• Proceedings of the Korea Water Resources Association Conference
• /
• 2020.06a
• /
• pp.170-170
• /
• 2020

Due to the development of technologies, complex computation of huge data set is possible with a prevalent personal computer. Therefore, machine learning methods have been widely applied in the hydrologic field such as regression-based regional frequency analysis (RFA). The main purpose of this study is to compare two frameworks of RFA based on the artificial neural network (ANN) models: quantile regression technique (QRT-ANN) and parameter regression technique (PRT-ANN). As an output layer of the ANN model, the QRT-ANN predicts quantiles for various return periods whereas the PRT-ANN provides prediction of three parameters for the generalized extreme value distribution. Rainfall gauging sites where record length is more than 20 years were selected and their annual maximum rainfalls and various hydro-meteorological variables were used as an input layer of the ANN model. While employing the ANN model, 70% and 30% of gauging sites were used as training set and testing set, respectively. For each technique, ANN model structure such as number of hidden layers and nodes was determined by a leave-one-out validation with calculating root mean square error (RMSE). To assess the performances of two frameworks, RMSEs of quantile predicted by the QRT-ANN are compared to those of the PRT-ANN.

• Structural Engineering and Mechanics
• /
• v.90 no.1
• /
• pp.97-106
• /
• 2024

This paper presents a significant contribution to aided design by conducting an analytical examination of geometric links with the aim of establishing criteria for assessing an analogy measure of the extrinsic geometric and kinematic characteristics of the Variable Compression Ratio (VCR) engine with a Geared Mechanism (GBCM) in comparison to the existing Fixed Compression Ratio (FCR) engine with a Standard-Reciprocating Crankshaft configuration. Employing a mechanical approach grounded in projective computational methods, a parametric study has been conducted to analyze the kinematic behavior and geometric transformations of the moving links. The findings indicate that in order to ensure equivalent extrinsic behavior and maintain consistent input-output performance between both engine types, precise adjustments of intrinsic geometric parameters are necessary. Specifically, for a VCR configuration compared to an FCR configuration, regardless of compression ratio and gearwheel radius, for the same crankshaft ratios and stroke lengths, it is imperative to halve lengths of connecting rods, and crank radius. These insights underscore the importance of meticulous parameter adjustment in achieving comparable performance across different engine configurations, offering valuable implications for design optimization.

• The Transactions of the Korean Institute of Electrical Engineers B
• /
• v.53 no.3
• /
• pp.188-195
• /
• 2004

This paper presents neural load torque observer that is used to deadbeat load torque observer and gain compensation by parameter estimator As a result, the response of the PMSM(permanent magnet synchronous motor) follows that nominal plant. The load torque compensation method is composed of a neural deadbeat observer To reduce the noise effect, the post-filter implemented by MA(moving average) process, is adopted. The parameter compensator with RLSM (recursive least square method) parameter estimator is adopted to increase the performance of the load torque observer and main controller The parameter estimator is combined with a high performance neural load torque observer to resolve the problems. The neural network is trained in on-line phases and it is composed by a feed forward recall and error back-propagation training. During the normal operation, the input-output response is sampled and the weighting value is trained multi-times by error back-propagation method at each sample period to accommodate the possible variations in the parameters or load torque. As a result, the proposed control system has a robust and precise system against the load torque and the Parameter variation. A stability and usefulness are verified by computer simulation and experiment.

• Journal of Communications and Networks
• /
• v.18 no.5
• /
• pp.688-698
• /
• 2016

In this paper, we investigate the achievable sum rate and energy efficiency of downlink massive multiple-input multiple-output antenna systems with zero-forcing precoding, by taking into account the randomness of user locations. Specifically, we propose two types of non-uniform user distributions, namely, center-intensive user distribution and edge-intensive user distribution. Based on these user distributions, we derive novel tight lower and upper bounds on the average sum rate. In addition, the impact of user distributions on the optimal number of users maximizing the sum rate is characterized. Moreover, by adopting a realistic power consumption model which accounts for the transmit power, circuit power and signal processing power, the energy efficiency of the system is studied. In particular, closed-form solutions for the key system parameters, such as the number of antennas and the optimal transmit signal-to-noise ratio maximizing the energy efficiency, are obtained. The findings of the paper suggest that user distribution has a significant impact on the system performance: for instance, the highest average sum rate is achieved with the center-intensive user distribution, while the lowest average sum rate is obtained with the edge-intensive user distribution. Also, more users can be served with the center-intensive user distribution.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.63 no.3
• /
• pp.376-382
• /
• 2014

E-Actuator is an essential part of an eVGT, it receives the command from the main ECU and controls the vane. An e-Actuator failure can cause an abrupt change in engine output and it may induce an accident. Therefore, it is required to detect anomalies in the e-Actuator in real time to prevent accidents. In this paper, an e-Actuator fault detection method using on-line parameter identification is proposed. To implement on-line fault detection algorithm, many constraints are considered. The test input and sampling rate are selected considering the constraints. And new recursive system identification algorithm is proposed which reduces the memory and MCU power dramatically. The relationship between the identified parameters and real elements such as gears, spring and motor are derived. The fault detection method using the relationship is proposed. The experiments with the real broken gears show the effectiveness of the proposed algorithm. It is expected that the real time fault detection is possible and it can improve the safety of eVGT system.

• Structural Engineering and Mechanics
• /
• v.28 no.5
• /
• pp.525-548
• /
• 2008

This article deals the theory for solving an inverse problem of plate structures using the frequency-domain information instead of classical time-domain delays or free vibration eigenmodes or eigenvalues. A reduced set of output parameters characterizing the defect is used as a regularization technique to drastically overcome noise problems that appear in imaging techniques. A deconvolution scheme from an undamaged specimen overrides uncertainties about the input signal and other coherent noises. This approach provides the advantage that it is not necessary to visually identify the portion of the signal that contains the information about the defect. The theoretical model for Quantitative nondestructive evaluation, the relationship between the real and ideal models, the finite element method (FEM) for the forward problem, and inverse procedure for detecting the defects are developed. The theoretical formulation is experimentally verified using dynamic responses of a steel plate under impact loading at several points. The signal synthesized by FEM, the residual, and its components are analyzed for different choices of time window. The noise effects are taken into account in the inversion strategy by designing a filter for the cost functional to be minimized. The technique is focused toward a exible and rapid inspection of large areas, by recovering the position of the defect by means of a single accelerometer, overriding experimental calibration, and using a reduced number of impact events.

• Journal of the Korea Safety Management & Science
• /
• v.3 no.3
• /
• pp.151-164
• /
• 2001

Currently, there are little evaluated model for each industry and systematic analysis for each item about current business web sites. And the approach way to improve it all depends on individual. This research is about two different points of view on I-Evaluation development as an approach method to provide analysis, evaluation, and guideline on business web sites. The one is about developing a working step and a site evaluation model that are necessary to improve the quality of site. The other is about a framework development to apply a feedback on site most rapidly and site optimization. The former is from the methodological point of view for I-Evaluation, and the latter is from the point of view for I-Evaluation Framework. In terms of methodology, developing site evaluation model and defining a working step belong to it. Site evaluation model means customizing each customers web site, using each evaluated scoring model which can be a standard for each industry to analyze a similar business web site. Defining a working step means defining input and output parameters for composed elements, working processes, and results analysis on an evaluated model. And also it includes building a working environments to automatic steps mentioned the above by clarifying them.

• Journal of Computational Design and Engineering
• /
• v.3 no.1
• /
• pp.1-13
• /
• 2016

The increase of consumer needs for quality metal cutting related products with more precise tolerances and better product surface roughness has driven the metal cutting industry to continuously improve quality control of metal cutting processes. In this paper, two different approaches are discussed. First, design of experiments (DOE) is used to determine the significant factors and then fuzzy logic approach is presented for the prediction of surface roughness. The data used for the training and checking the fuzzy logic performance is derived from the experiments conducted on a CNC milling machine. In order to obtain better surface roughness, the proper sets of cutting parameters are determined before the process takes place. The factors considered for DOE in the experiment were the depth of cut, feed rate per tooth, cutting speed, tool nose radius, the use of cutting fluid and the three components of the cutting force. Finally the significant factors were used as input factors for fuzzy logic mechanism and surface roughness is predicted with empirical formula developed. Test results show good agreement between the actual process output and the predicted surface roughness.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.11 no.7
• /
• pp.603-610
• /
• 2001

Industrial processes are normally operated by skilled humans who have the cumulative and logical information about the system. Fuzzy control has been investigated for many application. Intelligent control approaches based on fuzzy logic have a chance to include human thinking. This paper represents modeling approach based upon operators knowledge without mathematical model of the system and optimize the controller. The experimented system is constructed for sending a ball to the goal position using wind of two DC motors in the predefined path. A vision camera to mimic human eyes detects the ball position. The system used in this experiment could be hardly modeled by mathematical methods and ould not be easily controlled by conventional manners. The controller is designed based on the input-output data and experimental knowledge obtained by trials, and optimized under the predefined performance criterion. And this paper shows the data adaptation for changeable operating condition. When the system is driven in the abnormal condition with unconsidered noise, the new optimal operating parameters could be defined by adjusting membership functions. Thus, this technique could be applied in industrial fields.

• Journal of Communications and Networks
• /
• v.7 no.4
• /
• pp.450-461
• /
• 2005

In this paper, we derive analytical expressions for the exact pairwise error probability (PEP) of a space-time coded system operating over spatially correlated fast (constant over the duration of a symbol) and slow (constant over the length of a code word) fad­ing channels using a moment-generating function-based approach. We discuss two analytical techniques that can be used to evaluate the exact-PEPs (and therefore, approximate the average bit error probability (BEP)) in closed form. These analytical expressions are more realistic than previously published PEP expressions as they fully account for antenna spacing, antenna geometries (uniform linear array, uniform grid array, uniform circular array, etc.) and scattering models (uniform, Gaussian, Laplacian, Von-mises, etc.). Inclusion of spatial information in these expressions provides valuable insights into the physical factors determining the performance of a space-time code. Using these new PEP expressions, we investigate the effect of antenna spacing, antenna geometries and azimuth power distribution parameters (angle of arrival/departure and angular spread) on the performance of a four-state QPSK space-time trellis code proposed by Tarokh et al. for two transmit antennas.