• Steel and Composite Structures
• /
• v.8 no.5
• /
• pp.403-421
• /
• 2008

The design provisions for semi-rigid steel frames have been incorporated in codes of practice for steel structures. In order to do the same, it is necessary to know the experimental moment-relative rotation (M-${\theta}_r$) behaviour of beam-to-column connections. In spite of numerous publications and collection of several connection databases, there is no unified approach for the semi-rigid design of steel frames. Amongst the many connection models available, the Frye-Morris polynomial model, with its limitations reported in the literature, is simple to adopt at least for the linear design space. However this model requires more number of connection tests and regression analyses to make it a realistic prediction model. In this paper, 3D nonlinear finite element (FE) analysis of beam-column connection specimens, carried out using ABAQUS software, for evaluating the M-${\theta}_r$ behaviour of semi-rigid top and seat-angle (TSA) bolted connections are described. The finite element model is validated against experimental behaviour of the same connection with regard to their moment-rotation behaviour, stress distribution and mode of failure of the connections. The calibrated FE model is used to evaluate the performance of the Frye-Morris polynomial model. The results of the numerical parametric studies carried out using the validated FE model have been used in proposing modifications to the Frye-Morris model for TSA connection in terms of the powers of the size parameters.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.12
• /
• pp.1772-1781
• /
• 2017

In this study, the design methodology for alleviating the overfitting problem of Polynomial Neural Networks(PNN) is realized with the aid of two kinds techniques such as L2 regularization and Sum of Squared Coefficients (SSC). The PNN is widely used as a kind of mathematical modeling methods such as the identification of linear system by input/output data and the regression analysis modeling method for prediction problem. PNN is an algorithm that obtains preferred network structure by generating consecutive layers as well as nodes by using a multivariate polynomial subexpression. It has much fewer nodes and more flexible adaptability than existing neural network algorithms. However, such algorithms lead to overfitting problems due to noise sensitivity as well as excessive trainning while generation of successive network layers. To alleviate such overfitting problem and also effectively design its ensuing deep network structure, two techniques are introduced. That is we use the two techniques of both SSC(Sum of Squared Coefficients) and $L_2$ regularization for consecutive generation of each layer's nodes as well as each layer in order to construct the deep PNN structure. The technique of $L_2$ regularization is used for the minimum coefficient estimation by adding penalty term to cost function. $L_2$ regularization is a kind of representative methods of reducing the influence of noise by flattening the solution space and also lessening coefficient size. The technique for the SSC is implemented for the minimization of Sum of Squared Coefficients of polynomial instead of using the square of errors. In the sequel, the overfitting problem of the deep PNN structure is stabilized by the proposed method. This study leads to the possibility of deep network structure design as well as big data processing and also the superiority of the network performance through experiments is shown.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.27 no.6
• /
• pp.822-831
• /
• 2021

Although programs have been developed to evaluate the risk of dragging anchors, it is practically difficult for VTS(vessel traffic service) operators to calculate and evaluate these risks by obtaining input factors from anchored ships. Therefore, in this study, the gross tonnage (GT) that could be easily obtained from the ship by the VTS operators was set as an independent variable, and linear and nonlinear regression analyses were performed using the input factors as the dependent variables. From comparing the fit of the polynomial model (linear) and power series model (nonlinear), the power series model was evaluated to be more suitable for all input factors in the case of container ships and bulk carriers. However, in the case of tanker ships, the power supply model was suitable for the LBP(length between perpendiculars), width, and draft, and the polynomial model was evaluated to be more suitable for the front wind pressure area, weight of the anchor, equipment number, and height of the hawse pipe from the bottom of the ship. In addition, all other dependent variables, except for the front wind pressure area factor of the tanker ship, showed high degrees of fit with a coefficient of determination (R-squared value) of 0.7 or more. Therefore, among the input factors of the dragging anchor risk assessment program, all factors except the external force, seabed quality, water depth, and amount of anchor chain let out are automatically applied by the regression analysis model formula when only the GT of the ship is provided.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1231-1234
• /
• 2003

A sensor working in outdoor may generate some faulty signal owing to dust and high temperature. This paper describes an intelligent sensor system and controller which has a reconstruction mechanism for faulty signal. The faulty signals are dievided into two types as linear distortion and non linear distortion, respectively. The linear distorted signal is due to dust, and non linear distorted signal is due to physical breakdown of sensor or high temperature. These distorted signal have been reconstructed by the proposed method based on polynomial regression method and principal component analysis approach.. The proposed method has been applied to sun tracking system working in outdoor. For a robust and precision control of sun tracker, a fuzzy controller was also proposed. The fuzzy controller controls the tracker by using the collected sensor signal. The tolerance of the position control is within 1.5 degree. To show the validity of the developed system, some experiments in the field were illustrated.

• Journal of the Korean Society of Safety
• /
• v.33 no.4
• /
• pp.21-29
• /
• 2018

In many companies handling flammable liquids, explosion-proof electrical equipment have been installed according to the Korean Industrial Standards (KS C IEC 60079-10-1). In these standards, hazardous area for explosive gas atmospheres has to be classified by the evaluation of the evaporation rate of flammable liquid leakage. The evaporation rate is an important factor to determine the zones classification and hazardous area distance. However, there is no systematic method or rule for the estimation of evaporation rate in these standards and the first principle equations of a evaporation rate are very difficult. Thus, it is really hard for industrial workplaces to employ these equations. Thus, this problem can trigger inaccurate results for evaluating evaporation range. In this study, empirical models for estimating an evaporation rate of flammable liquid have been developed to tackle this problem. Throughout the sensitivity analysis of the first principle equations, it can be found that main factors for the evaporation rate are wind speed and temperature and empirical models have to be nonlinear. Polynomial regression is employed to build empirical models. Methanol, benzene, para-xylene and toluene are selected as case studies to verify the accuracy of empirical models.

• Journal of Welding and Joining
• /
• v.14 no.4
• /
• pp.33-42
• /
• 1996

In recent years there has been a significant growth in the use of the automated and/or robotic welding system, carried out as a means of improving productivity and quality, reducing product costs and removing the operator from tedious and potentially hazardous environments. One of the major difficulties with the automated and/or robotic welding process is the inherent lack of mathematical models for determination of suitable welding process parameters. Partial-penetration, single-pass bead-on-plate welds were fabricated in 12mm AS 1204 mild steel flats employing five different welding process parameters. The experimental results were used to develop three empirical equations: curvilinear; polynomial; and linear equations. The results were also employed to find the best mathematical equation under weld bend width to assist in the process control algorithms for the Gas Metal Arc Welding(GMAW) process and to correlate welding process parameters with weld bead width of bead-on-plates deposited. With the help of a standard statistical package program. SAS, multipe regression analysis was undertaken for investigating and modeling the GMAW process, and significance test techniques were applied for the interpretation of the experimental data.

• Textile Coloration and Finishing
• /
• v.21 no.5
• /
• pp.41-50
• /
• 2009

Design of experiments (DoE) concept was successfully applied to determine the optimum processing conditions that yield maximum % exhaustion for berberine interaction with synthetic tanning agent pretreated polyamide substrates. The potential of synthetic tanning agent to provide anionic sites on the polyamide for berberine interaction which is cationic in nature was tested to increase the % exhaustion of berberine in this article. Experiments were designed according to Central Composite Rotatable Design (CCRD). The three factors for synthetic tanning agent pretreatment and two factors for berberine interaction each at five different levels, including central and axial points were considered. Experiments were conducted in a laboratory scale infra-red treatment instrument according to CCRD. For each response, second order polynomial models were developed using multiple linear regression analysis incorporating linear, interactions and squared effects of all variables and then optimized. The significance of the mathematical model developed was ascertained using Excel regression (solver) analysis module. Analysis of variance (ANOVA) was performed to check the adequacy and accuracy of the fitted models. The response surfaces and contour maps showing the interaction of process variables were constructed. Applying Monte Carlo simulation, response surface and contour plots, optimum operating conditions were found and at this optimum point, % exhaustion of 81% and 74% respectively for synthetic tanning agent pretreatment and berberine interaction were observed and subsequently the results were experimentally investigated.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.5
• /
• pp.1021-1026
• /
• 2017

Mixture experiments are commonly encountered in many fields including food, chemical and pharmaceutical industries. In mixture experiments, measured response depends on the proportions of the components present in the mixture and not on the amount of the mixture. Statistical analysis of the data from mixture experiments has mainly focused on a continuous response variable. In the example of quantal response data in mixture experiments, however, the tumor incidence data have been analyzed in Chen et al. (1996) to study the effects of 3 dietary components on the expression of mammary gland tumor. In this paper, we compared the logistic regression models with linear predictors such as second degree Scheffe polynomial model, Becker model and Akay model in terms of classification accuracy.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.4
• /
• pp.692-700
• /
• 2017

In this paper, we propose the structural design of Interval Type-2 FCM based RBFNN. Proposed model consists of three modules such as condition, conclusion and inference parts. In the condition part, Interval Type-2 FCM clustering which is extended from FCM clustering is used. In the conclusion part, the parameter coefficients of the consequence part are estimated through LSE(Least Square Estimation) and WLSE(Weighted Least Square Estimation). In the inference part, final model outputs are acquired by fuzzy inference method from linear combination of both polynomial and activation level obtained through Interval Type-2 FCM and acquired activation level through Interval Type-2 FCM. Additionally, The several parameters for the proposed model are identified by using differential evolution. Final model outputs obtained through benchmark data are shown and also compared with other already studied models' performance. The proposed algorithm is performed by using Iris and Vehicle data for pattern classification. For the validation of regression problem modeling performance, modeling experiments are carried out by using MPG and Boston Housing data.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.20 no.3
• /
• pp.181-194
• /
• 2017

This study aims to develop a regression model for forest volume estimation using field-collected forest inventory information and airborne LiDAR data. The response variable of the model is forest stem volume, was measured by random sampling from each individual plot of the 30 circular sample plots collected in Bonghwa-gun, Gyeong sangbuk-do, while the predictor variables for the model are Height Percentiles(HP) and Height Bin(HB), which are metrics extracted from raw LiDAR data. In order to find the most appropriate model, the candidate models are constructed from simple linear regression, quadratic polynomial regression and multiple regression analysis and the cross-validation tests were conducted for verification purposes. As a result, $R^2$ of the multiple regression models of $HB_{5-10}$, $HB_{15-20}$, $HB_{20-25}$, and $HBgt_{25}$ among the estimated models was the highest at 0.509, and the PRESS statistic of the simple linear regression model of $HP_{25}$ was the lowest at 122.352. $HB_{5-10}$, $HB_{15-20}$, $HB_{20-25}$, and $HBgt_{25}-based$ models, thus, are comparatively considered more appropriate for Korean forests with complicated vertical structures.