煤矿复杂多源数据的混合相似性度量方法

A Hybrid Similarity Measurement for Complicated and Multi-Source Data in Coal Mine

随着煤矿产业自动化、电子化和信息化的发展,目前已有的数据描述方式不能有效刻画煤矿中多源、多维、动态的海量数据信息.针对该问题,首先将源于哲学范畴中的本体引入煤矿生产领域,以矿区监测位置本体作为语义模型与数值描述相结合,形成树状层次结构来描述煤矿复杂多源数据;然后,通过基于语义和数值的混合相似性度量方法计算煤矿数据的相似度;最后通过实验验证了本文所述方法的可行性和有效性.

With the rapid development of automatization; electrification, informatization in coal mine filed, the huge mining data from multi-source, multidimensional and dynamic, can not be depicted by the existing data description ways. Faced to the problem, ontology which originated from philosophy is first introduced into mining filed, and complicated and multi-source data in coal mine are described through the hierarchical tree structure formed by combinatiorl of semantic model and numerical description. Then, the similarity of complicated and multi-source data in coal mine is calculated by the hybrid measurement based on semantic and numerical similarity. The multidimensional attributes of mining data are considered in the hybrid measurement, as well as differences of location concepts. Finally, the high accuracy and good resolving power are validated through the experimental results.