Premium
Multi‐objective optimisation for design and operation of anaerobic digestion using GA‐ANN and NSGA‐II
Author(s) -
Huang Mingzhi,
Han Wei,
Wan Jinquan,
Ma Yongwen,
Chen Xiaohong
Publication year - 2016
Publication title -
journal of chemical technology and biotechnology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.64
H-Index - 117
eISSN - 1097-4660
pISSN - 0268-2575
DOI - 10.1002/jctb.4568
Subject(s) - mean squared error , biogas , bioreactor , artificial neural network , genetic algorithm , anaerobic digestion , sorting , effluent , biogas production , process engineering , computer science , mathematics , engineering , environmental engineering , chemistry , mathematical optimization , waste management , algorithm , statistics , methane , artificial intelligence , organic chemistry
BACKGROUND Due to the complexity of nonlinear and biochemical phenomena involved in anaerobic wastewater treatment units, efficient operation and control is limited and difficult. The objective of this study was to implement a new multi‐objective control strategy to simultaneously optimise the effluent chemical oxygen demand ( COD eff ) and the biogas flow rate ( Q gas ) in an anaerobic bioreactor using non‐dominated sorting genetic algorithms‐ II ( NSGA‐II ) and genetic algorithm–artificial neural network ( GA‐ANN ). The novel approach considered two operational objectives, i.e. control of the effluent quality and control of the maximum production rate of biogas, and took advantage of the difference between the dynamics of the liquid and gas phases using variables from both phases. RESULTS Based on the relatively lower values for the mean square error, the root mean square normalised error, and the mean absolute percentage error, as well as higher values for the correlation coefficient, GA‐ANN models achieve better performances than ANN models. Based on GA‐ANN and NSGA‐II , a more flexible and precise optimisation process was allowed, and an optimal balance between effluent quality and biogas flow rate of the anaerobic bioreactor can be achieved. CONCLUSION The proposed GA‐ANN models were shown to be capable of dynamically predicting COD eff and Q gas . Meanwhile the optimisation system developed may offer a useful tool for simulation, design, control and optimisation of anaerobic biodigesters. © 2014 Society of Chemical Industry