• Journal of The Korean Society of Agricultural Engineers
• /
• v.63 no.1
• /
• pp.103-116
• /
• 2021

Analysis of runoff is substantial for effective water management in the watershed. Runoff occurs by reaction of a watershed to the rainfall and has non-linearity and uncertainty due to the complex relation of weather and watershed factors. ANN (Artificial Neural Network), which learns from the data, is one of the machine learning technique known as a proper model to interpret non-linear data. The performance of ANN is affected by the ANN's structure, the number of hidden layer nodes, learning rate, and activation function. Especially, the activation function has a role to deliver the information entered and decides the way of making output. Therefore, It is important to apply appropriate activation functions according to the problem to solve. In this paper, ANN models were constructed to estimate runoff with different activation functions and each model was compared and evaluated. Sigmoid, Hyperbolic tangent, ReLU (Rectified Linear Unit), ELU (Exponential Linear Unit) functions were applied to the hidden layer, and Identity, ReLU, Softplus functions applied to the output layer. The statistical parameters including coefficient of determination, NSE (Nash and Sutcliffe Efficiency), NSEln (modified NSE), and PBIAS (Percent BIAS) were utilized to evaluate the ANN models. From the result, applications of Hyperbolic tangent function and ELU function to the hidden layer and Identity function to the output layer show competent performance rather than other functions which demonstrated the function selection in the ANN structure can affect the performance of ANN.

• International Journal of Contents
• /
• v.9 no.4
• /
• pp.11-15
• /
• 2013

"Protein Folding Problem" is considered to be one of the "Great Challenges of Computer Science" and prediction of disordered protein is an important part of the protein folding problem. Machine learning models can predict the disordered structure of protein based on its characteristic of "learning from examples". Among many machine learning models, we investigate the possibility of multilayer perceptron (MLP) as the predictor of protein disorder. The investigation includes a single hidden layer MLP, multi hidden layer MLP and the hierarchical structure of MLP. Also, the target node cost function which deals with imbalanced data is used as training criteria of MLPs. Based on the investigation results, we insist that MLP should have deep architectures for performance improvement of protein disorder prediction.

• Journal of the Korean Chemical Society
• /
• v.8 no.4
• /
• pp.179-182
• /
• 1964

The partition function of liquid ammonia was developed by applying the modified significant structure theory of liquid.$^5$ The molar volume, vapor pressure, vaporization entropy were calculated over a wide temperature range. The critical properties for the liquid were also calculated. And surface tensions of the liquid were calculated in a similar manner developed for water by Chang and Pak.$^{10}$ The results show good agreement with experimental observations.

• Proceedings of the Korea Committee for Ocean Resources and Engineering Conference
• /
• 2002.10a
• /
• pp.277-282
• /
• 2002

This study was performed to find out basic function of artificial upwelling structures. Generation of artificial upwelling current was affected by size of structures, incident current and porosity. when stratification parameters was about 3.0, relative height(hs/h) of structures was $0.125{\sim}0.15$, stable artificial upwelling current was generated in the back-side of structures. when porosity is lower than 50%, the effect of artificial upwelling structure was to be better than little by little.

• Journal of Fisheries and Marine Sciences Education
• /
• v.29 no.2
• /
• pp.436-446
• /
• 2017

This study was performed to verify the validity and utility of the Korean Behavior Rating Inventory of Executive Function-Preschool Version(BRIEF-P). The BRIEF-P is one of the most used instruments to measure Executive Function(EF) and a teacher-rating assessment method designed to measure the executive function of preschoolers(age 2-5). A total of 200 preschoolers(104 boys, 96 girls; recruited aged 3 to 5years) participated. Confirmatory factor analysis was carried out to evaluate the model structure. Correlation analysis was done to verify the concurrent validity. In the results, the model fit indexes were good with the five BRIEF-P subscales (that is, inhibit, shift, emotional control, working memory, plan/organize). The internal consistency of the BRIEF-P five subscales were supported. As a results, the Korean Behavior Rating Inventory of Executive Function-Preschool Version(BRIEF-P) was confirmed to be a reliable and valid assessment tool to measure executive function in preschoolers.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.20 no.12
• /
• pp.3782-3791
• /
• 1996

A great deal of effert has been invested in upgrading the performance and the efficiency of mechanical structures. Using experimental modal analysis(EMA) or finite element analysis(FEA) data of mechanical structures, this performance and efficiency can be effectively evaluated. In order to analyze complex structures such as automobiles and aircraft, for the sake of computing efficiency, the dynamic substructuring techniques that allow to predict the dynamic behavior of a structure based on that of the composing structures, are widely used. By llinking a modal model obtained from EMA and an analytical model obtained from FEA, the best conditioned structures can be desinged. In this paper, a new algorithm for structural dynamic modification-SRFSM (substructure response function sensitivity method) is proposed by linking frequency responce function synthesis and response function sensitivity. A mehtod to obtain response function sensitivity using direct derivative of mechanical impedance, is also used.

• Proceedings of the KSR Conference
• /
• 2004.10a
• /
• pp.157-161
• /
• 2004

Urban transit is system that offer target Performance and function because various lower part system such as vehicles, signaling, power supply, rail track is consisted as complex. It is important first of all that describe correctly system requirement until configuration of system, design, manufacture, installation and test verification so that the complicated urban transit system is to act safety and reliable and to display target Performance and function properly. The system requirement assist the system developer to comprehensive the performance and function of system in basic design. Therefore, In this paper wishes to study way to analyze system of system requirement and apply to Urban transit.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.212-212
• /
• 2015

So far, geomorphologic dispersion due to the heterogeneity characteristics of flow paths in a basin has been demonstrated as a major factor affecting to the hydrologic response function of a catchment. This effect has considered by many previous studies taking into account flow path length factors, especially in the application of width function. Based upon the analysis of topographic index, another important geomorphologic factor extracted from DEM data, this work presents a new factor named saturation to evaluate its effects to the formation of the well-known instantaneous unit hydrograph (IUH) in Nash model and drainage structure in a river basin. First, the geomorphologic parameters corresponding to different saturation conditions are computed from DEM data with the support of GIS software. Then, in the combination of hydrologic and geomorphologic data, effective rainfall in each saturation degree and the Nash parameters are calculated using excel. Finally, the verification process with direct runoff data is conducted using Fortran programming. This process is applied to five sub-watersheds in Bocheong catchment ($485.21km^2$) in Korea where the necessary data are available and believable. The results from this approach will improve researchers and students'understandings about the relationship between rainfall and runoff and its relation with drainage structure within a catchment.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.57 no.1
• /
• pp.20-24
• /
• 2008

A bid-based pool(BBP) model is representative of energy market structure in a number of restructured electricity markets. Supply function equilibrium(SFE) models of interaction better match what is explicitly required in the bid formats of typical BBP markets. Many of the results in the SFE literature involve restrictive parametrization of the bid cost functions. In the SFE models, two parameters, intercept and slope, are available for strategic bidding. This paper addresses the realistic competition format that players can choose both parameters arbitrarily. In a fixed demand function, equilibrium conditions for generation company's profit maximization have a degree of freedom, which induces multi-equilibrium. So it is hard to choose a convergent equilibrium. However, consideration of stochastic demand function makes the equilibrium conditions independent each other based on the amount of variance of stochastic demand function. This variance provides the bidding players with incentives to change the slope parameter from an equilibrium for a fixed demand function until the slope parameter equilibrium.

• Journal of Advanced Navigation Technology
• /
• v.11 no.1
• /
• pp.93-98
• /
• 2007

The importance of information security is increasing by the rapid development of the communication network. So, cryptosystems are used to solve these problems and securities of cryptosystems are dependent on keys. In this paper, we propose a key generation method which is based on cryptographically secure MD5 hash function. The basic structure of the MD5 hash function features is a repetitive structure which is processed in a block unit of 512 bits from inputs of limited length and generates a fixed output of 128 bits. The security of proposed method is based on the hash function and the proposed method can be also utilized for authentication algorithm or data encryption algorithm.