摘要
钻孔机器人钻进过程中,随钻信息的精确采集和处理对提高作业效率、降低安全隐患、优化钻孔技术具有显著的社会经济价值。基于砂岩、沉积岩、花岗岩、石英岩在不同钻压条件下的振动信号,通过时频处理提取关于振动信号的50个特征值,应用LDA+PCA的降维方法获取特征向量,建立钻头信号的“指纹”信息,通过CNN(Convolutional Neural Networks,卷积神经网络)对“指纹”进行识别预测,提出一种多源振动信号特征提取识别方法。实践结果表明,能够对钻头信号特征进行高精度识别,综合识别率为83.3%。为钻进介质种类识别提供可靠信息源,这对提高地下矿山掘进和开采效率及安全钻进工作具有指导意义。
In the drilling process of drilling robot, the accurate acquisition and processing of information while drilling has significant social and economic value for improving operation efficiency, reducing potential safety hazards, and optimizing drilling technology. Based on the vibration signals of sandstone, sedimentary rock, granite and quartzite under different WOB conditions, 50 eigenvalues of the vibration signals were extracted through time-frequency processing. The feature vector was obtained by using the dimension reduction method of LDA+PCA, and the “fingerprint” information of the bit signal was established. The “fingerprint”was identified and predicted by CNN(convolutional neural networks). A feature extraction and recognition method of multi-source vibration signals was proposed. The practice results show that the bit signal features can be recognized with high accuracy, and the comprehensive recognition rate is 83.3%. It provides a reliable information source for the identification of drilling media, guiding significance for improving the tunneling and mining efficiency of underground mines and safe drilling.
作者
李彦忠
胡坤
汪浅予
LI Yanzhong;HU Kun;WANG Qianyu(State Key Laboratory of Mining Response and Disaster Prevention and Control in Deep Coal Mines Anhui University of Science and Technology,Huainan 232001,China;School of Mechanical EngineeringAnhui University of Science and Technology,Huainan 232001,China)
出处
《煤矿机电》
2022年第2期1-5,16,共6页
Colliery Mechanical & Electrical Technology
基金
国家重点研发计划(2020YFB1314203)
安徽省重点研发计划(202004a070200043)。
