Modelling a rotating biological contactor treating heavy metal contaminated wastewater using artificial neural network

The performance of a continuously operated laboratory-scale rotating biological contactor (RBC) was assessed for the removal of heavy metals viz. Cu(II), Cd(II) and Pb(II) from synthetic wastewater using artificial neural networks (ANNs). The RBC was inoculated with Sulfate Reducing Bacteria consortium (predominantly Desulfovibrio species), and the performance was evaluated at different hydraulic retention times (HRTs) and inlet heavy metal concentrations. A feed-forward back-propagation neural network model was developed using 90 data sets obtained over a period of three months, to predict the removal of heavy metal (HMRE) and COD (CODRE). The predictive capability of the model was evaluated in terms of the coefficient of determination (R) and mean absolute percentage error between the model fitted and actual experimental data, whereas sensitivity analysis was performed on the input parameters by determining the absolute average sensitivity (AAS) values. The higher AAS value of the HRT compared with that of inlet heavy metal concentration suggested that the change of HRT has a significant influence on HMRE and CODRE. Overall, the results obtained from this study demonstrated that ANNs can efficiently predict RBC behaviour with regard to heavy metal and COD removal characteristics under the prevailing operational conditions.


GRAPHICAL ABSTRACT INTRODUCTION
Heavy metals are released into the environment from different industries, such as metallurgy, tanneries, mining, electroplating industries, etc. (Kikot et al. ). Heavy metals such as Cu, Cd, and Pb discharged from industrial wastewater are toxic at high concentration and, thus, pose serious risk to both human health and the environment. Hence, removal of heavy metals from wastewater before their discharge into the environment is mandatory (Kiran et al. ).
Compared with the physico-chemical methods, biological methods have been proven to be cost-effective and environmentally friendly technology for heavy metal removal from wastewater (Bai et al. ; Yang et al. a, b; Yajun et al. ). Different kinds of bioreactor systems have been employed to treat heavy metals present in wastewater. For example, wastewater containing heavy metals (e.g. Cd, Cu, Cr, Zn, Pb and Ni) were treated in suspended growth bioreactors, viz. Continuously Stirred Tank Reactors (Gola et al. ), and Membrane Bioreactor employed for the treatment of textile industry wastewater containing chromium (VI). On the other hand, attached growth bioreactors, viz. Packed Bed Bioreactor was employed for the treatment of acid mine drainage containing a variety of heavy metals (Dev et al. ), and Rotating Biological Contactors (RBCs) for heavy metal removal from synthetic wastewater (Kiran et al. a).
In the recent past, RBCs have received great attention due to their capability to treat various types of refractory wastewaters, e.g., coloured industry wastewater (Pakshirajan & Kheria ), agrochemical industry wastewater (Vasiliadou et al. ) and gold mine wastewater (Guadalima & Monteros ). The advantages of RBCs over other attached growth systems include but are not limited to: (1) handling specific contaminants viz. hydrocarbons, heavy metals, xenobiotics and pharmaceuticals/personal care products, (2) handling high organic loadings and resistance to toxic shocks, (3) requiring low land area, maintenance, energy and start-up costs, and (4) energy generation along with wastewater treatment (Hassard et al. ).
Recently, the application of artificial neural networks (ANNs) to model and predict the operational efficiency of various biological systems, including wastewater and waste gas treatment systems, has gained remarkable attention (Nair et  ). The primary advantage of ANN over phenomenological/conceptual models is that it does not require information about the complex nature of the underlying process to be explicitly described in mathematical form (Sahoo et al. ). The ANN model, in this study, was developed using the most widely employed feed-forward back-propagation (BP) algorithm, which uses a gradient descent procedure to minimize the objective function (Rumelhart et al. ). In other words, in the BP algorithm, the weights between input, output and hidden layers are modified and the process is repeated until the error between the output of the neural network and the desired output is minimized.
López et al. () used a two-stage biological waste gas treatment system consisting of a first-stage biotrickling filter (BTF) and second-stage biofilter (BF) for the removal of a gas-phase methanol (M), hydrogen sulphide (HS) and αpinene (P) mixture. The performance of the gas treatment system was modelled using two multi-layer perceptrons (MLPs) employing the back-propagation algorithm, in order to predict the removal efficiencies of methanol (RE M ), hydrogen sulphide (RE HS ) and α-pinene (RE P ). An MLP with the topology 3-4-2 was able to predict RE M and RE HS in the BTF, while a topology of 3-3-1 was able to approximate the RE P in the BF. In another study, a three- The ANN model used a three-layer feed-forward-based Levenberg-Marquardt algorithm with a topology 3-25-1 to predict the TCE removal with R 2 greater than 0.99 during the model training, validation, and testing. The performance of a laboratory-scale anaerobic bioreactor was modelled by using the feed-forward back-propagation algorithm to optimize the CH 4 content of 60%-70% in biogas obtained from digestion of the organic fraction of municipal solid waste (Nair et al. ). As evident from the aforementioned reports, ANNs have shown their capability to accurately predict the behaviour of various biological systems by establishing the correspondence between input and output domains. Thus, it is clear that ANN models can be developed for bioreactors with a definite objective and adequate training of time-series data collected from such reactors. However, modelling the performance of RBC or any other reactor for heavy metal removal from wastewater has not been reported so far. This is mainly important considering the non-biodegradable and toxic nature of heavy metals in wastewater such as acid mine drainage (AMD), characterized by its low pH and high sulphate content along with heavy metals.
Hence, this study focused on the feasibility of an ANN model to estimate, predict and simulate experimental results obtained from an RBC treating heavy-metal-contaminatedsynthetic wastewater. The objectives of this study were formulated as follows: (i) to create an MLP, by varying the internal network parameters, that would predict the HM RE in the RBC used for heavy metal removal from synthetic wastewater (Kiran et al. a), (ii) testing the developed model with data that was not presented to the ANN during training, (iii) to perform sensitivity analysis, analyze and determine the most influencing input parameters (HRT and inlet heavy metal concentrations) for each output (HM RE and COD RE ), and (iv) to emphasize the application of ANNs for control of bioreactor input parameters and to address the potential advantages of ANNs for predicting bioreactor performance. To the best of the authors' knowledge, this is the first report that has used ANN to model an RBC treating heavy-metal-contaminated wastewater.

Microorganism and chemicals
Mixed Sulphate Reducing Bacteria (SRB) consortium (predominantly consisting of Desulfovibrio species) used in the RBC was acquired from a laboratory-scale upflow anaerobic packed bed reactor treating sulphate-rich wastewater (Kiran et al. ). During the experiments, sulphate and COD concentration in the influent were adjusted to maintain a COD/SO 2À 4 ratio of 0.67 ± 0.08. The pH of the solution was adjusted to 7 using 1 N NaOH. The stock solutions of Cd(II), Cu(II), and Pb(II) prepared using Cd(NO 3 ) 2 ·4H 2 O, CuCl 2 ·2H 2 O, and PbNO 3 , respectively were procured from Alpha Chemika Co. Ltd, India, Euclid Co. Ltd, India, and Karni Chemicals Co. Ltd, India. All the chemicals used for experiments were of reagent grade and were used without further purification (Kiran et al. a).

Reactor set-up and operation
The operational protocol of the laboratory-scale anaerobic RBC (total working volume ¼ 3 L) made up of polymethyl methacrylate material is described elsewhere (Pakshirajan & Kheria ). The RBC consists of two identical stages connected in series. Each stage had a working volume of 1.5 L and was equipped with seven discs enclosed with polystyrene mesh and polyurethane foam in which the SRB consortium was immobilized. The disc diameter was 0.16 m with a thickness of 0.0056 m and they were spaced at 0.02 m distance.
The submerged surface was 40% (Kiran et al. a).
The RBC was operated under continuous mode and was maintained at a temperature of 25 ± 2 C. The suspended biomass was measured in terms of mixed liquor volatile suspended solids. Samples collected at regular intervals were centrifuged at 8000 × g for 5 min, and the supernatant obtained was analysed for metal, sulphate, sulphide and COD concentrations (Kiran et al. a). The metal stock solutions of Cd(II), Cu(II), and Pb(II) of 10,000 mg/l concentration each were prepared using Cd(NO 3 ) 2 ·4H 2 O, CuCl 2 ·2H 2 O, and PbNO 3 , respectively. The phase-wise (three phases) inlet concentrations of the metals Cd(II) and Pb(II) were chosen as 50, 75 and 90 mg/l, whereas for Cu(II), the concentrations were 100, 150 and 175 mg/l. All these inlet metal concentrations were chosen based on the studies conducted earlier using the same anaerobic biomass containing SRB (Kiran et al. , b). Reactor performance in terms of RE HM was evaluated at two different HRTs (24 and 48 h). Each experiment was carried out for a period until three steady-state values of effluent heavy metal concentration at the respective HRT were obtained.
All the results presented are averages of duplicate sample analyses. The combined effect of inlet metal and COD concentration and HRT on removal was examined by calculating the RE as given in Equation (1): where C i and C o are the inlet and outlet concentrations (mg/l), respectively.

Analytical methods
The heavy metal concentration in the samples was determined using an atomic absorption spectrometer (Varian, AA240, The Netherlands). The COD and sulphate concentration was determined by following the closed reflux method and standard turbidimetric method, respectively (APHA ). The dissolved sulphide concentration in the liquid samples was determined as per the method described elsewhere (Cord-Ruwisch ).

ANN model development
The  output layer (linear): where X ∧ is the normalized value, and X min and X max are the minimum and maximum values of X respectively.
From the steady-state RBC operation data, the set of 90 data points were divided into training and testing sets; 70% (N Tr : 63) of the data points were used for training the network while the remaining 30% (N Te : 27) were used for testing the developed model. An MLP was formulated to predict the output parameters, HM RE (%) and COD RE (%) using inlet concentration and HRT as the input parameters.   Table 2. The best network topology for the RBC was found to be 2-12-2 (i.e. two input parameters, 12 neurons in one hidden layer and two output parameters). After obtaining the best network topology for the RBC, the connection weights and  (Table 3).

RESULTS AND DISCUSSION
These connection weights determine which input neuron dominates the contribution to a specific hidden neuron, while the sign (þ, -) suggests the nature of correlation between an input to a neuron and the output from the neuron. Detailed information on the interpretation of connection weights for neural network models have been discussed in Garson (). The biases, which are essentially constant, are an additional input into the next layer and are not influenced by the previous layer but they do have outgoing connections with their own weights. The bias unit ensures that even when all the inputs are zeros, there will still be an activation in the neuron.    Cd (Figure 7(b)), and (iii) the allowable inlet heavy metal concentration can be up to 80 mg/l and HRT should be 48 h for Pb (Figure 7(c)).

Predictive capability of the developed model
Metal removal and COD removal were observed to be the maximum at HRT of 48 h rather than at 24 h. This is because a long HRT allows the SRB to reduce sulphate to a greater extent by efficient utilization of the carbon source, thereby precipitating the metals as their respective sulphide salts. Compared with cadmium and lead, the higher RE of copper at a higher inlet concentration range can be attributed to its low solubility product value with sulphide, i.e., copper sulphide is least soluble compared with cadmium sulphide and lead sulphide (Kiran et al. a). At a low initial metal concentration, precipitation of the insoluble metal due to sulphide produced by the SRB avoids any toxic effect of the metal on the SRB. However, high initial metal concentration level has resulted in a reduced activity of the SRB, leading to low sulphate reduction efficiency and metal removal (Kiran et al. b). At a low concentration combination of these metals, the removal of all the metals was the maximum,