Genetic Algorithm을 이용한 최적 인덱스 펀드 구축 방안의 모색

김길주(Kil-Joo Kim),안창호(Chang-Ho An) 융복합지식학회 2019 융복합지식학회논문지 Vol.7 No.2

H.Markowitz의 평균-분산 framework을 기초로 하는 기존의 포트폴리오 이론에서는 펀드 작성시의 목적함수는 위험의 최소화이며, 제약 조건은 일정 수준의 목표 수익이 되며, 펀드에 편입되는 종목은 사전에 주어져 있고, 각 종목의 편입 비율만 결정되는 형태이다. 그러나 종목 선택의 관점에서 펀드 작성은 조합 최적화 문제로 귀착된다. 즉, 조합 최적화 문제에서는 모든 해를 열거하고, 그 중에서 가장 최적인 해를 선택하는 것이 이상적이다. 실제로는 문제의 규모와 함께 조사하여야 하는 해의 개수가 급격히 증가하므로, 계산기로는 엄밀한 최적해를 발견한다는 것이 불가능하다. 펀드 작성은 조합 최적화 문제이며 그 최적해를 탐색하는 방법으로 유전적 알고리듬을 사용한다. 유전적 알고리듬은 자연계의 교차, 도태, 돌연변이를 컴퓨터로 실행하는 것으로 최적화 문제 중에서도 조합 최적화 문제에 효과적인 방법이다. 지금까지 대부분의 다목적 최적화는 목적함수의 1차원화와 단 목적 최적화의 반복에 의하여 해를 구하였지만, 본 연구에서는 유전적 알고리듬이 복수의 해 후보 집단을 구성하는 것을 이용하여 파레토 최적해 집단을 직접 찾는다는 점이 다르다. 또한 유전적 알고리듬을 사용하여 기존 펀드의 재조정도 가능하다. 더 나아가 평가 함수의 변경이나 재조정 방법을 개량하면 좀더 발전된 결과를 얻을 수 있다. In the traditional portfolio theory based on H.Markowitz’s mean-variance framework, the objective function in fund structuring is the minimization of risk, the constraints become the target return at a certain level, the stocks incorporated into the fund are given in advance, and only the proportion of stocks’ inclusion is determined. However, in terms of stock selection, the fund structuring comes down to the combinatorial optimization problem. In other words, it is ideal to choose the optimal solution by enumerating all solutions in the combinatorial optimization problem. In reality, with the scale of the problem, the number of solutions to be investigated increases dramatically, so it is impossible for calculators to find a rigorous optimal solution. Fund structuring is a combinatorial optimization problem and genetic algorithms are used as a way to search for the optimal solution. The genetic algorithm is an effective method for combinatorial optimization problem among the optimization problems by executing the crossover, selection and mutation of the natural systems by computer operations. Many conventional multi-objective optimization techniques have solved the problem by one-dimensional optimization of the objective function and repetition of the single-objective optimization. In this study, however, the pareto optimal solution group is directly obtained by using the genetic algorithm constituting a plurality of solution candidate groups is different. It is also possible to rebalancing existing funds using genetic algorithms. Furthermore, by improving the method of changing or re-adjusting the evaluation function, more advanced results can be obtained.

Genetic evolution vs. function approximation: Benchmarking algorithms for architectural design optimization

Thomas Wortmann 한국CDE학회 2019 Journal of computational design and engineering Vol.6 No.3

This article presents benchmark results from seven simulation-based problems from structural, building energy, and daylight optimization. Growing applications of parametric design and performance simula-tions in architecture, engineering, and construction allow the harnessing of simulation-based, or black-box, optimization in the search for less resource- and/or energy consuming designs. In architectural design optimization (ADO) practice and research, the most commonly applied black-box algorithms are genetic algorithms or other metaheuristics, to the neglect of more current, global direct search or model-based, methods. Model-based methods construct a surrogate model (i.e., an approximation of a fitness landscape) that they refine during the optimization process. This benchmark compares meta-heuristic, direct search, and model-based methods, and concludes that, for the given evaluation budget and problems, the model-based method (RBFOpt) is the most efficient and robust, while the tested genetic algorithms perform poorly. As such, this article challenges the popularity of genetic algorithms in ADO, as well as the practice of using them for one-to-one comparisons to justify algorithmic innovations.

Optimization for Disassemble Sequence Planning of Electromechanical Products during Recycling Process Based on Genetic Algorithms

Zhang Chunming 보안공학연구지원센터 2016 International Journal of Multimedia and Ubiquitous Vol.11 No.4

Disassemble and re-manufacturing is an important way to save energy. Sequence planning is the core issue of disassemble. In this study, disassemble model has been established based on the analysis of disassemble route planning and disassemble sequence. The disassemble sequence optimization based on genetic algorithms has been carried out. In the gear pump disassemble sequence, for example, an illustrative example, proposed in this study to verify the disassemble sequence planning based on electromechanical product recovery process optimization genetic.

트러스 구조의 확률적 탐색법을 이용한 최적설계

류미란,박정선 한국 항공대학교 항공산업기술연구소 2001 航空宇宙産業技術硏究所 硏究誌 Vol.11 No.-

전통적인 최적화 기법은 도함수를 구하여 최적화를 수행한다. 그러므로 이러한 전통적인 기법을 수많은 도함수를 구해야 한다. 따라서 국부 최적화에는 효과적이나 전역 최적화에는 어려움이 있다. 이러한 문제를 극복하기 위해 확률적 탐색 방법을 이용한다. 확률적인 탐색과정을 가지는 최적화 방법들로는 유전자 알고리즘, 모사풀림, 인공생명 알고리듬 등을 들 수 있다 이 알고리듬들 중 모사풀림을 사용하여 트러스 구조물을 최적화 하였다. Discrete optimization methods takes advantage of function values only, different form conventional optimization method using gradient information. The dicrete optimization methods do not need gradient calculations. The dicrete optimization methods are efficient to the global optimization. Genetic algorithm, simulated annealing and artificial life algorithm are optimization methods that have the stochastic search processes. In this paper, we studied these optimization methods. And using the simulated annealing, one of the these optimization methods, truss structures are optimized.

Design Optimization Methods for Electrical Machines: A Review

Omar Mohd. Fairoz Bin,Sulaiman Erwan Bin,Soomro Irfan Ali,Ahmad Md. Zarafi Bin,Aziz Roziah 대한전기학회 2023 Journal of Electrical Engineering & Technology Vol.18 No.4

Most of the appliances in industrial equipment and systems uses electric machines. They fill the various requirements for global sustainability not only physically or technologically but also environmentally. Therefore, progressively complex engineering domains and constraints are involved in the design optimization process such as electromagnetics, structural mechanics, and heat transfer. This paper aims to present a review of the design optimization methods for electrical machines, including design analysis methods and models, optimization models and algorithms. Several efficiency optimization methods are highlighted such as Gradient Based Algorithm, Tabu Search, Genetic Algorithm, Differential Evolution, Particle Swarm Optimization, Multi-objective Algorithm and Deterministic Optimization Method. Meanwhile, Deterministic Optimization Method has been presented on Field excitation, Permanent magnet and Hybrid excitation flux switching machines for the optimization. From the literature reviews, it is observed that DOM algorithms gained the best design technique for electric machines to produce optimal performances.

An Optimized Space Layout Process for Future Design Using Genetic Algorithms

Jisoo Kim,Yeonsook Hwang 한국주거학회 2015 한국주거학회 국제학술대회논문집 Vol.2015 No.4

The methodologies for space design have been diversified since the specialized approaches have been introduced to designers. For designing better plans, many stakeholders of buildings use quantitative approaches rather than intuitive. Genetic algorithm, on the other hand, has been studied because of its optimized model with numeric data. Therefore, this paper aims to propose effective and optimized space programming approaches, especially using specific genetic algorithms. The suggested optimized space layout planning process consists of three processes; modeling process, optimizing process and result visualization. In modeling process, we suggest the required information for finding optimized space layout process. Optimizing process, on the other hand, we separate the process to topology optimization and size optimization. In this process, function of spaces, minimum and maximum area are set to constraints of the algorithms so that optimized design plans can be shown depends on the constraints. Lastly, in result visualization, we suggest to convert the conceptual diagram to utilizable floor plan.

Channel geometry optimization using a 2D fuel cell model and its verification for a polymer electrolyte membrane fuel cell

Yang, W.J.,Wang, H.Y.,Kim, Y.B. Pergamon Press ; Elsevier Science Ltd 2014 INTERNATIONAL JOURNAL OF HYDROGEN ENERGY - Vol.39 No.17

This paper studies the optimization method of channel geometries for a proton exchange membrane fuel cell (PEMFC) using a genetic algorithm (GA). The channel and rib widths and channel height are selected as geometry variables. The fuel cell output power is chosen as the cost function for the optimization. In this paper, an in-house genetic algorithm is constructed, and the fuel cell output power is obtained using an interfacing program connected to a commercial computational fluid dynamics (CFD) tool, COMSOL, in a Matlab environment. The 2D PEMFC is used to calculate the performance cost function for computational time and cost. The calculated output power of the PEMFC is delivered to the in-house GA program to check for optimality. After the optimality is checked, the new geometry data is fed back to the COMSOL to calculate the PEMFC output power until the optimization process is finished. Experiments are conducted to support the optimized results using three different channel geometries: channel-to-rib width ratios of 0.5:1, 1:1, and 2:1. A full 3D PEMFC CFD model is constructed using COMSOL to support the 2D CFD optimization results. This paper shows the possibility of applying the geometry optimization process to sophisticated electrochemical reaction systems, such as a PEMFC, using a GA and a commercial CFD tool on the Matlab platform. The geometries and materials can be optimized using this approach to obtain the most efficient performance of an electrochemical system.

A Genetic Relation Algorithm with Guided Mutation for the Large-Scale Portfolio Optimization

YanChen,Chuan Yue,Shingo Mabu,Kotaro Hirasawa 제어로봇시스템학회 2009 제어로봇시스템학회 국제학술대회 논문집 Vol.2009 No.8

The survey of the relevant literature showed that the rehave been many studies for portfolio optimization problem and that the number of studies which have investigated the optimum portfoliousing evolutionary computation is quite high. But almost none of these studies deals with genetic relation algorithm(GRA). This study presents an approach to large-scale port folio optimization problem using GRA with a new operator, called guided mutation. In order to pickup the most efficient portfolio, GRA considers the correlation coefficient between stock brands as strength, which indicates there lation between nodes in each individual of GRA. Guided mutation generates off spring according to the average value of correlation coefficients in each individual. A genetic relation algorithm with guided mutation(GRA/G) for the portfolio optimization is proposed in this paper. Genetic network programming(GNP), which was proposed in our previous research, is used to validate the performance of the portfolio generated with GRA/G. The results show that GRA/Gap proach is successful in portfolio optimization.

Monarch butterfly optimization-based genetic algorithm operators for nonlinear constrained optimization and design of engineering problems

El-Shorbagy M A,Alhadbani Taghreed Hamdi 한국CDE학회 2024 Journal of computational design and engineering Vol.11 No.3

This paper aims to present a hybrid method to solve nonlinear constrained optimization problems and engineering design problems (EDPs). The hybrid method is a combination of monarch butterfly optimization (MBO) with the cross-over and mutation operators of the genetic algorithm (GA). It is called a hybrid monarch butterfly optimization with genetic algorithm operators (MBO-GAO). Combining MBO and GA operators is meant to overcome the drawbacks of both algorithms while merging their advantages. The self-adaptive cross-over and the real-valued mutation are the GA operators that are used in MBO-GAO. These operators are merged in a distinctive way within MBO processes to improve the variety of solutions in the later stages of the search process, speed up the convergence process, keep the search from getting stuck in local optima, and achieve a balance between the tendencies of exploration and exploitation. In addition, the greedy approach is presented in both the migration operator and the butterfly adjusting operator, which can only accept offspring of the monarch butterfly groups who are fitter than their parents. Finally, popular test problems, including a set of 19 benchmark problems, are used to test the proposed hybrid algorithm, MBO-GAO. The findings obtained provide evidence supporting the higher performance of MBO-GAO compared with other search techniques. Additionally, the performance of the MBO-GAO is examined for several EDPs. The computational results show that the MBO-GAO method exhibits competitiveness and superiority over other optimization algorithms employed for the resolution of EDPs.

Development of a Mutation Operator in a Real-Coded Genetic Algorithm for Bridge Model Optimization

김재천,한만석,신수봉 대한토목학회 2024 KSCE Journal of Civil Engineering Vol.28 No.5

A mutation operator in a real-coded genetic algorithm is developed and applied for efficient bridge-model optimization. A mutation operator that changes uniformly or dynamically with a crossover operator is proposed to address optimization problems. The performance of the combined genetic operators was verified using a variety of available test problems based on the convergence and search speed of the global optimal solution. It is shown that the genetic algorithm proposed in this study yields relatively better results than the available algorithms and is more effective in constrained optimization problems. The performance of the proposed genetic algorithm is also verified through a sample study using a field load test for the model optimization of an existing bridge.