首页 > 最新文献

Journal of Applied Geophysics最新文献

英文 中文
A multi-facies constrained geostatistical inversion for predicting fractured tight sandstone thin reservoirs: a case study in Western China 裂缝性致密砂岩薄储层的多相约束地质统计反演——以中国西部地区为例
IF 2.1 3区 地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY Pub Date : 2026-02-01 Epub Date: 2025-12-02 DOI: 10.1016/j.jappgeo.2025.106052
Sijia Li , Guangzhi Zhang , Zhenfeng Liu , Zhen Yang , Yuanyuan Tan , Jinghui Cui
This study tackles the persistent challenge of predicting thin, heterogeneous reservoirs within fault-fracture-modified tight sandstone formations by proposing a multi-facies constrained geostatistical inversion method. Traditional lithofacies-constrained geostatistical inversion methods fail to resolve high heterogeneity and accurately find low Poisson impedance “sweet spots”. To overcome these limitations, we redefine fault-fracture bodies as distinct geological facies and integrate them with lithofacies constraints within a Bayesian inversion framework to facilitate the building of the variogram with the aim of realizing partition modeling. Curvature and structural entropy are preferred as the tools to characterize the fault-fracture bodies to guide spatial modeling. Application in a western China basin demonstrates superior performance over conventional methods: the characterization accuracy of lateral heterogeneity is improved while preserving preferable vertical resolution, high-quality sandstone reservoirs are precisely positioned. The remarkable consistency between the predicted results and well log data serves as a robust validation of the proposed method's reliability. The success of this study reveals the potential of this method to conduct predictions on fractured tight sandstone thin reservoirs characterized by high lateral heterogeneity in China as well as other areas with similar geological backgrounds.
本研究通过提出一种多相约束的地质统计反演方法,解决了预测断层裂缝改造致密砂岩储层中薄而非均质储层的长期挑战。传统的岩相约束地球统计反演方法无法解决高非均质性问题,也无法准确找到低泊松阻抗的“甜点”。为了克服这些局限性,我们将断裂体重新定义为不同的地质相,并将其与岩相约束整合到贝叶斯反演框架中,以便于变异函数的建立,以实现分区建模。曲率和结构熵是表征断裂体的首选工具,可以指导空间建模。在中国西部盆地的应用表明,该方法优于常规方法:在保持较好垂向分辨率的同时,提高了横向非均质性表征精度,精确定位了优质砂岩储层。预测结果与测井数据之间的显著一致性,有力地验证了该方法的可靠性。该研究的成功表明,该方法在中国以及其他具有类似地质背景的地区具有高度横向非均质性的裂缝性致密砂岩薄储层预测中的潜力。
{"title":"A multi-facies constrained geostatistical inversion for predicting fractured tight sandstone thin reservoirs: a case study in Western China","authors":"Sijia Li ,&nbsp;Guangzhi Zhang ,&nbsp;Zhenfeng Liu ,&nbsp;Zhen Yang ,&nbsp;Yuanyuan Tan ,&nbsp;Jinghui Cui","doi":"10.1016/j.jappgeo.2025.106052","DOIUrl":"10.1016/j.jappgeo.2025.106052","url":null,"abstract":"<div><div>This study tackles the persistent challenge of predicting thin, heterogeneous reservoirs within fault-fracture-modified tight sandstone formations by proposing a multi-facies constrained geostatistical inversion method. Traditional lithofacies-constrained geostatistical inversion methods fail to resolve high heterogeneity and accurately find low Poisson impedance “sweet spots”. To overcome these limitations, we redefine fault-fracture bodies as distinct geological facies and integrate them with lithofacies constraints within a Bayesian inversion framework to facilitate the building of the variogram with the aim of realizing partition modeling. Curvature and structural entropy are preferred as the tools to characterize the fault-fracture bodies to guide spatial modeling. Application in a western China basin demonstrates superior performance over conventional methods: the characterization accuracy of lateral heterogeneity is improved while preserving preferable vertical resolution, high-quality sandstone reservoirs are precisely positioned. The remarkable consistency between the predicted results and well log data serves as a robust validation of the proposed method's reliability. The success of this study reveals the potential of this method to conduct predictions on fractured tight sandstone thin reservoirs characterized by high lateral heterogeneity in China as well as other areas with similar geological backgrounds.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"245 ","pages":"Article 106052"},"PeriodicalIF":2.1,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145737521","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Fracture detection using effective Azimuthal Young's modulus in terms of Fourier series 基于傅立叶级数的有效方位杨氏模量的裂缝检测
IF 2.1 3区 地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY Pub Date : 2026-02-01 Epub Date: 2025-11-17 DOI: 10.1016/j.jappgeo.2025.106033
Yuanyuan Tan , Guangzhi Zhang , Zhengqian Ma , Xingyao Yin
Natural fracture detection is a crucial task in exploratory geophysics. Nonetheless, it remains a challenging topic to predict fracture parameters with high precision from reflection seismic data due to the subtle seismic responses of fractures and the multi-parameter ill-posed inverse problem. Consequently, we propose a new method for predicting fracture parameters using the effective azimuthal Young's modulus in terms of Fourier series. Firstly, we derive an analytical expression of the effective azimuthal Young's modulus in terms of Fourier series for rocks with vertically aligned fractures, which isolates the fracture information from the rock matrix information. The second-order Fourier coefficient of the effective azimuthal Young's modulus (EFC) is proven to be a linear function of the fracture density. Additionally, numerical analysis reveals that the second-order EFC is always negative, which can be utilized to eliminate the 90-degree uncertainty in predicting fracture orientation. On this basis, we construct a method to estimate fracture orientation and density by introducing principal component analysis (PCA) and the discrete Fourier transformation (DFT). The effective azimuthal Young's modulus is inverted from the PCA processed azimuthal seismic data using Bayesian inversion theory. The effective azimuthal Young's modulus in the form of the truncated Fourier series can forecast the fracture orientation without the 90-degree uncertainty, where DFT is used to calculate EFC. Meanwhile, the fracture density can be directly anticipated using the second-order EFC. Ultimately, synthetic and field data examples proved the proposed method can effectively delineate the stratum with relatively developed fractures.
天然裂缝探测是勘探地球物理中的一项重要任务。然而,由于裂缝的地震反应非常微妙,且存在多参数不适定逆问题,因此利用反射地震数据进行裂缝参数的高精度预测仍然是一个具有挑战性的课题。因此,我们提出了一种利用有效方位角杨氏模量的傅立叶级数来预测裂缝参数的新方法。首先,我们推导了具有垂直排列裂缝的岩石有效方位杨氏模量的傅里叶级数解析表达式,将裂缝信息与岩石基质信息分离开来。证明了有效方位杨氏模量(EFC)的二阶傅里叶系数是裂缝密度的线性函数。此外,数值分析表明,二阶EFC总是负的,这可以用来消除预测裂缝方向的90度不确定性。在此基础上,引入主成分分析(PCA)和离散傅里叶变换(DFT),构建了裂缝方位和密度估计方法。利用贝叶斯反演理论,对主成分分析处理后的方位地震资料进行有效方位杨氏模量反演。截断傅立叶级数形式的有效方位角杨氏模量可以预测裂缝方向,而不需要90度的不确定性,其中DFT用于计算EFC。同时,利用二阶EFC可以直接预测裂缝密度。最终,综合和现场数据实例证明,该方法能有效圈定裂缝相对发育的地层。
{"title":"Fracture detection using effective Azimuthal Young's modulus in terms of Fourier series","authors":"Yuanyuan Tan ,&nbsp;Guangzhi Zhang ,&nbsp;Zhengqian Ma ,&nbsp;Xingyao Yin","doi":"10.1016/j.jappgeo.2025.106033","DOIUrl":"10.1016/j.jappgeo.2025.106033","url":null,"abstract":"<div><div>Natural fracture detection is a crucial task in exploratory geophysics. Nonetheless, it remains a challenging topic to predict fracture parameters with high precision from reflection seismic data due to the subtle seismic responses of fractures and the multi-parameter ill-posed inverse problem. Consequently, we propose a new method for predicting fracture parameters using the effective azimuthal Young's modulus in terms of Fourier series. Firstly, we derive an analytical expression of the effective azimuthal Young's modulus in terms of Fourier series for rocks with vertically aligned fractures, which isolates the fracture information from the rock matrix information. The second-order Fourier coefficient of the effective azimuthal Young's modulus (EFC) is proven to be a linear function of the fracture density. Additionally, numerical analysis reveals that the second-order EFC is always negative, which can be utilized to eliminate the 90-degree uncertainty in predicting fracture orientation. On this basis, we construct a method to estimate fracture orientation and density by introducing principal component analysis (PCA) and the discrete Fourier transformation (DFT). The effective azimuthal Young's modulus is inverted from the PCA processed azimuthal seismic data using Bayesian inversion theory. The effective azimuthal Young's modulus in the form of the truncated Fourier series can forecast the fracture orientation without the 90-degree uncertainty, where DFT is used to calculate EFC. Meanwhile, the fracture density can be directly anticipated using the second-order EFC. Ultimately, synthetic and field data examples proved the proposed method can effectively delineate the stratum with relatively developed fractures.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"245 ","pages":"Article 106033"},"PeriodicalIF":2.1,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145694551","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
An optimized physics-informed neural networks for modeling thermal stress around an open wellbore 优化的物理信息神经网络,用于模拟裸眼井筒周围的热应力
IF 2.1 3区 地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY Pub Date : 2026-02-01 Epub Date: 2025-12-11 DOI: 10.1016/j.jappgeo.2025.106062
Tuan Nguyen-Sy , Thi Loan Bui , Bao Viet Tran
This study introduces an optimized Physics-Informed Neural Networks (PINNs) for modeling thermal diffusion and resulted thermal stress around a wellbore, with applications in CO2 injection, geothermal energy, and black oil production. A semi-surrogate PINNs approach is developed by integrating synthetic data from closed-form solutions for short-term diffusion, significantly improving model accuracy in early diffusion regimes. The methodology employs advanced training techniques with Adam and L-BFGS optimizers to balance accuracy and efficiency. The parameterized PINNs model further extends the framework to accommodate varying diffusion coefficients, time scales, and nonlinear thermal behaviors. Validation against numerical methods demonstrates superior performance, particularly in long-term diffusion scenarios. This study provides a computationally efficient framework that is readily extendable to complex multi-physics scenarios, making it valuable for real-time applications in CO2 injection, geothermal energy, and related fields.
该研究引入了一种优化的物理信息神经网络(pinn),用于模拟井筒周围的热扩散和产生的热应力,并应用于二氧化碳注入、地热能源和黑油生产。通过整合短期扩散封闭解的合成数据,开发了一种半替代pinn方法,显著提高了早期扩散状态下的模型精度。该方法采用先进的训练技术与亚当和L-BFGS优化器平衡准确性和效率。参数化的PINNs模型进一步扩展了框架,以适应不同的扩散系数、时间尺度和非线性热行为。对数值方法的验证证明了优越的性能,特别是在长期扩散情况下。该研究提供了一个计算效率高的框架,可以很容易地扩展到复杂的多物理场景,使其在二氧化碳注入、地热能和相关领域的实时应用中具有价值。
{"title":"An optimized physics-informed neural networks for modeling thermal stress around an open wellbore","authors":"Tuan Nguyen-Sy ,&nbsp;Thi Loan Bui ,&nbsp;Bao Viet Tran","doi":"10.1016/j.jappgeo.2025.106062","DOIUrl":"10.1016/j.jappgeo.2025.106062","url":null,"abstract":"<div><div>This study introduces an optimized Physics-Informed Neural Networks (PINNs) for modeling thermal diffusion and resulted thermal stress around a wellbore, with applications in CO2 injection, geothermal energy, and black oil production. A semi-surrogate PINNs approach is developed by integrating synthetic data from closed-form solutions for short-term diffusion, significantly improving model accuracy in early diffusion regimes. The methodology employs advanced training techniques with Adam and L-BFGS optimizers to balance accuracy and efficiency. The parameterized PINNs model further extends the framework to accommodate varying diffusion coefficients, time scales, and nonlinear thermal behaviors. Validation against numerical methods demonstrates superior performance, particularly in long-term diffusion scenarios. This study provides a computationally efficient framework that is readily extendable to complex multi-physics scenarios, making it valuable for real-time applications in CO2 injection, geothermal energy, and related fields.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"245 ","pages":"Article 106062"},"PeriodicalIF":2.1,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145790707","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Acoustic emission responses and failure characteristics of rocks with varying rockburst tendencies under uniaxial loading 单轴加载下不同岩爆倾向岩石的声发射响应及破坏特征
IF 2.1 3区 地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY Pub Date : 2026-02-01 Epub Date: 2025-12-26 DOI: 10.1016/j.jappgeo.2025.106082
Xiaojun Wang , He Zhang , Xiao Feng , Qinglin Chen , Shirong Cao , Haowen Jiang , Jian liu
Understanding damage mechanisms in rocks exhibiting different rockburst tendencies is critical for monitoring rock damage at varied risk levels via acoustic emission (AE) technology. This study conducted rockburst tendency evaluation and uniaxial compression AE tests on limestone, granite, and red sandstone. Metallographic imaging and scanning electron microscopy (SEM) tests were also performed on fracture surfaces. AE responses and failure characteristics of rocks with different rockburst tendencies were analyzed. The parameter r (RA/AF) served as a rock damage index, with its Coefficient of Variation (CV) and AE b-value calculated. The results show that AE ring-down counts have a period of decline during the unstable crack propagation stage. Limestone and granite with rockburst tendencies fail under tension-shear coupling with mutual transitions between failure modes. The CV(r) shows a gradual decrease, stabilization, and significant increase. Red sandstone without rockburst tendencies mainly undergoes tensile action, and its CV(r) remains stable after decreasing. Compared with the traditional b-value, the CV(r) more effectively identifies failure progression, and its abrupt surge serves as a precursor for failure in rocks possessing rockburst tendency.
了解具有不同岩爆倾向的岩石的损伤机制对于利用声发射技术监测不同危险水平的岩石损伤至关重要。研究对石灰岩、花岗岩和红砂岩进行了岩爆倾向评价和单轴压缩声发射试验。对断口表面进行了金相成像和扫描电镜(SEM)测试。分析了不同岩爆倾向岩石的声发射响应和破坏特征。参数r (RA/AF)作为岩石损伤指标,计算其变异系数(CV)和AE b值。结果表明,在不稳定裂纹扩展阶段,声发射衰响次数呈下降趋势。具有岩爆倾向的灰岩和花岗岩在拉剪耦合作用下破坏,破坏模式相互转换。CV(r)呈逐渐下降、稳定和显著上升的趋势。无冲击地压倾向的红砂岩主要受拉伸作用,其CV(r)减小后保持稳定。与传统的b值相比,CV(r)能更有效地识别破坏进程,其突变波动是具有岩爆倾向的岩石破坏的前兆。
{"title":"Acoustic emission responses and failure characteristics of rocks with varying rockburst tendencies under uniaxial loading","authors":"Xiaojun Wang ,&nbsp;He Zhang ,&nbsp;Xiao Feng ,&nbsp;Qinglin Chen ,&nbsp;Shirong Cao ,&nbsp;Haowen Jiang ,&nbsp;Jian liu","doi":"10.1016/j.jappgeo.2025.106082","DOIUrl":"10.1016/j.jappgeo.2025.106082","url":null,"abstract":"<div><div>Understanding damage mechanisms in rocks exhibiting different rockburst tendencies is critical for monitoring rock damage at varied risk levels via acoustic emission (AE) technology. This study conducted rockburst tendency evaluation and uniaxial compression AE tests on limestone, granite, and red sandstone. Metallographic imaging and scanning electron microscopy (SEM) tests were also performed on fracture surfaces. AE responses and failure characteristics of rocks with different rockburst tendencies were analyzed. The parameter r (RA/AF) served as a rock damage index, with its Coefficient of Variation (CV) and AE b-value calculated. The results show that AE ring-down counts have a period of decline during the unstable crack propagation stage. Limestone and granite with rockburst tendencies fail under tension-shear coupling with mutual transitions between failure modes. The CV(r) shows a gradual decrease, stabilization, and significant increase. Red sandstone without rockburst tendencies mainly undergoes tensile action, and its CV(r) remains stable after decreasing. Compared with the traditional b-value, the CV(r) more effectively identifies failure progression, and its abrupt surge serves as a precursor for failure in rocks possessing rockburst tendency.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"245 ","pages":"Article 106082"},"PeriodicalIF":2.1,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145924829","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Retrieving reflections using passive seismic modeling: A case study based on passive seismic operation in Dehdasht Area, SW Zagros, Iran 利用被动地震模型反演反射波:以伊朗Zagros西南部Dehdasht地区被动地震作业为例
IF 2.1 3区 地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY Pub Date : 2026-02-01 Epub Date: 2025-12-23 DOI: 10.1016/j.jappgeo.2025.106076
Fatemeh Alsadat Tayeb Hosseini , Zaher-Hossein Shomali , Javad Jamali , Mohammad Reza Hatami
Passive seismic studies have made significant advancements in subsurface structure modeling. However, retrieving body-wave reflections remains challenging. This study aims to overcome this challenge through transient noise modeling and field-data application in the Dehdasht Area in southwestern Iran. In the synthetic modeling, noise sources were randomly distributed in both time and space to emulate the stochastic nature of passive seismic noise. Reflections were extracted using cross-correlation and cross-coherency techniques from both the synthetic model and the real Dehdasht data, and common virtual shot gathers were generated after suppressing surface waves to enhance reflection visibility. Even with a limited number of stations (13) and large inter-station spacing (2 km), The results demonstrate that passive seismic interferometry can successfully retrieve deep reflection waves (up to ∼7 s two-way travel time), largely independent of data acquisition geometry. Moreover, reflection hyperbolas in the virtual shot gathers were clearer when using the cross-coherency method compared to cross-correlation. A comparative analysis between common virtual shot gathers and corresponding active-source shots confirmed the consistency of retrieved reflections, highlighting the potential of passive seismic interferometry as a complementary tool to active source methods, particularly in areas with complex geological structures and high wave attenuation observed in active-source data and those identified in passive seismic recordings.Keywords: cross-coherency, cross-correlation, modeling, reflection, passive signals, virtual shot gather.
被动地震研究在地下构造建模方面取得了重大进展。然而,恢复体波反射仍然具有挑战性。本研究旨在通过伊朗西南部Dehdasht地区的瞬态噪声建模和现场数据应用来克服这一挑战。在综合建模中,噪声源在时间和空间上随机分布,以模拟被动地震噪声的随机性。利用互相关和交叉相干技术提取合成模型和真实Dehdasht数据的反射波,抑制表面波后生成共同的虚拟射击集,提高反射可见性。即使在台站数量有限(13个)和台站间距较大(2 km)的情况下,研究结果表明,被动地震干涉测量可以成功地检索深反射波(双向传播时间长达~ 7 s),在很大程度上与数据采集几何形状无关。此外,使用交叉相干方法相比于相互关联方法,虚拟镜头集中的反射双曲线更清晰。通过对比分析常见的虚拟拍摄集和相应的主动震源拍摄集,确认了反演反射的一致性,强调了被动地震干涉测量作为主动震源方法的补充工具的潜力,特别是在地质结构复杂、主动震源数据和被动地震记录中观测到的波衰减高的地区。关键词:交叉相干,互相关,建模,反射,无源信号,虚拟镜头采集。
{"title":"Retrieving reflections using passive seismic modeling: A case study based on passive seismic operation in Dehdasht Area, SW Zagros, Iran","authors":"Fatemeh Alsadat Tayeb Hosseini ,&nbsp;Zaher-Hossein Shomali ,&nbsp;Javad Jamali ,&nbsp;Mohammad Reza Hatami","doi":"10.1016/j.jappgeo.2025.106076","DOIUrl":"10.1016/j.jappgeo.2025.106076","url":null,"abstract":"<div><div>Passive seismic studies have made significant advancements in subsurface structure modeling. However, retrieving body-wave reflections remains challenging. This study aims to overcome this challenge through transient noise modeling and field-data application in the Dehdasht Area in southwestern Iran. In the synthetic modeling, noise sources were randomly distributed in both time and space to emulate the stochastic nature of passive seismic noise. Reflections were extracted using cross-correlation and cross-coherency techniques from both the synthetic model and the real Dehdasht data, and common virtual shot gathers were generated after suppressing surface waves to enhance reflection visibility. Even with a limited number of stations (13) and large inter-station spacing (2 km), The results demonstrate that passive seismic interferometry can successfully retrieve deep reflection waves (up to ∼7 s two-way travel time), largely independent of data acquisition geometry. Moreover, reflection hyperbolas in the virtual shot gathers were clearer when using the cross-coherency method compared to cross-correlation. A comparative analysis between common virtual shot gathers and corresponding active-source shots confirmed the consistency of retrieved reflections, highlighting the potential of passive seismic interferometry as a complementary tool to active source methods, particularly in areas with complex geological structures and high wave attenuation observed in active-source data and those identified in passive seismic recordings.Keywords: cross-coherency, cross-correlation, modeling, reflection, passive signals, virtual shot gather.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"245 ","pages":"Article 106076"},"PeriodicalIF":2.1,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145840195","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Deep learning high-precision geological anomaly identification method and application 深度学习高精度地质异常识别方法及应用
IF 2.1 3区 地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY Pub Date : 2026-02-01 Epub Date: 2025-12-11 DOI: 10.1016/j.jappgeo.2025.106064
Yiliang Luo , Gulan Zhang , Shiyun Ran , Xiangwen Li , Jing Duan , Chenxi Liang , Qihong Zhong , Jiawei Zhang , Caijun Cao
The popular seismic facies-guided high-precision geological anomaly identification method (FHGI) can minimize the impacts of the complexity of seismic data, the accuracy of horizon times (or depths) of the target horizons, and the space-variant seismic wavelet, thereby resulting in high-precision geological anomaly identification results; however, it still requires the horizon time information and has limitations in computational efficiency. In this paper, to achieve high-efficiency and high-precision geological anomaly identification without the horizon time information, we propose a deep learning high-precision geological anomaly identification method (HGIM). HGIM is composed of the flowchart of HGIM, the FHGI-based high-precision geological anomaly identification label automatic generation (FLG), the deep learning high-precision geological anomaly identification network (HGIN), and the loss function of HGIM. FLG aims to use the FHGI results and data augmentation to generate sufficient training data for HGIN; HGIN takes three-dimensional (3D) seismic data as its inputs, the corresponding geological anomaly labels obtained by FLG as its labels, and uses the 3D convolution kernel for high-precision geological anomaly identification; The loss function of HGIM aims to calculate the loss function which focuses on the geological anomalies. An actual 3D seismic data example demonstrates that HGIM has great potential as a technique for high-efficiency and high-precision geological anomaly identification.
目前流行的以地震相为导向的高精度地质异常识别方法(FHGI)可以最大限度地减少地震资料复杂性、目标层位层次(或深度)精度以及空间变地震小波的影响,从而获得高精度地质异常识别结果;然而,它仍然需要视界时间信息,并且在计算效率上存在局限性。为了在没有层位时间信息的情况下实现高效高精度的地质异常识别,提出了一种深度学习高精度地质异常识别方法(HGIM)。HGIM由HGIM流程图、基于fhgi的高精度地质异常识别标签自动生成(FLG)、深度学习高精度地质异常识别网络(HGIN)和HGIM的损失函数组成。FLG的目的是利用FHGI的结果和数据增强来生成足够的HGIN训练数据;HGIN以三维地震数据为输入,以FLG获得的相应地质异常标签为标签,利用三维卷积核进行高精度地质异常识别;HGIM的损失函数旨在计算以地质异常为中心的损失函数。实际的三维地震数据实例表明,HGIM技术在高效高精度地质异常识别方面具有很大的潜力。
{"title":"Deep learning high-precision geological anomaly identification method and application","authors":"Yiliang Luo ,&nbsp;Gulan Zhang ,&nbsp;Shiyun Ran ,&nbsp;Xiangwen Li ,&nbsp;Jing Duan ,&nbsp;Chenxi Liang ,&nbsp;Qihong Zhong ,&nbsp;Jiawei Zhang ,&nbsp;Caijun Cao","doi":"10.1016/j.jappgeo.2025.106064","DOIUrl":"10.1016/j.jappgeo.2025.106064","url":null,"abstract":"<div><div>The popular seismic facies-guided high-precision geological anomaly identification method (FHGI) can minimize the impacts of the complexity of seismic data, the accuracy of horizon times (or depths) of the target horizons, and the space-variant seismic wavelet, thereby resulting in high-precision geological anomaly identification results; however, it still requires the horizon time information and has limitations in computational efficiency. In this paper, to achieve high-efficiency and high-precision geological anomaly identification without the horizon time information, we propose a deep learning high-precision geological anomaly identification method (HGIM). HGIM is composed of the flowchart of HGIM, the FHGI-based high-precision geological anomaly identification label automatic generation (FLG), the deep learning high-precision geological anomaly identification network (HGIN), and the loss function of HGIM. FLG aims to use the FHGI results and data augmentation to generate sufficient training data for HGIN; HGIN takes three-dimensional (3D) seismic data as its inputs, the corresponding geological anomaly labels obtained by FLG as its labels, and uses the 3D convolution kernel for high-precision geological anomaly identification; The loss function of HGIM aims to calculate the loss function which focuses on the geological anomalies. An actual 3D seismic data example demonstrates that HGIM has great potential as a technique for high-efficiency and high-precision geological anomaly identification.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"245 ","pages":"Article 106064"},"PeriodicalIF":2.1,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145790706","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Simultaneously seismic dip estimation and random noise attenuation via dip-informed Radon dictionary 同时利用倾角通知氡字典进行地震倾角估计和随机噪声衰减
IF 2.1 3区 地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY Pub Date : 2026-02-01 Epub Date: 2025-12-16 DOI: 10.1016/j.jappgeo.2025.106059
Kuijie Cai, Xing Zhang, Yuqing Wang, Shujuan Wang
Seismic dip is a key attribute for subsurface interpretation and subsequent processing such as noise attenuation and seismic interpolation. However, existing estimation methods are often vulnerable to strong random noise and intersecting seismic events. Strong random noise can severely degrade data quality and complicate the interpretation of subsurface structures. We propose a Dip-Informed Radon Dictionary (DIRD) method that simultaneously estimates seismic dip and suppresses random noise within a unified iterative framework. The core idea is the joint estimation of dip values and the denoised seismic signals by linking them through a dip-informed Radon dictionary. Noise is attenuated via a sparse reconstruction algorithm based on the dip-informed Radon dictionary, while more accurate dip estimates are obtained during the iterative suppression process. The algorithm alternately refines the dip parameters and denoised signals via sparse optimization, which significantly improves its robustness to heavy random noise. Furthermore, the DIRD framework decomposes seismic patches into multiple dip components, providing a more accurate estimation for intersecting events within the same spatial region. Experiments on synthetic data with an input SNR of −3.99 dB show that the DIRD method achieves an output SNR of 3.27 dB, outperforming FXDECON(2.25 dB) and APF(2.50 dB). The DIRD method also demonstrates superior accuracy and robustness of dip estimation compared to nonlinear PWD, structure tensor, and Radon methods in terms of mean squared error, mean absolute error, and standard deviation.
地震倾角是地下解释和后续处理(如噪声衰减和地震插值)的关键属性。然而,现有的估计方法往往容易受到强随机噪声和地震事件交叉的影响。强随机噪声会严重降低数据质量,使地下结构的解释复杂化。提出了一种在统一迭代框架内同时估计地震倾角和抑制随机噪声的Radon字典(DIRD)方法。其核心思想是通过倾角信息Radon字典将倾角值和去噪地震信号联系起来,从而对倾角值和去噪地震信号进行联合估计。通过基于倾角通知Radon字典的稀疏重建算法衰减噪声,同时在迭代抑制过程中获得更准确的倾角估计。该算法通过稀疏优化交替对倾角参数和去噪信号进行细化,显著提高了算法对强随机噪声的鲁棒性。此外,DIRD框架将地震块分解为多个倾角分量,为同一空间区域内的相交事件提供了更准确的估计。在输入信噪比为- 3.99 dB的合成数据上进行的实验表明,DIRD方法的输出信噪比为3.27 dB,优于FXDECON(2.25 dB)和APF(2.50 dB)。与非线性PWD、结构张量和Radon方法相比,DIRD方法在均方误差、平均绝对误差和标准差方面也显示出更高的精度和鲁棒性。
{"title":"Simultaneously seismic dip estimation and random noise attenuation via dip-informed Radon dictionary","authors":"Kuijie Cai,&nbsp;Xing Zhang,&nbsp;Yuqing Wang,&nbsp;Shujuan Wang","doi":"10.1016/j.jappgeo.2025.106059","DOIUrl":"10.1016/j.jappgeo.2025.106059","url":null,"abstract":"<div><div>Seismic dip is a key attribute for subsurface interpretation and subsequent processing such as noise attenuation and seismic interpolation. However, existing estimation methods are often vulnerable to strong random noise and intersecting seismic events. Strong random noise can severely degrade data quality and complicate the interpretation of subsurface structures. We propose a Dip-Informed Radon Dictionary (DIRD) method that simultaneously estimates seismic dip and suppresses random noise within a unified iterative framework. The core idea is the joint estimation of dip values and the denoised seismic signals by linking them through a dip-informed Radon dictionary. Noise is attenuated via a sparse reconstruction algorithm based on the dip-informed Radon dictionary, while more accurate dip estimates are obtained during the iterative suppression process. The algorithm alternately refines the dip parameters and denoised signals via sparse optimization, which significantly improves its robustness to heavy random noise. Furthermore, the DIRD framework decomposes seismic patches into multiple dip components, providing a more accurate estimation for intersecting events within the same spatial region. Experiments on synthetic data with an input SNR of −3.99 dB show that the DIRD method achieves an output SNR of 3.27 dB, outperforming FXDECON(2.25 dB) and APF(2.50 dB). The DIRD method also demonstrates superior accuracy and robustness of dip estimation compared to nonlinear PWD, structure tensor, and Radon methods in terms of mean squared error, mean absolute error, and standard deviation.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"245 ","pages":"Article 106059"},"PeriodicalIF":2.1,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145790734","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Spatial distribution and geological significance of helium source rocks revealed by 3D magnetic inversion in the Huazhou-Huayin area, Weihe Basin 渭河盆地华州—华阴地区氦源岩三维磁反演空间分布及地质意义
IF 2.1 3区 地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY Pub Date : 2026-02-01 Epub Date: 2026-01-02 DOI: 10.1016/j.jappgeo.2025.106090
Xinlei Cai , Qianyi Li , Xuliang Feng , Yang Zhang , Zheng Li , Tong Luo , Mingpu Fan , Guoqiang Zhang , Mengyao Li
Helium, an important strategic rare gas resource, has a wide range of industrial applications and utilization value. Weihe Basin in central China exhibits considerable potential for the development of helium resource. The origin of helium is predominantly governed by helium source rocks. However, the distribution of helium source rocks in Weihe Basin remains unclear at least. The aeromagnetic survey is an essential method to find out the distribution characteristics of helium source rock. In 2018, Shaanxi Gas Group Co., Ltd. obtained the first helium exploration right of China in the Huazhou-Huayin area of Weihe Basin. Aeromagnetic surveys at a scale of 1:50,000 were conducted in the Huazhou-Huayin area, and yielding high-quality magnetic field data. Using the outcrop of granite pluton nearly study area as a reference, the residual magnetic anomalies potentially associated with helium source rocks were extracted from the magnetic anomalies reduced to the pole. Three-dimensional magnetic susceptibility inversion was performed to delineate the spatial distribution of these source rocks, and the results indicate that Huashan pluton is the primary helium source rock in the Huazhou-Huayin area. The Dafuyu granite body, locating in the west of Huashan pluton, extends approximately 5000 m northward beneath the thick sedimentary cover. The granites encountered in the boreholes may be sporadically distributed by analyzing the magnetic susceptibility and reflection features of two seismic profiles. From the perspective of the distribution feature for the helium source rocks, the future efforts should be focused on the central part of the study area, particularly in the regions where the Dafuyu granite body extends northward into Weihe Basin.
氦气是一种重要的战略性稀有气体资源,具有广泛的工业应用和利用价值。渭河盆地具有相当大的氦资源开发潜力。氦的成因主要受氦源岩的支配。然而,渭河盆地的氦源岩分布尚不清楚。航磁测量是了解氦源岩分布特征的重要手段。2018年,陕西燃气集团股份有限公司获得渭河盆地华州-华阴地区中国首个氦气勘探权。在华州-华阴地区开展了1:5万比例尺的航磁调查,获得了高质量的磁场数据。以近研究区花岗岩岩体露头为参照,从还原到极点的磁异常中提取可能与氦源岩相关的残余磁异常。通过三维磁化率反演,圈定了烃源岩的空间分布,结果表明华山岩体是华州-华阴地区的主力烃源岩。大富峪花岗岩体位于华山岩体西侧,在厚厚的沉积盖层下向北延伸约5000 m。通过对两条地震剖面的磁化率和反射特征分析,发现钻孔中花岗岩可能是零星分布的。从氦源岩的分布特征来看,今后的工作重点应放在研究区中部,特别是大富峪花岗岩体向北延伸至渭河盆地的区域。
{"title":"Spatial distribution and geological significance of helium source rocks revealed by 3D magnetic inversion in the Huazhou-Huayin area, Weihe Basin","authors":"Xinlei Cai ,&nbsp;Qianyi Li ,&nbsp;Xuliang Feng ,&nbsp;Yang Zhang ,&nbsp;Zheng Li ,&nbsp;Tong Luo ,&nbsp;Mingpu Fan ,&nbsp;Guoqiang Zhang ,&nbsp;Mengyao Li","doi":"10.1016/j.jappgeo.2025.106090","DOIUrl":"10.1016/j.jappgeo.2025.106090","url":null,"abstract":"<div><div>Helium, an important strategic rare gas resource, has a wide range of industrial applications and utilization value. Weihe Basin in central China exhibits considerable potential for the development of helium resource. The origin of helium is predominantly governed by helium source rocks. However, the distribution of helium source rocks in Weihe Basin remains unclear at least. The aeromagnetic survey is an essential method to find out the distribution characteristics of helium source rock. In 2018, Shaanxi Gas Group Co., Ltd. obtained the first helium exploration right of China in the Huazhou-Huayin area of Weihe Basin. Aeromagnetic surveys at a scale of 1:50,000 were conducted in the Huazhou-Huayin area, and yielding high-quality magnetic field data. Using the outcrop of granite pluton nearly study area as a reference, the residual magnetic anomalies potentially associated with helium source rocks were extracted from the magnetic anomalies reduced to the pole. Three-dimensional magnetic susceptibility inversion was performed to delineate the spatial distribution of these source rocks, and the results indicate that Huashan pluton is the primary helium source rock in the Huazhou-Huayin area. The Dafuyu granite body, locating in the west of Huashan pluton, extends approximately 5000 m northward beneath the thick sedimentary cover. The granites encountered in the boreholes may be sporadically distributed by analyzing the magnetic susceptibility and reflection features of two seismic profiles. From the perspective of the distribution feature for the helium source rocks, the future efforts should be focused on the central part of the study area, particularly in the regions where the Dafuyu granite body extends northward into Weihe Basin.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"245 ","pages":"Article 106090"},"PeriodicalIF":2.1,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145924746","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Instability and failure characteristics of surrounding rock and coal pillar of gob-side roadways under coupled longitudinal-transverse wave and stress fields during close-distance multi-seam mining 近距离多煤层开采中纵横波和应力场耦合作用下采空区巷道围岩及煤柱失稳破坏特征
IF 2.1 3区 地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY Pub Date : 2026-02-01 Epub Date: 2025-12-05 DOI: 10.1016/j.jappgeo.2025.106055
Hao Xu , Shengquan He , Feng Shen , Dazhao Song , Xueqiu He , Zhenlei Li , Majid Khan , Fanxiang Zhao
Under static-dynamic stress coupling in close-distance multi-seam mining, gob-side roadway surrounding rock and adjacent coal pillar are subjected to intense mine pressure. This study investigates coal-rock failure under coupled longitudinal-transverse wave and stress conditions. Microseismic monitoring, numerical simulations, and field measurements were conducted to show that microseismic events mainly cluster near the excavated coal seam, as well as adjacent roof and floor strata. The surrounding rock of gob-side roadway and the adjacent coal pillar (8 m wide) exhibit a higher microseismic event density compared to other areas. Under static loading, tensile failure initiates at the mid-height of the coal pillar. The roadway exhibits pronounced asymmetric deformation, with lateral displacement reaching 42 mm on the gob side and 10 mm on the coal side. Severe fragmentation occurs on the coal pillar side contributes to this asymmetric deformation. Under dynamic loading of longitudinal-transverse waves, the gob and fracture zones exhibit significantly higher attenuation than other strata. Meanwhile, surrounding rock masses and coal pillar structures show elevated dynamic responses compared to adjacent areas. The kinetic energy reaches its maximum during the longitudinal-transverse wave coupling stage, with the horizontal component exceeding the vertical component. Wave coupling intensifies asymmetric damage, leading to over 70 % of failure volume in the coal pillar. The pillar stress state transitions from compressive to tensile, with 95.8 % of the stored elastic energy released. Borehole imaging shows 7.79 m fracture depth on the pillar side and minimal damage on the coal side. The field observations confirm the reliability of numerical simulations. The analysis indicates that a remaining coal pillar above the studied coal seam causes stress concentration at the working face, with peak stress reaching 50 MPa. The combination effect of high static stress and dynamic disturbances generated by key stratum rupture serves as the main mechanism contributing to strong mine pressure behavior. This mechanism results in asymmetric roadway deformation and coal pillar instability. The findings provide a theoretical basis for optimizing support design and mitigating dynamic hazards in gob-side roadways under similar geological conditions.
近距离多煤层开采在动静应力耦合作用下,空侧巷道围岩及邻近煤柱承受着强烈的矿压。研究了煤岩在纵横波和应力耦合作用下的破坏规律。通过微震监测、数值模拟和现场实测表明,微震事件主要集中在采掘煤层附近,以及邻近的顶底板地层附近。空侧巷道围岩及相邻煤柱(8 m宽)微震事件密度高于其他区域。静载荷作用下,煤柱中高处开始出现拉破坏。巷道表现出明显的不对称变形,采空区侧移42 mm,煤区侧移10 mm。煤柱侧的严重破碎造成了这种不对称变形。在纵横波动载荷作用下,采空区和断裂带的衰减幅度明显高于其他地层。与此同时,围岩和煤柱结构相对于邻近区域表现出更高的动力响应。在纵-横波耦合阶段动能最大,水平分量大于垂直分量。波浪耦合加剧了非对称破坏,导致煤柱破坏体积的70%以上。矿柱应力状态由压向拉转变,储存的弹性能释放95.8%。钻孔成像显示矿柱侧裂缝深度为7.79 m,煤侧损伤最小。现场观测证实了数值模拟的可靠性。分析表明,研究煤层上方残余煤柱导致工作面应力集中,峰值应力达到50 MPa。高静应力与关键层破裂产生的动扰动共同作用是形成强矿压行为的主要机制。这一机制导致巷道不对称变形和煤柱失稳。研究结果为类似地质条件下采空区巷道支护优化设计和减轻动力灾害提供了理论依据。
{"title":"Instability and failure characteristics of surrounding rock and coal pillar of gob-side roadways under coupled longitudinal-transverse wave and stress fields during close-distance multi-seam mining","authors":"Hao Xu ,&nbsp;Shengquan He ,&nbsp;Feng Shen ,&nbsp;Dazhao Song ,&nbsp;Xueqiu He ,&nbsp;Zhenlei Li ,&nbsp;Majid Khan ,&nbsp;Fanxiang Zhao","doi":"10.1016/j.jappgeo.2025.106055","DOIUrl":"10.1016/j.jappgeo.2025.106055","url":null,"abstract":"<div><div>Under static-dynamic stress coupling in close-distance multi-seam mining, gob-side roadway surrounding rock and adjacent coal pillar are subjected to intense mine pressure. This study investigates coal-rock failure under coupled longitudinal-transverse wave and stress conditions. Microseismic monitoring, numerical simulations, and field measurements were conducted to show that microseismic events mainly cluster near the excavated coal seam, as well as adjacent roof and floor strata. The surrounding rock of gob-side roadway and the adjacent coal pillar (8 m wide) exhibit a higher microseismic event density compared to other areas. Under static loading, tensile failure initiates at the mid-height of the coal pillar. The roadway exhibits pronounced asymmetric deformation, with lateral displacement reaching 42 mm on the gob side and 10 mm on the coal side. Severe fragmentation occurs on the coal pillar side contributes to this asymmetric deformation. Under dynamic loading of longitudinal-transverse waves, the gob and fracture zones exhibit significantly higher attenuation than other strata. Meanwhile, surrounding rock masses and coal pillar structures show elevated dynamic responses compared to adjacent areas. The kinetic energy reaches its maximum during the longitudinal-transverse wave coupling stage, with the horizontal component exceeding the vertical component. Wave coupling intensifies asymmetric damage, leading to over 70 % of failure volume in the coal pillar. The pillar stress state transitions from compressive to tensile, with 95.8 % of the stored elastic energy released. Borehole imaging shows 7.79 m fracture depth on the pillar side and minimal damage on the coal side. The field observations confirm the reliability of numerical simulations. The analysis indicates that a remaining coal pillar above the studied coal seam causes stress concentration at the working face, with peak stress reaching 50 MPa. The combination effect of high static stress and dynamic disturbances generated by key stratum rupture serves as the main mechanism contributing to strong mine pressure behavior. This mechanism results in asymmetric roadway deformation and coal pillar instability. The findings provide a theoretical basis for optimizing support design and mitigating dynamic hazards in gob-side roadways under similar geological conditions.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"245 ","pages":"Article 106055"},"PeriodicalIF":2.1,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145737523","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Unexploded ordnance detection with deep learning: A LSTM-based approach with multi-factor analysis 基于深度学习的未爆弹药探测:基于lstm的多因素分析方法
IF 2.1 3区 地球科学 Q2 GEOSCIENCES, MULTIDISCIPLINARY Pub Date : 2026-02-01 Epub Date: 2025-12-15 DOI: 10.1016/j.jappgeo.2025.106068
Yizhou Lu , Bin Ma , Qingyun Wang , Jianxing Zhang , Lei Ai , Weifan Mao , Zhao Li , Zhuting Yu , Yinglin Guo
In recent years, persistent armed conflicts have posed a threat to the safety of people’s lives and property due to the presence of unexploded ordnance (UXO). The detection of UXO is an urgent yet challenging issue. When employing electromagnetic methods for subsurface UXO detection, it is difficult to distinguish between signals emitted by UXO and those from harmless clutter, leading to a high false alarm rate in UXO detection, which wastes manpower and financial resources. In this paper, we have implemented a UXO detection model based on Long Short-Term Memory (LSTM) networks, which learns pattern information from frequency-domain electromagnetic signals to differentiate between UXO, clutter, and background noise. Furthermore, we conducted orthogonal experiments across multiple levels for three critical factors – burial depth, sensor height, and UXO dip angle – to collect frequency-domain electromagnetic data using a customized ordnance model. The detection model excels on the dataset, achieving high accuracy and recall, which suggests that it can identify approximately all of UXOs while maintaining minimal false alarm rates. We conducted a multi-factor analysis of variance to systematically evaluate the impact of the key factors on detection performance. Our findings indicate that sensor height significantly influences the detection model’s performance, offering valuable insights and guidance for real-world UXO detection operations.
近年来,由于未爆弹药的存在,持续不断的武装冲突对人民的生命和财产安全构成威胁。未爆弹药的探测是一个紧迫而又具有挑战性的问题。采用电磁法探测地下未爆弹时,难以区分未爆弹发出的信号和无害杂波发出的信号,导致未爆弹探测虚警率高,浪费了人力和财力。在本文中,我们实现了一个基于长短期记忆(LSTM)网络的未爆弹检测模型,该模型从频域电磁信号中学习模式信息,以区分未爆弹、杂波和背景噪声。此外,我们对埋深、传感器高度和未爆弹药倾角这三个关键因素进行了多层次正交实验,利用定制的弹药模型收集频域电磁数据。该检测模型在数据集上表现出色,实现了较高的准确率和召回率,这表明它可以在保持最小误报率的同时识别出大约所有的未爆弹药。我们进行了多因素方差分析,系统地评估关键因素对检测性能的影响。我们的研究结果表明,传感器高度显著影响检测模型的性能,为现实世界的未爆弹药检测操作提供了有价值的见解和指导。
{"title":"Unexploded ordnance detection with deep learning: A LSTM-based approach with multi-factor analysis","authors":"Yizhou Lu ,&nbsp;Bin Ma ,&nbsp;Qingyun Wang ,&nbsp;Jianxing Zhang ,&nbsp;Lei Ai ,&nbsp;Weifan Mao ,&nbsp;Zhao Li ,&nbsp;Zhuting Yu ,&nbsp;Yinglin Guo","doi":"10.1016/j.jappgeo.2025.106068","DOIUrl":"10.1016/j.jappgeo.2025.106068","url":null,"abstract":"<div><div>In recent years, persistent armed conflicts have posed a threat to the safety of people’s lives and property due to the presence of unexploded ordnance (UXO). The detection of UXO is an urgent yet challenging issue. When employing electromagnetic methods for subsurface UXO detection, it is difficult to distinguish between signals emitted by UXO and those from harmless clutter, leading to a high false alarm rate in UXO detection, which wastes manpower and financial resources. In this paper, we have implemented a UXO detection model based on Long Short-Term Memory (LSTM) networks, which learns pattern information from frequency-domain electromagnetic signals to differentiate between UXO, clutter, and background noise. Furthermore, we conducted orthogonal experiments across multiple levels for three critical factors – burial depth, sensor height, and UXO dip angle – to collect frequency-domain electromagnetic data using a customized ordnance model. The detection model excels on the dataset, achieving high accuracy and recall, which suggests that it can identify approximately all of UXOs while maintaining minimal false alarm rates. We conducted a multi-factor analysis of variance to systematically evaluate the impact of the key factors on detection performance. Our findings indicate that sensor height significantly influences the detection model’s performance, offering valuable insights and guidance for real-world UXO detection operations.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"245 ","pages":"Article 106068"},"PeriodicalIF":2.1,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145840192","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
期刊
Journal of Applied Geophysics
全部 Acc. Chem. Res. ACS Applied Bio Materials ACS Appl. Electron. Mater. ACS Appl. Energy Mater. ACS Appl. Mater. Interfaces ACS Appl. Nano Mater. ACS Appl. Polym. Mater. ACS BIOMATER-SCI ENG ACS Catal. ACS Cent. Sci. ACS Chem. Biol. ACS Chemical Health & Safety ACS Chem. Neurosci. ACS Comb. Sci. ACS Earth Space Chem. ACS Energy Lett. ACS Infect. Dis. ACS Macro Lett. ACS Mater. Lett. ACS Med. Chem. Lett. ACS Nano ACS Omega ACS Photonics ACS Sens. ACS Sustainable Chem. Eng. ACS Synth. Biol. Anal. Chem. BIOCHEMISTRY-US Bioconjugate Chem. BIOMACROMOLECULES Chem. Res. Toxicol. Chem. Rev. Chem. Mater. CRYST GROWTH DES ENERG FUEL Environ. Sci. Technol. Environ. Sci. Technol. Lett. Eur. J. Inorg. Chem. IND ENG CHEM RES Inorg. Chem. J. Agric. Food. Chem. J. Chem. Eng. Data J. Chem. Educ. J. Chem. Inf. Model. J. Chem. Theory Comput. J. Med. Chem. J. Nat. Prod. J PROTEOME RES J. Am. Chem. Soc. LANGMUIR MACROMOLECULES Mol. Pharmaceutics Nano Lett. Org. Lett. ORG PROCESS RES DEV ORGANOMETALLICS J. Org. Chem. J. Phys. Chem. J. Phys. Chem. A J. Phys. Chem. B J. Phys. Chem. C J. Phys. Chem. Lett. Analyst Anal. Methods Biomater. Sci. Catal. Sci. Technol. Chem. Commun. Chem. Soc. Rev. CHEM EDUC RES PRACT CRYSTENGCOMM Dalton Trans. Energy Environ. Sci. ENVIRON SCI-NANO ENVIRON SCI-PROC IMP ENVIRON SCI-WAT RES Faraday Discuss. Food Funct. Green Chem. Inorg. Chem. Front. Integr. Biol. J. Anal. At. Spectrom. J. Mater. Chem. A J. Mater. Chem. B J. Mater. Chem. C Lab Chip Mater. Chem. Front. Mater. Horiz. MEDCHEMCOMM Metallomics Mol. Biosyst. Mol. Syst. Des. Eng. Nanoscale Nanoscale Horiz. Nat. Prod. Rep. New J. Chem. Org. Biomol. Chem. Org. Chem. Front. PHOTOCH PHOTOBIO SCI PCCP Polym. Chem.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1