Drilling Fluid Optimization Design System Based On Fusion Reasoning Model

The drilling fluid design is one important part of the drilling engineering design and is also important theoretical support to the drilling fluid field construction. According to current researches, the drilling fluid design mainly depends on specified staff through the analysis to relative data from the oil wells, experimental data and experience of the designers. The design outcomes are different so that the quality of the design is hard to be guaranteed and the formats don’t follow the same standard, which makes it hard to keep up with the international competence. As a result, to improve the design quality of drilling fluid, it’s common to introduce the computer and the AI systems to aid the design, aiming at solving the problems in traditional design. Meanwhile, with the rapid development of oil exploration technology and the increasing of oil demand, higher demands on drilling fluid are required by modern drilling technology. New types of drilling technology are continually developed and applied. In pursuit of high efficiency and low cost, the intelligent design and management of drilling fluid have been placed more and more emphasis on. Accordingly, it’s quite necessary to develop the software of modern drilling fluid design and managing drilling fluid data.To sum up, the drilling fluid optimization design system was developed to aid the drilling fluid design based on case-based reasoning, rule-based reasoning and support vector machine, improving the efficiency and the quality of the design. The system avoided the simplification of one models, and deviations of the models from the reality and fitted the experts’thoughts. The results of the essay could be summarized as follows:(1) The processes and theory of traditional drilling fluid design were researched and a computer aided drilling fluid design system similar to the experts’thoughts was developed.(2) Technical data of drilling fluid from big oilfields were collected and case bases of drilling fluid were established. The experience from experts in drilling fluid was collected and rule bases were established. Based on the laboratory experiment data and drilling fluid formulation, established prediction model of drilling fluid formulation combined with the support vector machine.(3) The principals, individual applications and comprehensive applications of case-based reasoning, rule-based reasoning and support vector machine were studied.(4) Drilling Fluid Optimization Design System was developed on the Visual Studio2010platform,vb.net and c++as programming language. The result of the system design and field application examples of five wells were compared, and found that the system better to complete the type selection and performance parameters design of drilling fluid.