基于分布式光纤声传感技术的水泥环流体泄漏工况诊断

doi:10.3969/j.issn.2096-1693.2025.02.031

石油科学通报 ›› 2025, Vol. 10 ›› Issue (6): 1330-1349. doi: 10.3969/j.issn.2096-1693.2025.02.031

基于分布式光纤声传感技术的水泥环流体泄漏工况诊断

李晓蓉¹^,^*(), 苏飞宇², 李仨兴², 赵杨³, 司晓宇¹, 冯永存²

1 中国石油大学(北京)安全与海洋工程学院，北京 102249
2 中国石油大学(北京)石油工程学院，北京 102249
3 中国石油大学(北京)非常规油气科学技术研究院，北京 102249

收稿日期:2024-10-08 修回日期:2025-01-16 出版日期:2025-12-30 发布日期:2025-12-30
作者简介:李晓蓉(1986年—)，副教授，博导，长期从事复杂油气井井壁稳定性和井筒完整性基础理论、方法和技术研究，主持国家自然科学基金项目等多项基础科研项目和工程技术服务项目，在《SPE Journal》等期刊发表科技论文30余篇。授权发明专利10余项。任中国岩石力学与岩石工程学会国际秘书处兼职副秘书长；国际岩石力学大会(IGS)论文委员会联合主席等，xiaorongli@cup.edu.cn。
基金资助:
国家自然科学基金面上项目“应力—腐蚀耦合作用下CCUS井水泥环密封失效机理研究”(5247040354);国家自然科学基金青年项目“分布式光纤声传感(DAS)监测水泥环完整性研究”(52004298)

Diagnosis of cement sheath fluid leakage based on distributed optical fiber sensing

LI Xiaorong¹^,^*(), SU Feiyu², LI Saxing², ZHAO Yang³, SI Xiaoyu¹, FENG Yongcun²

1 College of Safety and Ocean Engineering, China University of Petroleum, Beijing 102249, China
2 College of Petroleum Engineering, China University of Petroleum, Beijing 102249, China
3 Unconventional Oil and Gas Science and Technology Institute, China University of Petroleum, Beijing 102249, China

Received:2024-10-08 Revised:2025-01-16 Online:2025-12-30 Published:2025-12-30
Contact: * xiaorongli@cup.edu.cn

摘要/Abstract

摘要：

水泥环完整性对油气井安全高效生产至关重要。随着油田开发的深入，增产增注等作业可能导致水泥环产生微裂隙，削弱其封隔油气水层的能力。传统固井检测方法只能提供单点或瞬时状态，难以实现全井段的实时监测。分布式光纤声传感技术为水泥环泄漏的实时监测提供了全新方法。基于自主搭建的分布式光纤监测井筒完整性实验平台，开展了不同泄漏通道尺寸、位置和泄漏量下的水泥环失效监测实验，获取了光纤监测数据。在传统谱减降噪方法中引入过减因子，有效解决了谱减过程的负值问题，通过结合短时傅里叶变换和连续小波变换提取泄漏信号的时频域特征，从而确定泄漏位置。同时，通过计算光纤信号的功率谱分布和声压级，建立了泄漏流量与声波能量之间的关系。最后，提出了一种基于卷积神经网络(Convolutional Neural Networks, CNN)和双向门控循环单元(Bidirectional Gated Recurrent Unit, BiGRU)的特征识别方法，实现了对泄漏工况的精准分类。研究结果表明，改进的谱减法在低频区域(0~100 Hz)能有效抑制宽频带噪声和脉冲噪声，抑制幅度可达 100 dB 至 120 dB。泄漏流量与声波能量呈正相关，流量仅改变特征频率的峰值，不影响频率分布特征。构建的CNN-BiGRU模型能有效识别光纤信号的空间和时序特征，表现出较高的准确性和良好的泛化能力，为分布式声光纤监测水泥环完整性的数据解释提供了支持。

关键词: 分布式声光纤传感器, 水泥环完整性, 时频分析方法, CNN-BiGRU, 深度学习模型

Abstract:

The integrity of cement sheath plays a decisive role in ensuring the safe and efficient production of oil and gas wells. However, the micro-cracks inside the cement sheath make the interlayer isolation fail with the further development of the oil field. The state of cement sheath at a certain measuring point or at a certain moment is provided by traditional detection methods, which cannot meet the needs of the whole well section real-time monitoring. Distributed acoustic optical fiber sensing provides a new method for real-time monitoring of fluid leakage in micro-cracks of cement sheath. In this paper, the data of fluid leakage under different leakage sizes, leakage locations and leakage flow rates are obtained based on the distributed optical fiber monitoring wellbore integrity experimental platform. The over-subtraction factor is introduced into the traditional spectral subtraction noise reduction method, which effectively solves the negative value problem of the spectral subtraction process. The time-frequency domain characteristics of the leakage signal are extracted by combining the Short-Time Fourier Transform and the Continuous Wavelet Transform to determine the leakage location. Then, the relationship between fluid leakage flow and acoustic energy is determined by calculating the power spectrum distribution and sound pressure level of the signal. Finally, a feature recognition method based on Convolutional Neural Network (CNN) and Bidirectional Gated Recurrent Unit (BiGRU) is proposed to achieve accurate classification of leakage conditions. The results show that the improved spectral subtraction effectively suppresses broadband noise and impulse noise in the low frequency region (0-100 Hz), and the suppression amplitude reaches 100 dB to 120 dB. The acoustic energy captured by the fiber is positively correlated with the fluid leakage, and the flow rate does not change the frequency distribution characteristics, but only changes the peak value of the frequency. It is found that the CNN-BiGRU model has high accuracy and good generalization ability, which effectively identifies the spatial and temporal features of the signal. The information of interlayer isolation failure of cement sheath is effectively obtained by distributed acoustic sensor, which has certain guiding significance for cementing operation.

Key words: distributed acoustic sensor, cement sheath integrity, time-frequency analysis, CNN-BiGRU, deep learning model

中图分类号:

TE25
O439

李晓蓉, 苏飞宇, 李仨兴, 赵杨, 司晓宇, 冯永存. 基于分布式光纤声传感技术的水泥环流体泄漏工况诊断[J]. 石油科学通报, 2025, 10(6): 1330-1349.

LI Xiaorong, SU Feiyu, LI Saxing, ZHAO Yang, SI Xiaoyu, FENG Yongcun. Diagnosis of cement sheath fluid leakage based on distributed optical fiber sensing[J]. Petroleum Science Bulletin, 2025, 10(6): 1330-1349.

https://sykxtb.cup.edu.cn/CN/Y2025/V10/I6/1330

图/表 30

图1 井筒中泄漏的主要通道^[4]

Fig. 1 Main leakage channels in the wellbore^[4]

表1 全分布式光纤传感技术的类型和原理

Table 1 Types and principles of fully distributed optical fiber sensing technology

散射原理	解调技术	测量参数	测量长度	空间分辨率	精度	不足	应用
拉曼散射	ROTDR/DTS	温度	约10 km	约1 m	温度精度：0.1 ℃	空间分辨率低，米量级。	长距离温度测量
布里渊散射	BOTDA BOFDA	应变、温度	50~100 km	约0.2~1 m	应变精度：10^-5 温度精度：0.5 ℃	系统复杂，测量耗时长，测量频率慢。	长距离应变和温度测量
瑞利散射	DAS/OTDR	应变率	20~25 km	约1 m	应变精度：10^-8 低频：0.001 Hz	空间分辨率低，米量级。	长距离声波信号监测
瑞利散射	OFDR	应变、温度	约100 m	约1 mm	应变精度：10^-6 温度精度：0.1 ℃	测量距离短，测量频率较慢。	短距离高分辨率应变和温度测量

图2 测试管道上螺旋缠绕光纤的示意图^[13]

Fig. 2 Schematic diagram of spirally wound optical fiber on a test tube^[13]

图3 分布式光纤的安装方式^[21]

Fig. 3 Installation methods for distributed optical fibers^[21]

表2 光纤信号降噪处理方法对比

Table 2 Comparison of optical fiber signal noise reduction processing methods.

方法	特点	性能对比	适用条件
小波变换	通过正交小波变换获取信号的小波系数，确定合适的阈值，将小于阈值的小波系数处理掉，并进行信号重构。	有效地对非平稳信号进行降噪，需要选取合适参数才能获取良好的降噪效果，且计算复杂度较高。	适用于具有多尺度特性的非平稳信号
滑动平均法	对噪声点附近的采样点进行算数平均，作为该点光滑后的值。	计算过程简单，但对强噪声信号抑制能力差，会模糊信号的细节。	信号波动较小的情况
数字滤波器	输入、输出均为数字信号，通过一定运算关系，改变输入信号所含频率成分的相对比例，获取纯净信号。	能够有效去除特定频率范围的噪声，但对未知或非周期性噪声效果不佳。	已知高频/低频噪声信号频率范围
谱减法	利用含噪信号的幅度谱/能量谱与噪声的幅度谱/能量谱相减，得到估计的干净信号的幅度谱/能量谱。	对噪声估计值较为依赖，噪声估计不准确时可能导致信号失真，但在噪声平稳情况下表现良好。	噪声近似平稳，并具有可加性
自适应滤波	利用前一时刻已获得的滤波器参数，自动调节当前时刻的滤波器参数，以适应信号与噪声未知特性。	可以处理非线性和非平稳噪声，但需要较长时间的训练，实时性差。	噪声特性未知或存在随机性

图4 DAS数据采集系统

Fig. 4 Cquisition system for DAS data

图5 水泥环流体泄漏实验平台(左—DAS数据采集部分，中—光纤连接部分，右—水泥环完整性监测部分)

Fig. 5 Cement sheath fluid leakage experimental platform (Left: DAS data acquisition part Middle: fiber connection part. Right: cement sheath integrity monitoring part.)

表3 DAS监测水泥环流体泄漏实验材料

Table 3 DAS monitoring cement sheath fluid leakage experi-mental materials

类型	性能
单模光纤	前端光纤100 m，测试段20 m，尾端30 m
地层套管	外径150 mm×内径140 mm×高度1500 mm
表层套管	外径50 mm×内径46 mm×高度1500 mm
泄漏通道	直径为20 mm、25 mm和32 mm

图6 实验室井筒模型，包含预置泄漏通道(左—上通道25 mm下通道25 mm；右—上通道20 mm、下通道32 mm)

Fig. 6 Laboratory wellbore model with pre-set leakage path. (Left: upper channel 25 mm, lower channel 25 mm. Right: upper channel 20 mm, lower channel 32 mm.)

图7 分布式光纤传感监测原理^[34]
光相位与扰动事件产生的轴向应变成正比，其线性关系系数K_F与光纤的折射率和泊松比等性质有关^[35]。可通过记录瑞利背向散射光强度和相位来反映光纤的时空应变状态，揭示事件发生的信息。

Fig. 7 Monitoring principle of distributed optical fiber sensing^[34]

图8 时频域示意图^[37]

Fig. 8 Schematic diagram of time-frequency domain^[37]

图9 GRU 网络单元内部结构示意图

Fig. 9 Schematic diagram of the internal structure of GRU

图10 残差块结构示意图

Fig. 10 Schematic diagram of residual block structure

图11 CNN-BiGRU模型示意图

Fig. 11 Schematic diagram of CNN-BiGRU model

图12 流体泄漏工况监测流程图

Fig. 12 Flow chart of fluid leakage condition monitoring

图13 信号预处理和降噪效果图(32 mm泄漏通道数据)

Fig. 13 Signal pre-processing and noise reduction effect

图14 不同管径流体泄漏能量谱

Fig. 14 Energy spectrum of fluid leakage in different diameters

图15 不同管径流体泄漏短时傅里叶变换时频图谱

Fig. 15 Short-Time Fourier Transform spectra of fluid leakage with different diameters

图16 3种泄漏通道的光纤信号时频图

Fig. 16 Time-frequency diagram of optical fiber signals of three leakage channels

图17 经典混淆矩阵示意图

Fig. 17 Schematic diagram of the classic confusion matrix

图18 25 mm管径下不同流量流体泄漏光纤应变云图

Fig. 18 Optical fiber strain cloud diagram of different flow rate fluid leakage under 25 mm pipe diameter

图19 25 mm裂缝通道声压级随泄漏流量变化曲线

Fig. 19 Curve of sound pressure level of 25 mm crack channel changing with leakage flow

图20 20 mm和32 mm泄漏尺寸下流体泄漏光纤应变云图

Fig. 20 Strain cloud diagram of fluid leakage under different leakage sizes of 20 mm and 32 mm

图21 CNN-BiGRU模型的准确度变化曲线

Fig. 21 Accuracy curve of CNN-BiGRU model

图22 CNN-BiGRU模型的Loss值变化曲线

Fig. 22 Loss value change curve of CNN-BiGRU model

图23 CNN-BiGRU模型的混淆矩阵

Fig. 23 Confusion matrix of CNN-BiGRU model

图24 分类前、后K-Means聚类结果

Fig. 24 K-Means clustering results before and after classification

图25 CNN、CNN-GRU、CNN-BiGRU的训练和验证曲线

Fig. 25 Training and verification curves of CNN, CNN-GRU and CNN-BiGRU

图26 CNN和CNN-GRU的混淆矩阵

Fig. 26 Confusion matrix of CNN and CNN-GRU

表4 CNN、CNN-GRU、CNN-BiGRU的二级分类指标

Table 4 Secondary classification indicators of CNN, CNN-GRU and CNN-BiGRU

模型	准确率	精准率/%			灵敏度/%			特异度/%
模型	准确率	20 mm	25 mm	32 mm	20 mm	25 mm	32 mm	20 mm	25 mm	32 mm
CNN	92.90	90.48	100	88.29	87.96	100	90.74	95.37	100	93.98
CNN-GRU	94.44	92.45	100	90.91	90.74	100	92.59	96.30	100	95.37
CNN-BiGRU	97.84	94.69	100	99.03	99.07	100	94.44	97.22	100	99.54

参考文献 43

[1]	NORSOK.Standard D-010 well integrity in drilling and well operations (2nd edition)[M]. Oslo, Norway: Standards Norway, 1998.
[2]	GU C, FENG Y, MOSLEH M H, et al. Fluid-driven debonding of cement interfaces of an injector well: A coupled thermo-hydro-mechanical approach[J]. Rock Mechanics and Rock Engineering, 2024: 1-19.
[3]	GU C, SU F, FENG Y, et al. Investigation on radial crack of cement sheath during CO₂ injection using a TM-XFEM coupled model[C]// Arma US Rock Mechanics/Geomechanics Symposium. ARMA, 2023: ARMA-2023-0525.
[4]	CARROLL S, CAREY J W, DZOMBAK D, et al. Role of chemistry, mechanics, and transport on well integrity in CO₂ storage environments[J]. International Journal of Greenhouse Gas Control, 2016, 49: 149-160. doi: 10.1016/j.ijggc.2016.01.010 URL
[5]	谢荣华, 刘继生, 李晓伟. 固井质量评价技术发展及其对油田开发的影响[J]. 测井技术, 2019, 43(04): 339-343.
	[XIE R H, LIU J S, LI X W. Development of cementing quality evaluation technology and its influence on oilfield development[J]. Logging Technology, 2019, 43(04): 339-343.]
[6]	MEEK R, HULL R, WOLLER K, et al. Estimation of hydraulic fracture height and pressure deflation using a pulsed vertical seismic profile and a DAS fiber in the Midland basin[C]// SPE/AAPG/SEG Unconventional Resources Technology Conference. URTEC, 2019: D013S019R004.
[7]	JIN G, ROY B. Hydraulic-fracture geometry characterization using low-frequency DAS signal[J]. The Leading Edge, 2017, 36(12): 975-980. doi: 10.1190/tle36120975.1 URL
[8]	JOHANNESSEN K, DRAKELEY B, FARHADIROUSHAN M. Distributed acoustic sensing-a new way of listening to your well/reservoir[C]// SPE intelligent energy international conference and exhibition. SPE, 2012: SPE-149602-MS.
[9]	LUO B, JIN G, STANEK F. Near-field strain in distributed acoustic sensing-based microseismic observation[J]. Geophysics, 2021, 86(5): P49-P60.
[10]	ANWAR I, CAREY B, JOHNSON P, et al. Detecting and characterizing fluid leakage through wellbore flaws using fiber-optic distributed acoustic sensing[C]// ARMA US Rock Mechanics/Geomechanics Symposium. ARMA, 2022: ARMA-2022-0264.
[11]	RAAB T, REINSCH T, ALDAZ CIFUENTES S R, et al. Real-time well-integrity monitoring using fiber-optic distributed acoustic sensing[J]. SPE Journal, 2019, 24(05): 1997-2009. doi: 10.2118/195678-PA URL
[12]	郑元辽, 刘月明. 光纤传感监测压力管道泄漏技术进展综述[J]. 传感器与微系统, 2014, 33(9): 15-17.
	[ZHENG Y L, LIU Y M. A review of the progress of optical fiber sensing technology for monitoring pressure pipeline leakage[J]. Sensors and Microsystems, 2014, 33(9): 15-17.]
[13]	HUSSELS M T, CHRUSCICKI S, ARNDT D, et al. Localization of transient events threatening pipeline integrity by fiber-optic distributed acoustic sensing[J]. Sensors, 2019, 19(15): 3322. doi: 10.3390/s19153322 URL
[14]	TANAKA S, KOBAYASHI K, SHIMOMICHI T, et al. Optical fiber sensor cable for pipe thinning detection in high temperature environment conditions[C]// 19th International Conference on Optical Fibre Sensors. SPE, 2008, 7004: 490-493.
[15]	LIM K, WONG L, CHIU W K, et al. Distributed fiber optic sensors for monitoring pressure and stiffness changes in out-of-round pipes[J]. Struct Control Health Monit. 2016, 23, 303-314. doi: 10.1002/stc.1771 URL
[16]	付士崇, 张丹, 罗群. 基于光纤分布式声波传感的输气管道泄漏识别与定位方法[J]. 激光与光电子学进展, 2025, 62(3): 111-117.
	[FU S C, ZHANG D, LUO Q. Gas pipeline leakage identification and location method based on optical fiber distributed acoustic sensing[J]. Advances in Laser and Optoelectronics, 2025, 62(3): 111-117.]
[17]	STAJANCA P, CHRUSCICKI S, HOMANN T, et al. Detection of leak-induced pipeline vibrations using fiber optic distributed acoustic sensing[J]. Sensors, 2018, 18(9): 2841. doi: 10.3390/s18092841 URL
[18]	HORNMAN J C. Field trial of seismic recording using distributed acoustic sensing with broadside sensitive fibre‐optic cables[J]. Geophysical Prospecting, 2017, 65(1): 35-46. doi: 10.1111/gpr.2017.65.issue-1 URL
[19]	SANNI M, HVEDING F, KOKAL S, et al. Lessons learned from in-well fiber-optic DAS/DTS deployment[C]// SPE Annual Technical Conference and Exhibition. SPE, 2018: D021S016R007.
[20]	NING I L C, SAVA P. Multicomponent distributed acoustic sensing[M]// SEG Technical Program Expanded Abstracts 2016. Society of Exploration Geophysicists, 2016: 5597-5602.
[21]	MUNN J D, COLEMAN T I, PARKER B L, et al. Novel cable coupling technique for improved shallow distributed acoustic sensor VSPs[J]. Journal of Applied Geophysics, 2017, 138: 72-79. doi: 10.1016/j.jappgeo.2017.01.007 URL
[22]	刘志伟, 尚盈, 王晨, 等. 基于多维空间数据融合算法的分布式光纤振动管道泄漏监测技术[J]. 激光与光电子学进展, 2023, 60(09): 404-409.
	[LIU Z W, SHANG Y, WANG C, et al. Distributed optical fiber vibration pipeline leakage monitoring technology based on multi-dimensional spatial data fusion algorithm[J]. Advances in Laser and Optoelectronics, 2023, 60(09): 404-409.]
[23]	JIA M, SUN Z, DUAN C, et al. Pipeline leaks early warning based on distributed optical fiber sensing technology[J]. IEEE Photonics Technology Letters, 2024, 36(6): 397-400. doi: 10.1109/LPT.2024.3362913 URL
[24]	刘志坤, 唐小明, 朱洪伟. 基于改进谱减法的语音增强研究[J]. 计算机仿真, 2009, 26(06): 363-366.
	[LIU Z K, TANG X M, ZHU H W. Research on speech enhancement based on improved spectral subtraction[J]. Computer Simulation, 2009, 26(06): 363-366.]
[25]	张永康, 尚盈, 王晨, 等. 分布式光纤入侵信号检测与识别[J]. 光电工程, 2021, 48(03): 19-36.
	[ZHANG Y K, SHANG Y, WANG C, et al. Distributed optical fiber intrusion signal detection and recognition[J]. Optoelectronic Engineering, 2021, 48(03): 19-36.]
[26]	郑来芳, 张俊生, 梁海坚, 等. 基于时频混合特征提取算法的光纤传感信号识别研究[J]. 电子测量与仪器学报, 2020, 34(09): 153-159.
	[ZHENG L F, ZHANG J S, LIANG H J, et al. Recognition of optical fiber sensing signals based on time-frequency hybrid feature extraction algorithm[J]. Journal of Electronic Measurement and Instrumentation, 2020, 34(09): 153-159.]
[27]	WU G, QI J, ZHOU Y. Time-frequency analysis method: Research and prospect[J]. Journal of Image and Signal Processing, 2018, 7(1): 24-35. doi: 10.12677/JISP.2018.71003 URL
[28]	WU H, QIAN Y, ZHANG W, et al. Feature extraction and identification in distributed optical-fiber vibration sensing system for oil pipeline safety monitoring[J]. Photonic Sensors, 2017, 7: 305-310. doi: 10.1007/s13320-017-0360-1 URL
[29]	靳喜博, 刘琨, 江俊峰, 等. 基于卷积神经网络的多维度分布式光纤振动传感事件识别[J]. 光学学报, 2024, 44(01): 384-394.
	[JIN X B, LIU K, JIANG J F, et al. Multi-dimensional distributed optical fiber vibration sensing event recognition based on convolutional neural network[J]. Journal of Optics, 2024, 44(01): 384-394.] doi: 10.1007/s12596-015-0272-7 URL
[30]	周子纯, 刘琨, 江俊峰, 等. 基于卷积长短期记忆全连接深度神经网络的光纤振动传感事件识别[J]. 光学学报, 2021, 41(13): 159-167.
	[ZHOU Z C, LIU K, JIANG J F, et al. Optical fiber vibration sensing event recognition based on convolutional long short-term memory fully connected deep neural network[J]. Journal of Optics, 2021, 41(13): 159-167.]
[31]	黄翔东, 张皓杰, 刘琨, 等. 基于综合特征的光纤周界安防系统高效入侵事件识别[J]. 物理学报, 2017, 66(12): 207-216.
	[HUANG X D, ZHANG H J, LIU K, et al. Efficient intrusion event identification of fiber perimeter security system based on comprehensive characteristics[J]. Acta Physical Sinica, 2017, 66(12): 207-216.]
[32]	马愈昭, 张婷婷, 朱庆啸, 等. 基于ISSA和GA-BiLSTM神经网络的光纤周界入侵事件识别[J]. 激光与光电子学进展, 2024, 61(05): 151-161.
	[MA Y Z, ZHANG T T, ZHU Q X, et al. Identification of fiber perimeter intrusion events based on ISSA and GA-BiLSTM neural networks[J]. Advances in Laser and Optoelectronics, 2024, 61(05): 151-161.]
[33]	刘均荣, 韩艳慧, 王哲, 等. 基于分布式光纤声波监测数据和机器学习的井筒流体类型识别方法[J]. 中国石油大学学报(自然科学版), 2023, 47(03): 107-114.
	[LIU J R, HAN Y H, WANG Z, et al. Identification method of wellbore fluid type based on distributed optical fiber acoustic monitoring data and machine learning[J]. Journal of China University of Petroleum (Natural Science Edition) 2023, 47(03): 107-114.]
[34]	李晓蓉, 刘旭丰, 张毅, 等. 基于分布式光纤声传感的油气井工程监测技术应用与进展[J]. 石油钻采工艺, 2022, 44(3): 309-320.
	[LI X R, LIU X F, ZHANG Y, et al. Application and progress of oil and gas well engineering monitoring technology based on distributed optical fiber acoustic sensing[J]. Petroleum Drilling and Production Technology 2022, 44(3): 309-320.]
[35]	王毅. 分布式光纤振动信号测井数据处理方法研究及产量预测[D]. 长江大学, 2023.
	[WANG Y. Research on distributed optical fiber vibration signal logging data processing method and production prediction[D]. Yangtze University, 2023.
[36]	赵晓雷. 基于短时傅里叶变换的语音信号处理研究[J]. 舰船电子工程, 2018, 38(4): 19-22.
	[ZHAO X L. Research on speech signal processing based on short time Fourier transform[J]. Ship Electronic Engineering, 2018, 38(4): 19-22.]
[37]	KENWRIGHT B. Quaternion fourier transform for character motions[C]// Workshop on Virtual Reality Interaction and Physical Simulation (VRIPHYS). The Eurographics Association, 2015: 1-4.
[38]	BAYISSA W L, HARITOS N, THELANDERSSON S. Vibration-based structural damage identification using wavelet transform[J]. Mechanical Systems and Signal Processing, 2008, 22(5): 1194-1215. doi: 10.1016/j.ymssp.2007.11.001 URL
[39]	刘刚, 刘澎涛, 杨全枝, 等. 小波包变换在油井出砂信号特征识别中的应用[J]. 石油机械, 2013, 41(9): 80-83.
	[LIU G, LIU P T, YANG Q Z, et al. Application of wavelet packet transform in feature recognition of oil well sanding signal[J]. Petroleum Machinery, 2013, 41(9): 80-83.]
[40]	KINSLER L E, FREY A R, COPPENS A B, et al. Fundamentals of acoustics[M]. John Wiley & Sons, 2000.
[41]	CHO K, VAN MERRIËNBOER B, BAHDANAU D, et al. On the properties of neural machine translation: Encoder-decoder approaches[C]// Proceedings of SSST-8, Eighth Workshop on Syntax, Semantics and Structure in Statistical Translation. Doha, Qatar: Association for Computational Linguistics, 2014: 103-111.
[42]	HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2016: 770-778.
[43]	CHUNG J, GULCEHRE C, CHO K H, et al. Gated feedback recurrent neural networks[C]// International Conference on Machine Learning. PMLR, 2015: 2067-2075.

选择文件类型/文献管理软件名称

选择包含的内容

基于分布式光纤声传感技术的水泥环流体泄漏工况诊断

Diagnosis of cement sheath fluid leakage based on distributed optical fiber sensing

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 30

参考文献 43

相关文章 1

编辑推荐

Metrics

本文评价

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

基于分布式光纤声传感技术的水泥环流体泄漏工况诊断

Diagnosis of cement sheath fluid leakage based on distributed optical fiber sensing

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 30

参考文献 43

相关文章 1

编辑推荐

Metrics

本文评价