We applied multilayer perceptron (MLP) and radial basis function (RBF) neural network in upstream and downstream water quality stations of the Karaj Reservoir in Iran. For both neural networks, inputs were pH, turbidity, temperature, chlorophyll-a, biochemical oxygen demand (BOD) and nitrate, and
the output was dissolved oxygen (DO). We used an MLP neural network with two hidden layers, for upstream station 15 and 33 neurons in the first and second layers respectively, and for the downstream station, 16 and 21 neurons in the first and second hidden layer were used which had minimum amount of errors. For learning process 6-fold cross validation were applied to avoid over fitting. The best results acquired from RBF model, in which the mean bias error (MBE) and root mean squared error (RMSE) were 0.063 and 0.10 for the upstream station. The MBE and RSME were 0.0126 and 0.099 for the downstream station. The coefficient of determination (R2) between the observed data and the predicted data for upstream
and downstream stations in the MLP was 0.801 and 0.904, respectively, and in the RBF network were 0.962 and 0.97, respectively. The MLP neural network had acceptable results; however, the results of RBF network were more accurate. A sensitivity analysis for the MLP neural network indicated that temperature was the first parameter, pH the second and nitrate was the last factor affecting the prediction of DO concentrations. The results proved the workability and accuracy of the RBF model in the prediction of the DO.
water quality prediction; the MLP neural network; the RBF neural network; dissolved oxygen; Amir Kabir Reservoir
Civil Engineering Department at Kharazmi University, Mofateh St., Tehran, Iran
This study aimed at determining the treatability of high-strength wastewater (chemical oxygen demand, COD>4000 mg/L) using combined anaerobic-aerobic granular sludge in lagoon systems. The lagoon systems were simulated in laboratory-scale aerated and non-aerated batch processes inoculated with dried granular microorganisms at a dose of 0.4 g/L. In the anaerobic batch, a removal efficiency of 25% was not attained until the 12th day. It took 14 days of aerobic operation to achieve sCOD removal efficiency of 94% at COD:N:P of 100:4:1. The best removal efficiency of sCOD (96%) was achieved in the sequential
anaerobic-aerobic batch of 12 days and 2 days, respectively at COD:N:P ratio of 200:4:1. Sequential anaerobic-aerobic treatment can achieve efficient and cost effective treatment for high-strength wastewater in lagoon systems.
This study evaluated some markers of oxidative stress in the organs of African Catfish, Clarias gariepinus from Eleyele River in Oyo State, Nigeria. Clarias gariepinus (250 g-400 g) were collected from Eleyele River (a suspected polluted River) and Clarias gariepinus from a clean fish farm (Durantee fisheries) were used as the control. Levels of Malondialdehyde (index of lipid peroxidation), Glutathione (GSH) and activities of antioxidant enzymes- Superoxide dismutase (SOD), Catalase and Glutathione-STransferase (GST) were evaluated in the liver, kidney and gills of the fish. From the results, there were
significant (p<0.001) increases in malondialdehyde and GSH levels in the liver, kidney and gills of Clarias gariepinus from Eleyele River compared with control. The activity of GST increased significantly (p<0.05; p<0.001) in the liver and kidney of fish from Eleyele River compared with control. There was a significant decrease (p<0.05; p<0.001) in SOD activity in all the organs of Clarias gariepinus from Eleyele River compared with conrol and also a significant (p<0.001) decrease in catalase activity in the gills and kidney of the fish but catalase activity increased in the liver. Increase in lipid peroxidation and alterations in antioxidant status in Clarias gariepinus from Eleyele River show that the fish were under oxidative stress. These suggest that the River is polluted probably as a result of various wastes frequently discharged into the River. This could pose serious health risks to consumers of water and aquatic organisms from the River.
This study aimed at advanced oxidation of hetero tri-functional reactive dye Acid orange 7 using photo-Fenton conditions in a lab-scale experiment. Decolourisation of Acid Orange 7 dye by Fenton´s process was dependent on concentration of Hydrogen peroxide, Ferrous sulphate, pH, and contact time. A 23
factorial design was used to evaluate the effects of these key factors: pH, Fe(II), and H2O2 concentration, for a dye concentration of 50 mg/L with COD of 340 mg/L at pH 3.0. The response function was removal of colour under optimised conditions; pH 3.0, [Fe(II)] 40.83 mg/L, [H2O2] 4.97 mmol/L; 13.6 min. of treatment resulting in 100% colour removal. The final COD of treated wastewater was nil suggesting that AOP is a potentially useful process of color removal and dye degradation/mineralisation of effluent having AO7. Minimum contact time for complete decolourisation was at 5 mmol/l H2O2 concentration. Increase in FeSO4 (mg/l) concentration resulted in decrease of time for complete decolourisation. Box-Behnken Design was used to optimize the process variables. Maximum and minimum levels of pH (3-5), H2O2 (4-6 mmol/l), FeSO4 (30-46 mg/l) and contact time (5-15 minutes) were used. The statistical analysis revealed a value of 0.88 for coefficient of regression (R2) indicating a good fit of model. Calculated F-value was found higher than the tabulated value confirming to significance of the model. Based on student´s t-test, Ferrous sulphate, pH, and contact time have a positive effect on the percent decolourisation of Acid Orange 7.
Laboratory tests were conducted to evaluate the performance of waste rubber shreds in leachate collection layer of engineered landfills. The study found that waste rubber shreds layer in combination with a gravel layer can be of potential use in landfill drainage system. To study the performance, conventional gravel along with waste rubber shreds were used in different combinations (with total layer thickness = 500 mm) as leachate collection media. For the laboratory study poly vinyl chloride (PVC) pipes were used. The size range of waste rubber shreds used were 25 mm to 75 mm in length and width = 10 to 20 mm. The gravel size used in the leachate collection media is 10 mm to 20 mm size. Performance study of 7 Test Cols. with different combinations of waste rubber shreds and gravel bed thickness were studied to find out the best combination. The study found that the Test Col.-3 having waste rubber shreds thickness = 200 mm and gravel layer thickness = 300 mm gave the best results in terms of percentage removal in various physicochemical parameters present in the leachate. Further to find the best size rubber shreds three more Test Cols - 8, 9 and 10 were constructed having the rubber shreds and gravel layer ratio same as that of Test Col.-3 but having rubber shreds width = 10 mm, 15 mm and 20 mm respectively. Based on the results obtained using Test Cols. 8, 9 and 10 the study found that smaller size rubber shreds gave bests results in terms of improvement in various leachate parameters.