| The system of crude oil development is large and complex.It involves a large number of data which impacts the production of crude oil.At present,the management of oil field production is based on the production of crude oil,and oil production has become an important basis for measuring the level of oil development.In order to ensure the high and stable yield of crude oil production,the early warning system needs to be established.It could make early warning for the crisis of oil development.According to early warning information,the managers could change the old plan to ensure the healthy development of oil field development.In this thesis,we use the algorithm of association rules to complete early warning index selection process,and applying the support vector machine(SVM)prediction algorithm to predict the yield of crude oil,according to the actual demand of oil field the early warning methods based on the planning value is proposed.Firstly,this thesis has carried on the brief introduction to the theory of data mining technology and early warning.Then completing the demand analysis of the oil field development dynamic early warning system according to the theoretical research and the actual situation.Finally completing the system design and implementation according to the needs analysis.Then using the Java language and MySQL to complete the development of the system,get the oil field development dynamic early warning system based on data mining.The system includes functional modules of system management,user management,data report management,crude oil production forecast and warning subsystems.The SVM prediction algorithm is verified by the actual production data.The results show that the early warning system can complete the prediction and early warning of crude oil production.According to the early warning information the manager could make adjustments to the development plan.It could improve the production management level of oilfield crude oil. |
PDF Full Text Request