• Proceedings of the KIEE Conference
• /
• 2001.07d
• /
• pp.2030-2032
• /
• 2001

In recent years, many attempts have been made to predict the behavior of bonds, currencies, stock, or other economic markets. Most previous experiments used multilayer perceptrons(MLP) for stock market forecasting. The Kospi 200 Index is modeled using different neural networks and fuzzy system predictions. In this paper, a multilayer perceptron architecture, a dynamic polynomial neural network(DPNN) and a fuzzy system are used to predict the Kospi 200 index. The results of prediction is compared with the root mean squared error(RMSE) and the scatter plot. Results show that both networks can be trained to predict the index. And the fuzzy system is performing slightly better than DPNN and MLP.

• Proceedings of the KIEE Conference
• /
• 1996.07b
• /
• pp.1234-1236
• /
• 1996

In this paper, the trajectory needed to be tracked by the manipulator was defined in a new plot differently from conventional methods. And the trajectory provides Solution directly related to coordinates of output variables from the plant. So, it overcomes nonlinearity between joint and Cartesian coordinates in movement mode and it makes use of inverse Kinematics unnecessary, which was obstacle for real-time control. The 2-axis SCARA robot was modelled and simulation was performed to validate in this paper. And it proved this has better performance in rapidity and decrease of position-error, compared to the conventional FLCs.

• The Korean Journal of Applied Statistics
• /
• v.30 no.1
• /
• pp.195-202
• /
• 2017

We can sometimes find instability against numerical solutions in nonlinear regression. All iterative procedures in nonlinear regression require initial parameter values to be selected. Poor starting values may result in convergence to an unwanted stationary point of the error sum of squares surface. Starting values can sometimes cause the chaos effect in the nonlinear regression model. We can find the chaos phenomena with the convergence plot of starting values in the parameter space.

• Journal of the Korean Society for Precision Engineering
• /
• v.21 no.9
• /
• pp.158-166
• /
• 2004

A new method to measure residual unbalance of rotor - bearing system was proposed. The method which determine residual unbalance based on polar plot and an analytical method which calculates the residual unbalance of the rotor from the vibration response of the Jeffcott rotor are proposed in this study with respect to a real rotor system of which the residual unbalance is unknown. The unbalance eccentricity of the produced experimental model is 3.78 mil, developing the measurement method of the residual unbalance more convenient than the proposed method of ISO and API standard. The proposed method was experimentally compared with the ISO standard, and the two methods were exactly correspondent to each other within an error of 1%.

• Journal of Korean Society of Water and Wastewater
• /
• v.30 no.3
• /
• pp.253-261
• /
• 2016

A variety of methods for detecting the location of an underground water pipeline are being used across the world; the current main methods used in South Korea, however, have the problems of low precision and efficiency and the limitations in actual application. On this, this study developed locator capable of detecting the location of a water pipe by the use of an IMU sensor, and technology for using the extended karman filter to correct error in location detection and to plot the location on the coordinate system. This study carried out a tract test and a road test as basic experiments to measure the performance of the developed technology and equipment. As a result of the straight line, circular and ellipse track tests, the 1750 IMU sensor showed the average error of 0.08-0.11%; and thus it was found that the developed locator can detect a location precisely. As a result of the 859.6-m road test, it was found that the error was 0.31 m in case the moving rate of the sensor was 0.3-0.6 m/s; and thus it was judged that the equipment developed by this study can be applied to long-distance water pipes of over 1 km sufficiently. It is planned to evaluate its field applicability in the future through an actual pipe network pilot test, and it is expected that locator capable of detecting the location of a water pipe more precisely will be developed through research for the enhancement of accuracy in the algorithm of location detection.

• Journal of Korean Society of Forest Science
• /
• v.6 no.1
• /
• pp.24-26
• /
• 1967

1. The purpose of this experiment was to compose the precision of dot grid method and transects method with that of pianimeter method in area measurement. 2. The following conclusions were obtained through study on precision of area measurement of 20 plots. 3. The percentage error of each method was none significant in comparision with planimeter method. 4. The dot grid method gave 0.2% of overestimated value and the transects method 0.67% of underestimated value. 5. Accordingly, the dot grid method gave the better result than the transects method. 6. The transects method had small errors at the plots larger than 30ha. while it had big errors at plots smaller than 30ha. Therefore, it can be said that the transects method is suitable for area measurement of plots larger than 30 hectares.

• Structural Engineering and Mechanics
• /
• v.85 no.4
• /
• pp.469-484
• /
• 2023

A deep recursive bidirectional Cuda Deep Neural Network Long Short Term Memory (Bi-CuDNNLSTM) layer is recruited in this paper to predict the entire force time histories, and the corresponding hysteresis and backbone curves of reinforced concrete (RC) bridge piers using experimental fast and slow cyclic tests. The proposed stacked Bi-CuDNNLSTM layers involve multiple uncertain input variables, including horizontal actuator displacements, vertical actuators axial loads, the effective height of the bridge pier, the moment of inertia, and mass. The functional application programming interface in the Keras Python library is utilized to develop a deep learning model considering all the above various input attributes. To have a robust and reliable prediction, the dataset for both the fast and slow cyclic tests is split into three mutually exclusive subsets of training, validation, and testing (unseen). The whole datasets include 17 RC bridge piers tested experimentally ten for fast and seven for slow cyclic tests. The results bring to light that the mean absolute error, as a loss function, is monotonically decreased to zero for both the training and validation datasets after 5000 epochs, and a high level of correlation is observed between the predicted and the experimentally measured values of the force time histories for all the datasets, more than 90%. It can be concluded that the maximum mean of the normalized error, obtained through Box-Whisker plot and Gaussian distribution of normalized error, associated with unseen data is about 10% and 3% for the fast and slow cyclic tests, respectively. In recapitulation, it brings to an end that the stacked Bi-CuDNNLSTM layer implemented in this study has a myriad of benefits in reducing the time and experimental costs for conducting new fast and slow cyclic tests in the future and results in a fast and accurate insight into hysteretic behavior of bridge piers.

• Computers and Concrete
• /
• v.33 no.1
• /
• pp.55-75
• /
• 2024

The present study introduces an optimum performance soft computing model for predicting the compressive strength of high-performance concrete (HPC) by comparing models based on conventional (kernel-based, covariance function-based, and tree-based), advanced machine (least square support vector machine-LSSVM and minimax probability machine regressor-MPMR), and deep (artificial neural network-ANN) learning approaches using a common database for the first time. A compressive strength database, having results of 1030 concrete samples, has been compiled from the literature and preprocessed. For the purpose of training, testing, and validation of soft computing models, 803, 101, and 101 data points have been selected arbitrarily from preprocessed data points, i.e., 1005. Thirteen performance metrics, including three new metrics, i.e., a20-index, index of agreement, and index of scatter, have been implemented for each model. The performance comparison reveals that the SVM (kernel-based), ET (tree-based), MPMR (advanced), and ANN (deep) models have achieved higher performance in predicting the compressive strength of HPC. From the overall analysis of performance, accuracy, Taylor plot, accuracy metric, regression error characteristics curve, Anderson-Darling, Wilcoxon, Uncertainty, and reliability, it has been observed that model CS4 based on the ensemble tree has been recognized as an optimum performance model with higher performance, i.e., a correlation coefficient of 0.9352, root mean square error of 5.76 MPa, and mean absolute error of 4.1069 MPa. The present study also reveals that multicollinearity affects the prediction accuracy of Gaussian process regression, decision tree, multilinear regression, and adaptive boosting regressor models, novel research in compressive strength prediction of HPC. The cosine sensitivity analysis reveals that the prediction of compressive strength of HPC is highly affected by cement content, fine aggregate, coarse aggregate, and water content.

• Physical Therapy Rehabilitation Science
• /
• v.13 no.2
• /
• pp.143-151
• /
• 2024

Objective: This study aims to assess the test-retest reproducibility of the Short Form Berg Balance Scale (SF-BBS) and the Short Form Postural Assessment Scale for Stroke (SF-PASS) among chronic stroke survivors, focusing on their reliability for consistent measurements over time. Design: A cross-sectional study design Methods: Thirty chronic stroke survivors participated in this study, undergoing evaluations with SF-BBS and SF-PASS scales at two different points, separated by a seven-day interval. The analysis focused on test-retest reliability, employing statistical measures such as the Intra-Class Coefficient (ICC_2,1), Standard Error of Measurement (SEM), Minimal Detectable Change (MDC), and MDC%, the Bland-Altman plot to assess the limits of agreement and the extent of random measurement error. Results: The study found notable test-retest reproducibility for both SF-BBS and SF-PASS, with ICC values demonstrating strong reliability (0.932 to 0.941, with a confidence interval of 0.889 to 0.973). SEM values for SF-BBS and SF-PASS were reported as 1.34 and 0.61, respectively, indicating low measurement error. MDC values of 3.71 for SF-BBS and 1.69 for SF-PASS suggest that the scales have an acceptable level of sensitivity to change, with reliability metrics falling below 20% of the maximum possible score. Conclusions: The findings suggest that both SF-BBS and SF-PASS exhibit high intra-class correlation coefficients, indicating strong test-retest reliability. The SEM and MDC values further support the scales' reproducibility and reliability as tools for evaluating mobility and dynamic balance in chronic stroke survivors. Therefore, these scales are recommended for clinical use in this population, providing reliable measures for assessing progress in rehabilitation.

• Proceedings of the Korean Society of Near Infrared Spectroscopy Conference
• /
• 2001.06a
• /
• pp.4102-4102
• /
• 2001

In multivariate analysis, absorbance spectrum is measured over a band of wavelengths. One does not often pay attention to the size of this wavelength band. However, it is desirable that spectrum is measured at only necessary wavelengths as long as the acceptable accuracy of prediction can be met. In this paper, the method of selecting an optimal band of wavelengths based on the loading vector analysis was proposed and applied for determining total protein in human serum using near-infrared transmission spectroscopy and PLSR. Loading vectors in the full spectrum PLSR were used as reference in selecting wavelengths, but only the first loading vector was used since it explains the spectrum best. Absorbance spectra of sera from 97 outpatients were measured at 1530∼1850 nm with an interval of 2 nm. Total protein concentrations of sera were ranged from 5.1 to 7.7 g/㎗. Spectra were measured by Cary 5E spectrophotometer (Varian, Australia). Serum in the 5 mm-pathlength cuvette was put in the sample beam and air in the reference beam. Full spectrum PLSR was applied to determine total protein from sera. Next, the wavelength region of 1672∼1754 nm was selected based on the first loading vector analysis. Standard Error of Cross Validation (SECV) of full spectrum (1530∼l850 nm) PLSR and selected wavelength PLSR (1672∼1754 nm) was respectively 0.28 and 0.27 g/㎗. The prediction accuracy between the two bands was equal. Wavelength selection based on loading vector in PLSR seemed to be simple and robust in comparison to other methods based on correlation plot, regression vector and genetic algorithm. As a reference of wavelength selection for PLSR, the loading vector has the advantage over the correlation plot since the former is based on multivariate model whereas the latter, on univariate model. Wavelength selection by the first loading vector analysis requires shorter computation time than that by genetic algorithm and needs not smoothing.