基于ST-DBSCAN聚类算法的矿井冲击地压微震监测数据时空特性分析

Analysis of spatio-temporal characteristics of mine rock burst microseismic monitoring data based on ST-DBSCAN clustering algorithm

矿井冲击地压是地下矿山中的一种常见地质灾害,对矿山的安全生产和员工的人身安全造成威胁。矿井常规的数据分析多采用数理统计方法进行危险预警分析,具有一定的片面性和局限性,对高维的微震监测数据时空特性分析明显不足,数据分析缺乏实时性,预警分析对特定软件的依赖性过大。为了提高冲击地压的预警能力,提出了一种基于ST-DBSCAN时空聚类算法的方法来分析矿井冲击地压微震监测数据的时空特性,相对于传统聚类算法,该方法表现出更高的鲁棒性。通过矿井真实微震数据验证分析,证明了ST-DBSCAN时空聚类算法能够满足冲击地压微震监测数据时、空、强分析要求,可以识别和分类工作面回采过程中的应力增高区,能够为矿井冲击地压的预警和管理提供决策支持。

Mine rock burst is a common geological disaster in underground mines,which poses a threat to the safe production of mines and the personal safety of employees.Conventional data analysis in mines often uses mathematical and statistical methods for hazard warning analysis,which has certain one sidedness and limitations.There is a significant lack of analysis on the spatiotemporal characteristics of high-dimensional microseismic monitoring data,and data analysis lacks real-time performance.Warning analysis relies too heavily on specific software.In order to improve the early warning capability of rock burst,a method based on the ST-DBSCAN spatio-temporal clustering algorithm was proposed to analyze the spatio-temporal characteristics of the mine rock burst microseismic monitoring data,and the method showed higher robustness compared with the traditional clustering algorithm.Through the verification and analysis of real microseismic data in the mine,it was proved that the ST-DBSCAN spatio-temporal clustering algorithm can meet the requirements of analyzing the rock burst microseismic monitoring data in terms of time,space,and strength.It can identify and classify the stress concentration area in the process of mining back to the working face,and provide decision-making support for the early warning and management of the rock burst in the mine.