Neural network prediction of thermophilic (65°C) sulfidogenic fluidized‐bed reactor performance for the treatment of metal‐containing wastewater

Abstract

The performance of a fluidized-bed reactor (FBR) based sulfate reducing bioprocess was predicted using artificial neural network (ANN). The FBR was operated at high (65 degrees C) temperature and it was fed with iron (40-90 mg/L) and sulfate (1,000-1,500 mg/L) containing acidic (pH = 3.5-6) synthetic wastewater. Ethanol was supplemented as carbon and electron source for sulfate reducing bacteria (SRB). The wastewater pH of 4.3-4.4 was neutralized by the alkalinity produced in acetate oxidation and the average effluent pH was 7.8 +/- 0.8. The oxidation of acetate is the rate-limiting step in the sulfidogenic ethanol oxidation by thermophilic SRB, which resulted in acetate accumulation. Sulfate reduction and acetate oxidation rates showed variation depending on the operational conditions with the maximum rates of 1 g/L/d (0.2 g/g volatile solids (VS)/d) and 0.3 g/L/d (0.06 g/g VS/d), respectively. This study presents an ANN model predicting the performance of the reactor and determining the optimal architecture of this model; such as best back-propagation (BP) algorithm and neuron numbers. The Levenberg-Marquardt algorithm was selected as the best of 12 BP algorithms and optimal neuron number was determined as 20. The developed ANN model predicted acetate (R=0.91), sulfate (R=0.95), sulfide (R=0.97), and alkalinity (R=0.94) in the FBR effluent. Hence, the ANN based model can be used to predict the FBR performance, to control the operational conditions for improved process performance.

Artificial Intelligence-Based Prediction Models for Environmental Engineering

“…Results indicated that the developed ANN model showed a satisfactory match between the measured and the predicted concentrations of sulfate (R = 0.998), COD (R = 0.993), acetate (R = 0.976), and zinc (R = 0.827) in the CSTR effluent. Apart from the above-mentioned studies, several other successful ANN modeling studies [64][65][66][67][68][69] have recently been conducted, specifically in various parts of the field of wastewater engineering.…”

Prediction of ferric iron precipitation in bioleaching process using partial least squares and artificial neural network

“…Basic Yellow 28 by electrocoagulation process [18]. Sahinkaya and co-workers developed an artificial neural network model for estimation of the performance of a fluidized-bed reactor (FBR) based sulfate reducing bioprocess and control the operational conditions for improved process performance [19]. Sahinkaya also modeled the biotreatment of zinc-containing wastewater in a sulfidogenic CSTR by using artificial neural network [20].…”

Biologically Fe2+ oxidizing fluidized bed reactor performance and controlling of Fe3+ recycle during heap bioleaching: an artificial neural network-based model

“…The BP network is one of the several networks that is widely used for predicting the output and is successfully applied to a wide range of problems [6,7,9]. The best BP algorithm for the present application, with minimum training error (0.0489), is the Levenberg-Marquardt algorithm ( Table 2).…”

“…So far, there are several applications of ANN models in the engineering area. For example, Strike et al [1] used ANN to model H 2 S and NH 3 components of biogas from anaerobic digestion; Clair and Ehrman [2] used ANN to simulate the effect of climate change on discharge and the export of dissolved organic carbon and nitrogen from river basin; Nunnari et al [3] applied ANN to model air pollution; Cinar [4] used ANN to analyze the system behavior and to determine operational problems of a full-scale activated sludge wastewater treatment plant; Cinar et al [5] used ANN to evaluate the performance of a membrane bioreactor; Karaca and Ö zkaya [6] used ANN to predict the leachate quantity from a full-scale municipal solid waste landfill; Ozkaya et al [7] used ANN to estimate methane fraction in biogas from field-scale landfill; Holubar et al [8] used ANN to predict biogas production and composition; Sahinkaya et al [9] used ANN to determine the performance of high rate sulfidogenic fluidized bed reactor treating acidic metal containing wastewater, neural network as a tool. There are also other similar engineering applications of ANN.…”