Artificial Neural Network Modelling of Sequencing Batch Reactor Performance

Artificial Neural Network Modelling of Sequencing Batch Reactor Performance

Eldon R. Rene (University of La Coruña, Spain), Sung Joo Kim (University of Ulsan, South Korea), Dae Hee Lee (University of Ulsan, South Korea), Woo Bong Je (University of Ulsan, South Korea), Mirian Estefanía López (University of La Coruña, Spain) and Hung Suck Park (University of Ulsan, South Korea)
DOI: 10.4018/978-1-61350-116-0.ch019


Sequencing batch reactor (SBR) is a versatile, eco-friendly, and cost-saving process for the biological treatment of nutrient-rich wastewater, at varying loading rates. The performance of a laboratory-scale SBR was monitored to ascertain the chemical oxygen demand (COD) and total nitrogen (T-N) removals under four different operating conditions, by varying the operating time for the nitrification/denitrification steps, i.e., the cycle times. A multi-layered neural network was developed using COD, T-N, carbon to nitrogen ratio (C/N), aeration time, and mixed liquor suspended solids concentration (MLSS) data. This chapter compares the neural simulation results to the experimental results and extracts information on the significant factors affecting SBR performance. The application of artificial neural networks to biological processes such as SBR is a relatively new technique in wastewater and water quality management, and the results presented herein indicate the promising start of the adoption of computational science in this domain of research.
Chapter Preview


Background Information on Saline Wastewater

Water quality is continuously degrading as a result of the free flow of contaminants like suspended solids, organic wastes, bacteria, nitrates, phosphates and other recalcitrant chemicals. Considering a general case, as point source pollution, excessive amount of nitrates bearing wastewater are generated from pulp and paper, ammunition manufacturing facilities, besides being generated enormously as non-point source run-offs from the over application of nitrogen rich fertilizers (Metcalf and Eddy, 2003). When ingested, nitrate is converted to nitrite, which then reacts with haemoglobin in blood to form methemoglobin, a condition that severely affects the oxygen transport mechanism in blood.

Physical, chemical, and biological treatment systems have been successfully implemented in industrial facilities for environmental management, as an end-of-pipe treatment step, i.e., before ultimate discharge. The implementation and use of biological techniques, viz., aerobic and anaerobic, for treating toxic chemicals present in wastewater is exponentially growing at the industrial level, especially considering the degree of treatment accomplished at a low cost. Though biological systems have been used efficiently for handling high-strength wastewaters, there is very little information available about their effectiveness for handling saline wastewaters, especially those arising from fish cleaning industries (Soto et al., 2002; Chowdhury et al., 2010).

This study aims to address this issue, by proposing a sequencing batch reactor as the biological treatment step for the treatment of liquid waste generated from a fish market zone near the harbour in Ulsan city, South Korea. The saline rich wastewater generates mainly from fish cleaning, floor washing, raw material washing and from melted ice packs. Additionally, the characteristics of wastewater also depend on the seasons and the different varieties of fish caught over a particular period. This wastewater is likely to contain low levels of soluble protein, fish scales and high salt content. Generally, the most commonly used approach to discard this type of wastewater is to mix it with the municipal sewerage system, before treatment in a centralized wastewater treatment plant. However, discharging these wastes is not a simple task, as the magnitude depends principally on the waste volume, rate of discharge and the assimilatory capacity of the receiving medium. High salt content may pose severe adverse effects during combined treatment of these liquid wastes. High salt content (>1%) could cause plasmolysis and loss of activity of the biological cells and affects system performance (Uygur and Kargi, 2004). The various physical and chemical parameters such as solid content, pH, temperature, odor, organic matter such as biochemical oxygen demand, chemical oxygen demand, oil and grease content, nitrogen and phosphorous content should be evaluated before choosing an appropriate treatment system.

This chapter provides the reader an useful insight into the process fundamentals of sequencing batch reactor operation, and links the application of artificial intelligence, i.e., artificial neural networks, to model biological wastewater treatment process. It provides a description of neural networks, the importance of network parameters for neural modeling and sensitivity analysis. Anew, an explanation of the methodology adopted to collect the experimental data from the sequencing batch reactor, data analysis, and neural modeling results are also discussed from a practical view-point.


Sequencing Batch Reactor Operation

A sequencing batch reactor can be used to attain high treatment efficiencies, as secondary treatment systems, for simultaneous nitrification and denitrification in one reactor configuration, operated under different cycle times. From an application view-point, chemical oxygen demand, phosphate and nitrogen removal can be easily, yet collectively be achieved for both industrial and domestic wastewaters using sequencing batch reactors. A large number of studies have also been carried out using sequencing batch reactors to identify the possible mechanisms of organic and nitrogen removal under low/high saline conditions.

Key Terms in this Chapter

Sequencing Batch Reactor: A biological treatment system (bioreactor) having similar operational characteristics to that of an activated sludge process in environmental engineering applications. They are designed to operate under non steady state conditions, in a batch mode, with aeration and sludge settlement both occurring in the same reactor.

Nitrification: The process of biological conversion of ammonium to nitrate nitrogen, commonly used in wastewater engineering. Nitrification is a two step process. In the first step, Nitrosomonas convert ammonia and ammonium to nitrite, while in the second step, another bacterium, Nitrobacter converts nitrite to nitrate.

Back Propagation Learning: This learning rule is based on gradient descent. The connection weights between different layers are initialized with pseudo-random values and are changed in a direction that will reduce the error. When a learning pattern is clamped, the activation values are propagated to the output units, and the actual network output is compared with the desired output values.

Multilayer Perceptrons (MLPs): MLPs are feed forward neural networks trained with the standard back propagation algorithm. They are supervised networks, and hence they require a desired response to be trained. By proper training, they can transform input data into a desired response, so they are widely used for pattern classification and prediction purposes. With one or two hidden layers, they can approximate, virtually, any input output map.

Denitrification: The process of biological reduction of nitrate (NO3) to nitrogen gas (N2) by facultative heterotrophic bacteria. The facultative bacteria utilize oxygen by taking dissolved oxygen out of the water or by taking it off of nitrate molecules.

Artificial Neural Network: An artificial neural network is a system based on the functioning of biological neural networks, an emulation of the biological neural system.

Complete Chapter List

Search this Book: