SMART-Agua: a Hybrid Intelligent System for Diagnosis of Water Production Problem

This paper describes a hybrid intelligent system, called SMART-Agua, which is aimed to diagnose and give solution to water production problems in petroleum wells. Several innovative techniques like a multi-set based fuzzy algebra of qualitative uncertainties for expert system and perception-based data mining form the technological framework of the developed system. The diagnostic phase uses the expert knowledge and experience stored in the knowledge base in the form of fuzzy rules, to profile the nature and cause of the problem. The data mining tools optimize the evaluation of water production behavior and identify the interferences between wells based on production data and geologic model analysis. At the solution analysis phase, the well´s potential for additional productivity if the water cut is reduced is analyzed and solution plan is generated. SMART-Agua is currently at field testing phase in PEMEX, Mexican Oil Company. A case study and some preliminary results are discussed