• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.373-373
• /
• 2021

Rainfall forecasting is an important issue that is applied in many areas, such as agriculture, flood warning, and water resources management. In this context, this study proposed a statistical and machine learning-based forecasting model for monthly rainfall. The Bayesian Gaussian process was chosen to optimize the hyperparameters of the Stacked Long Short-term memory (SLSTM) model. The proposed SLSTM model was applied for predicting monthly precipitation of Seoul station, South Korea. Data were retrieved from the Korea Meteorological Administration (KMA) in the period between 1960 and 2019. Four schemes were examined in this study: (i) prediction with only rainfall; (ii) with deseasonalized rainfall; (iii) with rainfall and minimum temperature; (iv) with deseasonalized rainfall and minimum temperature. The error of predicted rainfall based on the root mean squared error (RMSE), 16-17 mm, is relatively small compared with the average monthly rainfall at Seoul station is 117mm. The results showed scheme (iv) gives the best prediction result. Therefore, this approach is more straightforward than the hydrological and hydraulic models, which request much more input data. The result indicated that a deep learning network could be applied successfully in the hydrology field. Overall, the proposed method is promising, given a good solution for rainfall prediction.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.391-393
• /
• 2021

In this paper, we introduce a long short-term memory (LSTM)-based fall detection system using TensorFlow that can detect falls occurring in the elderly in daily living. 3-axis accelerometer data are aggregated for fall detection, and then three types of parameter are calculated. 4 types of activity of daily living (ADL) and 3 types of fall situation patterns are classified. The parameterized data applied to LSTM. Learning proceeds until the Loss value becomes 0.5 or less. The results are calculated for each parameter θ, SVM, and GSVM. The best result was GSVM, which showed Sensitivity 98.75%, Specificity 99.68%, and Accuracy 99.28%.

• Smart Structures and Systems
• /
• v.31 no.1
• /
• pp.45-56
• /
• 2023

Vibration-based structural health monitoring is used to ensure the safety of structures by installing sensors in structures. The peak picking method, one of the applications of vibration-based structural health monitoring, is a method that analyze the dynamic characteristics of a structure using the peaks of the frequency response function. However, the results may vary depending on the person predicting the peak point; further, the method does not predict the exact peak point in the presence of noise. To overcome the limitations of the existing peak picking methods, this study proposes a new method to automate the modal analysis process by utilizing long short-term memory, a type of recurrent neural network. The method proposed in this study uses the time series data of the frequency response function directly as the input of the LSTM network. In addition, the proposed method improved the accuracy by using the phase as well as amplitude information of the frequency response function. Simulation experiments and lab-scale model experiments are performed to verify the performance of the LSTM network developed in this study. The result reported a modal assurance criterion of 0.8107, and it is expected that the dynamic characteristics of a civil structure can be predicted with high accuracy using data without experts.

• Smart Structures and Systems
• /
• v.32 no.6
• /
• pp.359-369
• /
• 2023

Servo-motor driven uniaxial shake tables have been widely used for education and research purposes in earthquake engineering. These shake tables are mostly displacement-controlled by a digital proportional-integral-derivative (PID) controller; however, accurate reproduction of acceleration time histories is not guaranteed. In this study, a control strategy is proposed and verified for uniaxial shake tables driven by a servo-motor. This strategy incorporates a deep-learning algorithm named Long Short-Term Memory (LSTM) network into a displacement PID feedback controller. The LSTM controller is trained by using a large number of experimental data of a self-made servo-motor driven uniaxial shake table. After the training is completed, the LSTM controller is implemented for directly generating the command voltage for the servo motor to drive the shake table. Meanwhile, a displacement PID controller is tuned and implemented close to the LSTM controller to prevent the shake table from permanent drift. The control strategy is named the LSTM-PID control scheme. Experimental results demonstrate that the proposed LSTM-PID improves the acceleration tracking performance of the uniaxial shake table for both bare condition and loaded condition with a slender specimen.

• International Journal of Highway Engineering
• /
• v.17 no.4
• /
• pp.135-141
• /
• 2015

PURPOSES: This study has been conducted to determine a homogeneous segment and integration to improve the efficiency of short-term traffic count. We have also attempted to reduce the traffic monitoring budget. METHODS: Based on the statistical approach, a homogeneous segment in the same road section is determined. Statistical analysis using t-test, mean difference, and correlation coefficient are carried out for 10-year-long (2004-2013) short-term count traffic data and the MAPE of fresh data (2014) are evaluated. The correlation coefficient represents a trend in traffic count, while the mean difference and t-score represent an average traffic count. RESULTS : The statistical analysis suggests that the number of target segments varies with the criteria. The correlation coefficient of more than 30% of the adjacent segment is higher than 0.8. A mean difference of 36.2% and t-score of 19.5% for adjacent segments are below 20% and 2.8, respectively. According to the effectiveness analysis, the integration criteria of the mean difference have a higher effect as compared to the t-score criteria. Thus, the mean difference represents a traffic volume similarity. CONCLUSIONS : The integration of 47 road segments from 882 adjacent road segments indicate 8.87% of MAPE, which is within an acceptable range. It can reduce the traffic monitoring budget and increase the count to improve an accuracy of traffic volume estimation.

• Structural Engineering and Mechanics
• /
• v.53 no.5
• /
• pp.881-896
• /
• 2015

In this study, a method for fatigue performance estimation of deepwater steel catenary riser (SCR) under short-term vortex-induced vibration was investigated for selected S-N curves. General tendency between S-N curve capacity and fatigue performance was analysed. SCRs are generally used to transport produced oil and gas or to export separated oil and gas, and are exposed to various environmental loads in terms of current, wave, wind and others. Current is closely related with VIV and it affects fatigue life of riser structures significantly. In this regards, the process of appropriate S-N curve selection was performed in the initial design stage based on the scale of fabrication-related initial imperfections such as welding, hot spot, crack, stress concentration factor, and others. To draw the general tendency, the effects of stress concentration factor (SCF), S-N curve type, current profile, and three different sizes of SCRs were considered, and the relationship between S-N curve capacity and short-term VIV fatigue performance of SCR was derived. In case of S-N curve selection, DNV (2012) guideline was adopted and four different current profiles of the Gulf of Mexico (normal condition and Hurricane condition) and Brazil (Amazon basin and Campos basin) were considered. The obtained results will be useful to select the S-N curve for deepwater SCRs and also to understand the relationship between S-N curve capacity and short-term VIV fatigue performance of deepwater SCRs.

• Korean Journal of Psychosomatic Medicine
• /
• v.6 no.2
• /
• pp.136-146
• /
• 1998

Objectives : This study was performed to evaluate the effects of PTSD on memory function, to investigate the difference of memory function between PTSD and non-PTSD patients, and to identify major variables correlated to PTSD scale and Memory Assessment Scale. Methods: The authors used PTSD-scale(Mississippi scale and Combat Exposure Scale) for measuring PTSD severity. And, Beck Depression Inventory was also used. Memory assessment scale was assessed by well trained psychologist. Thirty one Vietnam veterans who had been hospitalized were collected consecutively. These patients were evaluated by psychiatrists with interview and measurement for fifteen months since March, 1997. The collected data were analyzed by SPSS and the stastistic methods used for analysis Chi-square, t-test, and Pearson's correlation. Results : 1) There were significant differences in short-term memory and verbal memory between PTSD and non-PTSD in Vietnam veterans. 2) Mississippi scale and Combat Exposure Scale were negatively correlated to short-term memory and verbal memory(Pearson's correlation). 3) Religion status was a significant variable between PTSD and non-PTSD in Vietnam veterans. 4) There is no significant difference in visual memory and total memory scale between PTSD and Non-PTSD in Vietnam veterans Conclusions : Neuropsychological changes were found in the posttraumatic stress disorder. There were significant differences in short-term memory and verbal memory between PTSD and non-PTSD in Vietnam veterans. Mississippi scale and Combat Exposure Scale were negatively correlated to short-term memory and verbal memory. We suggest that neuropsychological test might be used for an objective assessment of patients with the combat related PTSD and be considered helpful in the assessment of patients with the diagnosis. And we also suggest rehabilitation strategies would be used to compensate for memory deficits in PTSD patients.

• Journal of Periodontal and Implant Science
• /
• v.41 no.5
• /
• pp.211-217
• /
• 2011

Purpose: The marginal bone levels around implants following restoration are used as a reference for evaluating implant success and survival. Two design concepts that can reduce crestal bone resorption are the microthread and platform-switching concepts. The aims of this study were to analyze the placement of microthreaded and platform-switched implants and their short-term survival rate, as well as the level of bone around the implants. Methods: The subjects of this study were 27 patients (79 implants) undergoing treatment with microthreaded and platform-switched implants between October 2008 and July 2009 in the Dental Hospital of Yonsei University Department of Periodon-tology. The patients received follow-up care more than 6 months after the final setting of the prosthesis, at which time periapical radiographs were taken. The marginal bone level was measured from the reference point to the lowest observed point of contact between the marginal bone and the fixture. Comparisons were made between radiographs taken at the time of fixture installation and those taken at the follow-up visit. Results: During the study period (average of 11.8 months after fixture installation and 7.4 months after the prosthesis delivery), the short-term survival rate of microthreaded and platform-switched implants was 100% and the marginal bone loss around implants was $0.16{\pm}0.08$ mm, the latter of which is lower than the previously reported values. Conclusions: This short-term clinical study has demonstrated the successful survival rates of a microthread and platform-switched implant system, and that this system is associated with reduced marginal bone loss.

• KDI Journal of Economic Policy
• /
• v.35 no.4
• /
• pp.95-136
• /
• 2013

Using Workplace Panel Survey of 2005, 2007 and 2009 waves, this study estimates the effects of trade unions on employment and the proportion of irregular workers, short-term and part-time workers, and agent temporary and outsourced workers. While the estimation result shows that the percentage of hired workers increases under union presence, these results seem to be contaminated with bias because the differences between unionized firms before union establishment and non-unionized firms are not completely controlled even after adjusting for observed characteristics. Meanwhile, unionized firms and non-unionized firms with grievance procedures employ higher proportion of irregular workers. The proportion of short-term and part-time workers increases only when they are entitled to join trade unions. These imply that the rise in the percentage of irregular workers due to unions and grievance procedures is attributed to the increase in the percentage of agent temporary and outsourced workers. Also, when short-term and part-time workers are allowed to join the union, the firm replaces agent temporary and outsourced workers with short-term and part-time workers, so that the proportion of irregular workers do not change.

• Journal of Korean Society of Transportation
• /
• v.27 no.4
• /
• pp.127-135
• /
• 2009

Most traffic information that drivers receive while driving are stored in their short-term memory and disappear within a few seconds. Contemporary modeling approaches using a dummy variable can't fully explain this phenomenon. As such, this study proposes to use utility functions of real-time in-vehicle traffic safety information (IVTSI), analyzing its safety impacts based on empirical data from an on-site driving experiment at signalized intersection approach with a limited visibility. For this, a driving stability evaluation model is developed based on driver's driving speed choice, applying an ordered probit model. To estimate the specified utility functions, the model simultaneously accounts for various factors, such as traffic operation, geometry, road environment, and driver's characteristics. The results show three significant facts. First, a normal density function (exponential function) is appropriate to explain the utility of IVTSI proposed under study over time. Second, the IVTSI remains in driver's short-term memory for up to nearly 22 second after provision, decreasing over time. Three, IVTSI provision appears more important than the geometry factor but less than the traffic operation factor.