Abstract

Current economic conditions have dictated that PSC operators maximize their production or accelerate the production of their reserves. One approach is to perform well stimulations. For decades, oil wells have been stimulated using various methods such as matrix acidizing and hydraulic fracturing. However, there have been advances in the stimulation processes both in knowledge and technology incorporated. One of the major PSC operators, Repsol-YPF Maxus SES, has claimed that the current success rate of their matrix acidizing practices is typically 30 to 40%. This is considered to be average. The success rate will rely on selection of the well candidates, acid types and additives. To improve the success rate, the use of a software package to perform the selection may not be sufficient anymore. The need to incorporate an integral approach using the database of experience, history and knowledge is recommended. The tool to accomplish this is called Knowledge Base System or KBS technology, considered "one of the applications of Artificial Intelligence in Expert Systems".

The KBS can then indicate what are the best fluids and procedures that can be designed for a certain matrix acidizing program.

Once formation damage is identified, analysis is performed to determine if matrix acidizing is the solution based on the experience and history. This knowledge will govern the fluid/acid selection (including fluid additives such as corrosion inhibitor, surfactant and iron-control agent) that relies on the formation characteristics, the reservoir fluid properties, and acid compatibility. The matrix acidizing procedure will depend upon the current formation pressure, temperature, and any limitations in injection rates.

This paper reports on development of a KBS using the Zelda B-7 well as a model for a feasibility (preliminary) study.

Recently, after this well was serviced to perform an ESP repair, it was returned to production at 250 to 400 BFPD with ±50% water cut (fluctuating) and the ESP frequently suffering from underload. The well is suspected of have a lack of fluid influx due to productivity impairment following the well service. The recommendation is to perform matrix acidizing, and the VP-Expert is then utilized as a modeling toll for the KBS development. The KBS modeling flowchart is outlined as follows: