• Proceedings of the Korea Water Resources Association Conference
• /
• 2020.06a
• /
• pp.183-183
• /
• 2020

Accurate quantitative precipitation estimation plays an important role in hydrological modelling and prediction. Instantaneous quantitative precipitation estimation (QPE) by utilizing the weather radar data is a great applicability for operational hydrology in a catchment. Previously, regression technique performed between reflectivity (Z) and rain intensity (R) is used commonly to obtain radar QPEs. A novel, recent approaching method which might be applied in hydrological area for QPE is Long Short-Term Memory (LSTM) Networks. LSTM networks is a development and evolution of Recurrent Neuron Networks (RNNs) method that overcomes the limited memory capacity of RNNs and allows learning of long-term input-output dependencies. The advantages of LSTM compare to RNN technique is proven by previous works. In this study, LSTM networks is used to estimate the quantitative precipitation from weather radar for an urban catchment in South Korea. Radar information and rain-gauge data are used to evaluate and verify the estimation. The estimation results figure out that LSTM approaching method shows the accuracy and outperformance compared to Z-R relationship method. This study gives us the high potential of LSTM and its applications in urban hydrology.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.26 no.4
• /
• pp.52-65
• /
• 1984

In general precise estimation of hourly of daily distribution of the long-term run-off should be very important in a design of source of irrigation. However, there have not been a satisfying method for forecasting of stationar'y long-term run-off in Korea. Solving this problem, this study introduces unit-hydrograph method frequently used in short-term run-off analysis into the long-term run-off analysis, of which model basin was selected to be Sumgin-river catchment area. In the estimation of effective rainfall, conventional method neglects the Soil moisture condition of catchment area, but in this study, the initial discharge (qb) occurred just before rising phase of the hydrograph was selected as the index of a basin soil moisture condition and then introduced as 3rd variable in the analysis of the reationship between cumulative rainfall and cumulative loss of rainfall, which built a new type of separation method of effective rainfall. In next step, in order to normalize significant potential error included in hydrological data, especially in vast catchment area, Snyder's correlation method was applied. A key to solution in this study is multiple correlation method or multiple regressional analysis, which is primarily based on the method of least squres and which is solved by the form of systems of linear equations. And for verification of the change of characteristics of unit hydrograph according to the variation of a various kind of hydrological charateristics (for example, precipitation, tree cover, soil condition, etc),seasonal unit hydrograph models of dry season(autumn, winter), semi-dry season (spring), rainy season (summer) were made respectively. The results obtained in this study were summarized as follows; 1.During the test period of 1966-1971, effective rainfall was estimated for the total 114 run-off hydrograph. From this estimation results, relative error of estimation to the ovservation value was 6%, -which is mush smaller than 12% of the error of conventional method. 2.During the test period, daily distribution of long-term run-off discharge was estimated by the unit hydrograph model. From this estimation results, relative error of estimation by the application of standard unit hydrograph model was 12%. When estimating by each seasonal unit bydrograph model, the relative error was 14% during dry season 10% during semi-dry season and 7% during rainy season, which is much smaller than 37% of conventional method. Summing up the analysis results obtained above, it is convinced that qb-index method of this study for the estimation of effective rainfall be preciser than any other method developed before. Because even recently no method has been developed for the estimation of daily distribution of long-term run-off dicharge, therefore estimation value by unit hydrograph model was only compared with that due to kaziyama method which estimates monthly run-off discharge. However this method due to this study turns out to have high accuracy. If specially mentioned from the results of this study, there is no need to use each seasonal unit hydrograph model separately except the case of semi-dry season. The author hopes to analyze the latter case in future sudies.

• Nuclear Engineering and Technology
• /
• v.55 no.5
• /
• pp.1630-1643
• /
• 2023

The correct situation awareness (SA) of operators is important for managing nuclear power plants (NPPs), particularly in accident-related situations. Among the three levels of SA suggested by Ensley, Level 3 SA (i.e., projection of the future status of the situation) is challenging because of the complexity of NPPs as well as the uncertainty of accidents. Hence, several prediction methods using artificial intelligence techniques have been proposed to assist operators in accident prediction. However, these methods only predict short-term plant status (e.g., the status after a few minutes) and do not provide information regarding the uncertainty associated with the prediction. This paper proposes an algorithm that can predict the multivariate and long-term behavior of plant parameters for 2 h with 120 steps and provide the uncertainty of the prediction. The algorithm applies bidirectional long short-term memory and an attention mechanism, which enable the algorithm to predict the precise long-term trends of the parameters with high prediction accuracy. A conditional variational autoencoder was used to provide uncertainty information about the network prediction. The algorithm was trained, optimized, and validated using a compact nuclear simulator for a Westinghouse 900 MWe NPP.

• Journal of Korean Society of Transportation
• /
• v.30 no.5
• /
• pp.43-59
• /
• 2012

It is generally assumed about trip distribution estimation model that growth factor model's estimation accuracy is higher than that of other models in short-term and that gravity model's estimation accuracy is higher than that of other models in long-term. For validation of such assumptions, this study compares estimation accuracies of each estimation model using 3year(1988, 1992, 2004) O-D tables from Daegu city. Each estimation model's accuracy were compared by mid-size and large-size zone as well as short-term and long-term target years. The results show that the trip distribution estimation model selection by usual assumption is not always right.

• Journal of Korean Society on Water Environment
• /
• v.22 no.6
• /
• pp.1088-1093
• /
• 2006

In this study, the method of annual sediment estimation for reservoir long-term operation is proposed. Long-term daily precipitation and evaporation are predicted by Markov Chain. Using these values, reservoir inflow is simulated by NWS-PC model. Reservoir sediment load is estimated by sediment rating relation curve which is observed. From the simulation results, it was found that each simulated value by Markov Chain and NWS-PC was well compared to the observed ones and also estimated reservoir sediment was appropriate to the compared values using empirical equations. It is thought that the proposed method for estimation of reservoir sediment can be useful used to operate the reservoir.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.26 no.3
• /
• pp.183-191
• /
• 2021

A battery management system (BMS) provides some functions for ensuring safety and reliability that includes algorithms estimating battery states. Given the changes caused by various operating conditions, the state-of-health (SOH), which represents a figure of merit of the battery's ability to store and deliver energy, becomes challenging to estimate. Machine learning methods can be applied to perform accurate SOH estimation. In this study, we propose a Long-Term Recurrent Convolutional Network (LRCN) that combines the Convolutional Neural Network (CNN) and Long Short-term Memory (LSTM) to extract aging characteristics and learn temporal mechanisms. The dataset collected by the battery aging experiments of NASA PCoE is used to train models. The input dataset used part of the charging profile. The accuracy of the proposed model is compared with the CNN and LSTM models using the k-fold cross-validation technique. The proposed model achieves a low RMSE of 2.21%, which shows higher accuracy than others in SOH estimation.

• Journal of Korea Water Resources Association
• /
• v.30 no.6
• /
• pp.721-729
• /
• 1997

Total scour depth at a bridge is comprised of three components: long-term changes, contraction scour and local scour. Therefore, the analysis of long-term bed elevation changes is very important in the estimation of total scour depth at bridge sites. In this research, long-term bed elevation changes at the Namhan River Bridge are analysed using CHARIMA and HEC-6 models. The results show that, for 5-year steady normal stream flow, the bed elevation is aggreded by 45cm for CHARIMA model but degraded by 5cm for HEC-6 model. For 5-year unsteady flow, the bed elevation is changed greatly and it has a great influence on the estimation of total scour depth. Therefore, to make a proper estimation of total scour depth, not only contraction scour and local scour, but also long-term bed elevation changes should be estimated precisely.

• The Journal of Society for e-Business Studies
• /
• v.15 no.1
• /
• pp.139-155
• /
• 2010

The purpose of this paper is to develop a framework of software cost estimation guidelines and to derive a mid-long term direction for development of the software cost estimation guidelines. In this paper, all the steps in the software life cycle are researched in the view of cost estimation, and current software cost estimation guidelines and models have been reviewed and analysed first. Second, a plan to separate unit cost per function point from standard procedure in current software cost estimation guidelines is presented to strengthen maket self-regulating function as a mid-long term developmental direction for software cost estimation guidelines. Third, construction of cost repository, making standard procedure for software cost estimation guidelines, development of various kinds of software cost estimation models, and a system for experts on software cost estimation are presented as the prerequisites for the future model framework of software cost estimation guidelines. Finally a roadmap for establishing the future model is proposed.

• Communications for Statistical Applications and Methods
• /
• v.24 no.4
• /
• pp.325-337
• /
• 2017

In this paper, we proposed a new long-term lifetime distribution with four parameters inserted in a risk competitive scenario with decreasing, increasing and unimodal hazard rate functions, namely the Weibull-Poisson long-term distribution. This new distribution arises from a scenario of competitive latent risk, in which the lifetime associated to the particular risk is not observable, and where only the minimum lifetime value among all risks is noticed in a long-term context. However, it can also be used in any other situation as long as it fits the data well. The Weibull-Poisson long-term distribution is presented as a particular case for the new exponential-Poisson long-term distribution and Weibull long-term distribution. The properties of the proposed distribution were discussed, including its probability density, survival and hazard functions and explicit algebraic formulas for its order statistics. Assuming censored data, we considered the maximum likelihood approach for parameter estimation. For different parameter settings, sample sizes, and censoring percentages various simulation studies were performed to study the mean square error of the maximum likelihood estimative, and compare the performance of the model proposed with the particular cases. The selection criteria Akaike information criterion, Bayesian information criterion, and likelihood ratio test were used for the model selection. The relevance of the approach was illustrated on two real datasets of where the new model was compared with its particular cases observing its potential and competitiveness.

• Korean Journal of Construction Engineering and Management
• /
• v.8 no.6
• /
• pp.207-215
• /
• 2007

The purpose of this study is to predict the long-term maintenance expense of BTL school projects which were ordered from the Ministry of Education and Human Resources Development and each Metropolitan and Province Office of Education. For conducting this study, the adapted research method includes a case study of BTL school projects ordered from Seoul Metropolitan Office of Education in 2006. After examination of initial investment based on each $school^{\circ}{\phi}s$ operation account, it estimates maintenance expense and long-term maintenance expense. Also it compare using two methods: one is the long-term maintenance expense estimation in apartment houses and the other is AEAM(annual equivalent amount method). The results of this study are as follows: 1) It is analyzed long-term maintenance expense rates of each BTL school. As a result, it is construction(14.0), civil(1.4%), mechanical(6.5%), equipment(6.5%), electronic(11.0%), fixture(5.1%) and the rest(1.0%). 2)It is applied using two methods: one is the long-term maintenance expense estimation in apartment houses and the other is AEAM. Finally, It is compared expense deflection per $100{\beta}{\ge}$ in each month.