Shih-Hsien Chang

Advanced reservoir simulation using soft computing

Reservoir simulation is a challenging problem for the oil and gas industry. A correctly calibrated reservoir simulator provides an effective tool for reservoir evaluation that can be used to obtain essential reservoir information. A long-standing problem in reservoir simulation is history matching, which is to find a suitable set of values for the simulators input parameters such that the simulator correctly predicts the fluid (oil, gas, water, etc.) outputs of the wells on the reservoir, over the time period of interest. Due to the sheer size of the problem, completely satisfactory results of history matching have been difficult and expensive to achieve. This paper presents a novel technique of using fuzzy control to solve history matching. Intended for implementation on a cluster of PCs, our technique aims not only to solve history matching faster, but also solves it at a lower cost. Preliminary results and ongoing work are described.

Characterization and multiphase equilibrium prediction of crude oil heavy components

For compositional simulation purposes, a new method has been developed for estimating the specific gravities of pseudo-components that are required to characterize the heavy components of crude oils. This method makes use of the always available data for the heavy components of crude oils : average molecular weight and specific gravity of the heptanes-plus (C 7+ ) fraction. To evaluate the new method and compare it with the existing estimation methods, a procedure has been implemented for the phase equilibrium predictions of crude oil/carbon dioxide systems without using any tuning techniques. The phase equilibrium predictions, based on the specific gravities estimated by the new method and the existing methods proposed by Whitson et al. (1989) and Yarborough (1979), were all reasonably accurate and of the same magnitude. The specific gravities estimated by the new method and the method proposed by Whitson et al. (1989) always provided similar predictions. The satisfactory results reported here also indicate that a simple truncated exponential distribution for the heavy components of crude oils is adequate.

Solving nonlinear engineering problems with the aid of neural networks

In this paper, a technique is presented for using neural networks as an aid for solving nonlinear engineering problems, which are encountered in optimization, simulations and modeling, or complex engineering calculations. Iterative algorithms are often used to find the solutions of such problems. For many large-scale engineering problems, finding good starting points for the iterative algorithms is the key to good performance. We describe using neural networks to select starting points for the iterative algorithms for nonlinear systems. Since input/output training data are often easily obtained from the problem description or from the system equations, a neural network can be trained to serve as a rough model of the underlying problem. After the neural network is trained, it is used to select starting points for the iterative algorithms. We illustrate the method with four small nonlinear equation groups, two real applications in petroleum engineering are also given to demonstrate the methods potential application in engineering.

Improved efficiency of miscible CO{sub 2} floods and enhanced prospects for CO{sub 2} flooding heterogeneous reservoirs

This project examines three major areas in which CO{sub 2} flooding can be improved: fluid and matrix interactions, conformance control, sweep efficiency, and reservoir simulation for improved oil recovery. This report discusses the activity during the calendar quarter covering October 1, 1999 through December 31, 1999 that covers mostly the second fiscal quarter of the projects third year. Injectivity experiments were performed on two Indian limestone cores. In tests on the first core, a variety of brine, CO{sub 2} WAG, and oil contaminant injection schemes indicated infectivity reduction due to phase conditions and contamination. The results are only quantitative because of plugging and erosion in the core. To date, tests on the second core have investigated the effects of long-term brine stability on the reduction of fluid-rock interaction, in order to quantify fluid effects on infectivity. The authors continue to develop a new approach in reservoir simulation to improve the history matching process on clusters of PCs. The main objective was to improve simulation of complex improved oil recovery methods, such as CO{sub 2}-foam for mobility control and sweep enhancements. Adsorption experiments using circulation and flow-through methods were used to determine the loss of surfactants for economic evaluation. A sacrificial agent, lignosulfonate, was used to reduce the adsorption of the primary foaming agent in both Berea sandstone and Indian limestone. The lignosulfonate has also shown a chromatograph effect, advancing more rapidly through the reservoir, thus initially adsorbing onto the rock before the primary foaming agent arrives. Therefore, considering the simplicity of operation and economics of reducing the cost of expensive surfactant to improve oil recovery, coinjection of lignosulfonate with the primary foaming agent might be a practical approach to consider for field application.

Improved Efficiency of Miscible CO{sub 2} Floods and Enhanced Prospects for CO{sub 2} Flooding Heterogeneous Reservoirs

This work will examine three major areas in which CO{sub 2} flooding can be improved: fluid and matrix interactions, conformance control/sweep efficiency, and reservoir simulation for improved oil recovery. The first full quarter of this project has been completed. We began examining synergistic affects of mixed surfactant versus single surfactant systems to enhance the properties of foams used for improving oil recovery in CO{sub 2} floods. The purpose is to reduce the concentration of surfactants or finding less expensive surfactants. Also, we are examining the effect of oil saturation on the development of foam in CO{sub 2}-surfactant solution systems. CO{sub 2} flooding of low permeability, vugular, and fracture reservoirs are another major thrust of this project. Work conducted this quarter involved simulating gravity stable floods using large core samples; results showed excellent recovery in a low permeability vugular core.

Improved Efficiency of Miscible C02 Floods and Enhanced Prospects for C02 Flooding Heterogeneous Reservoirs

Surfactant and foam properties have been evaluated at high pressure using the foam durability apparatus. For a number of surfactant solutions the interfacial tension with cense CO2, critical micelle concentrations, foaming ability, and foam stability were determined. Preliminary results show that these tests correlate well to predict surfactant properties and mobility in cores. Work has also restarted in the parallel-dual permeability system.