Improvement of Gauss-Newton method for parameter estimation through the use of information index

Kalogerakis, Nicolas; Luus, Rein

doi:10.1021/i100012a015

Cited by 23 publications

(8 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The thermal cracking and LTO reaction schemes were developed using experimental data from Hayashitani et al (1977) and Adegbesan et al (1987). Kinetic parameters of thermal cracking and LTO reactions were estimated by Belgrave et al (1990) using the method described by Kalogerakis and Luus (1983). Kinetic parameters for coke combustion reaction were obtained from the work of Thomas et al (1985).…”

Section: Belgraveõs Modelmentioning

confidence: 99%

Applicability of Kinetic Models for In Situ Combustion Processes with Different Oil Types

et al. 2016

View full text Add to dashboard Cite

The in situ combustion (ISC) process is of interest as an enhanced oil recovery method because it is an alternative to traditional steam-based processes for heavy oil and bitumen recovery. ISC is a technique applicable outside the window of reservoir conditions deemed appropriate for steam injection (such as deeper and thinner reservoirs). The process involves complex chemical reactions and physical recovery mechanisms, and predicting the likelihood of successful ISC in field applications remains challenging. This paper describes a numerical investigation of the capability of different ISC kinetic models to predict the combustion behaviors of different types of oils (light oil, heavy oil, and bitumen). Three kinetic models (of Coats, Crookston, and Belgrave) were selected from literature and compared using data from four published combustion-tube experiments. The comparison procedure is as follows:(1) validate the numerical modeling of each kinetic model by matching the selected experimental results or duplicating the numerical results found in published literature; (2) adjust fluid viscosities and densities to match the fluid properties of each experiment;and (3) use each validated kinetic model to predict the performance of the other experiments without further tuning the kinetic parameters. The knowledge derived from the experiments provides guidance for choosing the appropriate kinetic model when no other data are available and for the preliminary design and screening study of a potential ISC project.

show abstract

Section: Belgraveõs Modelmentioning

confidence: 99%

Applicability of Kinetic Models for In Situ Combustion Processes with Different Oil Types

et al. 2016

View full text Add to dashboard Cite

show abstract

“…In the case of overstepping, a suitable value for γ is obtained by a monotonicity test or by employing an optimal step size policy (Kalogerakis and Luus, 1983). Overstepping may occur when the parameter values are away from the optimum.…”

Section: Gauss-newton-marquardt Minimization Algorithmmentioning

confidence: 99%

Interaction Parameter Estimation in Cubic Equations of State Using Binary Phase Equilibrium and Critical Point Data

Englezos

Bygrave

Kalogerakis

1998

Ind. Eng. Chem. Res.

Self Cite

View full text Add to dashboard Cite

Two methods for the estimation of the interaction parameters in cubic equations of state by using the entire binary phase equilibrium database and the critical point locus, respectively, are presented. The solution of the optimization problem is accomplished in both methods by a Gauss-Newton-Marquardt minimization algorithm. The methods are computationally efficient and robust because they are based on implicit objective functions and hence avoid phase equilibrium or critical point calculations during the parameter optimization. The use of the entire phase equilibrium database and the critical locus can be a stringent test of the correlational ability of the equation of state. In the illustrative examples, the results were obtained by using the Peng-Robinson and the Trebble-Bishnoi equations of state with quadratic mixing rules and temperature-independent interaction parameters.

show abstract

“…The initial conditions are given by Eq. (12). At the observation times, the sensitivity coefficients are substituted into Eq.8 or 10.…”

Section: Caracotsios and Stewart Ls •mentioning

confidence: 99%

A Fully Implicit, Three-Dimensional, Three-Phase Simulator With Automatic History-Matching Capability

Tan¹,

Kalogerakis²

1991

All Days

View full text Add to dashboard Cite

The reservoir simulator described in this paper has the capability of automatically determining parameter values of reservoir properties in specified zones that will best match observed data. The observed data can be grid cell pressures, water-oil ratios, gas-oil ratios and flowing well bottom hole pressures. A modified Gauss-Newton approach is used that significantly reduces the computational time in generating new parameter estimates. As a test problem, the Second SPE Comparative Problem which is a numerically difficult three phase coning problem, was chosen. The proposed algorithm was able to converge to the original values of permeability and porosity from a wide range of initial guesses and the computational effort was equivalent to performing a few typical simulator runs.

show abstract

Improvement of Gauss-Newton method for parameter estimation through the use of information index

Cited by 23 publications

References 6 publications

Applicability of Kinetic Models for In Situ Combustion Processes with Different Oil Types

Applicability of Kinetic Models for In Situ Combustion Processes with Different Oil Types

Interaction Parameter Estimation in Cubic Equations of State Using Binary Phase Equilibrium and Critical Point Data

A Fully Implicit, Three-Dimensional, Three-Phase Simulator With Automatic History-Matching Capability

Contact Info

Product

Resources

About