,Jin, Seung-seop,Jung, Ho-Yeon,Kim, Sehoon,Lee, Jong-Jae,Jung, Hyung-Jo Techno-Press 2018 Structural Engineering and Mechanics, An Int'l Jou Vol.65 No.2

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.

BIM Mesh Optimization Algorithm Using K-Nearest Neighbors for Augmented Reality Visualization

Pa Pa Win Aung(빠 빠 윈 아웅),Lee, Donghwan(이동환),Park, Jooyoung(박주영),Cho, Mingeon(조민건),Park, Seunghee(박승희) 대한토목학회 2022 대한토목학회논문집 Vol.42 No.2

최근 BIM (Building Information Modeling)과 AR (Augmented Reality)을 결합한 실시간 시각화 기술이 건설관리 의사 결정 및 처리 효율성을 높이는 데 도움이 된다는 것을 보여주기 위한 다양한 연구가 활발히 진행되고 있다. 그러나, 대용량 BIM 데이터는 AR에 적용할 경우 데이터 전송 문제, 이미지 단절, 영상 끊김 등과 같은 다양한 문제가 발생함으로 3차원(3D) 모델의 메쉬 최적화를 통해 시각화의 효율성을 향상시켜야 한다. 대부분의 기존 메쉬 경량화 방법은 복잡하고 경계가 많은 3D 모델의 메쉬를 적절하게 처리할 수 없다. 이에 본 연구에서는 고성능 AR시각화를 위해 BIM 데이터를 재구성하기 위한 k-최근접이웃(KNN) 분류 프레임워크 기반 메쉬 경량화 알고리즘을 제안하였다. 제안 알고리즘은 선정된 BIM 모델을 삼각형 중심 개념 기반의 Unity C# 코드로 경량화하였고 모델의 데이터 세트를 활용하여 정점 사이의 거리를 정의할 수 있는 KNN로 분류되었다. 그 결과 전체 모델과 각 구조의 경량화 메쉬 점 및 삼각형 개수가 각각 약 56 % 및 약 42 % 감소됨을 확인할 수 있었다. 결과적으로, 원본 모델과 비교했을 때 경량화한 모델은 시각적인 요소 및 정보 손실이 없었고, 따라서, AR 기기 활용 시 고성능 시각화를 향상시킬 수 있을 것으로 기대된다. Various studies are being actively conducted to show that the real-time visualization technology that combines BIM (Building Information Modeling) and AR (Augmented Reality) helps to increase construction management decision-making and processing efficiency. However, when large-capacity BIM data is projected into AR, there are various limitations such as data transmission and connection problems and the image cut-off issue. To improve the high efficiency of visualizing, a mesh optimization algorithm based on the k-nearest neighbors (KNN) classification framework to reconstruct BIM data is proposed in place of existing mesh optimization methods that are complicated and cannot adequately handle meshes with numerous boundaries of the 3D models. In the proposed algorithm, our target BIM model is optimized with the Unity C# code based on triangle centroid concepts and classified using the KNN. As a result, the algorithm can check the number of mesh vertices and triangles before and after optimization of the entire model and each structure. In addition, it is able to optimize the mesh vertices of the original model by approximately 56 % and the triangles by about 42 %. Moreover, compared to the original model, the optimized model shows no visual differences in the model elements and information, meaning that high-performance visualization can be expected when using AR devices.

All-atom chain-building by optimizing MODELLER energy function using conformational space annealing

Joo, Keehyoung,Lee, Jinwoo,Seo, Joo-Hyun,Lee, Kyoungrim,Kim, Byung-Gee,Lee, Jooyoung Wiley Subscription Services, Inc., A Wiley Company 2009 Proteins Vol.75 No.4

<P>We have investigated the effect of rigorous optimization of the MODELLER energy function for possible improvement in protein all-atom chain-building. For this we applied the global optimization method called conformational space annealing (CSA) to the standard MODELLER procedure to achieve better energy optimization than what MODELLER provides. The method, which we call MODELLERCSA, is tested on two benchmark sets. The first is the 298 proteins taken from the HOMSTRAD multiple alignment set. By simply optimizing the MODELLER energy function, we observe significant improvement in side-chain modeling, where MODELLERCSA provides about 10.7% (14.5%) improvement for χ<SUB>1</SUB> (χ<SUB>1</SUB> + χ<SUB>2</SUB>) accuracy compared to the standard MODELLER modeling. The improvement of backbone accuracy by MODELLERCSA is shown to be less prominent, and a similar improvement can be achieved by simply generating many standard MODELLER models and selecting lowest energy models. However, the level of side-chain modeling accuracy by MODELLERCSA could not be matched either by extensive MODELLER strategies, side-chain remodeling by SCWRL3, or copying unmutated rotamers. The identical procedure was successfully applied to 100 CASP7 template base modeling domains during the prediction season in a blind fashion, and the results are included here for comparison. From this study, we observe a good correlation between the MODELLER energy and the side-chain accuracy. Our findings indicate that, when a good alignment between a target protein and its templates is provided, thorough optimization of the MODELLER energy function leads to accurate all-atom models. Proteins 2009. © 2008 Wiley-Liss, Inc.</P>

정적 부하 작업에서 EMG 모델과 세가지 최적화 모델을 이용한 척추 부하 평가

송영웅 ( Young Woong Song ),정민근 ( Min Keun Chung ) 한국산업위생학회 2005 한국산업보건학회지 Vol.15 No.1

This study investigated the spinal loads(L5/S1 disc compression and shear forces) predicted from four biomechanical models: one EMG model and three optimization models. Three objective functions used in the optimization models were to miminize 1) the cubed muscle forces: MF3, 2) the cubed muscle stress: MS3, 3) maximum muscle intensity: MI. Twelve healthy male subjects participated in the isometric voluntary exertion tests to six directions: flexion/extension, left/right lateral bending, clockwise/counterclockwise twist. EMG signals were measured from ten trunk muscles and spinal loads were assessed at 10, 20, 30, 40, 50, 60, 70, 80, 90%MVE(maximum voluntary exertion) in each direction. Three optimization models predicted lower L5/S1 disc compression forces than the EMG model, on average, by 31%(MF3), 27%(MS3), 8%(MI). Especially, in twist and extension, the differences were relatively large. Anterior-posterior shear forces predicted from optimization models were lower, on average, by 27%(MF3), 21%(MS3), 9%(MI) than by the EMG model, especially in flexion(MF3: 45%, MS3: 40%, MI: 35%). Lateral shear forces were predicted far less than anterior-posterior shear forces(total average=124 N), and the optimization models predicted larger values than the EMG model on average. These results indicated that the optimization models could underestimate compression forces during twisting and extension, and anterior-posterior shear forces during flexion. Thus, future research should address the antagonistic coactivation, one major reason of the difference between optimization models and the EMG model, in the optimization models.

지능형 전망모형을 결합한 로보어드바이저 알고리즘

김선웅(Sunwoong Kim) 한국지능정보시스템학회 2019 지능정보연구 Vol.25 No.2

Recently banks and large financial institutions have introduced lots of Robo-Advisor products. Robo-Advisor is a Robot to produce the optimal asset allocation portfolio for investors by using the financial engineering algorithms without any human intervention. Since the first introduction in Wall Street in 2008, the market size has grown to 60 billion dollars and is expected to expand to 2,000 billion dollars by 2020. Since Robo-Advisor algorithms suggest asset allocation output to investors, mathematical or statistical asset allocation strategies are applied. Mean variance optimization model developed by Markowitz is the typical asset allocation model. The model is a simple but quite intuitive portfolio strategy. For example, assets are allocated in order to minimize the risk on the portfolio while maximizing the expected return on the portfolio using optimization techniques. Despite its theoretical background, both academics and practitioners find that the standard mean variance optimization portfolio is very sensitive to the expected returns calculated by past price data. Corner solutions are often found to be allocated only to a few assets. The Black-Litterman Optimization model overcomes these problems by choosing a neutral Capital Asset Pricing Model equilibrium point. Implied equilibrium returns of each asset are derived from equilibrium market portfolio through reverse optimization. The Black-Litterman model uses a Bayesian approach to combine the subjective views on the price forecast of one or more assets with implied equilibrium returns, resulting a new estimates of risk and expected returns. These new estimates can produce optimal portfolio by the well-known Markowitz mean-variance optimization algorithm. If the investor does not have any views on his asset classes, the Black-Litterman optimization model produce the same portfolio as the market portfolio. What if the subjective views are incorrect? A survey on reports of stocks performance recommended by securities analysts show very poor results. Therefore the incorrect views combined with implied equilibrium returns may produce very poor portfolio output to the Black-Litterman model users. This paper suggests an objective investor views model based on Support Vector Machines(SVM), which have showed good performance results in stock price forecasting. SVM is a discriminative classifier defined by a separating hyper plane. The linear, radial basis and polynomial kernel functions are used to learn the hyper planes. Input variables for the SVM are returns, standard deviations, Stochastics %K and price parity degree for each asset class. SVM output returns expected stock price movements and their probabilities, which are used as input variables in the intelligent views model. The stock price movements are categorized by three phases; down, neutral and up. The expected stock returns make P matrix and their probability results are used in Q matrix. Implied equilibrium returns vector is combined with the intelligent views matrix, resulting the Black-Litterman optimal portfolio. For comparisons, Markowitz mean-variance optimization model and risk parity model are used. The value weighted market portfolio and equal weighted market portfolio are used as benchmark indexes. We collect the 8 KOSPI 200 sector indexes from January 2008 to December 2018 including 132 monthly index values. Training period is from 2008 to 2015 and testing period is from 2016 to 2018. Our suggested intelligent view model combined with implied equilibrium returns produced the optimal Black-Litterman portfolio. The out of sample period portfolio showed better performance compared with the well-known Markowitz mean-variance optimization portfolio, risk parity portfolio and market portfolio. The total return from 3 year-period Black-Litterman portfolio records 6.4%, which is the highest value. The maximum draw down is -20.8%, which is also the lowest value. Sharpe Ratio shows the highest

건강한 교회에서 교회 최적화 모형의 역할 연구

박관희 한국실천신학회 2013 신학과 실천 Vol.0 No.35

교회의 존재는 교인들로 하여금 “하나님 임재경험”을 통해 믿음과 순종의 삶을 살게 하는 것, 즉 하나님 나라의 왕적 통치 실현에 있다. 그래서 교회가 하나님의 나라를 지향하는 건강한 교회가 되려면 성령의 은혜, 목회자, 교인, 지역사회의 다양한 요소들을 고려해야만 한다. 이는 건강한 교회가 목회자나 교인들만의 노력으로 이루어지지 않기 때문이다. 그래서 건강한 교회가 되려면, 즉 하나님 임재경험과 하나님의 왕적 통치를 교회의 존재 목적에 적용하려면, 교회를 형성하는 요소들이 상호 협력하며 일체화가 되어야 한다. 이것을 설명해주는 개념이 바로 교회 최적화 모형이다. 최적화 모형(Optimization Model)은 최적화의 조건(교회건강변수)과 그에 따른 각 단계별 핵심사항(신앙생활변수)이, 가장 효과적이고 효율적인 방향으로 융합(목회철학변수)하여, 일체화된 상태(성령의 사역)를 추구하는 일련의 과정이다. 최적화의 주체는 “하나님의 주권과 섭리”에 근거한다. 그래서 최적화 모형은 가장 효과적이면서 효율적인 방향으로 융합하여 일체화된 상태를 지향하기 때문에, 절대적이라기보다는 상대적이다. 또한 최적화 모형은 크게 구성요소, 측정요소, 측정내용이 있다. 즉, 구성요소와 구성요소로부터 측정 가능한 개념으로 바꾼 측정요소와 측정내용으로 구성한다. 그래서 건강한 교회가 되기 위한 교회 최적화는 이와 같은 요소들을 균형 있게 설명하는 역할인 동시에, 개(個) 교회에 적용 가능하게 만드는 메커니즘(mechanism)이다. In this research, it has five main goals. First, it is to comprehend about healthy church correctly. Second, it is to consider and to compare church growth from the healthy church. Third, it is to observe the role of optimization model in the healthy church and the meaning of it. Fourth, it is to recognize a process of the church optimization from the healthy church in order to understand a condition of the church optimization and a stage of it. Finally, Korean Church gives an evidence for the growth and the development as the healthy church provides these factors through these data and results. As the healthy church kingly administers and serves God by experiencing the Presence of God, the church must consider intricate elements such as Holy Spirit, a pastor, a congregation, a local church, and a community. To be specific, Church Optimization Model means that the optimized condition, which a variable of a church health is equal to a variable of a measurement and a stage of the optimization is equal to a variable of the faithful life, is merged to the most effective and effectual way by a variable of a pastoral philosophy to create unification through Holy Spirit. In addition, the optimization model consists of combined elements such as a human, devotion, and innocence. It also forms a component of the Measurement that is equal to congregations, devotion, a purpose of visiting, and contents of a measurement. These factors are the same as optimization conditions and a church analysis which include visitors, a number of worshipping times, a number of visiting, and the amount of offering. Then, the optimization stage does not only go through a process for the congregation analysis that is equal to “pastoral image→visiting→settlement→nourishment→training→ministry,” but also go through a process for the pastoral analysis that is equal to “spirituality→pastoral leadership→worship,” “preaching→program,” and “church administration.” The subject of the optimization is based on the sovereignty and the providence of God. Subsequently, this model is relative, but no absolute. Overall, Church Optimization Model explains elements of the healthy church as above, and thus it can become a very useful mechanism that is available to the local church for applying to this model.

포트폴리오 구성을 위한 최적 시뮬레이션의 활용에 관한 연구

구기동 ( Ki Dong Koo ),이종구 ( Jong Gu Lee ) 한국경영공학회 2013 한국경영공학회지 Vol.18 No.2

This study was conducted for checking a developing process and a model of the Optimal Asset Allocation by the Mean-variance Model and the Optimal Simulation. Used mainly as an objective function, the Asset Allocation decides investment proportion by minimizing a risk and making a rate of return a constant. The model of Asset Allocation has developed from the Mean-variance Model, the linear model, the simulation model to the Optimal Simulation step by step. In the fist stage of combining the models, a parameter for simulation and optimization is estimated by creating input data. The second stage is that the Optimal Asset Allocation is produced by the quadratic programming. In the final step, the optimal range is calculated by the Optimal Simulation. And the optimal range is divided into the best case and the worst one. Each model draws an optimal investment proportion through an objective function. In a situation where a risk is the least, an investment proportion to a safe asset having low fluctuation appeared high independently of the model`s type. While a resource distribution is done by the Mean-variance Model, the Optimal Simulation suggested a resource distribution to risky asset under the same conditions. Therefore, it was found that there was a difference in distributing resource between the two models. It is showed that the Optimal Simulation can be used to select a concrete investment and to expect a range of a resource distribution among risky assets. The basis that chooses an asset achieving dominant result or minimizing risk by duration among individual assets can be offered. Because the domestic stock market has strong fluctuation, rather than the Mean-variance Model distributing resource conservatively, active method of resource distribution, the Optimal Simulation can make more profit. And the Optimal Simulation needs to be used actively in practical section because it has a strength the Mean-variance Model doesn`t have.