• 한국정보처리학회논문지
• /
• 제7권5호
• /
• pp.1401-1408
• /
• 2000

소프트웨어 개발 초기 단계의 문제점이 개발 후반부 산물의 품질에 심각한 영향을미치기 때문에 설계 명세를 이용하여 위험 부분을 예측하는위험도 예측 모델은 전체 시스템 개발비용을 낮추는 데 중요한 역할을 하고 있으며, 이러한 예측 모델은 결과 산물이 매우 크고 실행 정확성이 요구되는통신 소프트웨어 같은 실시간 시스템 설계에 더욱 필요하다. 판별분석, 인공신경망, 분류트리 등의 기법들을 이용한 모델들이 제안되었으나 이들은 결과에 대한 원인 분석의 어려움, 낮은 확장성 등의 문제점들을 지니고 있었다. 본 논문에서는 유전자 알고리즘에 의해 구축된 퍼지 규칙 베이스를 이용한 위험도 예측 모델을 제안한다. 제안 모델은 예측 결과에 대한 원인 분석이 용이하고 높은 확정성과 적용성을 지니고 규칙수에 대한 제안이 없다. 이러한 내부특성들 비교의 모의 실을 통한 예측 정확도 비교를 통해 제안 모델이 타 모델들보다 우수함을 보였다.

• 한국수자원학회논문집
• /
• 제53권6호
• /
• pp.395-408
• /
• 2020

본 연구논문은 북부아프리카의 알제리에 위치한 하천유역에서 다중선행일 유출량의 예측을 위하여 진화적 최적화기법과 k-fold 교차검증을 결합한 세 개의 서로 다른 기계학습 접근법 (인공신경망, 적응 뉴로퍼지 시스템, 그리고 웨이블릿 기반 신경망)을 개발하고 적용하는 것이다. 인공신경망과 적응 뉴로퍼지 시스템은 root mean squared error (RMSE), Nash-Sutcliffe efficiency (NSE), correlation coefficient (R), 그리고 peak flow criteria (PFC) 의 네 개의 통계지표를 기반으로 하여 모형의 훈련 및 테스팅 결과 유사한 모형수행결과를 나타내었다. 웨이블릿 기반 신경망모형은 하루선행일 테스팅의 결과 RMSE = 8.590 ㎥/sec 과 PFC = 0.252로 분석되어서 인공신경망의 RMSE = 19.120 ㎥/sec, PFC = 0.446 과 적응 뉴로퍼지 시스템의 RMSE = 18.520 ㎥/sec, PFC = 0.444 보다 양호한 결과를 나타내었고, NSE와 R의 값도 웨이블릿 기반 신경망모형이 우수한 것으로 나타났다. 그러므로 웨이블릿 기반 신경망은 알제리 세이보스 하천에서 다중선행일의 예측을 위하여 효율적인 도구로 사용할 수 있다.

• 대한교통학회지
• /
• 제22권2호
• /
• pp.77-86
• /
• 2004

데이터마이닝 (Data Mining)은 대용량의 데이터에 존재하는 관계, 패턴, 규칙 등을 효율적으로 탐색하여 이를 모형화함으로써, 유용한 정보로 추출 변환하는 일련의 과정이다. 특히 베이지안 망 (Bayesian Network)은 신경망, 유전자알고리즘 퍼지이론 등과 더불어 데이터마이닝의 중요한 기법 중의 하나로서 베이지안 통계 이론(Bayesian Statistics Theory)를 적용하여 변수들간의 확률적인 관계를 기호화함으로써, 설명변수들과 종속변수들간의 인과관계를 파악할 수 있다. 이 연구는 기존에 적용된 바가 없는 데이터마이닝의 베이지안 망을 이용하여 수도권 교통수단선택 모형을 구축한다. 2002년도 수도권 가구통행실태조사 자료의 사회 경제적 특성과 교통체계 특성을 반영하여 베이지안 망을 이용한 교통수단선택 모형을 설계 구축하여, 각 변수들간의 상관관계와 인과관계를 분석함으로써, 설명변수인 성과 연령의 구성비가 변하였을 때, 교통수단선택의 변화율(확률)을 예측한다. 이 연구를 통해 현실에서는 내재하나 설명변수간의 복잡한 상관성을 배제하고 설명변수들과 교통수단선택간의 단순한 직선관계를 가정하는 기존 교통수단선택 모형의 한계를 극복할 수 있는 가능성을 제시한다. 또한 선택되지 않은 교통수단에 대한 정보의 부족으로 인한 교통수단선택 모형 구축의 어려움을 극복한다. 또한 다양한 교통정책에 따른 교통수단선택의 변화를 실시간으로 시뮬레이션 할 수 있는 방법론을 개발한다.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 2005년도 춘계 학술발표회 논문집
• /
• pp.209-216
• /
• 2005

The controller that can control the smart base isolation system consisting of M damper and friction pendulum systems(FPS) is developed in this study. A fuzzy logic controller (FLC) is used to modulate the M damper force because the FLC has an inherent robustness and ability to handle non-linearities and uncertainties. A genetic algorithm (GA) is used for optimization of the FLC. When earthquake excitations are applied to the structures equipped with smart base isolation system, the relative displacement at the isolation level as well as the acceleration of the structure should be regulated under appropriate level. Thus, NSGA-II(Non-dominated Sorting Genetic Algorithm) is employed in this study as a multi-objective genetic algorithm to meet more than two control objectives, simultaneously. NSGA-II is used to determine appropriate fuzzy control rules as well to adjust parameters of the membership functions. Effectiveness of the proposed method for optimal design of the FLC is judged based on computed responses to several historical earthquakes. It has been shown that the proposed method can efficiently find Pareto optimal sets that can reduce both structural acceleration and base drift from numerical studies.

• 한국해양공학회지
• /
• 제28권5호
• /
• pp.404-410
• /
• 2014

The purpose of this study was to find the optimal arrangement of FPSO equipment in a module while considering the economic value and fire risk. We estimated the economic value using the pipe connections and pump installation cost in an HP (high pressure) gas compression module. The equipment risks were also analyzed using fire scenarios based on historical data. To consider the wind effect during a fire accident, fuzzy modeling was applied to improve the accuracy of the analysis. The objective functions consisted of the economic value and fire risk, and the constraints were the equipment maintenance and weight balance of the module. We generated a Pareto-optimal front group using a multi-objective GA (genetic algorithm) and suggested an equipment arrangement method that included the opinions of the designer.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2005년도 심포지엄 논문집 정보 및 제어부문
• /
• pp.6-8
• /
• 2005

In this paper, we discuss model identification of nonlinear data using GAs-based Fuzzy Polynomial Neural Networks(GAs-FPNN). Fuzzy Polynomial Neural Networks(FPNN) is proposed model based Group Method Data Handling(GMDH) and Neural Networks(NNs). Each node of FPNN is expressed Fuzzy Polynomial Neuron(FPN). Network structure of nonlinear data is created using Genetic Algorithms(GAs) of optimal search method. Accordingly, GAs-FPNN have more inflexible than the existing models (in)from structure selecting. The proposed model select and identify its for optimal search of Genetic Algorithms that are no. of input variables, input variable numbers and consequence structures. The GAs-FPNN model is select tuning to input variable number, number of input variable and the last part structure through optimal search of Genetic Algorithms. It is shown that nonlinear data model design using Genetic Algorithms based FPNN is more usefulness and effectiveness than the existing models.

• 제어로봇시스템학회논문지
• /
• 제13권4호
• /
• pp.344-351
• /
• 2007

In this paper, we propose an approach to design multi-fuzzy controllers for the superheat and the low pressure that have an influence on energy efficiency and stabilization of air conditioning system with multi-evaporators. Air conditioning system with multi-evaporators is composed of compressor, condenser, several evaporators and several expansion valves. It is quite difficult to control the air conditioning system because the change of the refrigerant condition give an impact on the overall air conditioning system. In order to solve the drawback, we design multi-fuzzy controllers which control simultaneously both three expansion valve and one compressor for the superheat and the low pressure of air conditioning system. The proposed multi fuzzy controllers are given as a kinds of controller types such as a simplified fuzzy inference type. Here the scaling factors of each fuzzy controller are efficiently adjusted by Hierarchical Fair Competition-based Genetic Algorithms. The values of performance index of the simulation results of the A company type compare with simulation results of simplified inference type.

• 제어로봇시스템학회논문지
• /
• 제6권3호
• /
• pp.273-283
• /
• 2000

In this paper we suggest an optimal design method of Fuzzy-Neural Networks(FNN) model for complex and nonlinear systems. The FNNs use the simplified inference as fuzzy inference method and Error Back Propagation Algorithm as learning rule. And we use a HCM(Hard C-Means) Clustering Algorithm to find initial parameters of the membership function. The parameters such as parameters of membership functions learning rates and momentum weighted value is proposed to achieve a sound balance between approximation and generalization abilities of the model. According to selection and adjustment of a weighting factor of an aggregate objective function which depends on the number of data and a certain degree of nonlinearity (distribution of I/O data we show that it is available and effective to design and optimal FNN model structure with a mutual balance and dependency between approximation and generalization abilities. This methodology sheds light on the role and impact of different parameters of the model on its performance (especially the mapping and predicting capabilities of the rule based computing). To evaluate the performance of the proposed model we use the time series data for gas furnace the data of sewage treatment process and traffic route choice process.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2007년도 심포지엄 논문집 정보 및 제어부문
• /
• pp.201-203
• /
• 2007

The paper concerns the successive optimization for structure and parameters of fuzzy inference systems that is based on parallel Genetic Algorithms (PGA) and information data granulation (IG). PGA is multi, population based genetic algorithms, and it is used tu optimize structure and parameters of fuzzy model simultaneously, The granulation is realized with the aid of the C-means clustering. The concept of information granulation was applied to the fuzzy model in order to enhance the abilities of structural optimization. By doing that, we divide the input space to form the premise part of the fuzzy rules and the consequence part of each fuzzy rule is newly' organized based on center points of data group extracted by the C-Means clustering, It concerns the fuzzy model related parameters such as the number of input variables to be used in fuzzy model. a collection of specific subset of input variables, the number of membership functions according to used variables, and the polynomial type of the consequence part of fuzzy rules, The simultaneous optimization mechanism is explored. It can find optimal values related to structure and parameter of fuzzy model via PGA, the C-means clustering and standard least square method at once. A comparative analysis demonstrates that the Dnmosed algorithm is superior to the conventional methods.

• 경영과학
• /
• 제24권2호
• /
• pp.157-176
• /
• 2007

Fuzzy cognitive map(FCM) has long been used as an effective way of constructing the human's decision making process explicitly. By taking advantage of this feature, FCM has been extensively used in providing what-if solutions to a wide variety of business decision making problems. In contrast, the goal-seeking analysis mechanism by using the FCM is rarely observed in literature, which remains a research void in the fields of FCM. In this sense, this study proposes a new type of the FCM-based goal-seeking analysis which is based on utilizing the genetic algorithm. Its main recipe lies in the fact that the what-if analysis as well as goal-seeking analysis are enabled very effectively by incorporating the genetic algorithm into the FCM-driven inference process. To prove the empirical validity of the proposed approach, valid questionnaires were gathered from a number of experts on software sales, and analyzed statistically. Results showed that the proposed approach is robust and significant.