Computers & Geosciences最新文献_第5页

Constructing three-dimension digital rock with porosity information constraint: A double-network-cycled style-based deep-learning approach 构建具有孔隙度信息约束的三维数字岩石：基于双网络循环样式的深度学习方法

IF 4.2 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences

Pub Date : 2024-10-19 DOI: 10.1016/j.cageo.2024.105741

Danping Cao , Shuai Hou , Zhiyu Hou

Understanding reservoir properties from geophysical responses necessitates the construction of accurate petrophysical templates based on adequate petrophysical data. However, engineering coring is often limited by complex subsurface conditions and high costs. Artificial intelligence (AI) techniques offer an efficient and economical way to synthesize digital samples. Nevertheless, traditional deep learning approaches may suffer from mode collapse, particularly when generating samples with complex structures from a limited number of training samples. To address this challenge, we propose novel Generative Adversarial Networks (GANs) to controllably generate 3D digital rock samples according to porosity distribution. By employing a style-transfer generator, multi-scale information of 3D digital rock is integrated into the generation process, effectively reducing the risk of mode collapse. Embedding the generator into a double-network-cycled framework further enhances the controllability of conditional information in the generated samples. Our analysis shows that the minimum error between the generated samples and the desired samples in terms of porosity is only 0.07%. A clear contrast is observed in morphological parameters, and differences in pore structure lead to significant variations in mechanical and hydraulic properties between original samples and synthetic samples with similar porosity. This indicates that the property contrast is likely caused by differences in pore structures rather than porosity. These findings will assist in future studies on the effect of pore structure on petrophysical properties and improve the utility of rock physics templates in geophysical inversion.

要从地球物理响应中了解储层属性，就必须根据充足的岩石物理数据构建准确的岩石物理模板。然而，工程岩心取样往往受限于复杂的地下条件和高昂的成本。人工智能（AI）技术为合成数字样本提供了一种高效、经济的方法。然而，传统的深度学习方法可能会出现模式崩溃，尤其是在从有限的训练样本生成具有复杂结构的样本时。为了应对这一挑战，我们提出了新颖的生成对抗网络（GANs），以根据孔隙度分布可控地生成三维数字岩石样本。通过使用样式转移生成器，三维数字岩石的多尺度信息被集成到生成过程中，从而有效降低了模式崩溃的风险。将生成器嵌入双网循环框架，进一步增强了生成样本中条件信息的可控性。我们的分析表明，生成的样本与所需样本在孔隙率方面的最小误差仅为 0.07%。在形态参数方面观察到了明显的对比，孔隙结构的差异导致原始样品与具有相似孔隙率的合成样品之间在机械和水力特性方面存在显著差异。这表明，性能对比可能是由孔隙结构的差异而不是孔隙率造成的。这些发现将有助于今后研究孔隙结构对岩石物理特性的影响，并提高岩石物理模板在地球物理反演中的实用性。

{"title":"Constructing three-dimension digital rock with porosity information constraint: A double-network-cycled style-based deep-learning approach","authors":"Danping Cao , Shuai Hou , Zhiyu Hou","doi":"10.1016/j.cageo.2024.105741","DOIUrl":"10.1016/j.cageo.2024.105741","url":null,"abstract":"<div><div>Understanding reservoir properties from geophysical responses necessitates the construction of accurate petrophysical templates based on adequate petrophysical data. However, engineering coring is often limited by complex subsurface conditions and high costs. Artificial intelligence (AI) techniques offer an efficient and economical way to synthesize digital samples. Nevertheless, traditional deep learning approaches may suffer from mode collapse, particularly when generating samples with complex structures from a limited number of training samples. To address this challenge, we propose novel Generative Adversarial Networks (GANs) to controllably generate 3D digital rock samples according to porosity distribution. By employing a style-transfer generator, multi-scale information of 3D digital rock is integrated into the generation process, effectively reducing the risk of mode collapse. Embedding the generator into a double-network-cycled framework further enhances the controllability of conditional information in the generated samples. Our analysis shows that the minimum error between the generated samples and the desired samples in terms of porosity is only 0.07%. A clear contrast is observed in morphological parameters, and differences in pore structure lead to significant variations in mechanical and hydraulic properties between original samples and synthetic samples with similar porosity. This indicates that the property contrast is likely caused by differences in pore structures rather than porosity. These findings will assist in future studies on the effect of pore structure on petrophysical properties and improve the utility of rock physics templates in geophysical inversion.</div></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"193 ","pages":"Article 105741"},"PeriodicalIF":4.2,"publicationDate":"2024-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142533978","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Research on elastic parameter inversion method based on seismic facies-controlled deep learning network 基于地震剖面控制深度学习网络的弹性参数反演方法研究

IF 4.2 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences

Pub Date : 2024-10-19 DOI: 10.1016/j.cageo.2024.105739

Xiaoyan Zhai , Gang Gao , Haojie Liu , Tengfei Chen

Deep learning has been widely applied in the field of geophysics, and existing research literature has demonstrated that intelligent geophysical inversion methods have high vertical resolution but low horizontal resolution. The reason lies in the fact that existing horizontal constraint methods mainly adopt convolutional models, without fully considering other prior information of seismic data. Within the same sedimentary unit, seismic response characteristics vary gradually due to similar lithology and geological characteristics. Therefore, the seismic facies information extracted from seismic data is integrated into deep learning network to enhance the horizontal prediction stability of the network. Firstly, according to the spatial and temporal characteristics of seismic data, a fusion network of three-dimensional convolutional neural network (3D-CNN), gated recurrent unit (GRU) and attention mechanism is established to improve the vertical resolution of inversion results. Then, seismic facies classification of the target layer is achieved by applying the K-means clustering method. Subsequently, to improve the horizontal resolution of the inversion results, seismic facies classification is transformed into temporal encoding data using the position coding theory in natural language processing, to form a seismic facies-controlled deep learning network. Finally, the deep learning network is trained and tested in the thin interlayer model and practical application adopting a semi-supervised learning method. The results indicate that incorporating seismic facies-controlled technology in the deep learning network can improve the horizontal resolution of the inversion results.

深度学习已广泛应用于地球物理领域，现有研究文献表明，智能地球物理反演方法具有较高的垂直分辨率，但水平分辨率较低。原因在于现有的水平约束方法主要采用卷积模型，没有充分考虑地震数据的其他先验信息。在同一沉积单元内，由于岩性和地质特征相似，地震响应特征也会逐渐发生变化。因此，将从地震数据中提取的地震面信息整合到深度学习网络中，以增强网络的水平预测稳定性。首先，根据地震数据的时空特征，建立三维卷积神经网络（3D-CNN）、门控递归单元（GRU）和注意力机制的融合网络，提高反演结果的垂直分辨率。然后，应用 K-means 聚类方法实现目标层的地震面分类。随后，为了提高反演结果的水平分辨率，利用自然语言处理中的位置编码理论将地震面分类转化为时间编码数据，形成地震面控制的深度学习网络。最后，采用半监督学习方法，在薄层间模型和实际应用中对深度学习网络进行了训练和测试。结果表明，在深度学习网络中加入地震面控制技术可以提高反演结果的水平分辨率。

{"title":"Research on elastic parameter inversion method based on seismic facies-controlled deep learning network","authors":"Xiaoyan Zhai , Gang Gao , Haojie Liu , Tengfei Chen","doi":"10.1016/j.cageo.2024.105739","DOIUrl":"10.1016/j.cageo.2024.105739","url":null,"abstract":"<div><div>Deep learning has been widely applied in the field of geophysics, and existing research literature has demonstrated that intelligent geophysical inversion methods have high vertical resolution but low horizontal resolution. The reason lies in the fact that existing horizontal constraint methods mainly adopt convolutional models, without fully considering other prior information of seismic data. Within the same sedimentary unit, seismic response characteristics vary gradually due to similar lithology and geological characteristics. Therefore, the seismic facies information extracted from seismic data is integrated into deep learning network to enhance the horizontal prediction stability of the network. Firstly, according to the spatial and temporal characteristics of seismic data, a fusion network of three-dimensional convolutional neural network (3D-CNN), gated recurrent unit (GRU) and attention mechanism is established to improve the vertical resolution of inversion results. Then, seismic facies classification of the target layer is achieved by applying the K-means clustering method. Subsequently, to improve the horizontal resolution of the inversion results, seismic facies classification is transformed into temporal encoding data using the position coding theory in natural language processing, to form a seismic facies-controlled deep learning network. Finally, the deep learning network is trained and tested in the thin interlayer model and practical application adopting a semi-supervised learning method. The results indicate that incorporating seismic facies-controlled technology in the deep learning network can improve the horizontal resolution of the inversion results.</div></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"193 ","pages":"Article 105739"},"PeriodicalIF":4.2,"publicationDate":"2024-10-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142533977","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Enhanced lithological mapping in arid crystalline regions using explainable AI and multi-spectral remote sensing data 利用可解释人工智能和多光谱遥感数据加强干旱结晶地区的岩性测绘

IF 4.2 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences

Pub Date : 2024-10-16 DOI: 10.1016/j.cageo.2024.105738

Hesham Morgan , Ali Elgendy , Amir Said , Mostafa Hashem , Wenzhao Li , Surendra Maharjan , Hesham El-Askary

Lithological classification is essential for understanding the spatial distribution of rocks, especially in arid crystalline areas. Artificial intelligence (AI) recent advancements with multi-spectral satellite imagery have been utilized to enhance lithological mapping in these areas. Here we employed different AI models namely, Support Vector Machine (SVM), Random Forest Classification (RFC), Logistic Regression, XGBoost, and K-nearest neighbors (KNN) for lithological mapping. This was followed by the application of explainable AI (XAI) for lithological discrimination (LD) which is still not widely explored. Based on the highest accuracy and F1 score of the previously mentioned models, RFC model outperformed all of them, and hence, it was integrated with XAI, using the SHapley Additive exPlanations (SHAP) method.

This approach successfully identified critical multi-spectral features for LD in arid crystalline zones when applied on the Landsat-8, Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), and SRTM-DEM datasets covering the Hammash and the Wadi Fatimah areas in Egypt and the Kingdom of Saudi Arabia, respectively. Field validation in the Hammash area confirmed the RFC model's efficacy, achieving a satisfactory 94% overall accuracy for 18 features. SHAP was able to identify the top ten features for proper LD over the Hammash area with 90.3% accuracy despite the complex nature of the ophiolitic mélange. For validation purposes, RCF was then utilized in the Wadi Fatimah region, using only the top 10 critical features rendered from the SHAP analysis. It performed well and had 93% accuracy. Notably, XAI/SHAP results indicated that elevation data, Landsat-8's Green Band (B3), and the two ASTER SWIR bands (B5 and B6) were essential and significant for identifying island arc rocks. Moreover, the SHAP model effectively delineated complex mélange matrices, primarily using ASTER SWIR band (B8). Our findings highlight the successful combination of RFC with XAI for LD and its potential utilization in similar arid crystalline environments worldwide.

岩性分类对于了解岩石的空间分布至关重要，尤其是在干旱的结晶地区。人工智能（AI）与多光谱卫星图像的最新进展已被用于加强这些地区的岩性绘图。在此，我们采用了不同的人工智能模型，即支持向量机（SVM）、随机森林分类（RFC）、逻辑回归（Logistic Regression）、XGBoost 和 K-nearest neighbors（KNN）来绘制岩性图。其次是应用可解释人工智能（XAI）进行岩性判别（LD），这种方法目前仍未得到广泛探索。这种方法应用于分别覆盖埃及哈马什和沙特阿拉伯王国 Wadi Fatimah 地区的 Landsat-8、高级星载热发射和反射辐射计（ASTER）以及 SRTM-DEM 数据集时，成功识别了干旱结晶区岩性判别的关键多光谱特征。在哈马什地区进行的实地验证证实了 RFC 模型的有效性，18 个地物的总体准确率达到 94%，令人满意。尽管蛇绿混杂岩的性质复杂，但 SHAP 能够以 90.3% 的准确率识别出哈马什地区适当 LD 的前十个特征。为了进行验证，我们在 Wadi Fatimah 地区使用了 RCF，仅使用了 SHAP 分析得出的前 10 个关键特征。该方法表现良好，准确率达到 93%。值得注意的是，XAI/SHAP 的结果表明，高程数据、Landsat-8 的绿波段（B3）和两个 ASTER SWIR 波段（B5 和 B6）对于识别岛弧岩石至关重要。此外，SHAP 模型主要利用 ASTER SWIR 波段（B8）有效地划分了复杂的混杂岩矩阵。我们的研究结果突显了 RFC 与 XAI 在 LD 方面的成功结合及其在全球类似干旱结晶环境中的潜在应用。

{"title":"Enhanced lithological mapping in arid crystalline regions using explainable AI and multi-spectral remote sensing data","authors":"Hesham Morgan , Ali Elgendy , Amir Said , Mostafa Hashem , Wenzhao Li , Surendra Maharjan , Hesham El-Askary","doi":"10.1016/j.cageo.2024.105738","DOIUrl":"10.1016/j.cageo.2024.105738","url":null,"abstract":"<div><div>Lithological classification is essential for understanding the spatial distribution of rocks, especially in arid crystalline areas. Artificial intelligence (AI) recent advancements with multi-spectral satellite imagery have been utilized to enhance lithological mapping in these areas. Here we employed different <span>AI</span> models namely, Support Vector Machine (SVM), Random Forest Classification (RFC), Logistic Regression, XGBoost, and K-nearest neighbors (KNN) for lithological mapping. This was followed by the application of explainable AI (XAI) for lithological discrimination (LD) which is still not widely explored. Based on the highest accuracy and F1 score of the previously mentioned models, RFC model outperformed all of them, and hence, it was integrated with XAI, using the SHapley Additive exPlanations (SHAP) method.</div><div>This approach successfully identified critical multi-spectral features for LD in arid crystalline zones when applied on the Landsat-8, Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER), and SRTM-DEM datasets covering the Hammash and the Wadi Fatimah areas in Egypt and the Kingdom of Saudi Arabia, respectively. Field validation in the Hammash area confirmed the RFC model's efficacy, achieving a satisfactory 94% overall accuracy for 18 features. SHAP was able to identify the top ten features for proper LD over the Hammash area with 90.3% accuracy despite the complex nature of the ophiolitic mélange. For validation purposes, RCF was then utilized in the Wadi Fatimah region, using only the top 10 critical features rendered from the SHAP analysis. It performed well and had 93% accuracy. Notably, XAI/SHAP results indicated that elevation data, Landsat-8's Green Band (B3), and the two ASTER SWIR bands (B5 and B6) were essential and significant for identifying island arc rocks. Moreover, the SHAP model effectively delineated complex mélange matrices, primarily using ASTER SWIR band (B8). Our findings highlight the successful combination of RFC with XAI for LD and its potential utilization in similar arid crystalline environments worldwide.</div></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"193 ","pages":"Article 105738"},"PeriodicalIF":4.2,"publicationDate":"2024-10-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142445319","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Fractal-based supervised approach for dimensionality reduction of hyperspectral images 基于分形的高光谱图像降维监督方法

IF 4.2 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences

Pub Date : 2024-09-30 DOI: 10.1016/j.cageo.2024.105733

Vanshika Gupta , Sharad Kumar Gupta , Amba Shetty

Dimensionality reduction is one of the most challenging and crucial issues apart from data mining, security, and scalability, which have retained much traction due to the ever-growing need to analyze the large volumes of data generated daily. Fractal Dimension (FD) has been successfully used to characterize data sets and has found relevant applications in dimension reduction. This paper presents an application of the FD Reduction (FDR) Algorithm on geospatial hyperspectral data, examining its usefulness for data sets with a relatively high embedding dimension. We examine the algorithm at two levels. First is the conventional FDR approach (unsupervised) at the image level. Alternatively, we propose a pixel-level supervised approach for band reduction based on time-series complexity analysis. Techniques for determining an optimal intrinsic dimension for the dataset using these two techniques are examined. We also develop a parallel GPU-based implementation for the unsupervised image-level FDR algorithm, reducing the run-time by nearly 10 times. Furthermore, both approaches use a support vector machine classifier to compare the classification performance of the original and reduced image obtained.

降维是除数据挖掘、安全性和可扩展性之外最具挑战性和关键性的问题之一，由于分析每天产生的大量数据的需求日益增长，降维问题一直备受关注。分形维度（FD）已成功用于描述数据集的特征，并在维度缩减中找到了相关应用。本文介绍了分形维度缩减（FDR）算法在地理空间高光谱数据中的应用，研究了该算法对嵌入维度相对较高的数据集的实用性。我们从两个层面对该算法进行了研究。首先是图像层面的传统 FDR 方法（无监督）。另外，我们还提出了一种基于时间序列复杂性分析的像素级有监督波段缩减方法。我们还研究了使用这两种技术确定数据集最佳内在维度的技术。我们还为无监督图像级 FDR 算法开发了基于 GPU 的并行实施方案，将运行时间缩短了近 10 倍。此外，这两种方法都使用支持向量机分类器来比较原始图像和缩小图像的分类性能。

{"title":"Fractal-based supervised approach for dimensionality reduction of hyperspectral images","authors":"Vanshika Gupta , Sharad Kumar Gupta , Amba Shetty","doi":"10.1016/j.cageo.2024.105733","DOIUrl":"10.1016/j.cageo.2024.105733","url":null,"abstract":"<div><div>Dimensionality reduction is one of the most challenging and crucial issues apart from data mining, security, and scalability, which have retained much traction due to the ever-growing need to analyze the large volumes of data generated daily. Fractal Dimension (FD) has been successfully used to characterize data sets and has found relevant applications in dimension reduction. This paper presents an application of the FD Reduction (FDR) Algorithm on geospatial hyperspectral data, examining its usefulness for data sets with a relatively high embedding dimension. We examine the algorithm at two levels. First is the conventional FDR approach (unsupervised) at the image level. Alternatively, we propose a pixel-level supervised approach for band reduction based on time-series complexity analysis. Techniques for determining an optimal intrinsic dimension for the dataset using these two techniques are examined. We also develop a parallel GPU-based implementation for the unsupervised image-level FDR algorithm, reducing the run-time by nearly 10 times. Furthermore, both approaches use a support vector machine classifier to compare the classification performance of the original and reduced image obtained.</div></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"193 ","pages":"Article 105733"},"PeriodicalIF":4.2,"publicationDate":"2024-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142424552","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Calculating sensitivity or gradient for geophysical inverse problems using automatic and implicit differentiation 利用自动微分和隐式微分计算地球物理逆问题的灵敏度或梯度

IF 4.2 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences

Pub Date : 2024-09-29 DOI: 10.1016/j.cageo.2024.105736

Lian Liu , Bo Yang , Yi Zhang , Yixian Xu , Zhong Peng , Dikun Yang

Automatic differentiation (AD) is a valuable computing technique that can automatically calculate the derivative of a function. Using the chain rule and algebraic manipulations, AD can save significant human effort by eliminating the need for theoretical derivations, coding, and debugging. Most importantly, it guarantees accurate derivatives, making it a popular choice for many non-linear optimization problems. However, its use in the geophysical inversion has been limited due to difficulties in differentiating the linear-equations solution, which cannot be explicitly defined as an elementary function. To address this issue, we employ an improved AD scheme using implicit differentiation (ADID) that creates a new AD operator that customizes the standard AD scheme to function more efficiently. We demonstrate the effectiveness and validity of ADID using a toy example and compare it with the widely used adjoint equation (AE) approach in a synthetic 2D magnetotelluric (MT) problem. ADID is highly versatile and compatible and can be easily implemented for similar geophysical problems. Finally, we show how ADID can be integrated into 3D MT and 3D direct current resistivity (DC) inversions.

自动微分（AD）是一种有价值的计算技术，可以自动计算函数的导数。自动微分利用链式法则和代数运算，无需进行理论推导、编码和调试，从而大大节省了人力。最重要的是，它能保证导数的精确性，因此成为许多非线性优化问题的首选。然而，由于线性方程解的微分困难，无法明确定义为基本函数，它在地球物理反演中的应用受到了限制。为了解决这个问题，我们采用了一种使用隐式微分的改进 AD 方案（ADID），它创建了一个新的 AD 算子，可定制标准 AD 方案，使其更有效地发挥作用。我们用一个玩具示例证明了 ADID 的有效性和正确性，并将其与广泛使用的邻接方程 (AE) 方法在合成二维磁辐射 (MT) 问题中进行了比较。ADID 具有很强的通用性和兼容性，可轻松用于类似的地球物理问题。最后，我们展示了如何将 ADID 集成到三维 MT 和三维直流电阻率 (DC) 反演中。

{"title":"Calculating sensitivity or gradient for geophysical inverse problems using automatic and implicit differentiation","authors":"Lian Liu , Bo Yang , Yi Zhang , Yixian Xu , Zhong Peng , Dikun Yang","doi":"10.1016/j.cageo.2024.105736","DOIUrl":"10.1016/j.cageo.2024.105736","url":null,"abstract":"<div><div>Automatic differentiation (AD) is a valuable computing technique that can automatically calculate the derivative of a function. Using the chain rule and algebraic manipulations, AD can save significant human effort by eliminating the need for theoretical derivations, coding, and debugging. Most importantly, it guarantees accurate derivatives, making it a popular choice for many non-linear optimization problems. However, its use in the geophysical inversion has been limited due to difficulties in differentiating the linear-equations solution, which cannot be explicitly defined as an elementary function. To address this issue, we employ an improved AD scheme using implicit differentiation (ADID) that creates a new AD operator that customizes the standard AD scheme to function more efficiently. We demonstrate the effectiveness and validity of ADID using a toy example and compare it with the widely used adjoint equation (AE) approach in a synthetic 2D magnetotelluric (MT) problem. ADID is highly versatile and compatible and can be easily implemented for similar geophysical problems. Finally, we show how ADID can be integrated into 3D MT and 3D direct current resistivity (DC) inversions.</div></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"193 ","pages":"Article 105736"},"PeriodicalIF":4.2,"publicationDate":"2024-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142357131","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Improving lithofacies prediction in lacustrine shale by combining deep learning and well log curve morphology in Sanzhao Sag, Songliao Basin, China 结合深度学习和测井曲线形态改进中国松辽盆地三兆沙砾岩湖相页岩岩性预测

IF 4.2 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences

Pub Date : 2024-09-29 DOI: 10.1016/j.cageo.2024.105735

Xiaozhuo Wu , Hao Xu , Haiyan Zhou , Lan Wang , Pengfei Jiang , Heng Wu

The accurate identification of shale lithofacies is crucial for characterizing the hydrocarbon potential of lacustrine shale oil reservoirs. Petrophysical logging, serving as an effective tool for acquiring subsurface lithofacies information, provides a convenient and reliable lithofacies identification solution. Deep learning technology, capable of adapting to the nonlinearity and non-stationarity inherent in geological statistics, exhibits unique advantages in conventional reservoir lithofacies prediction. However, the lithofacies of lacustrine shale formations undergo rapid spatial and temporal changes, rendering lithofacies prediction more complex compared to conventional reservoirs. In this study, the lower Qingshankou member in the Sanzhao Sag was selected as the research target, and the Deep Residual Shrinkage Network (DRSN), known for its ability to handle complex nonlinear relationships and mitigate the effects of noisy data through residual connections and shrinkage mechanisms, was employed as a deep learning framework for predicting lithofacies in lacustrine shale formations for the first time. Well logging data, including natural gamma ray (GR), acoustic (AC), deep investigate double lateral resistivity log (RD), shallow investigate double lateral resistivity log (RS), and corrected compensated neutron log (CNC), were used as input features for the model. The results indicate that the DRSN model achieves an accuracy of 76.3% in predicting lithofacies in lacustrine shale formations. However, the DRSN model still exhibits shortcomings in capturing lithofacies change information. To enhance the model's ability to identify lithofacies change interfaces, this study further explicitly introduces Well Logging Curve Morphological Features (WLCM) as additional features and establishes a recognition method combining DRSN with WLCM. The combined DRSN-WLCM model was validated using a separate test dataset, demonstrating an improved accuracy of 85.5%, using the five well logging attributes and the derivative of the AC as inputs. Furthermore, the study reveals the lithofacies spatial distribution characteristics of the lower Qingshankou member in the Sanzhao Sag. This method can be widely applied to lithofacies delineation in lacustrine shale formations and similar stratigraphic units.

准确识别页岩岩性对于确定湖相页岩油藏的碳氢潜力至关重要。岩石物理测井作为获取地下岩性信息的有效工具，提供了便捷可靠的岩性识别解决方案。深度学习技术能够适应地质统计中固有的非线性和非平稳性，在常规储层岩性预测中表现出独特的优势。然而，湖相页岩地层的岩性会发生快速的时空变化，使得岩性预测与常规储层相比更为复杂。本研究以三条嵯峨岩层中的青山口下统为研究对象，首次采用了深度残余收缩网络（DRSN）作为深度学习框架来预测湖相页岩层的岩性，DRSN以其能够处理复杂的非线性关系、通过残余连接和收缩机制减轻噪声数据的影响而著称。测井数据包括天然伽马射线（GR）、声波（AC）、深部调查双侧向电阻率测井（RD）、浅部调查双侧向电阻率测井（RS）和校正补偿中子测井（CNC），这些数据被用作模型的输入特征。结果表明，DRSN 模型预测湖相页岩地层岩性的准确率达到 76.3%。然而，DRSN 模型在捕捉岩性变化信息方面仍存在不足。为了提高该模型识别岩性变化界面的能力，本研究进一步明确引入了测井曲线形态特征（WLCM）作为附加特征，并建立了 DRSN 与 WLCM 相结合的识别方法。使用单独的测试数据集对 DRSN-WLCM 组合模型进行了验证，结果表明，使用五种测井属性和 AC 的导数作为输入，准确率提高了 85.5%。此外，该研究还揭示了三条嵯峨下部青山口岩层的岩性空间分布特征。该方法可广泛应用于湖相页岩地层及类似地层单元的岩性划分。

{"title":"Improving lithofacies prediction in lacustrine shale by combining deep learning and well log curve morphology in Sanzhao Sag, Songliao Basin, China","authors":"Xiaozhuo Wu , Hao Xu , Haiyan Zhou , Lan Wang , Pengfei Jiang , Heng Wu","doi":"10.1016/j.cageo.2024.105735","DOIUrl":"10.1016/j.cageo.2024.105735","url":null,"abstract":"<div><div>The accurate identification of shale lithofacies is crucial for characterizing the hydrocarbon potential of lacustrine shale oil reservoirs. Petrophysical logging, serving as an effective tool for acquiring subsurface lithofacies information, provides a convenient and reliable lithofacies identification solution. Deep learning technology, capable of adapting to the nonlinearity and non-stationarity inherent in geological statistics, exhibits unique advantages in conventional reservoir lithofacies prediction. However, the lithofacies of lacustrine shale formations undergo rapid spatial and temporal changes, rendering lithofacies prediction more complex compared to conventional reservoirs. In this study, the lower Qingshankou member in the Sanzhao Sag was selected as the research target, and the Deep Residual Shrinkage Network (DRSN), known for its ability to handle complex nonlinear relationships and mitigate the effects of noisy data through residual connections and shrinkage mechanisms, was employed as a deep learning framework for predicting lithofacies in lacustrine shale formations for the first time. Well logging data, including natural gamma ray (GR), acoustic (AC), deep investigate double lateral resistivity log (RD), shallow investigate double lateral resistivity log (RS), and corrected compensated neutron log (CNC), were used as input features for the model. The results indicate that the DRSN model achieves an accuracy of 76.3% in predicting lithofacies in lacustrine shale formations. However, the DRSN model still exhibits shortcomings in capturing lithofacies change information. To enhance the model's ability to identify lithofacies change interfaces, this study further explicitly introduces Well Logging Curve Morphological Features (WLCM) as additional features and establishes a recognition method combining DRSN with WLCM. The combined DRSN-WLCM model was validated using a separate test dataset, demonstrating an improved accuracy of 85.5%, using the five well logging attributes and the derivative of the AC as inputs. Furthermore, the study reveals the lithofacies spatial distribution characteristics of the lower Qingshankou member in the Sanzhao Sag. This method can be widely applied to lithofacies delineation in lacustrine shale formations and similar stratigraphic units.</div></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"193 ","pages":"Article 105735"},"PeriodicalIF":4.2,"publicationDate":"2024-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142424553","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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 oil production prediction approach based on variational mode decomposition and ensemble learning model 基于变异模式分解和集合学习模型的石油产量预测方法

IF 4.2 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences

Pub Date : 2024-09-24 DOI: 10.1016/j.cageo.2024.105734

Junyi Fang , Zhen Yan , Xiaoya Lu , Yifei Xiao , Zhen Zhao

Well production forecasting can provide scientific guidance for oilfield production and management, which is an indispensable part of the oilfield development process. In this study, the daily oil production data from oil wells are first decomposed into components with different frequencies by variational mode decomposition (VMD), which is usually used to process complex time series. The new features obtained from decomposition and other filtered features are then used as input data and for training and forecasting of GRU, TCN and Transformer models respectively. In the end, the three models are integrated as base learners using the Blending method, which specifically involves using the predicted outputs of the three models as new inputs to the RBFNN for training and realizing the final predictions. The VMD-Blending model was compared with traditional models based on the production dynamics data of three production wells in an oil field in the Tarim area, China. The result shows that VMD can effectively improve the prediction effect of the base learners, and the prediction effect of these models is further improved after Blending integration, and all of their prediction indexes are significantly better than those of the base learners and the traditional SVM and RNN models. The proposed VMD-Blending model has a well performance in the task of well capacity prediction and is an accurate and effective method for oil production prediction.

油井产量预测可以为油田生产和管理提供科学指导，是油田开发过程中不可或缺的一部分。在本研究中，首先通过变异模态分解（VMD）将油井的日产量数据分解为不同频率的成分，这种方法通常用于处理复杂的时间序列。然后将分解得到的新特征和其他过滤特征作为输入数据，分别用于 GRU、TCN 和 Transformer 模型的训练和预测。最后，使用混合法将这三个模型整合为基础学习器，具体来说，就是将这三个模型的预测输出作为 RBFNN 的新输入，用于训练和实现最终预测。基于中国塔里木地区某油田三口生产井的生产动态数据，将 VMD-Blending 模型与传统模型进行了比较。结果表明，VMD 能有效提高基础学习器的预测效果，而这些模型的预测效果在经过 Blending 集成后得到进一步提高，其各项预测指标均明显优于基础学习器和传统 SVM、RNN 模型。所提出的 VMD-Blending 模型在油井产能预测任务中表现良好，是一种准确有效的石油产量预测方法。

{"title":"An oil production prediction approach based on variational mode decomposition and ensemble learning model","authors":"Junyi Fang , Zhen Yan , Xiaoya Lu , Yifei Xiao , Zhen Zhao","doi":"10.1016/j.cageo.2024.105734","DOIUrl":"10.1016/j.cageo.2024.105734","url":null,"abstract":"<div><div>Well production forecasting can provide scientific guidance for oilfield production and management, which is an indispensable part of the oilfield development process. In this study, the daily oil production data from oil wells are first decomposed into components with different frequencies by variational mode decomposition (VMD), which is usually used to process complex time series. The new features obtained from decomposition and other filtered features are then used as input data and for training and forecasting of GRU, TCN and Transformer models respectively. In the end, the three models are integrated as base learners using the Blending method, which specifically involves using the predicted outputs of the three models as new inputs to the RBFNN for training and realizing the final predictions. The VMD-Blending model was compared with traditional models based on the production dynamics data of three production wells in an oil field in the Tarim area, China. The result shows that VMD can effectively improve the prediction effect of the base learners, and the prediction effect of these models is further improved after Blending integration, and all of their prediction indexes are significantly better than those of the base learners and the traditional SVM and RNN models. The proposed VMD-Blending model has a well performance in the task of well capacity prediction and is an accurate and effective method for oil production prediction.</div></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"193 ","pages":"Article 105734"},"PeriodicalIF":4.2,"publicationDate":"2024-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142357130","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

MAMCL: Multi-attributes Masking Contrastive Learning for explainable seismic facies analysis MAMCL：用于可解释地震剖面分析的多属性屏蔽对比学习

IF 4.2 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences

Pub Date : 2024-09-24 DOI: 10.1016/j.cageo.2024.105731

Long Han , Xinming Wu , Zhanxuan Hu , Jintao Li , Huijing Fang

Seismic facies analysis is crucial in hydrocarbon exploration and development. Traditional machine learning approaches typically require manual selection of attributes and lack interpretability analysis. We propose an interpretable framework, multi-attribute masking contrastive learning (MAMCL), designed to adaptively select, explore and aggregate seismic attributes for seismic facies analysis. The MAMCL framework includes a depthwise CNN module for feature extraction and an iTransformer module for feature aggregation. Based on the assumption that different attributes computed on the same seismic sample imply common information associated with the same geologic facies, we formulate an unsupervised strategy of contrastive learning to pre-train the MAMCL framework for refining the attributes. This pre-training method encourages the network to extract and integrate highly correlated attribute features by enhancing the expression of commonalities within the same sample, and implicitly increase the distance between features of different categories by differentiating the expressions of different samples. Ultimately, these refined features only need to be input into a simple clustering algorithm, such as K-Means, to achieve seismic facies classification. MAMCL requires no labels or manual selection of attributes and can utilize the self-attention mechanism of iTransformer to compute adaptive attribute weights, facilitating interpretability analysis. We applied MAMCL framework to both unlogged turbidite channel systems in Canterbury Basin, New Zealand, and logged Chengdao area in Bohai Bay Basin, China, achieving reliable classification results and providing interpretability analysis.

地震剖面分析在油气勘探和开发中至关重要。传统的机器学习方法通常需要人工选择属性，缺乏可解释性分析。我们提出了一种可解释性框架--多属性掩蔽对比学习（MAMCL），旨在为地震剖面分析自适应地选择、探索和聚合地震属性。MAMCL 框架包括一个用于特征提取的深度 CNN 模块和一个用于特征聚合的 iTransformer 模块。基于对同一地震样本计算的不同属性意味着与同一地质面相关的共同信息这一假设，我们制定了一种无监督的对比学习策略，对 MAMCL 框架进行预训练，以完善属性。这种预训练方法通过增强同一样本内共性的表达，鼓励网络提取和整合高度相关的属性特征，并通过区分不同样本的表达，隐式地增加不同类别特征之间的距离。最终，只需将这些细化特征输入 K-Means 等简单聚类算法，即可实现地震剖面分类。MAMCL 不需要标签或人工选择属性，并可利用 iTransformer 的自注意机制计算自适应属性权重，从而促进可解释性分析。我们将 MAMCL 框架应用于新西兰坎特伯雷盆地的未测井浊积岩河道系统和中国渤海湾盆地的测井成岛地区，取得了可靠的分类结果，并提供了可解释性分析。

{"title":"MAMCL: Multi-attributes Masking Contrastive Learning for explainable seismic facies analysis","authors":"Long Han , Xinming Wu , Zhanxuan Hu , Jintao Li , Huijing Fang","doi":"10.1016/j.cageo.2024.105731","DOIUrl":"10.1016/j.cageo.2024.105731","url":null,"abstract":"<div><div>Seismic facies analysis is crucial in hydrocarbon exploration and development. Traditional machine learning approaches typically require manual selection of attributes and lack interpretability analysis. We propose an interpretable framework, multi-attribute masking contrastive learning (MAMCL), designed to adaptively select, explore and aggregate seismic attributes for seismic facies analysis. The MAMCL framework includes a depthwise CNN module for feature extraction and an iTransformer module for feature aggregation. Based on the assumption that different attributes computed on the same seismic sample imply common information associated with the same geologic facies, we formulate an unsupervised strategy of contrastive learning to pre-train the MAMCL framework for refining the attributes. This pre-training method encourages the network to extract and integrate highly correlated attribute features by enhancing the expression of commonalities within the same sample, and implicitly increase the distance between features of different categories by differentiating the expressions of different samples. Ultimately, these refined features only need to be input into a simple clustering algorithm, such as K-Means, to achieve seismic facies classification. MAMCL requires no labels or manual selection of attributes and can utilize the self-attention mechanism of iTransformer to compute adaptive attribute weights, facilitating interpretability analysis. We applied MAMCL framework to both unlogged turbidite channel systems in Canterbury Basin, New Zealand, and logged Chengdao area in Bohai Bay Basin, China, achieving reliable classification results and providing interpretability analysis.</div></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"193 ","pages":"Article 105731"},"PeriodicalIF":4.2,"publicationDate":"2024-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142322571","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Estimation of electrical conductivity models using multi-coil rigid-boom electromagnetic induction measurements 利用多线圈刚性导波电磁感应测量估算导电率模型

IF 4.2 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences

Pub Date : 2024-09-23 DOI: 10.1016/j.cageo.2024.105732

Maria Carrizo Mascarell, Dieter Werthmüller, Evert Slob

Electromagnetic induction measurements from multi-coil configuration instruments are used to obtain information about the electrical conductivity distribution in the subsurface. The resulting inverse problem might not have a unique and stable solution. In that case, a local inversion method can be trapped in a local minimum and lead to an incorrect solution. In this study, we evaluate the well-posedness of the inverse problem for two and three-layered electrical conductivity models. We show that for a two-layered model, uniqueness is ensured only when both in-phase and quadrature data are available from the measurements. Results from a Gauss–Newton inversion and a lookup table demonstrate that the solution space is convex. Furthermore, we demonstrate that for even a simple three-layered model, the data contained in such measurements are insufficient to reach a correct or stable solution. For models with more than 2 layers, independent prior information is necessary to solve the inverse problem. The insights from the numerical examples are applied in a field case.

多线圈配置仪器的电磁感应测量用于获取地下电导率分布信息。由此产生的反演问题可能没有唯一且稳定的解。在这种情况下，局部反演方法可能会陷入局部最小值而导致错误的解。在本研究中，我们评估了两层和三层电导率模型的反演问题的好求解性。我们发现，对于双层模型，只有当测量数据中同时存在同相数据和正交数据时，才能确保唯一性。高斯-牛顿反演和查找表的结果表明，解空间是凸的。此外，我们还证明，即使是简单的三层模型，这些测量数据也不足以得出正确或稳定的解。对于两层以上的模型，独立的先验信息是解决逆问题的必要条件。从数值示例中获得的启示将应用于一个实地案例。

{"title":"Estimation of electrical conductivity models using multi-coil rigid-boom electromagnetic induction measurements","authors":"Maria Carrizo Mascarell, Dieter Werthmüller, Evert Slob","doi":"10.1016/j.cageo.2024.105732","DOIUrl":"10.1016/j.cageo.2024.105732","url":null,"abstract":"<div><div>Electromagnetic induction measurements from multi-coil configuration instruments are used to obtain information about the electrical conductivity distribution in the subsurface. The resulting inverse problem might not have a unique and stable solution. In that case, a local inversion method can be trapped in a local minimum and lead to an incorrect solution. In this study, we evaluate the well-posedness of the inverse problem for two and three-layered electrical conductivity models. We show that for a two-layered model, uniqueness is ensured only when both in-phase and quadrature data are available from the measurements. Results from a Gauss–Newton inversion and a lookup table demonstrate that the solution space is convex. Furthermore, we demonstrate that for even a simple three-layered model, the data contained in such measurements are insufficient to reach a correct or stable solution. For models with more than 2 layers, independent prior information is necessary to solve the inverse problem. The insights from the numerical examples are applied in a field case.</div></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"193 ","pages":"Article 105732"},"PeriodicalIF":4.2,"publicationDate":"2024-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142326942","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Sample size effects on landslide susceptibility models: A comparative study of heuristic, statistical, machine learning, deep learning and ensemble learning models with SHAP analysis 样本量对滑坡易感性模型的影响：启发式、统计、机器学习、深度学习和集合学习模型与 SHAP 分析的比较研究

IF 4.2 2区地球科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Computers & Geosciences

Pub Date : 2024-09-13 DOI: 10.1016/j.cageo.2024.105723

Shilong Yang , Jiayao Tan , Danyuan Luo , Yuzhou Wang , Xu Guo , Qiuyu Zhu , Chuanming Ma , Hanxiang Xiong

In landslide susceptibility assessment (LSA), inventory incompleteness impacts the accuracy of different models to varying degrees. However, this area remains under-researched. This study investigated six LSA models from heuristic, statistical, machine learning and ensemble learning models (analytical hierarchy process (AHP), frequency ratio (FR), logistic regression (LR), Keras based deep learning (KBDL), XGBoost, and LightGBM) across six different sample sizes (100%, 90%, 75%, 50%, 25%, and 10%). Results revealed that XGBoost and LightGBM consistently outperformed other models across all sample sizes. The LR and KBDL models followed, while FR model was the most affected by sample size variations. AHP, an empirical model, remained unaffected by sample size. Through SHapley Additive exPlanations (SHAP) analysis, elevation, NDVI, slope, land use, and distance to roads and rivers emerged as pivotal indicators for landslide occurrences in the study area, suggesting that human activities significantly influence these events. Five time-varying indicators regarding human activity and climate validated this inference, which provides a new method to identify landslide triggering factors, especially in areas of intense human activity. Based on the findings, a comprehensive framework for LSA is proposed to assist landslide managers in making informed decisions. Future research should focus on expanding model diversity to address the effects of sample size, enhancing the adaptability of the LSA framework, deepening the analysis of human activity impacts on landslides using explainable machine learning techniques, addressing temporal inventory incompleteness in LSA, and critically evaluating model sensitivity to sample size variations across multiple disciplines.

在滑坡易发性评估（LSA）中，清单的不完整性会在不同程度上影响不同模型的准确性。然而，这一领域的研究仍然不足。本研究调查了六种不同样本量（100%、90%、75%、50%、25% 和 10%）的启发式、统计、机器学习和集合学习模型（分析层次过程 (AHP)、频率比 (FR)、逻辑回归 (LR)、基于 Keras 的深度学习 (KBDL)、XGBoost 和 LightGBM）中的六种 LSA 模型。结果显示，在所有样本量下，XGBoost 和 LightGBM 的表现始终优于其他模型。LR 和 KBDL 模型紧随其后，而 FR 模型受样本量变化的影响最大。经验模型 AHP 则不受样本量的影响。通过 SHapley Additive exPlanations（SHAP）分析，海拔、NDVI、坡度、土地利用以及与道路和河流的距离成为研究区域滑坡发生的关键指标，这表明人类活动对这些事件有重大影响。有关人类活动和气候的五个时变指标验证了这一推论，为识别滑坡诱发因素，尤其是人类活动频繁地区的滑坡诱发因素提供了一种新方法。根据研究结果，提出了一个全面的山体滑坡评估框架，以帮助山体滑坡管理者做出明智的决策。未来的研究应侧重于扩大模型的多样性以解决样本大小的影响，增强 LSA 框架的适应性，利用可解释的机器学习技术深化人类活动对滑坡影响的分析，解决 LSA 中时间清单的不完整性，以及批判性地评估模型对跨学科样本大小变化的敏感性。

{"title":"Sample size effects on landslide susceptibility models: A comparative study of heuristic, statistical, machine learning, deep learning and ensemble learning models with SHAP analysis","authors":"Shilong Yang , Jiayao Tan , Danyuan Luo , Yuzhou Wang , Xu Guo , Qiuyu Zhu , Chuanming Ma , Hanxiang Xiong","doi":"10.1016/j.cageo.2024.105723","DOIUrl":"10.1016/j.cageo.2024.105723","url":null,"abstract":"<div><p>In landslide susceptibility assessment (LSA), inventory incompleteness impacts the accuracy of different models to varying degrees. However, this area remains under-researched. This study investigated six LSA models from heuristic, statistical, machine learning and ensemble learning models (analytical hierarchy process (AHP), frequency ratio (FR), logistic regression (LR), Keras based deep learning (KBDL), XGBoost, and LightGBM) across six different sample sizes (100%, 90%, 75%, 50%, 25%, and 10%). Results revealed that XGBoost and LightGBM consistently outperformed other models across all sample sizes. The LR and KBDL models followed, while FR model was the most affected by sample size variations. AHP, an empirical model, remained unaffected by sample size. Through SHapley Additive exPlanations (SHAP) analysis, elevation, NDVI, slope, land use, and distance to roads and rivers emerged as pivotal indicators for landslide occurrences in the study area, suggesting that human activities significantly influence these events. Five time-varying indicators regarding human activity and climate validated this inference, which provides a new method to identify landslide triggering factors, especially in areas of intense human activity. Based on the findings, a comprehensive framework for LSA is proposed to assist landslide managers in making informed decisions. Future research should focus on expanding model diversity to address the effects of sample size, enhancing the adaptability of the LSA framework, deepening the analysis of human activity impacts on landslides using explainable machine learning techniques, addressing temporal inventory incompleteness in LSA, and critically evaluating model sensitivity to sample size variations across multiple disciplines.</p></div>","PeriodicalId":55221,"journal":{"name":"Computers & Geosciences","volume":"193 ","pages":"Article 105723"},"PeriodicalIF":4.2,"publicationDate":"2024-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142242003","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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