• Journal of Korean Society of Water and Wastewater
• /
• v.21 no.6
• /
• pp.689-699
• /
• 2007

The accurate estimation of water pipe deterioration is indispensable to prevent pipe breakage and manage in advance. In this study, corrosion of water pipe is adopted, which is relatively underestimated although it takes most part of deteriorating pipeline. Predicting corrosion rate and corrosion depth of a pipe can make an increase the life span of the pipeline, which is laid under the ground according to characteristics of soil and water corrosion. For the purpose, mathematical models that can presume nominal depth through estimation of pit corrosion and corrosion rate is introduced. As comparison of results with conventional methods in other foreign countries, it is evaluated that the external corrosion depth is estimated less than the models, proposed by other researchers and the internal corrosion rate was processed faster than the external corrosion rate.

• Journal of Korean Society on Water Environment
• /
• v.29 no.5
• /
• pp.625-629
• /
• 2013

The objective of this study is to develop and evaluate simple and multiple regression models for Total Organic Carbon (TOC) concentration estimation in stream flow. For development (using water quality data in 2012) and evaluation (using water quality data in 2011) of regression models, we used water quality data from downstream of Yeongsan river basin during 2011 and 2012, and correlation analysis between TOC and water quality parameters was conducted. The concentrations of TOC were positively correlated with Chemical Oxygen Demand (COD), Biochemical Oxygen Demand (BOD), TN (Total Nitrogen), Water Temperature (WT) and Electric Conductivity (EC). From these results, simple and multiple regression models for TOC estimation were developed as follows : $TOC=0.5809{\times}BOD+3.1557$, $TOC=0.4365{\times}COD+1.3731$. As a result of the application evaluation of the developed regression models, the multiple regression model was found to estimate TOC better than simple regression models.

• Journal of Environmental Impact Assessment
• /
• v.14 no.4
• /
• pp.157-164
• /
• 2005

Existing models that predict of Daily water supply include statistical models and neural network model. The neural network model was more effective than the statistical models. Only neural network model, which predict of Daily water supply, is focused on estimation of the operational control. Neural network model takes long learning time and gets into local minimum. This study proposes Neuro Genetic hybrid model which a combination of genetic algorithm and neural network. Hybrid model makes up for neural network's shortcomings. In this study, the amount of supply, the mean temperature and the population of the area supplied with water are use for neural network's learning patterns for prediction. RMSE(Root Mean Square Error) is used for a MOE(Measure Of Effectiveness). The comparison of the two models showed that the predicting capability of Hybrid model is more effective than that of neural network model. The proposed hybrid model is able to predict of Daily water, thus it can apply real time estimation of operational control of water works and water drain pipes. Proposed models include accidental cases such as a suspension of water supply. The maximum error rate between the estimation of the model and the actual measurement was 11.81% and the average error was lower than 1.76%. The model is expected to be a real-time estimation of the operational control of water works and water/drain pipes.

• Journal of Korean Society of Water and Wastewater
• /
• v.10 no.2
• /
• pp.88-103
• /
• 1996

The objective of this study is to develop stochastic and deterministic models which could be used to synthesize water application time series. Adaptive models using mulitivariate ARIMA(Transfer Function Model) are developed for daily urban water use forecasting. The model considers several variables on which water demands is dependent. The dynamic response of water demands to several factors(e.g. weekday, average temperature, minimum temperature, maximum temperature, humidity, cloudiness, rainfall) are characterized in the model by transfer functions. Daily water use data of Kumi city in 1992 are employed for model parameter estimation. Meteorological data of Seonsan station are utilized to input variables because Kumi has no records about the meteorological factor data.To determine the main factors influencing water use, autocorrelogram and cross correlogram analysis are performed. Through the identification, parameter estimation, and diagnostic checking of tentative model, final transfer function models by each month are established. The simulation output by transfer function models are compared to a historical data and shows the good agreement.

• Journal of Korean Society on Water Environment
• /
• v.31 no.3
• /
• pp.272-285
• /
• 2015

Accurate assessment of chlorophyll-a (Chl-a) concentrations in inland waters using remote sensing is challenging due to the optical complexity of case 2 waters. and the inherent optical properties (IOPs) of natural waters are the most significant factors affecting light propagation within water columns, and thus play indispensable roles on estimation of Chl-a concentrations. Despite its importance, no IOPs retrieval model was specifically developed for inland water bodies, although significant efforts were made on oceanic inversion models. So we have applied and validated a recently developed Red-NIR three-band model and an IOPs Inversion Model for estimating Chl-a concentration and deriving inland water IOPs in Lake Uiam. Three band and IOPs based Chl-a estimation model accuracy was assessed with samples collected in different seasons. The results indicate that this models can be used to accurately retrieve Chl-a concentration and absorption coefficients. For all datasets the determination coefficients of the 3-band models versus Chl-a concentration ranged 0.65 and 0.88 and IOPs based model versus Chl-a concentration varied from 0.73 to 0.83 respectively. and Comparison between 3-band and IOPs based models showed significant performance with decrease of root mean square error from 18% to 33.6%. The results of this study provides the potential of effective methods for remote monitoring and water quality management in turbid inland water bodies using hyper-spectral remote sensing.

• Journal of Korean Society on Water Environment
• /
• v.21 no.6
• /
• pp.575-583
• /
• 2005

In this study the flow curve estimation is discussed using TANK model which is one of hydrologic models. The main interest is the accuracy of TANK model parameter estimation with respect to the sampling frequency of input data. For doing this, input data with various sampling frequencies is used to estimate model parameters. As a result, in order to generate relatively accurate flow curve, it is recommendable to measure stream flow at least every 8 days.

• Journal of Korean Society of Water and Wastewater
• /
• v.19 no.4
• /
• pp.395-403
• /
• 2005

It is very important to forecast water supply for reasonal operation and management of water utilities. In this paper, water supply forecasting models using artificial intelligence are developed. Artificial intelligence models shows better results by using Temperature(t), water supply discharge (t-1) and water supply discharge (t-2), which are expressed by neural network(LMNNWS; Levenberg-Marquardt Neural Network for Water Supply, MDNNWS; MoDular Neural Network for Water Supply) and neuro fuzzy(ANASWS; Adaptive Neuro-Fuzzy Inference Systems for Water Supply). ANFISWS model which is applied for water supply forecasting shows stable application to the variable water supply data. As results, MDNNWS model shows the highest overall accuracy among proposed water supply forecasting models and the lowest estimation error with the order of ANFISWS, LMNNWS model.

• Journal of Korean Society on Water Environment
• /
• v.40 no.4
• /
• pp.179-194
• /
• 2024

Biophysicochemical processes in water environments and treatment systems have been great concerns of engineers and scientists for controlling the fate and transport of contaminants. These processes are practically formulated as mathematical models written in coupled differential equations. However, because these process-based mathematical models consist of a large number of model parameters, they are complicated in analytical or numerical computation. Users need to perform substantial trials and errors to achieve the best-fit simulation to measurements, relying on arbitrary selection of fitting parameters. Therefore, this study adopted a Bayesian calibration method to estimate best-fit model parameters in a systematic way and evaluated the applicability of the calibration method to biophysicochemical processes of water environments and treatment systems. The Bayesian calibration method was applied to the microbial growth-decay kinetics and flocculation kinetics, of which experimental data were obtained with batch kinetic experiments. The Bayesian calibration method was proven to be a reasonable, effective way for best-fit parameter estimation, demonstrating not only high-quality fitness, but also sensitivity of each parameter and correlation between different parameters. This state-of-the-art method will eventually help scientists and engineers to use complex process-based mathematical models consisting of various biophysicochemical processes.

• Journal of Korean Society on Water Environment
• /
• v.33 no.3
• /
• pp.326-333
• /
• 2017

The lakes and reservoirs have been reported as important sources of carbon emissions to the atmosphere in many countries. Although field experiments and theoretical investigations based on the fundamental gas exchange theory have proposed the quantitative amounts of Net Atmospheric Flux (NAF) in various climate regions, there are still large uncertainties at the global scale estimation. Mechanistic models can be used for understanding and estimating the temporal and spatial variations of the NAFs considering complicated hydrodynamic and biogeochemical processes in a reservoir, but these models require extensive and expensive datasets and model parameters. On the other hand, data driven machine learning (ML) algorithms are likely to be alternative tools to estimate the NAFs in responding to independent environmental variables. The objective of this study was to develop random forest (RF) and multi-layer artificial neural network (ANN) models for the estimation of the daily $CO_2$ NAFs in Daecheong Reservoir located in Geum River of Korea, and compare the models performance against the multiple linear regression (MLR) model that proposed in the previous study (Chung et al., 2016). As a result, the RF and ANN models showed much enhanced performance in the estimation of the high NAF values, while MLR model significantly under estimated them. Across validation with 10-fold random samplings was applied to evaluate the performance of three models, and indicated that the ANN model is best, and followed by RF and MLR models.

• Korean Journal of Environmental Agriculture
• /
• v.33 no.1
• /
• pp.37-43
• /
• 2014

BACKGROUND: Water quality data are collected less frequently than flow data because of the cost to collect and analyze, while water quality data corresponding to flow data are required to compute pollutant loads or to calibrate other hydrology models. Regression models are applicable to interpolate water quality data corresponding to flow data. METHODS AND RESULTS: A regression model was suggested which is capable to consider flow and time variance, and the regression model coefficients were calibrated using various measured water quality data with genetic-algorithm. Both LOADEST and the regression using genetic-algorithm were evaluated by 19 water quality data sets through calibration and validation. The regression model using genetic-algorithm displayed the similar model behaviors to LOADEST. The load estimates by both LOADEST and the regression model using genetic-algorithm indicated that use of a large proportion of water quality data does not necessarily lead to the load estimates with smaller error to measured load. CONCLUSION: Regression models need to be calibrated and validated before they are used to interpolate pollutant loads, as separating water quality data into two data sets for calibration and validation.