Terng-Jou Wan

Using fuzzy inference system to improve neural network for predicting hospital wastewater treatment plant effluent

Abstract In this study, three types of adaptive neuro fuzzy inference system (ANFIS) and artificial neural network (ANN) were employed to predict suspended solids (SS eff ) and chemical oxygen demand (COD eff ) in the effluent from a hospital wastewater treatment plant. The results indicated that ANFIS statistically outperforms ANN in terms of effluent prediction. The minimum mean absolute percentage errors of 11.99% and 12.75% for SS eff and COD eff could be achieved using ANFIS. The maximum values of correlation coefficient for SS eff and COD eff were 0.75 and 0.92, respectively. The minimum mean square errors of 0.17 and 19.58, and the minimum root mean square errors of 0.41 and 4.42 for SS eff and COD eff could also be achieved. ANFISs architecture consists of both ANN and fuzzy logic including linguistic expression of membership functions and if–then rules, so it can overcome the limitations of traditional neural network and increase the prediction performance.

Evaluation of five antibiotic resistance genes in wastewater treatment systems of swine farms by real-time PCR

Antibiotics are widely used in livestock for infection treatment and growth promotion. Wastes from animal husbandry are a potential environmental source of antibiotic-insensitive pathogens, and the removal efficiency of the resistance genotypes in current wastewater treatment plants (WWTPs) is unknown. In this study, quantitative PCR was used for evaluating antibiotic resistance genes in wastewater treatment processes. Six wastewater treatment plants in different swine farms were included in this study, and five antibiotic resistance genes (ARGs) were tested for each treatment procedure. All of the tested ARGs including tetA, tetW, sulI, sulII, and blaTEM genes were detected in six swine farms with considerable amounts. The results showed that antibiotic resistance is prevalent in livestock farming. The ARG levels were varied by wastewater treatment procedure, frequently with the highest level at anaerobic treatment tank and lowest in the activated sludge unit and the effluents. After normalizing the ARG levels to 16S rRNA gene copies, the results showed that ARGs in WWTP units fluctuated partly with the quantity of bacteria. Regardless of its importance in biodegradation, the anaerobic procedure may facilitate bacterial growth thus increasing the sustainability of the antibiotic resistance genotypes. After comparing the copy numbers in influx and efflux samples, the mean removal efficiency of ARGs ranged between 33.30 and 97.56%. The results suggested that treatments in the WWTP could partially reduce the spread of antibiotic-resistant bacteria, and additional procedures such as sedimentation may not critically affect the removal efficiency.

Comparisons of grey and neural network prediction of industrial park wastewater effluent using influent quality and online monitoring parameters

In this study, Grey model (GM) and artificial neural network (ANN) were employed to predict suspended solids (SSeff) and chemical oxygen demand (CODeff) in the effluent from a wastewater treatment plant in industrial park of Taiwan. When constructing model or predicting, the influent quality or online monitoring parameters were adopted as the input variables. ANN was also adopted for comparison. The results indicated that the minimum MAPEs of 16.13 and 9.85% for SSeff and CODeff could be achieved using GMs when online monitoring parameters were taken as the input variables. Although a good fitness could be achieved using ANN, they required a large quantity of data. Contrarily, GM only required a small amount of data (at least four data) and the prediction results were even better than those of ANN. Therefore, GM could be applied successfully in predicting effluent when the information was not sufficient. The results also indicated that these simple online monitoring parameters could be applied on prediction of effluent quality well.

Using magnetic seeds to improve the aggregation and precipitation of nanoparticles from backside grinding wastewater

Backside grinding (BG) wastewater treatment typically requires large quantities of chemicals, i.e. polyaluminum chloride (PAC) coagulant and produces considerable amounts of sludge, increasing the loading and cost of subsequent sludge treatment and disposal processes. This study investigated the effects of the addition of magnetic seeds (FeO*Fe(2)O(3)) of selected particle sizes and of optimized combinations of magnetic seeds and PAC on the aggregation of silica nanoparticles from BG wastewater and on the sedimentation time at various pH values. The results show that the turbidity of BG wastewater was significantly reduced by the magnetic aggregation treatment. The dosage of PAC combined with 2.49gL(-1) or 1.24gL(-1) of magnetic seeds was reduced by 83% (from 60 to 10mgL(-1)) compared to the conventional process of using only PAC as a coagulant. The turbidity of the BG wastewater, initially 1900-2500NTU, could also be successfully decreased about to 23NTU by the addition of 3.74gL(-1) magnetite (FeO*Fe(2)O(3)) only at pH 5 with an applied magnetic field of 1000G. Different coagulation conditions using magnetic seeds combined with coagulant resulted in different aggregation performances. The treatment performance was more effective by using two-stage dosing, in which magnetic seeds and PAC were added separately, than that with one-stage dosing, where the magnetic seeds and PAC were added simultaneously during rapid mixing. The two-stage dosing allowed for a reduction in the optimum dosage of magnetic seeds from 3.74gL(-1) to 2.49gL(-1) or 1.24gL(-1) without affecting performance when coupled with 0.01gL(-1) of PAC coagulant. The developed method effectively reduced the production of waste sludge.