The Research And Application Of Monitoring Information Of Coal Mine Workers Based On Data Mining

Coal production enterprises has always attached great importance to safety of coal mine workers. These companies have invested a lot of manpower and material resources to build a variety of mine shaft personnel monitoring systems, such as personnel positioning system, iris recognition system and so on. With these systems, managers can know the number of mine workers, the activities of mine workers track and the distribution of mine workers in a timely manner, in order to make reasonable scheduling management. But in actual use, due to.the vulnerability management and human factors, it will make errors, omissions occur and the wrong situation reported. At the same time, there may be a variety of monitoring systems for mine workers in a coal mine, it can not objectively reflect the real situation of mine workers to consider in isolation monitoring information for each system.With long-term use of these monitoring systems, it has accumulated a large amount of data recorded information on the activities of mine workers. It can be found in activity patterns of mine workers to analysis and mining these vast amounts of historical data, so that can develop more accurate alarm rules and find personnel safety timely.This paper pretreated personnel positioning system historical data, formed a movement trajectory of underground workers and adopted PAM clustering algorithm based on the edit distance on movement trajectory, got a typical path sets of underground workers; Meanwhile, this paper fused and analysis historical data of the iris-attendance system and the personnel positioning system, formulated abnormal alarm rules of in out well, reduced the unreliability of previous rules based on user experience; Finally, developing analysis alarm system of underground workers, using database technology to get real-time positioning data and iris recognition data, and integrating two kinds of data in order to analyze real-time data and real-time alarming for violation of the rules of personnel monitoring data based on standard path and the rules of differences between the two systems.