基于PCA&mdash;SVDD方法的钻头异常钻进识别

PDF(1793 KB)

振动与冲击 ›› 2015, Vol. 34 ›› Issue (13) : 158-162.

论文

基于PCA—SVDD方法的钻头异常钻进识别

刘刚1,刘闯1,夏向阳1,裴重潋1蔡鹏2，赵少伟3

作者信息 +

Bit abnormal drilling condition recognition based on PCA – SVDD

Liu Gang1，Liu Chuang1，Xia Xiangyang1，Pei Chonglian1,CAI Peng2, ZHAO Shao wei3

Author information +

文章历史 +

摘要

为了快速识别钻头异常钻进情况，达到实时监测丛式井防碰需求，根据钻头振动信号的幅值归一化频数统计特征，提出一种基于PCA—SVDD的钻头异常钻进识别模型。提取钻头正常钻进信号的每一帧数据的归一化频数特征，该特征与波形的真实幅值大小无关，适合不同工作情况，将提取的特征应用PCA方法降维处理得到钻头正常钻进的特征向量作为训练样本，建立基于PCA—SVDD的钻头异常钻进诊断模型，通过现场数据检验，证明该方法可以有效、快速地识别钻头异常钻进情况。

Abstract

In order to detect bit abnormal drilling condition quickly and monitor the cluster wells timely, establish bit abnormal drilling recognition model based on PCA—SVDD, according to the characteristics of normalized amplitude frequency of bit vibration signals. Extract the normalized frequency characteristics, which was irrelevant to real amplitude value, of each frame of the bit normal drilling signal, which can adapt to every different situations. Taking the feature vector of signals in normal working conditions whose dimension was reduced by PCA as training samples, a PCA—SVDD bit abnormal drilling diagnostic model was established .The result of real experiment data showed that this method can detect abnormal condition effectively.

导出引用

刘刚1,刘闯1,夏向阳1,裴重潋1蔡鹏2，赵少伟3. 基于PCA—SVDD方法的钻头异常钻进识别[J]. 振动与冲击, 2015, 34(13): 158-162

Liu Gang1，Liu Chuang1，Xia Xiangyang1，Pei Chonglian1,CAI Peng2, ZHAO Shao wei3. Bit abnormal drilling condition recognition based on PCA – SVDD[J]. Journal of Vibration and Shock, 2015, 34(13): 158-162

参考文献

[1]刘刚，陈超，蔡鹏，等.井眼防碰技术在南海油田W9H的应用[J].科学技术与工程，2012，12(26)：6601-6604.
Liu Gang，Chen Chao， Cai Peng ，et al. The application of anti-collision monitoring technique in the well W9H of the South China Sea oilfield[J]Science Technology and Engineering,，2012，12(26)：6601-6604.
[2]刘刚，陈超，陈全枝，等.井眼防碰监测中的特征信号特征识别[J].石油机械，2012，40（8）：76:79.
Liu Gang，Chen Chao，Yang Quanzhi，et al. Feature recognition of risk signal in borehole anti-collision monitoring[J].China Petroleum Machinery，2012，40（8）：76:79.
[3]刘刚，孙金，何保生，等.定向井防碰地面监测系统设计及现场试验[J].石油钻探技术，2012,40（1）：7-11.
Liu Gang，Sun Jing，He Baosheng，et al. Design and field test of surface monitoring system for directional wells anti-collision[J].Petroleum Drilling Techniques，2012，40（8）：76:79.
[4]Deng Yongjun, Wang Wei，Qian Chengchun, et al. Boundary-processing technique in EMD method and Hilbert transform[J]. Chinese Science Bulletin，2001，46（11）：954-961.
[5]杜陈艳，张榆锋，杨平，等.经验模态分解边缘抑制方法综述[J].仪器仪表学报，2009，30（1）：56-59.
Du Chenyan，Zhang Yufeng，Yang Ping，et al. Approaches for the end effect restraint of empirical mode decomposition algorithm[J]. Chinese Journal of Scientific Instrument，2009，30（1）：56-59.
[6]Qi Keyu，He Zhengjia，Zi Yangang. Cosine window-based boundary processing method for EMD and its application in rubbing fault diagnosis[J].Mechanical Systems and Signal Processing，2007，21（7）：2750-2760.
[7]Jolliffe I T. Principal component analysis[M].New York: Springer Verlag,1989.
[8]Wang S W，Cai J T. Sensor-fault detection ，diagnosis and estimation for centrifugal chiller systems using principal-component analysis method[J].Applied Energy，2005,82（3）：197-213.
[9]Tax D M J，Duin R P W. Support vector domain description[J].Pattern Recognition Letters，1999，20（11/13）:1191-1199.
[10] Tax D M J，Duin R P W. Support vector data description[J].Machine Learning，2004，54（1）：45-66.
[11]Vapnik V N. The nature of statistical learning theory[M].New York: Springer Verlag，1995.
[12]Cho H W，Jeong M K，Kwon Y. Support vector data description for calibration monitoring of remotely located micro-robotic system[J].Journal of Manufacturing systems，2006，25（3）：196-208.
[13]祝志博，王培良，宋执环.基于PCA-SVDD的故障检测和自学习辨识[J].浙江大学学报（工学版），2010，44（4）：652-657.
Zhu Zhibo，Wang Peiliang，Song Zhihuan. PCA-SVDD based on fault detection and self-learning identification，Journal of Zhejiang University（Engineering Science），2010，44（4）：652-657.
[14]章林柯，何琳，阎兆立，等.声学故障的SVDD评估方法研究[J].声学技术，2011，30（4）：82-85.
Zhang Linke，He lin，Yan Zhaoli，et al. Evaluation of acoustic fault based on the support vector data description method[J].Technical Acoustics，，2011，30（4）：82-85.
[15]蔡金燕，杜敏杰.多分类SVDD混叠域识别新方法与故障诊断应用[J].航天控制，2012，30（6）：83-88.
Cai Jinyan，Du Minjie. A novel approach to discrimination the overlap region of multi-class classification SVDD and fault diagnosis application[J].Aerospace Control，2012，30（6）：83-88.
[16] 戚元华，林伟国，吴海燕. 基于时域统计特征的天然气管道泄漏检测方法[J].石油学报，2013,34（6）:1195-1199.
Qi Yuanhua，Lin Weiguo，Wu Haiyan. A leak detection method for natural gas pipelines based on time-domain statistic features[J]. Acta Petrolei Sinica，2013，34（6）：1195-1199.