The Research Into Volcanic Rock Fracturing Design And Construction System Based On Time Series Similarity Method

The volcanic rock fracturing design and construction system is constructed to design volcanic rock fracturing solution and on-site diagnostics and the choice and decision of construction control. A lot of design and construction data has been accumulated in the system. How to withdraw potential and valuable knowledge from a great deal of data to improve the efficiency has already become an urgent problem now. And most of data is fracturing curve, that is time series.This thesis focuses on finding the sub-series that are similar to the given time series. The volcanic rock fracturing design and construction system is designed and implemented in the thesis.The former Data Mining emphasizes on artificial participation, which causes the handicraft of Data Mining, the complicated handle of Data Mining, the low efficiency and low intelligence in the process of Data Mining. In order to implement the volcanic rock fracturing design and construction system better, through the combination of Agent and Data Mining technology, a Data Mining Model based on Multi-Agent is proposed in the thesis. Such problems as the organizational structure of the model, the design of various Agents, Agent collaboration and Agent communication language based on KQML are also studied and solved. This model uses the four-tier model to work together, that is, user interface level, middle management level and service provision level and data level. Every algorithm Agent can work parallel in service provision level, and the efficiency and accuracy of data mining have been improved. This model has the abilities to interactive with the environment and to integrate with data mining algorithm. Meanwhile, intelligence, initiactive and extendibility are embodied in the model.Some running results of the system are given in this thesis. The system works well.