• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2002.12a
• /
• pp.447-450
• /
• 2002

본 논문에서는 통합 관리 시스템을 이용하여 현재 김해에 있는 플랜트의 공정상태를 모니터링하고, 퍼지와 신경망을 이용하여 플랜트로부터 얻을 수 있는 데이터를 모델링 함으로써 공정의 상태를 미리 예측이 가능하도록 구성하였다. 이러한 예측된 모델을 이용하여 제어 전략을 설계함으로써 최적화된 제어 파라미터를 찾을 수 있으며, 특히 비정상적으로 들어오는 데이터에 대해서는 플랜트의 공정상태를 진단할 수 있는 가능성을 보이고 있다. 또한 기존의 통합 관리 시스템을 관리자라면 누구나 장소에 구애받지 않고 플랜트의 공정상태를 모니터링 할 수 있도록 웹을 이용하여 개발함으로써 플랜트의 능률성과 효율성 향상 및 인건비 절약 등 플랜트의 효율을 개선시킬 수 있다 즉, 네트워크가 구축되어 있는 곳이면 어디서나 모니터링이 가능하도록 시스템을 설계하였다.

• Proceedings of the IEEK Conference
• /
• 2009.05a
• /
• pp.381-383
• /
• 2009

본 논문에서는 대표적인 시스템 모델링 도구중의 하나인 RBF 뉴럴 네트워크(Radial Basis Function Neural Network)를 설계하고 모델을 최적화하기 위하여 최적화 알고리즘인 PSO(Particle Swarm Optimization) 알고리즘을 이용하였다. 즉, 모델의 최적화에 주요한 영향을 미치는 모델의 파라미터들을 PSO 알고리즘을 이용하여 동정한다. 제안된 RBF 뉴럴 네트워크는 은닉층에서의 활성함수로서 일반적으로 많이 사용되어지는 가우시안 커널함수를 사용한다. 더 나아가 모델의 최적화를 위하여 각 커널함수의 중심값은 HCM 클러스터링에 기반을 두어 중심값을 결정하고, PSO 알고리즘을 통하여 가우시안 커널함수의 분포상수, 은닉층에서의 노드 수 그리고 다수의 입력을 가질 경우 입력의 종류를 동정한다. 제안한 모델의 성능을 평가하기 위해 Mackey-Glass 시계열 공정 데이터를 적용하였으며 제안된 모델의 근사화와 일반화 능력을 분석한다.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.8
• /
• pp.39-47
• /
• 1999

This paper describes on-line retrainable RBF network in order to control the coiling temperature for a hot coil strip at Pohang Iron & Steel Company(POSCO). The proposed neural network can be used for improving conventional rule-based lookup table, which generates a heat transmission coefficient. To cope with time-varying characteristics of hot coil process, additional synaptic weights for on-line retraining purposes are introduced to hidden-to-output weights of conventional RBF network. Those weights are locally adjusted to newly incoming test data while preserving old information trained with off-line past data. Hence the effect of catastrophic interference can be greatly alleviated with the proposed network. In addition, rejection scheme is introduced for reliability concerns. From the experimental results applied to the actual process, it is noticed that overall control performance represents about 2.2% increase compared to the conventional one.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.7
• /
• pp.846-851
• /
• 2005

In this paper, we propose Genetic Algorithms(GAs)-based Optimized Polynomial Neural Networks(PNN). The proposed algorithm is based on Group Method of Data Handling(GMDH) method and its structure is similar to feedforward Neural Networks. But the structure of PNN is not fixed like in conventional neural networks and can be generated in a dynamic manner. As each node of PNN structure, we use several types of high-order polynomial such as linear, quadratic and modified quadratic, and it is connected as various kinds of multi-variable inputs. The conventional PNN depends on the experience of a designer that select the number of input variables, input variable and polynomial type. Therefore it is very difficult to organize optimized network. The proposed algorithm leads to identify and select the number of input variables, input variable and polynomial type by using Genetic Algorithms(GAs). The aggregate performance index with weighting factor is proposed as well. The study is illustrated with tile NOx omission process data of gas turbine power plant for application to nonlinear process. In the sequel the proposed model shows not only superb predictability but also high accuracy in comparison to the existing intelligent models.

• Journal of the Korea Computer Industry Society
• /
• v.2 no.9
• /
• pp.1163-1174
• /
• 2001

A general goal of the ATM(Asynchronous Transfer Mode) network is to support connections across various networks. ABR service using EPRCA(Enhanced Proportional Rate Control Algorithm) switch controls traffics in ATM network. EPRCA switch, traffic control method uses variation of the ACR(Allowed Cell Rate) to enhance the utilization of the link bandwidth. However, in ABR(Available Bit Rate) service, different treatments are offered according to different RTTs(Round Trip Times) of connections. To improve the above unfairness, this paper presents ABR DELAY mechanism, in which three reference parameters for cell delay are defined, and reflect on the messages of RM(Resource Management) cells. To evaluate our mechanism, we compare the fairness among TCP connections between ABR DELAY mechanism and ABR RRM mechanism. And also we execute simulations on a simple ATM network model where six TCP connections and a background traffic with different RTTs share the bandwidth of a bottleneck link. The simulation results, based on TCP goodput and efficiency, clearly show that ABR DELAY mechanism improves the fairness among TCP connections.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.12 no.1
• /
• pp.62-67
• /
• 2011

Grinding process is executed in the final machining stage to meet the quality requirements. In generally the ground surface of workpiece is affected by dressing condition as well as grinding condition. In order to estimate the roughness of workpiece, the several roughness models have been researched. These models defined the specific parameters and considered the several parameters which affect to roughness as multiply relationship among them. However, the multiply relationship among parameters is not enough to show the complicated grinding mechanism. Therefore, the neural network algorithm is used in this paper to predict the ground roughness for the plunge grinding. The proposed structure is composed of the initial roughness as well as final roughness model. The input parameters of proposed neural network are referred with the existing roughness model's. The performance of the proposed model is verified through experiments.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1994.10a
• /
• pp.743-747
• /
• 1994

A process planning system that generates alternative process plans offers multiple process plans for a part, thereby provides the flexibility to cope with the changes in shop floor status. In this paper, we introduce the concept of process net as a model for the generation of alternative process plans. We also show the usefulness of process net model by implementing the developed system to generate alternative process plans for rotational parts.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04a
• /
• pp.325-327
• /
• 2004

P2P 서비스는 서버를 거치지 않고 네트워크에 연결된 컴퓨터간의 직접적인 교환에 의해 자원과 서비스를 공유할 수 있는 방식으로 기존의 클라이언트/서버 방식과 달리 관련 프로그램만으로 자원에 대한 공유 및 교환이 가능하다. 이러한 P2P의 특징을 이용하여 서버가 중재자 역할을 하는 전자상거래 방식에 P2P 기술을 적용하여 구매자와 판매자의 직접적인 통신에 의해 상거래가 가능한 새로운 모델을 제안하고자 한다. 제안 모델은 구매자와 판매자간의 디지털 컨텐츠 교환과 지불에 대해 공정성을 보장해 주고, 컨텐츠를 사용한 구매자들로부터 받은 평판(reputation)값을 통하여 판매자가 제공하는 컨텐츠에 대한 신뢰성 또한 보장할 수 있다.

• Journal of the Korea Society for Simulation
• /
• v.28 no.2
• /
• pp.107-117
• /
• 2019

Because shipbuilding is single-product production with limited resources, production management technology is essential to manage the resources effectively and maximize the productivity of ship-process. Therefore, many shipbuilding companies are conducting research on ship production plan and process considering various constraints in the field by applying modeling and simulation. However, it is difficult to provide accurate production plan on sudden schedule and process changes, and to understand the interconnectivity between the processes that produce blocks in existing research. In addition, there are many differences between the production planning and field planning because detailed processes and quantity of blocks can not be considered. In this research, we propose the integrated process-based modeling method considering process-operation sequences, BOM(Bill of materials) and resource constraints of all the scheduled blocks in the indoor system. Through the integrated process-based model, it is easy for the user to grasp the assembly relationship, workspace and preliminary relationship of assembly process between the blocks in indoor system. Also, it is possible to obtain the overall production plan that maximizes resource efficiency without the separate simulation and resource modeling procedures because resource balancing that considers the amount of resource quantity shared in the indoor system is carried out.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.11a
• /
• pp.121-124
• /
• 2008

2000 년대 들어 방송통신융합의 흐름에 따라 등장한 IT 기반의 Non Linear 제작공정은 2005 년까지 IMX-50, DV25/50 포맷 등 50Mbps 급 이하의 동영상 Data 를 중심으로 영상 콘텐츠 제작을 위해 부분적인 프로그램 장르에 국한하여 구축, 운용되어 왔다. 최근 초고속 네트워크를 통해 대용량의 고해상도 영상데이터를 안정적으로 수용 처리하는, 상대적으로 저렴하면서 효과적인 기능을 보유한 컴퓨터 기술 기반 단위 Application 들이 등장함에 따라 고해상도 프로그램 제작을 지향하는 Contents 생산기지들을 중심으로 IT 기반 제작공정으로의 전환과 차세대 제작시스템으로 HD 급 NPS 구축 필요성이 대두되었다. 본 논문에서는 IT 기반의 방송 시스템 구축의 단초로서 최초로 HD 급 대용량 구축모델의 프로토 타입을 설계하고 발전 로드맵을 분석, 최적화를 위한 방안을 제시한다.