Systems modelling using genetic programming

https://doi.org/10.1016/S0098-1354(97)87659-4Get rights and content

Abstract

In this contribution, a Genetic Programming (GP) algorithm is used to develop empirical models of chemical process systems. GP performs symbolic regression, determining both the structure and the complexity of a model. Initially, steady-state model development using a GP algorithm is considered, next the methodology is extended to the development of dynamic input-output models. The usefulness of the technique is demonstrated by the development of inferential estimation models for two typical processes: a vacuum distillation column and a twin screw cooking extruder.

Reference (17)

  • GaniR.C. et al.

    A Generalised Model for a Distillation Columns — I

    Comp. Chem. Eng.

    (1986)
  • BettenhausenK.D. et al.

    Self-organising modelling of biotechnological batch and fed-batch fermentations

  • BettenhausenK.D. et al.

    Self-organising structured modelling of a biotechnological fed-batch fermentation by means of genetic programming

  • ElseyJ. et al.

    Dynamic modelling of a cooking extruder

  • FrohlichJ. et al.

    Extended and Generalised Genetic Programming for Function Analysis

    (1996)
  • HinchliffeM.P. et al.

    Modelling Chemical Process Systems using a Multi-Gene Genetic Programming Algorithm

  • HowardE. et al.

    Genetic programming as a means of assessing and reflecting chaos

  • IbaH. et al.

    Genetic programming using local hill-climbing

There are more references available in the full text version of this article.

Cited by (44)

  • Applying genetic programming in estimation of frost layer thickness on horizontal and vertical plates at ultra-low temperature

    2021, International Journal of Refrigeration
    Citation Excerpt :

    Related work was done by Kobayashi et al. (2011), who used GP for developing a feedback controller. Willis et al. (1997) used GP to model an input-output chemical process. Recently, Hosseini et al. (2020a, 2020b) and Moradkhani et al. (2020) developed general correlations for estimating the heat transfer coefficient and pressure drop in condensers using GP.

  • A Genetic Programming Approach for Construction of Surrogate Models

    2019, Computer Aided Chemical Engineering
    Citation Excerpt :

    Another (deterministic-based) approach to symbolic regression can be found in Cozad and Sahinidis (2018). In the context of Process Systems Engineering, GP has been applied to the obtention of dynamic models for a binary distillation column (Willis et al., 1997) and an extruder (Hinchliffe and Willis, 2003), process models of wastewater treatment reactors (Dürrenmatt and Gijer, 2012) and heat transfer correlations (Cai et al., 2006). Although the existence of these previous works, GP is not yet regarded as a tool commonly used for the generation of surrogate models.

  • An application of evolutionary system identification algorithm in modelling of energy production system

    2018, Measurement: Journal of the International Measurement Confederation
    Citation Excerpt :

    Models build on only the given input–output continuous data are known as regression models. The methods such as regression analysis, response surface methodology (RSM), partial least square regression, genetic programming (GP), artificial neural network (ANN), fuzzy logic (FL), M5- prime (M5′), support vector regression (SVR), adaptive neuro-fuzzy inference systems (ANFIS), etc. can be applied [2–7] to formulate these models. The models build must not only accurate predict the system output but shall also satisfy the system constraints.

  • Model development based on evolutionary framework for condition monitoring of a lathe machine

    2015, Measurement: Journal of the International Measurement Confederation
    Citation Excerpt :

    If there is a tie among the runs, the model of a given run with the lowest number of nodes (lower complexity) is chosen. The new evolutionary framework is implemented by modifying the software GPTIPS [29,30] code and developing its graphical interface (Fig. 7) for user friendly purpose. The parameter settings of MGGP and EN-MGGP are adjusted using a trial-and-error approach (Fig. 7) and based on the study conducted on applications of evolutionary algorithms in modelling the industrial processes [31–36].

View all citing articles on Scopus
View full text