Neural Network Based Model Predictive Control. Piche, S., Keeler, J., Martin, G., Boe, G., Johnson, D., & Gerules, M. In Advances in Neural Information Processing Systems, volume 12, 1999. MIT Press.
Neural Network Based Model Predictive Control [link]Paper  abstract   bibtex   
Model Predictive Control (MPC), a control algorithm which uses an optimizer to solve for the optimal control moves over a future time horizon based upon a model of the process, has become a stan(cid:173) dard control technique in the process industries over the past two decades. In most industrial applications, a linear dynamic model developed using empirical data is used even though the process it(cid:173) self is often nonlinear. Linear models have been used because of the difficulty in developing a generic nonlinear model from empirical data and the computational expense often involved in using non(cid:173) linear models. In this paper, we present a generic neural network based technique for developing nonlinear dynamic models from em(cid:173) pirical data and show that these models can be efficiently used in a model predictive control framework. This nonlinear MPC based approach has been successfully implemented in a number of indus(cid:173) trial applications in the refining, petrochemical, paper and food industries. Performance of the controller on a nonlinear industrial process, a polyethylene reactor, is presented.
@inproceedings{piche_neural_1999,
	title = {Neural {Network} {Based} {Model} {Predictive} {Control}},
	volume = {12},
	url = {https://proceedings.neurips.cc/paper/1999/hash/db957c626a8cd7a27231adfbf51e20eb-Abstract.html},
	abstract = {Model  Predictive  Control  (MPC),  a  control  algorithm which  uses  an  optimizer  to solve  for  the optimal  control  moves  over  a  future  time horizon based upon a model of the process, has become a stan(cid:173) dard control technique  in  the process industries over  the  past  two  decades.  In  most  industrial  applications,  a  linear  dynamic  model  developed using empirical data is  used even though the process it(cid:173) self is often nonlinear.  Linear models have been used because of the  difficulty  in  developing  a  generic  nonlinear  model  from  empirical  data and  the  computational expense  often involved  in  using  non(cid:173) linear models.  In  this  paper,  we  present  a  generic neural  network  based technique for developing nonlinear dynamic models from em(cid:173) pirical  data and show that these  models  can be efficiently  used  in  a  model predictive control framework.  This nonlinear MPC based  approach has been successfully implemented in a  number of indus(cid:173) trial  applications  in  the  refining,  petrochemical,  paper  and  food  industries.  Performance of the controller on  a  nonlinear industrial  process, a  polyethylene reactor, is  presented.},
	urldate = {2022-09-01},
	booktitle = {Advances in {Neural} {Information} {Processing} {Systems}},
	publisher = {MIT Press},
	author = {Piche, Stephen and Keeler, James and Martin, Greg and Boe, Gene and Johnson, Doug and Gerules, Mark},
	year = {1999},
	keywords = {/unread, ⛔ No DOI found},
}

Downloads: 0