论文信息 - Wastewater neutralisation control using a neural network based model predictive controller

Wastewater neutralisation control using a neural network based model predictive controller

The purpose of neutralization is to adjust the pH value to meet the requirements of the different processing units in the wastewater treatment system. However, there are many control problems in the industrial wastewater neutralization because of the nonlinearity of the titration curve, variable buffering, and changes in loading. In order to solve these problems, many authors have proposed different solutions. One of the alternative control schemes for the pH control of chemical process is the use of model predictive control (MPC). A neural network based MPC (NNMPC) is developed in this paper. In order to study the effectiveness of NNMPC, a simulation of wastewater neutralisation process was chosen as a test case.

Lin-En Kuo | S. S. Melsheimer | S. Melsheimer | L. Kuo

[1] M. Nazmul Karim,et al. Model-predictive pH control using real-time NARX approach , 1994 .

[2] Dale E. Seborg,et al. Nonlinear internal model control strategy for neural network models , 1992 .

[3] P. Werbos,et al. Long-term predictions of chemical processes using recurrent neural networks: a parallel training approach , 1992 .

[4] Sheng Chen,et al. Orthogonal least squares methods and their application to non-linear system identification , 1989 .

[5] Dale E. Seborg,et al. Modelling and Self-Tuning Control of a Multivariable pH Neutralization Process Part I: Modelling and Multiloop Control , 1989, 1989 American Control Conference.

[6] John Moody,et al. Fast Learning in Networks of Locally-Tuned Processing Units , 1989, Neural Computation.

[7] D. Costello. Evaluation of model-based control techniques for a buffered acid-based reaction system : Process operation and control , 1994 .