A general hybrid semi-parametric process control framework

Stosch, Moritz von; Oliveira, Rui; Peres, J.; Azevedo, S. Feyo de

doi:10.1016/j.jprocont.2012.05.004

Cited by 19 publications

(16 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the contrary, the nonlinearity and adaptivity natures of neural network enable its online tuning ability in representing time-varying nonlinear system dynamics. However, the representing capability of neural network depends on the learning of data, and cannot include any physical representation of object process directly (von Stosch et al, 2012), thus will result in limitation in extrapolation performance.…”

Section: Hybrid Tidal Prediction Methodsmentioning

confidence: 99%

A hybrid real-time tidal prediction mechanism based on harmonic method and variable structure neural network

Yin

Wang

2015

Engineering Applications of Artificial Intelligence

View full text Add to dashboard Cite

keywords:Tidal prediction Hybrid model Variable neural network Harmonic method Sliding data window a b s t r a c t Accurate real time tidal prediction is essential for human activities in coastal and marine fields. Tidal changes are influenced not only by periodic revolutions of celestial bodies but also by time-varying meteorological factors. For accurate real-time tidal prediction, a hybrid prediction mechanism is constructed by taking both advantages of harmonic analysis and neural network. In the proposed mechanism, conventional harmonic analysis is employed for representing the influences of celestial factors; and neural network is used for representing the nonlinear influences of meteorological factors. Furthermore, to represent time-varying tidal dynamics influenced by meteorological factors, a variable neural network is real-time constructed with the neurons and the connecting parameters are adaptively adjusted based on a sliding data window (SDW). The hybrid prediction method uses only the latest short-period data to generate predictions sequentially. Hourly tidal data measured at four American tidal stations are used to validate the effectiveness of the hybrid sequential tidal prediction model. Simulation results of tidal prediction demonstrate that the proposed model can generate accurate short-term prediction of tidal levels at very low computational cost.

show abstract

Section: Hybrid Tidal Prediction Methodsmentioning

confidence: 99%

A hybrid real-time tidal prediction mechanism based on harmonic method and variable structure neural network

Yin

Wang

2015

Engineering Applications of Artificial Intelligence

View full text Add to dashboard Cite

show abstract

“…Convergence and success of the identification depend on the initialized parameter values (Kahrs & Marquardt, 2008;von Stosch, Oliveira, Peres, & Feyo de Azevedo, 2012). Relatively small weight values are preferential (because generally inheriting better generalization capabilities), wherefore the initialization values of the parameters are usually constrained, e.g.…”

Section: General Remarks About the Identificationmentioning

confidence: 99%

“…For closed-loop control, there are two possible ways to exploit the hybrid process model (von Stosch et al, 2012), namely (i) by employing a control structure that directly uses the hybrid process model equations for the calculation of the control action; or (i) by the application of the hybrid semi-parametric model for the controller tuning.…”

Section: Controlmentioning

confidence: 99%

“…Whenever the process model equations are invertible, i.e. an analytical explicit expression can be obtained through manipulation, direct Feedback Linearizing Control (FLC) (Bazaei & Majd, 2003), Generic Model Control (Abonyi, Madar, & Szeifert, 2007;von Stosch et al, 2012;Xiong & Jutan, 2002) or Model (Adaptive) Reference Control (von Stosch et al, 2012) schema can be applied. These methods can account for nonlinearities and are relatively computationally inexpensive.…”

Section: Hybrid Semi-parametric Model Based Controller Structuresmentioning

confidence: 99%

“…The hybrid process model can also be exploited to tune any kind of standard controller both, prior to application (off-line) (Georgieva & de Azevedo, 2009;Georgieva & Feyo de Azevedo, 2007;Schubert et al, 1994a;von Stosch et al, 2012); and under control (on-line) (Andrasik, Meszaros, & de Azevedo, 2004;Chen et al, 2004;Patnaik, 2003Patnaik, , 2004Patnaik, , 2008Patnaik, , 2010Schubert et al, 1994a;Wei, Hussain, & Wahab, 2007;Zhang et al, 2006). The utilization of a hybrid semiparametric model for controller tuning bears some advantages, because (i) the hybrid semi-parametric model might be identified from standard operational data, but the controllers can be tuned considering set-point changes and load disturbances, both scenarios in which the model might have to extrapolate (von Stosch et al, 2012); and (ii) the coupling of different control inputs and/or variables can be captured with little effort by the hybrid semi-parametric model and subsequently considered during the tuning (multivariate control) (von Stosch et al, 2012).…”

Section: Hybrid Semi-parametric Model Based Controller Tuningmentioning

confidence: 99%

See 2 more Smart Citations

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

Stosch¹,

Oliveira²,

Peres³

et al. 2014

Computers & Chemical Engineering

Self Cite

313

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