Search

Objectives : In this study, we select input factors for machine learning models to predict dissolved oxygen (DO) in Gyeongan Stream and compare results of performance evaluation indicators to find the optimal model.Methods : The water quality data from the specific points of Gyeongan Stream were collected between January 15, 1998 and December 30, 2019. The pretreatment data were divided into train and test data with the ratio of 7:3. We used random forest (RF), artificial neural network (ANN), convolutional neural network (CNN), and gated recurrent unit (GRU) among machine learning. RF and ANN were tested by both random split and time series data, while CNN and GRU conducted the experiment using only time series data. Performance evaluation indicators such as square of the correlation coefficient (R2), root mean square error (RMSE), and mean absolute error (MAE) were used to compare the optimal results for the models.Results and Discussion : Based on the RF variable importance results and references, water temperature, pH, electrical conductivity, PO4-P, NH4-N, total phosphorus, suspended solids, and NO3-N were used as input factors. Both RF and ANN performed better with time series data than random split. The model performance was good in order of RF > CNN > GRU > ANN.Conclusions : The eight input factors (water temperature, pH, electrical conductivity, PO4-P, NH4-N, total phosphorus, suspended solids, and NO3-N) were selected for machine learning models to predict DO in Gyeongan Stream. The best model for DO prediction was the RF model with time series data. Therefore, we suggest that the RF with the eight input factors could be used to predict the DO in streams.

Objective:Coagulation of Moringa oleifera seeds was studied to determine the optimal conditions in treating a turbid water.Method:Jar-test experiments were performed at various rapid mixing (90-250 rpm; 1-2 min), slow mixing (10-60 rpm; 10-50 min), initial turbidity (100-500 NTU), coagulant doses (200-1,500 mg/L), and pH (2-12) in synthetic wastewater. Response surface analysis was used to optimize the coagulation in jar-test experiments.Results and Discussion:The optimal conditions for turbidity removal with M. oleifera were 200 rpm of rapid-mixing speed for 1 min, 40 rpm of slow-mixing speed for 20 min, and pH 8. The coagulation efficiency was found to be dependent on initial turbidity of water samples. Highest turbidity removals (96.5%) were obtained for water with very high initial turbidity (500 NTU). The effects of initial turbidity and coagulant dose were successfully analyzed by the response surface analysis. Within the design boundaries, turbidity removal was more sensitive to changes in initial turbidity than to changes in coagulant dose.Conclusions:The quadratic model between initial turbidity and coagulant for turbidity removal was shown. M. oleifera can be used as a potential coagulant especially for high turbidity water.

Objective : Present research discussed the utilization of pumpkin-seed residue (PSR) after oil extraction with methanol as an adsorbent for methylene blue (MB) removal from aqueous solution.Method : The experiment was carried out to evaluate the influence of PSR adsorbent dose (7.5-25 g/L), initial MB concentration (25-200 mg/L), contact time (30-120 min), pH (3-11), and temperature (293-333 K). Adsorption isotherms were modeled with the Langmuir, Freundlich, and Temkin isotherms. The kinetic data were analyzed using pseudo-first-order and pseudo-second-order models.Results and Discussion : A pseudo-equilibrium state was reached within 30 min of contact time at low initial MB concentration (25-50 mg/L) and 90 min at high concentration (100-200 mg/L). Increasing pH and temperature caused an increase in adsorption capacity. Thermodynamic studies demonstrated that the adsorption process was spontaneous with Gibb’s free-energy values ranging between -15.78 to -13.87 kJ/mol and endothermic with an enthalpy value of 0.011 kJ/mol. The adsorption equilibrium data fitted well with the Freundlich adsorption isotherm. The maximum monolayer adsorption capacity was 20.33 mg/g. Tempkin isotherm model clarified that the heat of sorption process was 6.28 J/mol. The adsorption kinetics was found to follow the pseudo-second order kinetics model and its rate constant was 0.002-0.278 g/mg･min.Conclusions : Findings of the present study indicated that the PSR can be successfully used for removal of MB from aqueous solution. Therefore, the PSR was shown to have good potential as a biosorbent for MB removal.

Coagulation of titanium tetrachloride (TiCl4) and polytitanium tetrachloride (PTC) was studied at various conditions in treating wastewater that contained phosphorus (P). Jar-test experiments were performed at various chemical doses (0.02–0.49 mM), alkalinities (50 ≤ AT ≤ 150 mg CaCO3/L of NaHCO3), pH levels (3.0–8.5), and OH/Ti ratio (0.3 ≤ B ≤ 3.0) to determine the conditions at which P (2 mg/L) removal was most efficient. The TiCl4 concentration ([TiCl4]) required to achieve 98–99% P removal was 0.25 mM at 50 mg CaCO3/L, 0.31 mM at 100 mg CaCO3/L, and 0.49 mM at 150 mg CaCO3/L. Response surface analysis predicted that P removal would reach 100% at controlled pH = 3.5 with minimum [TiCl4] = 0.21 mM. PTC removed P most effectively at B = 0.3, and the optimal dose at B = 0.3 was 0.27 mM. Laboratory results and a pilot experiment will help to optimize use of Ti-based coagulants to remediate wastewater that contains P.

We investigated the effects of initial pH (2≤pH0≤6) and sewage sludge compost content (5≤[SSC]≤25 g/kg) on leaching characteristics of lead (Pb) and zinc (Zn) in contaminated field soil. pH0 and [SSC] significantly affected the leaching of Pb and Zn in soils contaminated with them. The pH in the solution increased as reaction time and [SSC] increased. The leached amounts of Pb and Zn were highest at pH0=2 and increased with reaction time. As [SSC] increased, the leached amount of Pb decreased (50.4 mg/kg at control condition ([SSC]=0 g/kg); 22.9 mg/kg at [SSC]=25 g/kg at pH0=2) and the leached amount of Zn increased (20.1 mg/kg at [SSC]=0 g/kg; 31.7 mg/kg at [SSC]=25 g/kg at pH0=2). The change increased as pH0 decreased. Within the design boundaries, minimum leaching of Pb (14.7 mg/kg) occurred at pH0=5.1 and [SSC]=25 g/kg, and minimum leaching of Zn (5.0 mg/kg) occurred at pH0=5.1 and [SSC]=5 g/kg.

We experimentally determined the effects of microwave-assisted persulfate oxidation (MAPO) on absorbance spectra, specific ultraviolet absorbance at a wavelength of 254 nm (SUVA254), and color in treating landfill leachate. The effects of treatment temperature (60≤T≤90 °C) and sodium persulfate (SPS) concentration (0.05≤[SPS]≤0.3 M) were characterized. An absorbance band in the visible region disappeared after MAPO, but the absorbance in UV region remained. SUVA254 of treated leachate varied from 2.3 to 14.7 L/mg·m. Color number (CN) removal of 99% (CN=0.004 cm−1) was achieved within 30min with MAPO at 80 °C and [SPS]=0.2 M. As T and [SPS] increased, CN and total organic carbon decreased, but SUVA254 increased.

This study evaluates the efficacy of titanium tetrachloride (TiCl 4 ) on phosphorus (P) removal in synthetic wastewater. Jar test experiments were performed at various TiCl 4 concentration (0.25-0.59 mM), and intensities of slow (30-60 rpm) and rapid (100-250 rpm) mixings to determine the conditions at which P removal was most efficient. The P-removal efficiency was highest (about 99%) at TiCl 4 concentration ([TiCl 4 ])=0.39 mM with rapid-mixing intensity=100 rpm and slow-mixing intensity=30 rpm. The slow-mixing intensity was more sensitive than the rapid-mixing intensity to the P removal efficiency when [TiCl 4 ] was low (0.25≤[TiCl 4 ]≤0.27 mM). Key Words : Coagulation, Mixing Intensity, pH, Phosphorus, Titanium Tetrachloride 요약: 본 연구는 TiCl 4 의 농도(0.25-0.59 mM)와 급속교반속도(100-250 rpm), 완속교반속도(30-60 rpm)의 변화가 인 제거효율에 미치는 영향을 알아보고자 하였다. 응집제 농도가 0.25≤[TiCl 4 ]≤0.39 mM 범위에서 인 제거 효율은 TiCl 4 의 농도가 증가할수록 증가하였으며 처리후 인의 농도는 0.2 mg/L 이하였다. 응집제 농도 0.39 mM, 급속교반속도 100 rpm, 완속교반속도 30 rpm에서 인 제거효율이 약 99%로 가장 높았다. TiCl

This study evaluates the efficacy of titanium tetrachloride (TiCl4) and titanium sulfate [Ti(SO4)2] on phosphorus (P) removal in synthetic wastewater compared with aluminum sulfate [Al2(SO4)3]. Jar test experiments were performed at various chemical doses and initial pH (pHi) levels to determine the conditions at which P removal was most efficient. The doses with 99% P removal were 0.31 to 0.34 mM for TiCl4, 0.15 mM for Ti(SO4)2, and 0.19 mM for Al2(SO4)3 as Al at 2 mg P/L, 100 mg CaCO3/L, and pHi = 7.0 to 8.0. The optimum pHi was 7 for TiCl4, 5.5 for Ti(SO4)2, and 7 for Al2(SO4)3. The P-removal efficiencies were Ti(SO4)2 > Al2(SO4)3 > TiCl4 at pHi 5.5, and TiCl4 ≈ Al2(SO4)3 > Ti(SO4)2 at pHi 7 and 8.5. P-removal ability of TiCl4 decreased as alkalinity increased. Quadratic models successfully approximated the response surface for P removal efficiency with simultaneous changes in pHi and chemical concentrations [TiCl4 and Al2(SO4)3].