• Journal of the Korean Data and Information Science Society
• /
• v.14 no.4
• /
• pp.845-852
• /
• 2003

Forecasting exchange rate movements is a challenging task since exchange rates impact world economy and determine value of international investments. In particular, Korean-U.S. exchange rate behavior is very important because of strong Korean and U.S. trading relationship. Neural networks models have been used for short-term prediction of exchange rate movements. Least squares support vector machine (LS-SVM) is used widely in real-world regression tasks. This paper describes the use of LS-SVM for short-term and long-term prediction of Korean-U.S. exchange rate.

• Journal of the Korea Concrete Institute
• /
• v.12 no.4
• /
• pp.69-78
• /
• 2000

Creep of concrete is the most dominating factor affecting time-dependent deformations of concrete structures. Especially, creep deformation for design and construction in prestressed concrete structures should be predicted accurately because of its close relation with the loss in prestree of prestressed concrete structures. Existing creep-prediction models for special applications contain several impractical factors such as the lack ok accuracy, the requirement of long-term test and the lack of versatility for change in material properties, ets., which should be improved. In order to improve those drawbacks, a methodology to modify the creep-prediction equation specified in current Korean concrete structures design standard (KCI-99), which underestimates creep of concrete and does not consider change of condition in mixture design, is proposed. In this study, short-term creep tests were carried out for early-age concrete within 28 days after loading and their test results on influencing factors in the equation are analysed. Then, the prediction equation was modified by using the early-age creep test results. The modified prediction equation was verified by comparing their results with results obtained from long-term creep test.

• Korean Journal of Construction Engineering and Management
• /
• v.22 no.3
• /
• pp.21-30
• /
• 2021

This study proposes a 5-month rebar price prediction method using the recursive prediction method of deep learning. This approach predicts a long-term point in time by repeating the process of predicting all the characteristics of the input data and adding them to the original data and predicting the next point in time. The predicted average accuracy of the rebar prices for one to five months is approximately 97.24% in the manner presented in this study. Through the proposed method, it is expected that more accurate cost planning will be possible than the existing method by supplementing the systematicity of the price estimation method through human experience and judgment. In addition, it is expected that the method presented in this study can be utilized in studies that predict long-term prices using time series data including building materials other than rebar.

• Proceedings of the Korea Concrete Institute Conference
• /
• 1998.10a
• /
• pp.462-467
• /
• 1998

A rational method for prediction of long-term deflections of reinforced concrete beams under sustained loads was proposed. Strain and stress distributions of uncracked and fully cracked sections after creep and shrinkage were determined from the requirements of strain compatibility and force equilibrium of a section, and then long-term deflections were calculated from the section analysis results. In fully cracked section analysis, noncoincidence of the neutral axis of strain and the neutral axis of stress after creep and shrinkage was taken into account. The accuracy of the proposed method was verified by comparison with several experimental measurements of beam deflections. The proposed approximate procedure gave the better predictions than the existing approximate methods. At the same time, the proposed method also retained simplicity of the calculation, since maximum long-term deflection could be obtained without tedious integration of the curvatures.

• Geomechanics and Engineering
• /
• v.36 no.1
• /
• pp.83-97
• /
• 2024

It is important to make reasonable prediction about the long-term deformation of high rockfill geostructures. However, the deformation is usually underestimated using the rockfill parameters obtained from laboratory tests due to different size effects, which make it necessary to identify parameters from in-situ monitoring data. This paper proposes a novel hybrid back-analysis method with a modified objective function defined for the time-dependent back-analysis problem. The method consists of two stages. In the first stage, an improved weighted average method is proposed to quickly narrow the search region; while in the second stage, an adaptive response surface method is proposed to iteratively search for the satisfactory solution, with a technique that can adaptively consider the translation, contraction or expansion of the exploration region. The accuracy and computational efficiency of the proposed hybrid back-analysis method is demonstrated by back-analyzing the long-term deformation of two high embankments constructed for airport runways, with the rockfills being modeled by a rheological model considering the influence of stress states on the creep behavior.

• Steel and Composite Structures
• /
• v.30 no.6
• /
• pp.535-550
• /
• 2019

Prestressed concrete (PC) bridges using corrugated steel webbing have emerged as one of the most promising forms of steel-concrete composite bridge. However, their long-term behavior is not well understood, especially in the case of large-span bridges. In order to study the time-dependent performance, a large three-span PC bridge with corrugated steel webbing was compared to a similar conventional PC bridge to examine their respective time-dependent characteristics. In addition, a three-dimensional finite element method with step-by-step time integration that takes into account cantilever construction procedures was used to predict long-term behaviors such as deflection, stress distribution and prestressing loss. These predictions were based upon four well-established empirical creep prediction models. PC bridges with a corrugated steel web were observed to have a better long-term performance relative to conventional PC bridges. In particular, it is noted that the pre-cambering for PC bridges with a corrugated steel web could be smaller than that of conventional PC bridges. The ratio of side-to-mid span has great influence on the long-term deformation of PC bridges with a corrugated steel web, and it is suggested that the design value should be between 0.4 and 0.6. However, the different creep prediction models still showed a weak homogeneity, thus, the further experimental research and the development of health monitoring systems are required to further progress our understanding of the long-term behavior of PC bridges with corrugated steel webbing.

• Journal of Positioning, Navigation, and Timing
• /
• v.8 no.4
• /
• pp.209-214
• /
• 2019

Precise point positioning (PPP) requires precise orbit and clock products. International GNSS service (IGS) real-time service (RTS) data can be used in real-time for PPP, but it may not be possible to receive these corrections for a short time due to internet or hardware failure. In addition, the time required for IGS to combine RTS data from each analysis center results in a delay of about 30 seconds for the RTS data. Short-term orbit prediction can be possible because it includes the rate of correction, but the clock correction only provides bias. Thus, a short-term prediction model is needed to preidict RTS clock corrections. In this paper, we used a long short-term memory (LSTM) network to predict RTS clock correction for three minutes. The prediction accuracy of the LSTM was compared with that of the polynomial model. After applying the predicted clock corrections to the broadcast ephemeris, we performed PPP and analyzed the positioning accuracy. The LSTM network predicted the clock correction within 2 cm error, and the PPP accuracy is almost the same as received RTS data.

• Journal of Advanced Navigation Technology
• /
• v.23 no.6
• /
• pp.561-569
• /
• 2019

Solar photovoltaic can provide electrical energy with only radiation, and its use is expanding rapidly as a new energy source. This study predicts the short and long-term PV power generation using actual converter output data of photovoltaic system. The prediction algorithm uses multiple linear regression, support vector machine (SVM), and deep learning such as deep neural network (DNN) and long short-term memory (LSTM). In addition, three models are used according to the input and output structure of the weather element. Long-term forecasts are made monthly, seasonally and annually, and short-term forecasts are made for 7 days. As a result, the deep learning network is better in prediction accuracy than multiple linear regression and SVM. In addition, LSTM, which is a better model for time series prediction than DNN, is somewhat superior in terms of prediction accuracy. The experiment results according to the input and output structure appear Model 2 has less error than Model 1, and Model 3 has less error than Model 2.

• Journal of information and communication convergence engineering
• /
• v.5 no.2
• /
• pp.171-176
• /
• 2007

To give a guarantee a consistently high level of quality and reliability of Telematics traffic service, traffic flow forecasting is very important issue. In this paper, we proposed an adaptable integrated prediction model to predict the traffic flow in the future. Our model combines two methods, short-term prediction model and long-term prediction model with different combining coefficients to reflect current traffic condition. Short-term model uses the Kalman filtering technique to predict the future traffic conditions. And long-term model processes accumulated speed patterns which means the analysis results for all past speeds of each road by classifying the same day and the same time interval. Combining two models makes it possible to predict future traffic flow with higher accuracy over a longer time range. Many experiments showed our algorithm gives a better precise prediction than only an accumulated speed pattern that is used commonly. The result can be applied to the car navigation to support a dynamic shortest path. In addition, it can give users the travel information to avoid the traffic congestion areas.

• Journal of Korean Society of societal Security
• /
• v.1 no.3
• /
• pp.51-57
• /
• 2008

The long-term behaviors of prestressed concrete cable-stayed bridges are considerably influenced by the time dependant material characteristics such as creep and shrinkage. This study investigated the influences of the change of relative humidity by application of the CEB-FIP model and ACI model, which are generally used in the prediction of long-term behavior of concrete structures. In case of the moment of girder, CEB-FIP model predicted a bigger effect of relative humidity change than the ACI model. Furthermore, the effect was significant. Also, the long-term behaviors between these models were different each other even under the same material condition. Therefore, the prediction of the long-term behavior should be compensated after comparative analysis with the results of material tests of each construction site and between the different models.