基于矿用可穿戴设备与智能分析算法的数据处理技术研究

Research on data processing technology based on mining wearable equipment and intelligent analysis algorithm

针对采矿作业人员的身体健康监测问题,文中提出了一种基于矿用穿戴设备智能数据分析的采矿作业人员健康状态评估方法。该方法结合了多源数据融合方法和支持向量机(SVM)算法,对矿用可穿戴设备采集的数据进行预处理;在数据分析层面进行多源数据融合;采用小波变换方法提取数据特征参数,并将特征参数作为支持向量机算法的输入;计算出采矿作业人员的健康评估结果。仿真算例表明,文中所提的SVM算法相比于BP神经网络算法,在准确性、计算时间和收敛速度方面具有更优的性能;同时所提采矿作业人员的健康状态评估方法融合了多种特征数据,得到的健康状态评估结果更为精确。

Aiming at the problem of physical health detection of mining workers,this paper proposes an evaluation method of mining workers’ health status based on intelligent data analysis of mining wearable devices. This method combines the multi-source data fusion method and Support Vector Machine(SVM)algorithm. The data collected by the wearable equipment for mining is preprocessed;secondly,the multisource data fusion is carried out at the data analysis level. The wavelet transform method is used to extract the data feature parameters,and the feature parameters are taken as the input of support vector machine algorithm. The health of mining workers is calculated Health assessment results. The simulation results show that compared with BP neural network algorithm,the proposed SVM algorithm has better performance in accuracy,calculation time and convergence speed;at the same time,the proposed health state assessment method for mining workers integrates a variety of characteristic data,and the health status assessment results are more accurate.