• Journal of Korean Society of Environmental Engineers
• /
• v.35 no.12
• /
• pp.877-882
• /
• 2013

The application of ultrafiltration (UF) and microfiltration (MF) membranes has been increased for drinking water purification. The advantages of UF/MF membrane process compared to conventional treatment processes are stable operation under varying feed water quality, smaller construction area, and automatic operation. Most membrane treatment plants are designed with polymeric membranes. Recently, some studies suggested that the process of treating surface water with ceramic membranes is competitive to the application of polymeric membranes. Higher water flux, less frequent cleaning, and much longer lifetime are the advantages of ceramic membrane comparing to polymeric membrane. Therefore, this research focused on the application of ceramic MF membrane pilot plant at the slow sand filtration plant. The ceramic membrane pilot plant has three trains that used raw water and sand filtered water as a feed water, respectively. For optimizing the pilot plant process, the coagulation with PACl coagulant was used as a pretreatment of ceramic membrane process. In addition, CEB (Chemical Enhanced Backwash) process using $H_2SO_4$ and NaOCl was used for 1.5 days, respectively. The experimental results showed that applying the optimum coagulant dose before membrane filtration showed enhancing membrane fluxes for both raw water and sand filtered water. Also, when using raw water as a feed of membrane, minimum fouling rate was 2.173 kPa/cycle with 25 mg/L of PACl and when using sand filtered water, the minimum fouling rate was 0.301 kPa/cycle with 5 mg/L of PACl.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.47 no.3
• /
• pp.68-74
• /
• 2024

This study develops a model to determine the input rate of the chemical for coagulation and flocculation process (i.e. coagulant) at industrial water treatment plant, based on real-world data. To detect outliers among the collected data, a two-phase algorithm with standardization transformation and Density-Based Spatial Clustering of Applications with Noise (DBSCAN) is applied. In addition, both of the missing data and outliers are revised with linear interpolation. To determine the coagulant rate, various kinds of machine learning models are tested as well as linear regression. Among them, the random forest model with min-max scaled data provides the best performance, whose MSE, MAPE, R² and CVRMSE are 1.136, 0.111, 0.912, and 18.704, respectively. This study demonstrates the practical applicability of machine learning based chemical input decision model, which can lead to a smart management and response systems for clean and safe water treatment plant.

• Journal of Korean Society of Water and Wastewater
• /
• v.21 no.2
• /
• pp.243-252
• /
• 2007

Effects of alum dosage on the particle growth were investigated by monitoring particle counts in a rapid mixing process. Kaolin was used for turbid water sample and several other chemicals were added to adjust pH and ionic strength. The range of velocity gradient and mixing time applied for rapid mixing were $200{\sim}300sec^{-1}$ and 30~180 sec, respectively. Particle distribution in the synthetic water sample was close to the natural water where their turbidity was same. The number of particles in the range of $10.0{\sim}12.0{\mu}m$ increased rapidly with rapid mixing time at alum dose of 20mg/L, however, the number of $8.0{\sim}9.0{\mu}m$ particles increased at alum dose of 50mg/L. The number of $14.0{\sim}25.0{\mu}m$ particles at alum dose of 20mg/L was 10 times higher than them at alum dose of 50mg/L. Dominant particle growth was monitored at the lower alum dose than the optimum dose from a jar test at an extended rapid mixing time(about 120 sec). The number of $8.0{\sim}14.0{\mu}m$ particles was lower both at a higher alum doses and higher G values. At G value of $200sec^{-1}$ and at alum dose of 10-20mg/L, residual turbidity was lower as the mixing time increased. But at alum dose above 40mg/L and at same G value, lower residual turbidity occurred in a short rapid mixing time. Low residual turbidity at G value of $300sec^{-1}$ occurred both at lower alum doses and at shorter mixing time comparing to the results at G value of $200sec^{-1}$.

• Journal of Korean Society of Environmental Engineers
• /
• v.38 no.8
• /
• pp.420-427
• /
• 2016

This study was conducted on the sediment remediation using micro-bubble to remove fine particles. For this study, characteristics of contamination and release in sediment were analyzed. And then, the characteristics of bubbles on removal efficiency was investigated at various operation conditions. In particle size distribution of the sediment used for the study, the proportion of clay and silt (<0.075 mm) was about 7.7%, sand (0.075~4.75 mm) was about 67.8%, and gravel (${\geq}4.75$) was 24.5%. Total nitrogen (TN) and total phosphorus (TP) of the sediment were 2,790~3,260, 261~311 mg/kg respectively. Ignition loss and water content were 4.1~9.6, 32.9~53.2% respectively. In analysis of removal efficiency according to operation conditions of micro-bubble, it was the highest when operation condition is pressure 6 atm, pressurized water ratio 30%, and coagulant dosage 15 ppm. At the time, the sediment's removal efficiency was 19.9%. Accordingly removal efficiency of TN and TP were 21.4, 22.6% respectively. Finally a research was found that fine particles in sediment were almost removed by micro-bubble, which led to decrease nutrients' release at about 20.1~64.3% in comparison to sediment including lots of fine particles.