Intelligent predictive control of nonlinear processes using neural networks

Nørgaard, Magnus; Sörensen, Paul; Poulsen, Niels Kjølstad; Ravn, Ole; Hansen, Lars Kai

doi:10.1109/isic.1996.556218

Cited by 23 publications

(15 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The obtained controller is an easy-to-implement one. As far as we know, only two experimental studies of GPC -on pneumatic cylinders and not on PAMs-have been carried out so far [12] [13]. In [12], the model used is a linear one which limits greatly the performances of the controller.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A predictive robust cascade position-torque control strategy for Pneumatic Artificial Muscles

Chikh¹,

Poignet

Pierrot³

et al. 2010

Proceedings of the 2010 American Control Conference

View full text Add to dashboard Cite

Abstract-A robust cascade strategy is proposed and tested on an electropneumatic testbed for parallel robotic applications. It is applied on Pneumatic Artificial Muscles (PAMs). Nonlinear models are developed and presented. By specifying the pressure average between the two muscles, it is possible to control the torque by controlling the pressure in each muscle. A constrained LMI based H∞ controller is synthesized for the pressure inner loop. A computed torque is applied on the torque loop and a Generalized Predictive Controller (GPC) controller is then applied to control position of the muscles. Experimental results show the feasibility of the control strategy and good performances in terms of robustness and dynamic tracking.

show abstract

Section: Introductionmentioning

confidence: 99%

“…In [12], the model used is a linear one which limits greatly the performances of the controller. In [13], the authors used a model estimation based on neural network theory [13]. No application of GPC basing on an explicit nonlinear model has been found in literature.…”

Section: Introductionmentioning

confidence: 99%

A predictive robust cascade position-torque control strategy for Pneumatic Artificial Muscles

Chikh¹,

Poignet

Pierrot³

et al. 2010

Proceedings of the 2010 American Control Conference

View full text Add to dashboard Cite

show abstract

“…This technique is termed instantaneous linearization technique. It has been successfully applied in many control applications [14][15][16].…”

Section: Introductionmentioning

confidence: 99%

Adaptive and Predictive Control of Liquid‐Liquid Extractors Using Neural‐Based Instantaneous Linearization Technique

Mjalli

2006

Chem Eng & Technol

View full text Add to dashboard Cite

Nonlinearity of the extraction process is addressed via the application of instantaneous linearization to control the extract and raffinate concentrations. Two feed-forward neural networks with delayed inputs and outputs were trained and validated to capture the dynamics of the extraction process. These nonlinear models were then adopted in an instantaneous linearization algorithm into two control algorithms. The self-tuning adaptive control strategy was compared to an approximate model predictive control in terms of set point tracking capability, efficiency and stability. For the case of large, abrupt set point changes, the performance of the self-tuning algorithm was poor, especially for the raffinate control. The approximate model predictive control strategy was superior to the self-tuning control in terms of its ability to force the output to following the set point trajectory efficiently with smooth controller moves.

show abstract

“…Common methods of approximation include using multiple linear models to span the operating space. 17 For a more complete review of nonlinear control topics, see ref 18 and the references therein.…”

Section: Introductionmentioning

confidence: 99%

Computationally Efficient Neural Predictive Control Based on a Feedforward Architecture

Kuure-Kinsey

Cutright

Bequette

2006

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

A new strategy for integrating system identification and predictive control is proposed. A novel feedforward neural-network architecture is developed to model the system. The network structure is designed so that the nonlinearity can be mapped onto a linear time-varying term. The linear time-varying model is augmented with a Kalman filter to provide disturbance rejection and compensation for model uncertainty. The structure of the model developed lends itself naturally to a neural predictive control formulation. The computational requirements of this strategy are significantly lower than those using the nonlinear neural network, with comparable control performance, as illustrated on a challenging nonlinear chemical reactor and a multivariable process, each with both nonminimum and minimum phase behavior. PreludeWarren Seider has been a pioneer in the use of nonlinear models for optimization and control. While model predictive control (MPC), as initially applied in the oil and petrochemical industries, was based on linear convolution coefficient-based models, Warren recognized that there was no inherent limitation to linear models. His publications with David Brengel were among the first to use nonlinear models as the basis for control computations in a model predictive control framework. He also used nonlinear MPC as a basis for integrating design and control, recognizing that the limitations to linear controller performance could impose unnecessary limitations to the process design. One of the most impressive attributes of Professor Seider is his ability to bring novel design and control concepts directly into the undergraduate classroom. His process design textbook with Professors Seader and Lewin makes these research results accessible to undergraduate students, many of whom will have the chance to implement these techniques early in their industrial career. It is with pleasure that we contribute a paper on a new technique to incorporate a neural-network-based model into a model predictive control strategy to this special issue dedicated to Professor Seider. IntroductionThere has been a trend toward specialty chemicals manufacturing, with production plants that frequently change operating conditions to meet varying product requirements for different consumers. Linear control strategies, which work quite well in maintaining a plant at a given steady state, may fail when required to operate over a wide range of conditions. To develop appropriate control strategies, accurate nonlinear models of the processes must first be developed. The use of fundamental firstprinciples models has been applied to control specialty and batch chemical processes. 1,2 First-principles models can give physical insights into the system. This physical insight allows for more meaningful interpretation of control system behavior, and a first principle's basis often means the model is accurate over a wider range of input/output space. However, not all systems lend themselves to fundamental models without considerable time and effort. Thou...

show abstract

Intelligent predictive control of nonlinear processes using neural networks

Cited by 23 publications

References 9 publications

A predictive robust cascade position-torque control strategy for Pneumatic Artificial Muscles

A predictive robust cascade position-torque control strategy for Pneumatic Artificial Muscles

Adaptive and Predictive Control of Liquid‐Liquid Extractors Using Neural‐Based Instantaneous Linearization Technique

Computationally Efficient Neural Predictive Control Based on a Feedforward Architecture

Contact Info

Product

Resources

About