## Operating Regime Based Process Modeling and Identification (1994)

Venue: | COMPUTERS AND CHEMICAL ENGINEERING |

Citations: | 26 - 12 self |

### BibTeX

@ARTICLE{Johansen94operatingregime,

author = {Tor A. Johansen and Bjarne A. Foss},

title = {Operating Regime Based Process Modeling and Identification },

journal = {COMPUTERS AND CHEMICAL ENGINEERING},

year = {1994},

volume = {21},

pages = {159--176}

}

### Years of Citing Articles

### OpenURL

### Abstract

This paper presents a non-linear modeling framework that supports model development "in between" empirical and mechanistic modeling. A model is composed of a number of local models valid in different operating regimes. The local models are combined by smooth interpolation into a complete global model. It is illustrated how different kinds of empirical and mechanistic knowledge and models can be combined with process data within this framework. Furthermore, we describe a flexible computer aided modeling tool that supports modeling within this framework. Simple but illustrative examples from chemical engineering are used to highlight the flexibility and power of the framework.

### Citations

3093 | Fuzzy sets
- Zadeh
- 1965
(Show Context)
Citation Context ...based design of the i-functions appears to be considerably more transparent and attractive from an engineering point of view than the other approaches. Takagi and Sugeno (1985) suggested a fuzzy set (=-=Zadeh 1965-=-) based technique for combining in a smooth fashion a set of linear models into a non-linear model. It30 CHAPTER 2. INPUT/OUTPUT MODELING turns out that if the operating regimes Zi are viewed as fuzz... |

1282 |
Spline Models for Observational Data
- Wahba
(Show Context)
Citation Context ...and local polynomial models (Pottmann, Unbehauen and Seborg 1993) are also described in the literature. Clearly, piecewise polynomials are closely related to the spline-based modeling approach, e.g. (=-=Wahba 1990-=-). The main di erence lies in the way splines handles multi-dimensionality. While piecewise polynomials are multi-dimensional by nature, splines are not, and multi-dimensionality istypically introduce... |

805 |
Estimation of Dependencies Based on Empirical Data
- Vapnik
- 1982
(Show Context)
Citation Context ...er sets, and take into account the convergence rate, cf. (5.24). Unfortunately, accurate estimates of the convergence rates are not easily attainable. There are some convergence results available in (=-=Vapnik 1982-=-, Pollard 1984), but they su er from two major drawbacks: They are potentially very conservative, and applicable for independent stochastic processes only. Itis not clear how to extend these results t... |

724 | Hierarchical mixtures of experts and EM algorithm - Jordan, Jacobs - 1994 |

553 | Outline of a new approach to the analysis of complex systems and decision processes - Zadeh - 1973 |

424 | Smoothing Noisy Data with Spline Functions - Craven, Wahba - 1978 |

383 | Empirical Model-Building and Response Surfaces - Box - 1987 |

188 | Predicting the Future: Connectionist Approach - Weigend, Huberman, et al. - 1990 |

165 | A fuzzy-logic-based approach to qualitative modeling - Sugeno, Yasukawa - 1993 |

139 | Fitting autoregressive models for prediction - Akaike - 1969 |

108 | Fuzzy basis functions, universal approximation, and orthogonal leastsquares learning - Wang, Mendel - 1992 |

106 |
On the input-output stability of time-varying nonlinear feedback systems part I: Conditions derived using concepts of loop gain, conicity and positivity
- Zames
- 1966
(Show Context)
Citation Context ...de ne the linear time-invariant system Mj(q ;1 )+mH2(q ;1 ) n(t) =VR(q ;1 )n(t)+K 0 3 The initial condition is chosen as n(0) = n(0). It follows that n(t) n(t) for all t 0. By the Small Gain Theorem (=-=Zames 1966-=-), it follows that n is a bounded sequence, since VjjR(q ;1)jj1 <1. Hence, n is also a bounded sequence, and it follows from the de nition of n that y and u are bounded sequences. We get for t 0 n(t) ... |

77 |
Le cons sur la Theorie Mathematique de la Lutte popur la Vie. Paris: Gauthier Villare
- Volterra
- 1931
(Show Context)
Citation Context ...as illustrated in Fig. 3.3a, where each closed cycle corresponds to di erent initial conditions. We see that the behavior of this model is qualitatively similar to the classical Lotka-Volterra model (=-=Volterra 1931-=-) _x1 = b11x1 ; b12x1x2 _x2 = ;b21x2 + b22x1x2 which has trajectories as illustrated in Fig. 3.3b, with b11 = b12 = b21 = b22 =1. Now, these two models have similar behavior, but the underlying model ... |

60 | Constructing NARMAX models using ARMAX models - Johansen, Foss - 1993 |

36 | Fuzzy identi cation of systems and its applications to modeling and control - Takagi, Sugeno - 1985 |

35 | Pantelides, Modeling of combined discrete/continuous processes - Barton, C - 1994 |

35 | Modeling chemical processes using prior knowledge and neural networks - Thompson, Kramer - 1994 |

32 |
Construction of fuzzy models through clustering techniques
- Yoshinari, Pedrycz, et al.
- 1993
(Show Context)
Citation Context ...geno 1985). In that case, it is the fuzzy inference mechanism that implicitly gives an interpolation. Structure identi cation algorithms based on clustering (Bezdek et al. 1981a, Bezdek et al. 1981b, =-=Yoshinari et al. 1993-=-, Yager and Filev 1993, Nakamori and Ryoke 1994) and a heuristic local search algorithm (Sugeno and Kang 1988) have been proposed in this context. In this case, the b- and i-functions are interpreted ... |

24 | Nonlinear Predictive Control Using Local Models – Applied to a Batch Fermentation Process,” Control Engineering Practice - Foss, Johansen, et al. - 1995 |

19 | Fuzzy modeling and control of multilayer incinerator." Fuizy Sets and Systerns - Sugeno, Kang - 1936 |

18 | Identi cation - Soderstrom, Stoica - 1989 |

17 | Tools for semi-physical modelling - Lindskog, Ljung - 1995 |

12 | Structure identi cation of fuzzy model - Sugeno, Kang - 1988 |

10 | Dynamic modeling and reaction invariant control of pH. Chemical Engineering Science - Gustafsson, Waller - 1983 |

9 | State-space modeling using operating regime decomposition and local models - Johansen - 1993 |

9 | A constructive learning algorithm for local model networks - Murray-Smith, Gollee - 1994 |

9 | Construction of composite models from observed data - Skeppstedt, Ljung, et al. - 1992 |

9 | The conceptual side of mode switching - So¨derman, Top, et al. - 1993 |

7 | Modelling and self-tuning control of a multivariable pH neutralization process, part I: Modelling and multiloop control - Hall, Seborg - 1989 |

6 |
Transient performance and robustness of direct adaptive control
- Ydstie
- 1992
(Show Context)
Citation Context ...rties and a robust adaptive control system. These include di erent variations of -modi cation, -modi cation, and parameter projection (Narendra and Annaswamy 1989, Ioannou and Datta 1991, Datta 1993, =-=Ydstie 1992-=-). A.5 Closed Loop Stability With the feedback linearizing control structure, it is well known that certain states may be unobservable (Byrnes and Isidori 1984, Monaco and Normand-Cyrot 1987). In orde... |

4 | The Exploration ,and Exploitation of Response Surfaces: An Example of the Link Between the Fitted Surface and the Basic Mechanism of the System - Youle - 1955 |

4 | Kinetic analysis of gluconic acid production by pseudomonas ovalis - Ghose, Ghosh - 1976 |

4 | Identi cation of non-linear system structure and parameters using regime decomposition - Johansen - 1995 |

4 | Physiological state control of fermentation processes', Biotechnology and Bioengineering - Konstantinov - 1989 |

3 | H1 control of nonlinear processes using multiple linear models - Banerjee, Arkun, et al. - 1995 |

3 | Issues in nonlinear stochastic grey-box identi cation - Graebe - 1994 |

3 | Local learning for local model networks - Johansen - 1995 |

3 | A new design of a fuzzy model predictive control system for nonlinear processes - Nakamori, Suzuki, et al. - 1992 |

3 | A hybrid neural network * "rst principles approach to process modelling. A.I.Ch.E - Psichogios, Ungar - 1992 |

3 |
Grey-box modelling and identi cation using physical knowledge and bayesian techniques', Automatica 29
- Tulleken
- 1993
(Show Context)
Citation Context ...physical interpretation. Physically based constraints that ensure the model is consistent with certain aspects of the physical reality, and prior knowledge about the parameters, can also be utilized (=-=Tulleken 1993-=-). In another approach, Lindskog and Ljung (1994) illustrates how simple mechanistic knowledge can be used to nd adequate non-linearities in a semi-empirical model. Kramer, Thompson and Phagat (1992) ... |

3 |
Robustness of model reference adaptive controllers: Input-Output properties
- Tsakalis
- 1992
(Show Context)
Citation Context ...Time-varying Non-linear Systems The application of input/output stability theory (Desoer and Vidyasagar 1975), and in particular weighted L2-norms in the continuous-time case (Ioannou and Datta 1991, =-=Tsakalis 1992-=-), and weighted l2-norms in the discrete time case (Datta 1993), has turned out to be attractive for the analysis of stability and robustness of adaptive control loops based on linear models. The purp... |

3 | System identi cation in the behavioral framework via risk minimisation - Weyer, Williamson, et al. - 1995 |

2 | Supervisory Control of Mode-Switch Processes - Hilhorst - 1992 |

2 |
Uni ed structure and parameter identi cation of fuzzy models
- Yager, Filev
- 1993
(Show Context)
Citation Context ...e, it is the fuzzy inference mechanism that implicitly gives an interpolation. Structure identi cation algorithms based on clustering (Bezdek et al. 1981a, Bezdek et al. 1981b, Yoshinari et al. 1993, =-=Yager and Filev 1993-=-, Nakamori and Ryoke 1994) and a heuristic local search algorithm (Sugeno and Kang 1988) have been proposed in this context. In this case, the b- and i-functions are interpreted as membership function... |

1 | Integrating neural networks with rst-principles knowledge for bioreactor modeling and control - Aoyama - 1993 |

1 | The fundamentals of modelling and identi cation - Bohlin - 1989 |

1 | The experimental study of physical mechanismns - Box - 1965 |

1 | A mathematical analogue of electrically stimulated muscle using local model networks - Gollee, Hunt, et al. - 1994 |

1 | A predicting system based on combining an adaptive predictor and a knowledge base as applied to a blast furnace - Jian - 1993 |

1 | Orbit - user's guide and reference - Johansen - 1995 |