Combining neural and conventional paradigms for modelling,prediction and control

Abstract. Neural network based experimental and hybrid approaches to modeling of processes are presented. A hybrid modeling, combining parametric model with radial basis function network, is proposed. The parametric model is used for principal modeling of the process and the radial basis function network is applied for nonlinear error correction. Experimental modeling can be improved by selecting as model inputs only variables with high predictive importance. Two feature selection methods are presented: analysis of mutual information, and genetic algorithm based feature selection. The methods proposed are applied to modeling of the liquid-phase methanol synthesis. Analytical, experimental and hybrid approaches are applied and results demonstrate that a hybrid modeling approach, exploiting available analytical knowledge and experimental data, can considerably outperform a purely analytical approach.

show abstract

“…of K neurons. A Gaussian function can be applied for the activation function of neurons: (2) .. " ... ,.'…”

Section: Radial Basis Function Networkmentioning

confidence: 99%

“…The analytical and experimental modeling approaches can also be combined in a hybrid manner, thus exploiting analytical knowledge and experimental information obtained by measurement. Several possibilities of combining neural networks with a priori knowledge have been suggested elsewhere [2].…”

Section: Introductionmentioning

confidence: 99%

Hybrid modeling of kinetics for methanol synthesis

Potočnik

Grabec

Šetinc

et al. 2003

Soft Computing Approaches in Chemistry

View full text Add to dashboard Cite

show abstract

“…2, where structure A is referred to as parallel and structures B and C are called serial, sequential, cascade or consecutive. These structures are theoretically addressed in Agarwal (1997) considering that the white box would represent mechanistic information, and the black box consists of a nonparametric model. However, in the serial case, the order of the black and the white model might not be inter-changeable.…”

Section: How To Arrange the Models? Hybrid Semi-parametric Model Strumentioning

confidence: 99%

Hybrid semi-parametric modeling in process systems engineering: Past, present and future

Stosch¹,

Oliveira²,

Peres³

et al. 2014

Computers & Chemical Engineering

338

201

View full text Add to dashboard Cite

Hybrid semi-parametric models consist of model structures that combine parametric and nonparametric submodels based on different knowledge sources. The development of a hybrid semi-parametric model can offer several advantages over traditional mechanistic or data-driven modeling, as reviewed in this paper. These advantages, such as broader knowledge base, transparency of the modeling approach and cost-effective model development, have been widely recognized, not only in academia but also in the industry.In this paper, the most common hybrid semi-parametric modeling and parameter identification techniques are revisited. Applications in the areas of (bio)chemical engineering for process monitoring, control, optimization, scaleup and model-reduction are reviewed. It is outlined that the application of hybrid semi-parametric techniques does not automatically lead into better results but that rational knowledge integration has potential to significantly improve model-based process operation and design.

show abstract

“…Applications of generalised and specialised learning architectures are presented with the goal of inverting the plant dynamics. The neurocomputing techniques exploited in Jacobson and Reynolds (1993) have their roots in the pioneering work of Narendra and Parthasarathy (1990) and relevant applications are seen later on in Agarwal (1997). The study of active laminar flow control by Fan et al (1993) showed that a properly trained NN can establish complex nonlinear relationships between multiple inputs and outputs which are peculiar to an active flow control system.…”

Section: Introductionmentioning

confidence: 99%

Neural network-based modelling of subsonic cavity flows

Efe

Debiasi

Yan

et al. 2008

International Journal of Systems Science

View full text Add to dashboard Cite

A fundamental problem in the applications involved with aerodynamic flows is the difficulty in finding a suitable dynamical model containing the most significant information pertaining to the physical system. Especially in the design of feedback control systems, a representative model is a necessary tool constraining the applicable forms of control laws. This article addresses the modelling problem by the use of feedforward neural networks (NNs). Shallow cavity flows at different Mach numbers are considered, and a single NN admitting the Mach number as one of the external inputs is demonstrated to be capable of predicting the floor pressures. Simulations and real time experiments have been presented to support the learning and generalization claims introduced by NN-based models.

show abstract

Combining neural and conventional paradigms for modelling,prediction and control

Cited by 68 publications

References 41 publications

Hybrid modeling of kinetics for methanol synthesis

Hybrid modeling of kinetics for methanol synthesis

Hybrid semi-parametric modeling in process systems engineering: Past, present and future

Neural network-based modelling of subsonic cavity flows

Contact Info

Product

Resources

About