随着国家经济建设不断深入，地质环境问题日益严峻，地质灾害时有发生，为实现人与自然和谐共生以及城镇的可持续发展，开展区域地质灾害特征与风险评价研究意义重大。本文以陕北黄土高原中部的子长市作为研究区域，从地形地貌、地质环境、生态环境、人类工程活动等方面出发，分析区域地质灾害特征，基于极限平衡法与强度折减法对区内典型滑坡稳定性及失稳运动特征进行分析。选取坡度、岩土体类型、土地利用类型、归一化植被指数、年均降雨量等9个评价因子构建地质灾害危险性评价指标，利用传统信息量模型、组合赋权模型、机器学习模型3种方法比较分析获得了地质灾害区域危险性区划结果；选取人口、道路、耕地等4类评价因子，将层次分析与模糊综合评判进行组合并对子长市地质灾害易损性进行了评价，最后基于地质灾害的危险性与易损性评价结果开展了子长市的地质灾害风险评价。论文主要成果如下： （1）子长市现发育崩塌和滑坡2种地质灾害，共计105处。各灾害隐患多分布于河流两岸、黄土堆积层较厚以及人类工程活动强烈区域。区内的厚层黄土覆盖以及破碎的地形为地质灾害发育和形成提供了有利的地形地貌条件，而集中性降雨和人类工程活动强度则为诱发灾害的主要原因。 （2）以研究区内的新寨河村滑坡为典型滑坡，根据极限平衡法中的传递系数法和基于FLAC3D的强度折减法，建立了滑坡体上3处剖面评价模型，分别对自然和暴雨两种工况下的滑坡稳定性进行了分析，结果表明：该滑坡在暴雨工况下稳定性较差，滑坡整体上表现为蠕动-拉裂变形的滑动形式。 （3）通过信息量模型、引入距离函数的组合赋权模型与遗传算法优化支持向量机模型3类方法分别对子长市地质灾害危险性进行了综合评价，并利用受试者工作特征曲线对3类评价结果进行精度检验，表明机器学习模型能够更好的处理评价因子间的非线性关系，综合评价结果更为准确。 （4）基于模糊综合评判与层次分析组合方法展开地质灾害的易损性评价，根据“风险性=危险性×易损性”获得了研究区地质灾害的风险评价结果，其中极高、高风险区集中于研究区中部，占全区总面积的8.56%；中风险区占17.89%，主要沿河流、道路分布；低、极低风险区占73.55%，主要分布于研究区西北一带。

As China vigorously develops economic construction, geological environment problems are becoming more and more serious, and geological disasters occured from time to time. In order to make human and nature can coexist harmoniously and maintain the sustainable development of the city, it’s of great significance to study the characteristics and risk evaluation of geological hazards. In this paper, zichang city is taken as the research area, which is located in the northern of Shaanxi province. The characteristics of regional geological hazards are analyzed from the aspects of topography, geological environment, ecological environment and human engineering activities. The stability and instability movement characteristics of typical landslides in the region are analyzed based on limit equilibrium method and strength reduction method. Nine types of evaluation factors, including slope, rock and soil mass type, land use type, normalized difference vegetation index and average annual rainfall, were selected to construct risk evaluation indexes. The results of regional risk zoning were obtained by comparing and analyzing the three methods: traditional information volume model, combined weight model and machine learning model. Four evaluation factors, including population, road, farmland and GDP, were selected to evaluate the vulnerability of geological disasters in Zichang city by using the combination method of analytic hierarchy process and Fuzzy comprehensive evaluation method. Finally, according to above results, the risk assessment in Zichang city was completed. The main achievements of this paper are as follows:

(1) There are 105 geological disasters including collapses and landslides in Zichang city. They are mainly distributed in the areas along the rivers, thick loess deposits and strong human engineering activities. The thick loess cover and broken topography in the region provide favorable topographic and geomorphic conditions for the development and formation of geological disasters, while the concentrated rainfall and the intensity of human engineering activities are the main reasons for the occurrence of geological disasters.

(2) According to the landslide survey report, xinzhai river village landslide stability is studied, and based on limit equilibrium method and strength subtraction in FLAC^3D, three sections on landslide evaluation model is established, The stability of the landslide is analyzed under two conditions and the result shows that the stability of this landslide is worse with heavy rainfall. As a whole, the landslide is a sliding form of creep - pull - crack deformation.

(3) Through the information model, the introduction of distance function combination empowerment model with genetic algorithm optimization support vector machine model of three kinds of methods for long, made an evaluation of the geological disaster risk, and by using ROC curve accuracy test of three kinds of evaluation results, it shows that machine learning model can better deal with the nonlinear relation between evaluation factors. The risk assessment results can be also more accurate than others.

(4) The vulnerability assessment of Zichang city was completed based on fuzzy comprehension evaluation method and analytic hierarchy process models, and then according to the formulas of "Risk= Hazard×Vulnerability" ,we can get the geological disaster risk assessment result. The extremely high and high risk districts(8.56%) are focus in the central of Zichang city; the medium risk area (17.89%) are mainly distributed along the rivers and roads; low and very low risk areas accounted for 73.55%, mainly distributed in the northwest of Zichang city.

[1] 2020年中国生态环境状况公报(摘录)[J].环境保护,2021,49(11):47-68.

[2] 徐继维,张茂省,范文. 地质灾害风险评估综述[J].灾害学,2015,30(4):130-134.

[3] 卢全中,彭建兵,赵法锁. 地质灾害风险评估（价）研究综述[J].灾害学, 2013, 18(4): 59-63.

[4] 何倩. 彬长地区地质灾害特征及风险评价[D].西安科技大学,2018.

[5] 刘小青. 北川羌族自治县地质灾害风险评价[D].西南交通大学,2019.

[6] Brabb E E, PanPeyan E H, Bonilla M G. Map of Landslide Susceptibility in San Mateo Country, California[M]. US Geological Survey, 1972: 4-9.

[7] 董颖. 地质灾害风险评估理论与实践[M].北京:地质出版社,2010.

[8] 刘正刚. 日本滑坡治理主导思想及统计分析[J].铁道工程学报,1998,15(2): 81-91.

[9] Campbell Russell H, Bemlknopf Richard L. Forecasting the spatial distribution of landslide risk Abstracts with Programs[J]. Geological society of America, 1991, 23(5):145-149.

[10] 唐亚明,张茂省,李政国,等. 国内外地质灾害风险管理对比及评述[J].西北地质,2015(2): 238-246.

[11] Landslide hazard and risk assessment mapping of mountainous terrains-a case study from Kumaun Himalaya, India[J]. Engineering Geology, 1996,43(43):237-246.

[12] 张永波等. 地质灾害信息系统的设计与开发[M]. 北京: 地质出版社, 2001.

[13] C. J. Van Westen. and J. B Alzate Bonilla. Mountain hazard analysis using a PC based GIS. In:Proceedings 6th international congress International Association of Engineering Geology (IAEG): 1990, 8(2).

[14] Mejia-Navarro M, Wohl E E. Geological hazard and risk evaluation using GIS: methodology and model applied to Medellin, Colombia[J]. Bulletin of the association of engineering geologists, 1994, 31(4): 459-481.

[15] Anbalagan R, Singh B. Landslide hazard and risk assessment mapping of mountainous terrains—a case study from Kumaun Himalaya, India[J]. Engineering Geology, 1996, 43(4): 237-246.

[16] Lee S, Min K. Statistical analysis of landslide susceptibility at Yongin, Korea[J]. Environmental geology, 2001, 40(9): 1095-1113.

[17] Aleotti P, Chowdhury R. Landslide hazard assessment: summary review and new perspectives[J]. BullEng Geol Env, 1999(58):21-44.

[18] Saha A K, Gupta R P, Arora M K. GIS-based landslide hazard zonation in the Bhagirathi (Ganga) valley, Himalayas[J]. International journal of remote sensing, 2002, 23(2): 357-369.

[19] Ohlmacher G C, Davis J C. Using multiple logistic regression and GIS technology to predict landslide hazard in northeast Kansas, USA[J]. Engineering geology, 2003, 69(3-4): 331-343.

[20] M. Fall, R. Azzam, C. Noubactep. A multi-method approach to study the stability of natural slopes andlandslide susceptibility mapping[J]. Engineering Geology, 2006, 82: 241-263.

[21] Arnous M O, Green D R. GIS and remote sensing as tools for conducting geo-hazards risk assessment along Gulf of Aqaba coastal zone, Egypt[J]. Journal of Coastal Conservation, 2011, 15(4): 457-475.

[22] Bednarik M, Yilmaz I, Marschalko M. Landslide hazard and risk assessment: a case study from the Hlohovec–Sered’landslide area in south-west Slovakia[J]. Natural hazards, 2012, 64(1): 547-575.

[23] Promper C, Glade T. Multilayer-exposure maps as a basis for a regional vulnerability assessment for landslides: applied in Waidhofen/Ybbs, Austria[J]. Natural Hazards, 2016, 82(1): 111-127.

[24] Mallick J, Singh R K, AlAwadh M A, et al. GIS-based landslide susceptibility evaluation using FCE-AHP multi-criteria decision-making techniques in the Abha Watershed, Saudi Arabia[J]. Environmental Earth Sciences, 2018, 77(7): 1-25.

[25] 张晓东. 基于遥感和GIS的宁夏盐池县地质灾害风险评价研究[D].中国地质大学(北京),2018.

[26] 张业成, 张春山, 张梁. 中国地质灾害系统层次分析与综合灾度计算[J]. 中国地质科学院院报, 1993(Z1): 139-154.

[27] 唐川, 刘洪江. 泥石流堆积扇危险度分区定量评价研究[J]. 土壤侵蚀与水土保持学报, 1997(3): 64-71.

[28] 汪华斌, 吴树仁, 汪稔. 长江三峡库区滑坡灾害危险性评价[J]. 长江流域资源与环境, 1998(2): 91-97.

[29] 吴益平,唐辉明,葛修润.BP模型在区域滑坡灾害风险预测中的应用[J].岩土力学,2005(9):1409-1413.

[30] 张春山,何淑军,辛鹏等.陕西省宝鸡市渭滨区地质灾害风险评价[J].地质通报,2009,28(8):1053-1063.

[31] 吴森,张占成,周光红等.基于贡献率模型的汶川县滑坡灾害的易损性评价[J].三峡大学学报(自然科学版),2013,35(3):69-74.

[32] 孙冉,王成都,夏哲兵等.基于AHP-信息量法的费县地质灾害风险评价[J].环境科学与技术,2015,38(S1):430-435.

[33] 赵良军,李虎,刘玉锋等.新疆伊犁果子沟地质灾害风险评价及其致灾因子[J].干旱区研究,2017,34(3):693-700.

[34] Xue Q, ZHANG M, Gao B, et al. Risk assessment of geological hazards in Suide city, Shaanxi Province[J]. Journal of Engineering Geology, 2018, 26(3): 711-719.

[35] 周超,常鸣,徐璐等.贵州省典型城镇矿山地质灾害风险评价[J].武汉大学学报(信息科学版),2020,45(11):1782-1791.

[36] 罗路广,裴向军,谷虎等.基于GIS的“8.8”九寨沟地震景区地质灾害风险评价[J].自然灾害学报,2020,29(3):193-202.

[37] 杨淞月,翁茂芝,田华等.湖北省鹤峰县城区滑坡地质灾害风险评价[J].安全与环境工程,2021,28(1):144-155.

[38] 王东. 黄土地区滑坡地质灾害监测预警系统研究[D].北京交通大学,2009.

[39] 张茂省,孙传尧,校培喜等.延安市宝塔区地质灾害详细调查示范[J].西北地质,2007(2):29-55.

[40] 王恬. 子长县城空间形态优化研究[D].西安建筑科技大学,2017.

[41] 高元. 耦合于分形地貌的陕北河流交叉处城镇空间形态研究[D].西安建筑科技大学,2015.

[42] 惠鑫. 陕北黄土高原区地质灾害与降雨关系研究[D].长安大学,2010.

[43] 高姣姣. 高精度无人机遥感地质灾害调查应用研究[D].北京交通大学,2010.

[44] 苗雨. 基于GIS和Logistic模型的地质灾害危险性区划研究[D].长安大学,2010.

[45] 薛强. 延安宝塔区地质灾害危险性评价研究[D].西安科技大学,2008.

[46] 郑海军,孙羽,李晓雄.陕西省延川县永坪地区水文地质条件分析[J].中国非金属矿工业导刊,2018(S1):19-21+33.

[47] 杨少辉. 黄土丘陵区地质灾害发育特征及规律研究[D].长安大学,2010.

[48] 唐斌. ID3算法在地质灾害易发性评价中的应用[D].长安大学,2013.

[49] 子长市统计局.子长市2020年国民经济和社会发展统计公报[R].2021

[50] DZT0261-2014，滑坡崩塌泥石流灾害调查规范（1:50000）[S].中华人民共和国国土资源部，2014.

[51] 任亚龙,史伟宏,王佳武等.富县滑坡灾害发育特征和分布规律[J].陕西地质,2017,35(01):73-78.

[52] 贾晓强. 屏山县东池庄滑坡的稳定性分析和数值模拟[D].西南科技大学,2021.

[53] 马正徐. 毛坞滑坡体稳定性分析与评价[D].安徽理工大学,2020.

[54] 邓恩松. 基于FLAC3D数值模拟的伊犁野杏林黄土滑坡稳定性分析[D].新疆大学,2018.

[55] 中华人民共和国国土资源部.GB/T 32864-2016滑坡防治工程勘察规范[S].北京：中国标准出版社，2016.

[56] 徐文刚,余旭荣,年廷凯等.基于FLAC~(3D)的三维边坡稳定性强度折减法计算效率改进算法及其应用[J].吉林大学学报(地球科学版),2021,51(5):1347-1355.

[57] 王学志,翟诚,毕重等.FLAC3D在边坡稳定分析中的应用[J].吉林水利,2010(6):39-41.

[58] Zhou C, Yin K, Cao Y, et al. Application of time series analysis and PSO–SVM model in predicting the Bazimen landslide in the Three Gorges Reservoir, China[J]. Engineering geology, 2016, 204: 108-120.

[59] 肖华瓶. 基于蚁群算法的支持向量机分类方法研究及其应用[D].成都理工大学,2020.

[60] 唐跃,徐曲,柯波等.基于交叉验证的矿岩爆破块度SVM模型优选研究[J].爆破,2018,35(3):74-79.

[61] Li X Z, Kong J M. Application of GA–SVM method with parameter optimization for landslide development prediction[J]. Natural Hazards and Earth System Sciences, 2014, 14(3): 525-533.

[62] 姚李. 基于GIS的大渡河流域地质灾害危险性评价[D].成都理工大学,2020.

[63] 管新邦. 云南省滑坡地质灾害危险性评价研究[D].中国矿业大学(北京),2018.

[64] 刘宇恒, 邓辉, 熊倩莹. 基于层次分析法的茂县斜坡地质灾害易发性评价[J].长江科学院院报, 2017, 34(5): 31-35.

[65] 张晓东,刘湘南,赵志鹏等.基于层次分析法的盐池县地质灾害危险性评价[J].国土资源遥感,2019,31(3):183-192.

[66] 胡涛. 贵州省思南县地质灾害危险性评价研究[D].中国地质大学,2020.

[67] 赵晓燕,谈树成,李永平.基于斜坡单元与组合赋权法的东川区地质灾害危险性评价[J].云南大学学报(自然科学版),2021,43(2):299-305.

[68] 张泽的,刘东,张皓然等.基于PSO-AHP与粗集理论组合赋权的灌溉用水效率评价[J].节水灌溉,2018(10):59-63+67.

[69] 江强强,方堃,章广成.基于新组合赋权法的地质灾害危险性评价[J].自然灾害学报,2015,24(3):28-36.

[70] 韩玲,张庭瑜,张恒.基于IOE和SVM模型的府谷镇滑坡易发性分区[J].水土保持研究,2019,26(3):367-372.

[71] 周昊,陈斌,孟美丽.基于SVM-ABC模型的混凝土配合比优化[J].水电能源科学,2021,39(6):127-130.

[72] 贺玉琪,王栋,王远坤.BRR-SVR月降水量预测优化模型[J].水利学报,2019,50(12):1529-1537.

[73] Su H, Li X, Yang B, et al. Wavelet support vector machine-based prediction model of dam deformation[J]. Mechanical Systems and Signal Processing, 2018, 110: 412-427.

[74] 胡鹏,文章,胡新丽等.基于遗传算法-支持向量机的滑坡渗透系数反演[J].水文地质工程地质,2021,48(4):160-168.

[75] Chen Y, Ma H. Research on fault diagnosis of shearer rocker arm gear box based on PSO-SVM[J]. Coal Mine Machinery, 2015, 36(10): 303-306.

[76] Ji W, Liu D, Meng Y, et al. A review of genetic-based evolutionary algorithms in SVM parameters optimization[J]. Evolutionary Intelligence, 2021, 14(4): 1389-1414.

[77] Ali L, Wajahat I, Golilarz N A, et al. LDA–GA–SVM: improved hepatocellular carcinoma prediction through dimensionality reduction and genetically optimized support vector machine[J]. Neural Computing and Applications, 2021, 33(7): 2783-2792.

[78] Huang S, Zheng X, Ma L, et al. Quantitative contribution of climate change and human activities to vegetation cover variations based on GA-SVM model[J]. Journal of Hydrology, 2020, 584: 124687.

[79] Bahadure N B, Ray A K, Thethi H P. Image analysis for brain tumour detection using GA-SVM with auto-report generation technique[J]. International Journal of Biomedical Engineering and Technology, 2020, 32(3): 246-266.

[80] 谷艳昌,吴云星,黄海兵等.基于遗传算法优化支持向量机的大坝安全性态预测模型[J].河海大学学报(自然科学版),2020,48(5):419-425.

[81] 程彩凤,孙祥娥.基于社会情感优化算法的支持向量机参数选择[J].现代电子技术,2019,42(12):108-111+116.

[82] 刘希亮,赵学胜,陆锋等.基于GA-SVM的露天矿抛掷爆破抛掷率预测[J].煤炭学报,2012,37(12):1999-2005.

[83] 张向营,张春山,孟华君等.基于GIS和信息量模型的京张高铁滑坡易发性评价[J].地质力学学报,2018,24(1):96-105.

[84] 任敬,范宣梅,赵程等.贵州省都匀市滑坡易发性评价研究[J].水文地质工程地质,2018,45(5):165-172.

[85] 张晓东,刘湘南,赵志鹏等.信息量模型、确定性系数模型与逻辑回归模型组合评价地质灾害敏感性的对比研究[J].现代地质,2018,32(3):602-610.

[86] 张以晨,秦胜伍,翟健健等.基于信息量的长白山地区泥石流易发性评价[J].水文地质工程地质,2018,45(2):150-158.

[87] 冯满. 基于机器学习模型的大宁县滑坡灾害危险性评价[D].长安大学,2020.

[88] Hauke J, Kossowski T. Comparision of Values of Pearson’s and Spearman’s Cprrelation Coefficientson the Same Sets of Data[J].Quaestiones Geographicae, 2011, 30(2): 87-93.

[89] 杜常见,易庆林,周宝等.基于GIS和加权信息量的三峡库区云阳县滑坡灾害易发性评价[J].三峡大学学报(自然科学版),2017,39(2):48-53.

[90] 王哲,易发成.基于层次分析法的绵阳市地质灾害易发性评价[J].自然灾害学报,2009,18(1):14-23.

[91] 崔玲玲,许金兰,徐岗等.融合双特征图信息的图像显著性检测方法[J].中国图象图形学报,2018,23(4):583-594.

[92] 蔡向阳,铁永波.我国山区城镇地质灾害易损性评价研究现状与趋势[J].灾害学,2016,31(4):200-204.

[93] 郝文辉,贡长青,王伟等.秦皇岛市地质灾害土地资源易损性评价[J].地质灾害与环境保护,2011,22(3):56-60.

[94] 张晓东. 基于遥感和GIS的宁夏盐池县地质灾害风险评价研究[D].中国地质大学(北京),2018.

[95] 于洋,杨瑞金,高卫恒.多级模糊评估的输变电设备状态评价体系研究[J].云南电力技术,2017,45(1):62-66.

[96] 李云杰,屈飞行,杨梓俪.层次分析法在线路工程地质灾害危险性评价中的应用[J].甘肃水利水电技术,2020,56(4):33-36.

[97] 白月峰. 基于模糊综合评判法的河道治理工程质量评价研究[D].大连理工大学,2019.

[98] 上官超超. 合阳县地质灾害风险评价[D].长安大学,2017.

[99] 郑建仙.水利水电项目的质量控制应用研究——以乌江构皮滩水电站为例[J].中国设备工程,2018(24):149-150.

[100] 邹浩,陈金国,李庆伟等.大别山世界地质公园区域地质灾害风险分析[J].资源环境与工程,2021,35(5):641-651.

[101] 豆敬翔. 基于GIS的平武县地质灾害危险性评价研究[D].成都理工大学,2012.

[102] 李春燕,孟晖,张若琳等.基于承灾体易损性的县域单元地质灾害风险评估[J].地质通报,2021,40(9):1547-1559.