• Transactions on Control, Automation and Systems Engineering
• /
• v.3 no.1
• /
• pp.43-50
• /
• 2001

In this paper, we propose the design procedure of advance Polynomial Neural Networks(PNN) architecture for optimal model identification of complex and nonlinear system. The proposed PNN architecture is presented as the generic and advanced type. The essence of the design procedure dwells on the Group Method of Data Handling(GMDH). PNN is a flexible neural architecture whose structure is developed through learning. In particular, the number of layers of the PNN is not fixed in advance but is generated in a dynamic way. In this sense, PNN is a self-organizing network. With the aid of three representative numerical examples, compari-sons show that the proposed advanced PNN algorithm can produce the model with higher accuracy than previous other works. And performance index related to approximation and generalization capabilities of model is evaluated and also discussed.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.60 no.5
• /
• pp.913-920
• /
• 2011

Distribution systems for operating the various systems should be operated organically. Distribution systems for organic operations extend the functionality of the system and system integration for the extension is doing. However, in systems integration for each unique non-standard definition of linkage and exchange systems are used. Non-standard definition of a unique linkage and exchange systems in the maintenance and expansion will result in many negative results. To solve these problems, the system and the exchange of information about the links between systems should define standards. IEC (International Electrotechnical Commission) power to solve these problems in the field of the Common Information Model (Common Information Model, CIM) standard (IEC61970, IEC61968) is defined as the transmission and distribution system integration and standardization for the exchange of information. In this paper, standardized system for building CIM-based distribution automation system should describe the process of building. In addition, the process of establishing a common information model needed (CIM) defined and associated services for a defined set of models is proposed in the standardization process.

• Journal of Korean Institute of Industrial Engineers
• /
• v.39 no.5
• /
• pp.403-411
• /
• 2013

The quality of the products/processes has been improved remarkably since robust design (RD) methodology is applied into the practice manufacturing processes. A model building method based on the dual responses methods for multiple and time oriented responses on a drug development process is employed in this paper instead of the previous methods that handle the static nature of data and single response. Subsequently, the optimal solutions of a multiple and time series RD problem are obtained by using the proposed augmented weighted Tchebycheff method that has a significant flexibility on assigning weights. Finally, a pharmaceutical case study associated with a generic drug development process is conducted in order to illustrate the efficient optimal solutions from the proposed model.

• Proceedings of the KIEE Conference
• /
• 2000.11d
• /
• pp.806-808
• /
• 2000

We propose a new Dynamic Multilayer Perceptorn(DMP) architecture for optimal model identification of complex and nonlinear system in this paper. The proposed DMP scheme is presented as the generic and advanced type based on the GMDH(Group Method of Data Handling) method for the limitation of GMDH under only two system input variables. It is worth stressing that the number of the layers and the nodes in each layer of the DMP are not predetermined, unlike in the case of the popular multilayer perceptron structure, but these are generated in a dynamic manner. The experimental part of the study comes with representative nonlinear static system. Comparative analysis is included and shows that a new DMP can produce the model with higher accuracy than previous other works.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2002.05a
• /
• pp.134-138
• /
• 2002

The human faces do not have distinct features unlike other general objects. In general the features of eyes, nose and mouth which are first recognized when human being see the face are defined. These features have different characteristics depending on different human face. In this paper, We propose a face recognition algorithm using the hidden Markov model(HMM). In the preprocessing stage, we find edges of a face using the locally adaptive threshold scheme and extract features based on generic knowledge of a face, then construct a database with extracted features. In training stage, we generate HMM parameters for each person by using the forward-backward algorithm. In the recognition stage, we apply probability values calculated by the HMM to input data. Then the input face is recognized by the euclidean distance of face feature vector and the cross-correlation between the input image and the database image. Computer simulation shows that the proposed HMM algorithm gives higher recognition rate compared with conventional face recognition algorithms.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.9
• /
• pp.97-106
• /
• 1996

In this paper, we propose a face recognition algorithm using the hidden markov model and neural networks (HMM-NN). In the preprocessing stage, we find edges of a face using the locally adaptive threshold (LAT) scheme and extract features based on generic knowledge of a face, then construct a database with extracted features. In the training stage, we generate HMM parameters for each person by using the forward-backward algorithm. In the recognition stage, we apply probability vlaues calculated by the HMM to subsequent neural networks (NN) as input data. Computer simulation shows that the proposed HMM-NN algorithm gives higher recognition rate compared with conventional face recognition algorithms.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.16 no.4
• /
• pp.241-257
• /
• 2004

Global prognostic models based on NCOM(NCAR CSM Ocean Model) of NCAR which is generic from Bryan-Cox-Semtner model are established to study the ocean circulation in the neighboring seas of Korean peninsula. The model domain covers areas from $80.6{^\circ}S~88.6{^\circ}N$in meridional direction and the vertical water column is divided into 15 levels taking enhanced grid resolution of $0.3^\circ$ around Korean peninsula. Island option is used for 22 islands to simulate inshore circulation by hole-relaxation method and the restart hydrographic data are taken from NCAR(1998) CSM model that has been run for 300 years. The wind stress data are taken from Choi et al. (2002). Based on the model results, circulation patterns in the NW Pacific and global oceans are investigated. Volume transports calculated at five straits in the neighboring seas of Korean peninsula are compared with the results from Choi et al. (2002) and other observed data.

• Nuclear Engineering and Technology
• /
• v.50 no.1
• /
• pp.88-96
• /
• 2018

Concerns over reliability assessments of the main components in nuclear power plants (NPPs) related to aging and continuous operation have increased. The conventional reliability assessment for main components uses experimental correlations under general conditions. Most NPPs have been operating in Korea for a long time, and it is predictable that NPPs operating for the same number of years would show varying extent of aging and degradation. The conventional reliability assessment does not adequately reflect the characteristics of an individual plant. Therefore, the reliability of individual components and an individual plant was estimated according to operating data and conditions. It is essential to reflect aging as a characteristic of individual NPPs, and this is performed through prognostics. To handle this difficulty, in this paper, the general path model/Bayes, a data-based prognostic method, was used to update the reliability estimated from the generic database. As a case study, the authors consider the aging for steam generator tubes in NPPs and demonstrate the suggested methodology with data obtained from the probabilistic algorithm for the steam generator tube assessment program.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.6
• /
• pp.1431-1440
• /
• 1998

Video data are retrieved and stored in various compressed forms according to their characteristics, In this paper, we present a generic data model that captures the structure of a video document and that provides a means for indexing a video stream, Using this model, we design and implement CVIMS (the MPEG-2 Compressed Video Information Management System) to store and retrieve video documents, CVIMS extracts I-frames from MPEG-2 files, selects key-frames from the I -frames, and stores in database the index information such as thumbnails, captions, and picture descriptors of the key-frames, And also, CVIMS retrieves MPEG- 2 video data using the thumbnails of key-frames and v31ious labels of queries.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.7 no.2
• /
• pp.184-195
• /
• 1995

A theoretical one-dimensional model for the charging process in stratified thermal storage tanks is established presuming that the fluid ensuing from the tank inlet creates a perfectly mixed, layer above the thermocline. Both the generic and asymptotic closed-form solutions are obtained via the Laplace transformation. The asymptotic solution describes the nature of the charging pertaining to the case of no thermal diffusion, whereas the generic solution is of practical importance to understand the role of operating parameters on the stratification. The present model is validated through comparison with available experimental data, where they agree well with each other within a reasonable limit. An interpretation of the exact solution entails two important features associated with the charging process. The first is that an in-crease in the mixing depth $h_m$ causes a relatively slow temperature rise in the perfectly mixed region, but on the other hand it results in a faster decay of the overall temperature gradient across the thermocline. Next is the predominance of the mixing depth in the presence of the prefectly mixed region. In such a case the effect of the Peclet number is marginal and there-fore the thermal characteristics are solely dependent on the mixing depth paticularly for large $h_m$. The Peclet number affects significantly only for the case without mixing. Variation of the storage efficiency in response to the change in the mass flow rate agrees favorably with the published experimental results, which confirms the utility of the present study.