A Method Of Predicting The Future Reservoir Performance Of A Gas Reservoir Using A Digital Computer

Abstract

Introduction The natural-gas industry in the United States had its beginning in 1826 when natural gas was used for lighting the city of Fredonia, N. Y. The first natural-gas pipeline was a 25-mile wooden line constructed from hollow logs connecting West Bloomfield and Rochester, N. Y. There are no records indicating that a gas deliverability study was ever made here to determine the economics of installing this line, but it is recorded that an attempt was made to determine the capacity of the wells by measuring the time required to fill a large balloon. Thus, since the inception of the gas industry, engineers and operators have endeavored to develop easier methods for making deliverability studies. A gas deliverability study may be considered a projection of the annual rates at which volumes of gas reserves may be received into a gathering system. Such a study may cover a 20-year period and involve calculations on scores of wells. Studies have been and are being made manually - the repetitive calculations requiring thousands of man-hours. Manually, these calculations involve the use of curves and tables. The reading of these through many repeated calculations is slow and lends itself to inaccuracy if not to error. Because deliverability studies are required by the Federal Power Commission when gas transmission companies propose expansion, are used internally by companies for cash forecasts and budgets, are necessary in evaluating proposed line extensions into new gas supply areas and are foundations for gas-property financing, we, as petroleum consultants, are frequently asked to perform such studies. Because of the widely different possible uses of such studies and the infinite number of variables involved such as allowable, price escalation, producing practices, contract variables, field rules, etc., it is impractical to develop a program so general as to handle every and all situations. Likewise, it has proved inconvenient to completely rewrite a new program for each situation. THE NEW PROGRAM The program I offer for your inspection is one which covers several situations and one which I feel can be modified to fit almost all conditions. It is divided into blocks. The applicable ones for a specific problem provide a complete calculating procedure. A flow diagram for the blocks is presented in Figs. 6 and 7. The procedure has been programed in FOR-TRANSIT language and can handle up to 20 well reservoirs on the IBM 650 and 50 well reservoirs on the IBM 704. A larger system might store all the program blocks and utilize internal switching to develop the complete calculating procedure for a greater number of wells; however, the number 20 is not too restrictive, because similar wells can be lumped together in groups and handled as one well. The program considers allowables set by regulatory bodies or will generate an allowable based on the well's absolute open-flow potential. In the past, manual calculations to eliminate trial-and-error solutions have been based upon increments of reservoir pressure or increments of production. An engineer would arbitrarily assign an increment and solve for the resulting well deliverabilities. The well deliverabilities then were obtained related to production or pressure increments, but not in terms of time.