• Journal of the Korean Institute of Intelligent Systems
• /
• v.5 no.4
• /
• pp.41-55
• /
• 1995

In this paper, an optimal idenfication method using fuzzy-neural networks is proposed for modeling of nonlinear complex systems. The proposed fuzzy-neural modeling implements system structure and parameter identification using the intelligent schemes together wlth optimization theory, linguistic fuzzy implication rules, and neural networks(NNs) from input and output data of processes. Inference type for this fuzzy-neural modeling is presented as simplified inference. To obtain optimal model, the learning rates and momentum coefficients of fuzzy-neural networks(FNNs) are tuned automatically using improved modified complex method and modified learning algorithm. For the purpose of its application to nonlinear processes, data for route choice of traffic problems and those for activateti sluge process of sewage treatment system are used for the purpose of evaluating the performance of the proposed fuzzy-neural network modeling. The results show that the proposed method can produce the intelligence model with higher accuracy than other works achieved previously.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.641-644
• /
• 2019

블록체인은 탈중앙화된 신뢰 기반 분산 데이터베이스로 높은 신뢰성과 보안성을 제공하지만 기존의 블록체인들은 확장성이 떨어진다는 문제를 지닌다. 이 문제를 해결하기 위해 기존의 방법들은 소수의 대표노드들을 선출하여 합의 과정을 간소화 하려 하였다. 그러나 이러한 시도는 대표 노드를 선출하기 위해 지분 기반 투표 방식을 사용하기 때문에 많은 지분을 가진 노드들에게 권한이 집중될 수 있다는 한계를 갖는다. 본 연구는 이러한 한계점을 해결한 대표노드 선출/해임 모델을 소개한다. 제안 방법은 Raft 의 투표 알고리즘을 확장하여 대표노드의 공정한 선출과 대표노드의 부적절한 행위를 예방한다. 제안 방법은 모델 검증을 통해 도달 가능성, 안전성, 활동성이 확인되었다.

• Journal of KIISE:Databases
• /
• v.36 no.4
• /
• pp.292-305
• /
• 2009

Wireless sensor networks have a lot of application areas such as industrial process control, machine and resource management, environment and habitat monitoring. One of the main objects of using wireless sensor networks in these areas is the event detection. To detect events at a user's request, we need a join processing between sensor data and the predicates of the events. If there are too many predicates of events compared with a node's capacity, it is impossible to store them in a node and to do an in-network join with the generated sensor data This paper proposes a predicate-merge based in-network join approach to efficiently detect multiple events, considering the limited capacity of a sensor node and many predicates of events. It reduces the number of the original predicates of events by substituting some pairs of original predicates with some merged predicates. We create an estimation model of a message transmission cost and apply it to the selection algorithm of targets for merged predicates. The experiments validate the cost estimation model and show the superior performance of the proposed approach compared with the existing approaches.

• Journal of Internet of Things and Convergence
• /
• v.5 no.2
• /
• pp.67-72
• /
• 2019

As Internet voting can take place regardless of a voter's location, the participation rate of the voters would be increased and economic costs will be reduced. But the drawback of it is that all participants have to trust the election management server. If the server colludes with the specific candidate, the other candidates cannot prove rigged election. In addition, majority of researches on Internet voting are mainly focused on the voting restricted by the region and the country. Thus, it's not appropriate for the election in Internet community such as YouTube channels. As the Internet community is composed of members from all around the world, the new type of voting model is needed. In this study, we propose the smart contract based Internet voting model applicable on the blockchain network. The proposed smart contract model consists of candidate registration, voter registration, voting and counting stages. In the proposed model, anonymity of the voter is assured in the voter registration and voting stages, and all candidates can confirm the fairness of the election in the counting stage.

• Korean Journal of Construction Engineering and Management
• /
• v.16 no.6
• /
• pp.53-62
• /
• 2015

Time-Cost Trade-Off (TCTO) model is an important model in the construction project planning and control area. Two types of Existing TCTO model, continuous and discrete TCTO model, have been developed by researchers. However, Using only one type of model has a limitation to represent a realistic crash scenario of activities in the project. Thus, this paper presents a comprehensive TCTO model that combines a continuous and discrete model. Additional advanced features for non-linear relationship, incentive, and liquidated damage are included in the TCTO model. These features make the proposed model more applicable to the construction project. One CPM network with 6 activities is used to explain the proposed model. The model found an optimal schedule for the example to satisfy all the constraints. The results show that new model can represent more flexible crash scenario in TCTO model.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.39 no.3
• /
• pp.18-31
• /
• 2002

In this study, we introduce a concept of advanced neurofuzzy polynomial networks(ANFPN), a hybrid modeling architecture combining neurofuzzy networks(NFN) and polynomial neural networks(PNN). These networks are highly nonlinear rule-based models. The development of the ANFPN dwells on the technologies of Computational Intelligence(Cl), namely fuzzy sets, neural networks and genetic algorithms. NFN contributes to the formation of the premise part of the rule-based structure of the ANFPN. The consequence part of the ANFPN is designed using PNN. At the premise part of the ANFPN, NFN uses both the simplified fuzzy inference and error back-propagation learning rule. The parameters of the membership functions, learning rates and momentum coefficients are adjusted with the use of genetic optimization. As the consequence structure of ANFPN, PNN is a flexible network architecture whose structure(topology) is developed through learning. In particular, the number of layers and nodes of the PNN are not fixed in advance but is generated in a dynamic way. In this study, we introduce two kinds of ANFPN architectures, namely the basic and the modified one. Here the basic and the modified architecture depend on the number of input variables and the order of polynomial in each layer of PNN structure. Owing to the specific features of two combined architectures, it is possible to consider the nonlinear characteristics of process system and to obtain the better output performance with superb predictive ability. The availability and feasibility of the ANFPN are discussed and illustrated with the aid of two representative numerical examples. The results show that the proposed ANFPN can produce the model with higher accuracy and predictive ability than any other method presented previously.

• Review of KIISC
• /
• v.22 no.5
• /
• pp.62-72
• /
• 2012

스마트그리드는 전력시스템과 이를 제어하기 위한 통신 인프라를 중심으로 다양한 시스템이 서로 통합되는 종합적인 플랫폼으로 이해할 수 있다. 기존에 각기 독립적으로 운영되는 시스템과 통신 인프라가 통합되기 시작하면서 다양한 상호작용이 파생되고 그로 인해 잠재적인 보안 측면의 위험성도 커지게 된다. 전통적인 전력시스템은 폐쇄적인 SCADA 네트워크를 기반으로 운영되었기 때문에 최소한의 보안강도가 보장되었지만, 스마트그리드 하에서는 개방형 통신망과 연계되면서, 기존의 사이버 보안 위협들이 전력시스템으로 유입하게 된다. 기존의 IT 시스템과는 달리 전력시스템과 같은 제어시스템은 물리적 작동과 공정이 수반되기 때문에 새로운 위험이 발생하기도 하고 기존의 위험이 증폭되기도 한다. 전력시스템에서는 가용성이 그 무엇보다 중요하기 때문에, 스마트그리드 체제하에서의 다양한 위협요인을 미리 파악하고 이에 대비한 계획을 수립함으로써, 그러한 가용성의 수준을 유지할 필요가 있다. 이를 위해서는 기존의 사이버 위협이 어떠한 경로를 통해 전력시스템에 영향을 미치게 되며 그로 인한 잠재적 위험이 얼마나 되는가를 평가할 필요가 있다. 그러나 스마트그리드는 아직까지 구축중인 미래형 시스템이고 누적된 과거 데이터가 없기 때문에 가상의 하드웨어 기반 테스트베드 내지 소프트웨어 기반의 시뮬레이션 모델을 통해 이를 사전적으로 테스트할 필요가 있다. 또한 스마트그리드는 서로 다른 IT 시스템과 물리적 설비들이 결합되는 복잡한 시스템이라는 측면에서, 잠재적으로 발생 가능한 다양한 위험을 분석하고 평가할 수 있는 모델의 수립이 요구된다. 본고에서는 그러한 CPS 기반 시뮬레이션 모델에 대한 현재의 연구동향을 검토하고, 향후 실질적으로 구현하기 위한 방안을 제안하고자 한다.

• Journal of Korean Institute of Industrial Engineers
• /
• v.23 no.4
• /
• pp.669-686
• /
• 1997

In this research, an object-oriented Petri net model for representing a flexible process plan is proposed, which is hierarchically multi-faceted for supporting planning, scheduling, and shop floor control functions. The multi-faceted process plan model consists of the following: a) an object model which represents on object-oriented data model, b) a static model which represents a process flow model with process alternatives, and c) a dynamic model which represents a process activity model with resources alternatives, of a flexible process plan. Petri nets allow the static and the dynamic process plan models to be represented in a unified formalism with an ease of model transformation. The multi-faceted process plan model suggested in this paper, is illustrated with a prismatic port in comprehensive detail.

• Proceedings of the KIEE Conference
• /
• 2005.07d
• /
• pp.2873-2875
• /
• 2005

The optimal structure of the conventional Fuzzy Polynomial Neural Networks (FPNN)[3] depends on experience of designer. For the conventional Fuzzy Polynomial Neural Networks, input variable number, number of input variable, number of Membership Functions(MFs) and consequence structures are selected through the experience of a model designer iteratively. In this paper, we propose the new design methodology to find the optimal structure of Fuzzy Polymomial Neural Network by using Genetic Algorithms(GAs)[4, 5]. In the sequel, It is shown that the proposed Advanced Genetic Algorithms based Fuzzy Polynomial Neural Network(Advanced GAs-based FPNN) is more useful and effective than the existing models for nonlinear process. We used Medical Imaging System(MIS)[6] data to evaluate the performance of the proposed model.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.11a
• /
• pp.248-251
• /
• 2020

전 반도체 제조 및 검사 공정 과정을 자동화하는 스마트 팩토리의 실현에 있어 제품 검수를 위한 검사 장비는 필수적이다. 하지만 딥 러닝 모델 학습을 위한 데이터 처리 과정에서 엔지니어가 전체 웨이퍼 영상에 대하여 결함 항목 라벨을 매칭하는 것은 현실적으로 불가능하기 때문에 소량의 라벨 (labeled) 데이터와 나머지 라벨이 없는 (unlabeled) 데이터를 적절히 활용해야 한다. 또한, 웨이퍼 영상에서 결함이 발생하는 빈도가 결함 종류별로 크게 차이가 나기 때문에 빈도가 적은 (minor) 결함은 잡음처럼 취급되어 올바른 분류가 되지 않는다. 본 논문에서는 소량의 라벨 데이터와 대량의 라벨이 없는 데이터를 동시에 활용하면서 결함 사이의 발생 빈도 불균등 문제를 해결하는 점진적 데이터 평준화 (progressive pseudo-labeling balancer)를 제안한다. 점진적 데이터 평준화를 이용해 분류 네트워크를 학습시키는 경우, 기존의 테스트 정확도인 71.19%에서 6.07%-p 상승한 77.26%로 약 40%의 라벨 데이터가 추가된 것과 같은 성능을 보였다.