Abstract:

The application of artificial intelligence to well test interpretation is a new subject, The study of it is at the initial stage both at home and abroad. This paper describes the development of an Expert System prototype for the reservoir model identification in well test interpretation (RMIESWT), The model chosen by computer is based on the pressure derivative curve, and simulates the visual diagnosis performed by a human expert. A part of the reasoning involved in such a diagnosis uses a symbolic representation of the derivative curve，which，in the case of a human expert，is built almost unconsciously, Therefore，it is a very good simulation technique of artificial intelligence. Besides test data, the system seeks additional information available from other sources, such as drilling，production, well logging，coring，fluid sampling，enhanced recovery treatment and geology，to identify a reservoir model, which is particularly available to the incomplete test data. The expert system tool CM.1 is chosen as an expert system environment, which is a current tool，based on rules. The knowledge base is organized in a hierarchy of sub-problems whose logical relations are clear and easy to understand. Any rule and the certainty factor of the rule can feasibly deleted, inserted and changed in the knowledge base. During the consultation session between a user and the expert system, all the displays are in a way of the menu and in Chinese, The system is easily acceptable for the users in oilfield.

Key words: Artificial intelligence, Expert system, Application of well test, Model identification