基于Bayes-GTDM的滑坡多因素危险性评价研究---以十堰市武当山区片岩滑坡群为例

Study on multi-factor hazard susceptibility assessment of landslides based on Bayes-GTDM: A case study of phyllite landslide group in the Wudang Mountain area of Shiyan

摘要

本文为解决滑坡危险性评价中因参数不确定性和小样本条件导致的适应性差、分类不稳等问题,提出一种融合贝叶斯推理与灰靶决策(grey target decision making,GTDM)的多因素评价模型(Bayes-GTDM)。在传统灰靶决策的基础上,引入贝叶斯后验更新机制,实现指标权重的动态修正与危险等级的自适应划分。以湖北省十堰市武当山区片岩滑坡群为研究对象,构建了涵盖地形、岩性、水文和人类活动等11项因子的评价指标体系,并结合实地调查和数值仿真对模型进行实证检验,同时与人工神经网络(artificial neural network,ANN)、自适应神经模糊推理系统(adaptive neuro-fuzzy inference system,ANFIS)以及粒子群优化(particle swarm optimization,PSO)的反向传播(back propagation,BP)神经网络(PSO-BP)等模型进行了对比分析。Bayes-GTDM模型在小样本条件下的分类准确率达到100%,平均误差较传统模型降低约20%,并在所有6个验证样本中均与实测危险等级一致,显著优于其他方法。在临界等级判别中,该模型能够有效减少不确定性,使危险分区结果与实地调查高度吻合。Bayes-GTDM模型通过融合概率推理与靶向评价,突破了因子独立性和权重静态性限制,提升了复杂山区滑坡灾害的智能识别能力。

Abstract

To address the challenges of poor adaptability and unstable classification in landslide susceptibility assessment caused by parameter uncertainty and small-sample conditions, this paper proposes a multi-factor evaluation model (Bayes-GTDM) that integrates Bayesian inference with grey target decision making (GTDM). On the basis of traditional GTDM, a Bayesian posterior update mechanism is introduced to dynamically adjust indicator weights and achieve adaptive classification of hazard levels. Taking the phyllite landslide group in the Wudang Mountain area of Shiyan City, Hubei Province as the case study, an evaluation index system comprising 11 factors, including topography, lithology, hydrology, and anthropogenic activities, is constructed. The model is empirically validated through field investigation and numerical simulation, and further compared with artificial neural network (ANN), adaptive neuro-fuzzy inference system (ANFIS), and particle swarm optimization back propagation neural network (PSO-BP) models. The results demonstrate that the Bayes-GTDM model achieved 100% classification accuracy under small-sample conditions, reduced the average error by approximately 20% compared with conventional models, and produced hazard levels fully consistent with the field-verified classifications of all six validation samples. In particular, the model effectively reduced uncertainty in boundary classifications and generated susceptibility zonation results that were highly consistent with field observations. The Bayes-GTDM model, by integrating probabilistic reasoning with target-oriented evaluation, overcomes the limitations of factor independence assumptions and static weighting, thereby enhancing the intelligent identification of landslide hazards in complex mountainous regions.

关键词

滑坡危险性评价 / 灰靶决策 / 贝叶斯推理 / 多因素耦合 / 武当山区

Key words

landslide susceptibility assessment / grey target decision making / Bayesian inference / multifactor coupling / Wudang Mountain

参考文献

[1] EMBERSO N R, KIRSCHBAUM D, STANLEY T. New global characterisation of landslide exposure[J]. Natural Hazards and Earth System Sciences, 2020, 20(12): 3413-3424.

[2] 冯显杰, 李益敏, 邓选伦, 等 . 高山峡谷地区地质灾害易发性评价:以怒江州为例[J]. 河南理工大学学报(自然科学版), 2024, 43(3): 70-80.

[3] FENG Xianjie, LI Yimin, DENG Xuanlun, et al. Susceptibility evaluation of geological disasters in the Alpine and canyon areas: A case study of Nujiang Prefecture[J]. Journal of Henan Polytechnic University(Natural Science), 2024, 43(3): 70-80. (in Chinese)

[4] 古腾飞, 段平, 王伟鹏, 等 . 滑坡易发性评价方法综述[J]. 自然灾害学报, 2024, 33(6): 1-16.

[5] GU Tengfei, DUAN Ping, WANG Weipeng, et al. Review of methods for landslide susceptibility assessment[J]. Journal of Natural Disasters, 2024, 33(6): 1-16. (in Chinese)

[6] 何雨枫, 丁明涛, 段钰, 等 . 2000-2019年四川省滑坡灾害时空分布特征及影响因素分析[J]. 自然灾害学报, 2024, 33(6): 17-26.

[7] HE Yufeng, DING Mingtao, DUAN Yu, et al. Analysis of spatial-temporal distribution characteristics and influence factors of landslides in Sichuan Province from 2000 to 2019[J]. Journal of Natural Disasters, 2024, 33(6): 17-26. (in Chinese)

[8] 林沛文, 何书, 鲜木斯艳·阿布迪克依木, 等 . 基于逻辑回归-模糊层次分析模型的公路沿线滑坡易发性评价[J]. 自然灾害学报, 2024, 33(6): 27-36.

[9] LIN Peiwen, HE Shu, ABUDIKE YIMU XMSY, et al. Landslide susceptibility assessment along highways based on LR-FAHP model[J]. Journal of Natural Disasters, 2024, 33(6): 27-36. (in Chinese)

[10] 李成林, 刘严松, 赖思翰, 等 . 基于逻辑回归和支持向量机耦合模型的滑坡易发性分析[J]. 自然灾害学报, 2024, 33(2): 75-86.

[11] LI Chenglin, LIU Yansong, LAI Sihan, et al. Landslide susceptibility analysis based on the coupling model of logistic regression and support vector machine[J]. Journal of Natural Disasters, 2024, 33(2): 75-86. (in Chinese)

[12] 方然可, 刘艳辉, 黄志全 . 基于机器学习的区域滑坡危险性评价方法综述[J]. 中国地质灾害与防治学报, 2021, 32(4): 1-8.

[13] FANG Ranke, LIU Yanhui, HUANG Zhiquan. A review of the methods of regional landslide hazard assessment based on machine learning[J]. The Chinese Journal of Geological Hazard and Control, 2021, 32(4): 1-8. (in Chinese)

[14] 黄发明, 胡松雁, 闫学涯, 等 . 基于机器学习的滑坡易发性预测建模及其主控因子识别[J]. 地质科技通报, 2022, 41(2): 79-90.

[15] HUANG Faming, HU Songyan, YAN Xueya, et al. Landslide susceptibility prediction and identification of its main environmental factors based on machine learning models[J]. Bulletin of Geological Science and Technology, 2022, 41(2): 79-90. (in Chinese)

[16] 高彩云, 崔希民 . 基于多因素加权灰靶决策模型的滑坡灾害危险性评价[J]. 中南大学学报(自然科学版), 2016, 47(2): 524-530.

[17] GAO Caiyun, CUI Ximin. Landslide risk assessment based on multi-index weighted grey target decision model[J]. Journal of Central South University(Science and Technology), 2016, 47(2): 524-530. (in Chinese)

[18] 王春燕, 王力 . 灰色关联的模糊综合评价法在三峡库区巴东段滑坡危险性评价中的应用[J]. 水电能源科学, 2019, 37(4): 131-135.

[19] WANG Chunyan, WANG Li. Application of grey relational fuzzy comprehensive evaluation method to landslide risk assessment in Badong section of Three Gorges Reservoir area[J]. Water Resources and Power, 2019, 37(4): 131-135. (in Chinese)

[20] 张向营, 张春山, 孟华君, 等 . 基于Random Forest和AHP的贵德县北部山区滑坡危险性评价[J]. 水文地质工程地质, 2018, 45(4): 142-149.

[21] ZHANG Xiangying, ZHANG Chunshan, MENG Huajun, et al. Landslide hazard evaluation in the northern mountainous area of Guide County based on Random Forest and AHP[J]. Hydrogeology & Engineering Geology, 2018, 45(4): 142-149. (in Chinese)

[22] 邢昭, 孟小军, 袁晶晶, 等 . 信息量支持下SVM-GBDT模型的滑坡危险性评价[J]. 科学技术与工程, 2025, 25(7): 2712-2720.

[23] XING Zhao, MENG Xiaojun, YUAN Jingjing, et al. Landslide hazard assessment based on SVM-GBDT and gradient boosting decision tree model supported by information quantity[J]. Science Technology and Engineering, 2025, 25(7): 2712-2720. (in Chinese)

[24] TANG L, YU X Y, JIANG W W, et al. Comparative study on landslide susceptibility mapping based on unbalanced sample ratio[J]. Scientific Reports, 2023, 13: 5823.

[25] LU Z Y, LIU G Y, SONG Z H, et al. Advancements in technologies and methodologies of machine learning in landslide susceptibility research: Current trends and future directions[J]. Applied Sciences, 2024, 14(21): 9639.

[26] FENG W, TANG Y M, HONG B. Landslide hazard assessment methods along fault zones based on multiple working conditions: A case study of the Lixian-Luojiabu fault zone in Gansu Province(China)[J]. Sustainability, 2022, 14(13): 8098.

[27] 刘思峰 . 灰色系统理论及其应用[M].10版. 北京: 科学出版社, 2024.

[28] LIU Sifeng. Grey system theory and its applications[M]. 10th ed. Beijing: Science Press, 2024. (in Chinese)

[29] RONG G Z, LI K W, HAN L N, et al. Hazard mapping of the rainfall-landslides disaster chain based on GeoDetector and Bayesian network models in Shuicheng County, China[J]. Water, 2020, 12(9): 2572.

[30] VOGEL K, RIGGELSEN C, KORUP O, et al. Bayesian network learning for natural hazard analyses[J]. Natural Hazards and Earth System Sciences, 2014, 14(9): 2605-2626.

[31] MAHLER C F, VARANDA E, DE OLIVEIRA L C D. Analytical model of landslide risk using GIS[J]. Open Journal of Geology, 2012, 2(3): 182-188.

[32] ARGYRIOU A V, POLYKRETIS C, TEEUW R M, et al. Geoinformatic analysis of rainfall-triggered landslides in Crete(Greece)based on spatial detection and hazard mapping[J]. Sustainability, 2022, 14(7): 3956.

[33] 黄润秋 . 20世纪以来中国的大型滑坡及其发生机制[J]. 岩石力学与工程学报, 2007, 26(3): 433-454.

[34] HUANG Runqiu. Large-scale landslides and their sliding mechanisms in China since the 20th century[J]. Chinese Journal of Rock Mechanics and Engineering, 2007, 26(3): 433-454. (in Chinese)

[35] LI S L, XU Q, TANG M G, et al. Characterizing the spatial distribution and fundamental controls of landslides in the Three Gorges Reservoir area, China[J]. Bulletin of Engineering Geology and the Environment, 2019, 78(6): 4275-4290.

[36] ROUAINIA M, HELM P, DAVIES O, et al. Deterioration of an infrastructure cutting subjected to climate change[J]. Acta Geotechnica, 2020, 15(10): 2997-3016.

[37] 张江伟, 李小军, 迟明杰, 等 . 滑坡灾害的成因机制及其特征分析[J]. 自然灾害学报, 2015, 24(6): 42-49.

[38] ZHANG Jiangwei, LI Xiaojun, CHI Mingjie, et al. Analysis of formation mechanism and characteristics of landslide disasters[J]. Journal of Natural Disasters, 2015, 24(6): 42-49. (in Chinese)

[39] 黄润秋, 戚国庆 . 非饱和渗流基质吸力对边坡稳定性的影响[J]. 工程地质学报, 2002, 10(4): 343-348.

[40] HUANG Runqiu, QI Guoqing. The effect of unsaturated soil suction on slope stability[J]. Journal of Engineering Geology, 2002, 10(4): 343-348. (in Chinese)

[41] 张茂省, 李同录 . 黄土滑坡诱发因素及其形成机理研究[J]. 工程地质学报, 2011, 19(4): 530-540.

[42] ZHANG Maosheng, LI Tonglu. Triggering factors and forming mechanism of loess landslides[J]. Journal of Engineering Geology, 2011, 19(4): 530-540. (in Chinese)

[43] 孙红月 . 含碎石粘性土滑坡的成因机理与防治对策[D]. 杭州: 浙江大学, 2005.

[44] SUN Hongyue. Formation mechanism and control measures of the pebbly clay landslide[D]. Hangzhou: Zhejiang University, 2005. (in Chinese)

[45] 崔鹏, 林勇明 . 自然因素与工程作用对山区道路泥石流、滑坡形成的影响[J]. 灾害学, 2007, 22(3): 11-16.

[46] CUI Peng, LIN Yongming. Influences of natural and engineering factors on formation of debris flow and landslide along mountainous roads[J]. Journal of Catastrophology, 2007, 22(3): 11-16. (in Chinese)

[47] LI Y R, MAO J R, XIANG X Q, et al. Factors influencing development of cracking-sliding failures of loess across the eastern Huangtu Plateau of China[J]. Natural Hazards and Earth System Sciences, 2018, 18(4): 1223-1231.

[48] 黄润秋, 向喜琼, 巨能攀 . 我国区域地质灾害评价的现状及问题[J]. 地质通报, 2004, 23(11): 1078-1082.

[49] HUANG Runqiu, XIANG Xiqiong, JU Nengpan. Assessment of China’s regional geohazards: Present situation and problems[J]. Geological Bulletin of China, 2004, 23(11): 1078-1082. (in Chinese)

[50] HU H Y, KE H, ZHANG X Y, et al. Research on a multi-dimensional indicator assessment model for evaluating landslide risk near large Alpine Reservoirs[J]. Applied Sciences, 2024, 14(12): 5201.

[51] 唐亚明, 薛强, 毕俊擘, 等 . 陕北黄土崩塌灾害风险评价指标体系构建[J]. 地质通报, 2012, 31(6): 979-988.

[52] TANG Yaming, XUE Qiang, BI Junbo, et al. The construction of factors for assessing the risk of collapse at loess slopes in northern Shaanxi Province[J]. Geological Bulletin of China, 2012, 31(6): 979-988. (in Chinese)

[53] 牟萍 . 三峡库区滑坡涌浪灾害风险评价及指标体系研究[D]. 重庆: 重庆交通大学, 2021.

[54] MU Ping. Study on risk evaluation and index system of landslide-generated impulse wave disaster in the Three Gorges Reservoir Area[D]. Chongqing: Chongqing Jiaotong University, 2021. (in Chinese)

[55] 王佳佳, 殷坤龙, 肖莉丽 . 基于GIS和信息量的滑坡灾害易发性评价:以三峡库区万州区为例[J]. 岩石力学与工程学报, 2014, 33(4): 797-808.

[56] WANG Jiajia, YIN Kunlong, XIAO Lili. Landslide susceptibility assessment based on GIS and weighted information value: A case study of Wanzhou District, Three Gorges Reservoir[J]. Chinese Journal of Rock Mechanics and Engineering, 2014, 33(4): 797-808. (in Chinese)

基金资助

南水北调水源区公路工程智慧化与绿色化建造关键技术研究项目(CSCI-2024-Z-13)