Optimization of Biogas Production Process in a Wastewater Treatment Plant

Sludge is a byproduct of wastewater processing suitable for biogas production. The biogas consisting of about 60% methane can be used to generate electricity and heat. A data-driven approach for optimization of biogas production process in a wastewater treatment plant is presented. The process model is developed using routinely collected data categorized as controllable and uncontrollable variables. Process temperature, total solids, volatile solids, and pH value were selected as controllable variables for the model. The uncontrollable variables include sludge flow rate, organic load, and detention time. A multi-layer perceptron neural network is applied to construct the optimization model. Due to high computational complexity of the model, a particle swarm optimization algorithm is used to maximize the biogas production by finding the optimal settings of controllable variables. The model solution has resulted in a 20.8% increase of the biogas production under optimized operation conditions.

[1]  Mauricio G. C. Resende,et al.  Greedy Randomized Adaptive Search Procedures , 1995, J. Glob. Optim..

[2]  Celso C. Ribeiro,et al.  Greedy Randomized Adaptive Search Procedures , 2003, Handbook of Metaheuristics.

[3]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[4]  Andrew Kusiak,et al.  Wind farm power prediction: a data‐mining approach , 2009 .

[5]  Lothar Thiele,et al.  A Preference-Based Evolutionary Algorithm for Multi-Objective Optimization , 2009, Evolutionary Computation.

[6]  Zbigniew Michalewicz,et al.  Evolutionary Algorithms in Engineering Applications , 1997, Springer Berlin Heidelberg.

[7]  O Bernard,et al.  Robust regulation of anaerobic digestion processes. , 2003, Water science and technology : a journal of the International Association on Water Pollution Research.

[8]  Olivia Parr Rud,et al.  Data Mining Cookbook: Modeling Data for Marketing, Risk, and Customer Relationship Management , 2000 .

[9]  Dan Braha.  Data mining for design and manufacturing: methods and applications , 2001 .

[10]  G. Lettinga,et al.  Anaerobic digestion and wastewater treatment systems , 2004, Antonie van Leeuwenhoek.

[11]  Joo-Hwa Tay,et al.  Neural fuzzy modeling of anaerobic biological wastewater treatment systems , 1999 .

[12]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1989, Math. Control. Signals Syst..

[13]  G. A. Ekama,et al.  A steady state model for anaerobic digestion of sewage sludges , 2006 .

[14]  R. Braun,et al.  Advanced controlling of anaerobic digestion by means of hierarchical neural networks. , 2002, Water research.

[15]  H. Siegrist,et al.  The IWA Anaerobic Digestion Model No 1 (ADM1). , 2002, Water science and technology : a journal of the International Association on Water Pollution Research.

[16]  Neslihan Seçkin,et al.  Modeling flood discharge at ungauged sites across Turkey using neuro-fuzzy and neural networks , 2011 .

[17]  DebK.,et al.  A fast and elitist multiobjective genetic algorithm , 2002 .

[18]  Mehmet Cakmakci,et al.  Adaptive neuro-fuzzy modelling of anaerobic digestion of primary sedimentation sludge , 2007, Bioprocess and biosystems engineering.

[19]  Riccardo Poli,et al.  Particle swarm optimization , 1995, Swarm Intelligence.

[20]  Ian H. Witten,et al.  Data mining in bioinformatics using Weka , 2004, Bioinform..

[21]  Rimvydas Simutis,et al.  Bioprocess optimization and control: Application of hybrid modelling , 1994 .