Neural Network-based Model Predictive Control with CPSOGSA for SMBR Filtration

This paper presents the development of neural network based model predictive control (NNMPC) for controlling submerged membrane bioreactor (SMBR) filtration process.The main contribution of this paper is the integration of newly developed soft computing optimization technique name as cooperative hybrid particle swarm optimization and gravitational search algorithm (CPSOGSA) with the model predictive control. The CPSOGSA algorithm is used as a real time optimization (RTO) in updating the NNMPC cost function. The developed controller is utilized to control SMBR filtrations permeate flux in preventing flux decline from membrane fouling. The proposed NNMPC is comparedwith proportional integral derivative (PID) controller in term of the percentage overshoot, settling time and integral absolute error (IAE) criteria. The simulation result shows NNMPC perform better control compared with PID controller in term measured control performance of permeate flux.

[1]  José Ferrer,et al.  Instrumentation, control, and automation for submerged anaerobic membrane bioreactors , 2015, Environmental technology.

[2]  Junfei Qiao,et al.  Model predictive control of dissolved oxygen concentration based on a self-organizing RBF neural network , 2012 .

[3]  Vincenza Calabrò,et al.  Design and tuning of feedback controllers: effects on proteins ultrafiltration process modeled by a hybrid system , 2011 .

[4]  Hossein Nezamabadi-pour,et al.  GSA: A Gravitational Search Algorithm , 2009, Inf. Sci..

[5]  Wolfgang Marquardt,et al.  Run‐to‐run control of membrane filtration processes , 2007 .

[6]  Q. Henry Wu,et al.  MCPSO: A multi-swarm cooperative particle swarm optimizer , 2007, Appl. Math. Comput..

[7]  Sanjay Sharma,et al.  A wavelet neural network based non-linear model predictive controller for a multi-variable coupled tank system , 2015, Int. J. Autom. Comput..

[8]  Ignasi Rodríguez-Roda,et al.  Automatic control systems for submerged membrane bioreactors: a state-of-the-art review. , 2012, Water research.

[9]  Junzhi Yu,et al.  Neural-Network-Based Nonlinear Model Predictive Control for Piezoelectric Actuators , 2015, IEEE Transactions on Industrial Electronics.

[10]  Pierre Le-Clech,et al.  Fouling in membrane bioreactors used in wastewater treatment , 2006 .

[11]  Shafishuhaza Sahlan,et al.  Fouling control strategy for submerged membrane bioreactor filtration processes using aeration airflow, backwash, and relaxation: a review , 2016 .