Drilling Fluid Optimization Based On Knowledge Engineering System

Now knowledge has become a key resource of enterprise, so it is very important for all of enterprises to collect and use it. Drilling is a knowledge-intensive job, and the design of drilling fluid needs insistently knowledge management. Many oil fields have tried to develop some systems to help engineer design drilling fluid, but these systems emphasize concept reasoning and only get a instructive conclusion. At the same time, with the advancement of the technology of drilling fluid, these systems have not been suit for the change on the representation of knowledge and the management of knowledge base, and it can't meet the requirement of oilfields.This thesis is to help a engineer to design a proper prescription based on past knowledge. Because coarse knowledge seems disordering, methodology called CommonKADS is used to instruct people to analyze the drilling fluid, and to set up the knowledge model, deducing model and task model. Knowledge, Therefore, this model is implemented in several logical layers. When one layer changes, it affects other layers at minimal degree. The representation of knowledge is making the knowledge stored properly in computer. In this thesis, a combination of frameworks and production is used to represent the drilling fluid knowledge, for it is able to express the diversities and heuristics of the drilling fluid knowledge. A normal match-arithmetic usually works inefficiently, so this thesis puts forward to a new match-arithmetic, which matches the model with a batch of data according to the priority. Based on these researches, this thesis develops a two-tier system for the design of drilling fluid system.