| تعداد نشریات | 282 |
| تعداد نشریات سازمان | 80 |
| تعداد شمارهها | 3,231 |
| تعداد مقالات | 35,794 |
| تعداد مشاهده مقاله | 54,046,501 |
| تعداد دریافت فایل اصل مقاله | 93,209,435 |
پهنهبندی و پیشبینی خطر رخداد سیل با استفاده از مدلهای یادگیری عمیق در آبخیز قرهسو استان گلستان | ||
| پژوهش های آبخیزداری | ||
| مقاله 3، دوره 39، شماره 3 - شماره پیاپی 152، مهر 1405، صفحه 22-44 اصل مقاله (1.45 M) | ||
| نوع مقاله: پژوهشی | ||
| شناسه دیجیتال (DOI): 10.22092/wmrj.2026.371561.1648 | ||
| نویسندگان | ||
| کورش شیرانی* 1؛ محمدعلی صدری2 | ||
| 1دانشیار پژوهشکده حفاظت خاک و آبخیزداری، سازمان تحقیقات، آموزش و ترویج کشاورزی، تهران، ایران | ||
| 2کارشناس ارشد دانشگاه آزاد اسلامی لنجان اصفهان، ایران | ||
| چکیده | ||
| مقدمه و هدف سیل یکی از ویرانگرترین بلاهای طبیعی با پیامدهای اجتماعی، اقتصادی و زیستمحیطی است. ازاینرو، روشهای یادگیری ماشین برای مدلسازی و پیشبینی آن توسعهداده شدهاند. هدف این پژوهش، پهنهبندی و پیشبینی خطر رخداد سیل با استفاده از سه مدل یادگیری عمیق شامل مدلهای شبکه عصبی بازگشتی واحد دروازهدار (GRU)، حافظه کوتاه-بلندمدت (LSTM) و شبکههای پیچشی زمانی (TCN) و معرفی مناسبترین آنها در آبخیز قرهسو استان گلستان بود. مهمترین نوآوری این پژوهش مقایسه نتایج آن با نتایج دیگر پژوهشهای مرتبط با پهنهبندی خطرها در داخل کشور، بهکارگیری مدلهای جدید یادگیری عمیق و درنظرگرفتن بیشترین عاملهای سیلخیزی برای تعیین کارآمدترین مدل و افزایش دقت نقشههای پیشبینی رخداد سیل بود. مواد و روشها منطقه پژوهشی آبخیز قرهسو گرگان انتخاب شد. سپس، نقشههای عاملهای مؤثر جمعآوری و آمادهسازی شد. در این پژوهش 16 عامل بهعنوان متغیرهای مستقل و اطلاعات دریافتشده از سازمان مدیریت منابع آب وزارت نیرو و بازدیدهای میدانی بهعنوان متغیر وابسته، انتخاب شدند. سپس، نقشة پراکنش رخداد سیل تهیه شد. برای مدلسازی و پیشبینی خطر رخداد سیل، نقشه پراکنش رخداد سیل به دو دستة نقاط آزمایشی (30%) و آموزشی (70%) تقسیم شد. پس از اجرای مدلهای TCN، LSTM و GRU، نقشههای بهدست آمده از پهنهبندی خطر رخداد سیل به پنج ردة خیلیکم، کم، متوسط، زیاد و خیلیزیاد طبقهبندی شدند. سرانجام، دقت طبقهبندی و اعتبارسنجی نقشههای پهنهبندی ارزیابی، خطر رخداد سیل پیشبینی و مناسبترین مدل انتخاب شد. نتایج و بحث برای تهیة نقشة پهنهبندی رخداد سیل، دادههای مکانی 59 نقطه سیل (متغیر وابسته) بر اساس اطلاعات موجود و بررسی تصویرهای راداری در بازه زمانی قبل و بعد از سیلاب با استفاده نرمافزار گوگلارث، جمعآورری شد. در هر یک از این نقاط رخداد سیل در دوره 20 ساله (1403-1383) با 10 بار تناوب رخداد سیل ثبت و گزارششده بود. از این تعداد 70% بهعنوان دادههای آموزشی و 30% بهعنوان دادههای آزمایشی بهترتیب برای اجرا و اعتبارسنجی مدل، بهطور تصادفی انتخاب و تقسیم شدند. در مرحله بعد، از 16 عامل در قالب گروههای محیطی، آبشناسی، پستیبلندی آبخیز بههمراه دادههای اقلیمی (بهعنوان متغیر مستقل) و برای پهنهبندی و مدلسازی استفاده شد. دقت طبقهبندی مدلها با استفاده از دو شاخص نسبت فراوانی (FR) و مساحت سلول هسته (SCAI) ارزیابی شد. نتایج نشان داد که بیشتر نقاط سیل در ردههای پرخطر (زیاد و خیلیزیاد) بودند و بیشترین سهم از مساحت منطقه نیز مربوط به این پهنهها بود. افزون بر این، بر اساس مساحت زیر منحنیهای عملگر نسبی و هزینه مشخص شد کارآمدترین و مناسبترین و اقتصادیترین مدل در مقایسه با مدلهای LSTM و GRU مدل TCN بود. در میان مدلهای بررسیشده بیشترین و کمترین مساحت زیر منحنیهای عملگر نسبی و هزینه مربوط به مدل TCNبهترتیب 0/92 و 0/08 بود. نتایج این پژوهش نشان داد بر پایة مدل TCN، 45% از مساحت منطقه در رده خطر خیلیزیاد در برگیرندة 83% از رخدادهای سیل بودند. همچنین، بر پایة مدل LSTM، 23% از مساحت منطقه در برگیرندة 60% از رخدادهای سیل و بر پایة مدل GRU، 10% از مساحت منطقه در برگیرندة 4% از رخدادهای سیل بودند. از نقشههای پهنهبندی خطر رخداد سیل بهدست آمده در این پژوهش میتوان بهعنوان مبنایی برای برنامهریزی و مدیریت بحران ناشی از رخداد سیل، کاهش خطر سیل و مدیریت آن بهره برد. افزون بر این، بر اساس عملکرد عالی هر سه مدل TCN، LSTM و GRU، نتایج این پژوهش برای توسعه طرحهای آتی سازمانهای گوناگون و فعال در بسیاری از کشورهای در حال توسعه لازم و ضروری است. نتیجهگیری و پیشنهادها پس از اجرای هر سه مدل TCN، LSTM و GRU، نتایج شاخص AUC-ROC، در هر دو مرحله آموزش و آزمون نشاندهندة عملکرد عالی مدلهای نامبرده بود. نقشههای تهیهشده از پهنهبندی خطر رخداد سیل با کاربرد هر سه مدل، نشاندهنده دقت زیاد طبقهبندی پهنهها و توزیع مناسب نقاط سیلابی در ردههای زیاد و خیلیزیاد بود. برازش مدل TCN در پیشبینی رخداد سیل در مرحله آزمون در مقایسه با دو مدل دیگر، بهتر بود. ارزیابی دقت طبقهبندی و اعتبارسنجی مدلها نشان داد دو مدل TCN و LSTM با داشتن آستانه طبقات مناسب در طبقهبندی، اولویت کاربرد آنها برای پهنهبندی و پیشبینی خطر رخداد سیل بیشتر است. سرانجام در این پژوهش، اندازههای شاخص مساحت زیر منحنی ارزش یا هزینه برای مدلهای TCN، LSTM و GRU بهترتیب 0/08، 0/10 و 0/11 بهدست آمد. پیشنهاد میشود بر اساس اهمیت ویژگیهای ریختسنجی در پهنهبندی نقشههای خطر رخداد سیل، در پژوهشهای آینده از این سنجهها همراه با دیگر سنجههای متداول در تهیه نقشههای پهنهبندی استفاده شود. افزون بر این، بر پایة نتایج این پژوهش پیشنهاد میشود از روشهای یادگیری عمیق و ترکیب آن با یکدیگر استفاده شود. همچنین پیشنهاد میشود بهرهگیری از روشهای ترکیبی نوین (هیبریدی) و بهینهسازهای مناسب در دیگر آبخیزهای استان گلستان، بررسی و ارزیابی شود. | ||
| کلیدواژهها | ||
| پهنهبندی؛ پیشبینی؛ خطر؛ سیل؛ قرهسو؛ مدلهای یادگیری عمیق | ||
| عنوان مقاله [English] | ||
| Flood Hazard Zonation and Prediction using Deep Learning Models in the Qarasu Watershed, Golestan Province | ||
| نویسندگان [English] | ||
| Kourosh Shirani1؛ Mohamad Ali Sadri2 | ||
| 1Associate Professor, Soil Conservation and Watershed Management Research Institute, Agricultural Research, Education and Extension Organization (AREEO), Tehran, Iran | ||
| 2M.Sc., Islamic Azad University, Lenjan, Isfahan, Iran | ||
| چکیده [English] | ||
| Introduction and Goal Flooding is one of the most destructive natural disasters with social, economic and environmental consequences. Therefore, machine learning methods have been developed to model and predict it. The purpose of this research was to zone and predict flood hazard using three deep learning models, including Gated Recurrent Unit (GRU), Long Short-Term Memory (LSTM) and Temporal Convolutional Network (TCN) models, and to introduce the most suitable ones in the Qarasu watershed of Golestan Province. The most important innovation of this research was comparing its results with the results of other research related to susceptibility zonation within the country, using new deep learning models, and considering the most flood factors to determine the most efficient model and increase the accuracy of flood prediction maps. Materials and Methods The research area of the Qarasu watershed in Gorganwas was selected. Then maps of the effective factors were collected and prepared. In this study, 16 factors were selected as independent variables and information received from the Ministry of Energy and field visits were selected as dependent variable. Then, a flood event distribution map was prepared. To model and predict flood susceptibility, the flood event distribution map was divided into two groups of experimental points (30%) and training points (70%). After running the TCN, LSTM and GRU models, the maps obtained from flood hazard zonation maps were classified into five classes: very low, low, moderate, high and very high. Finally, the classification accuracy and validation of the zonation and prediction maps were evaluated the flood susceptibility was predicted, and the most appropriate model was selected. Results and Discussion To prepare a flood event zonation map, spatial data of 59 flood occurrence points (dependent variable) were collected using available information and by analyzing radar images acquired before and after flood events in Google Earth. Each of these locations experienced 10 recorded and documented flood events during the 20- year study period (2004–2024). Of the total dataset, 70% were randomly selected and divided as training data and 30% as test data for model implementation and validation, respectively. In the next step, 16 factors in the form of environmental groups, hydrology, watershed topography along with climatic data (as independent variables) for zonation and modeling. The classification accuracy of the models was evaluated using two indices, frequency ratio (FR) and kernel cell area (SCAI). The results showed that most flood areas were in the high-hazard class (high and very high) and the largest share of the area of the region was also related to these areas. In addition, based on the area under the relative operator curves and cost, it was determined that the most efficient, appropriate, and cost-effective model was the TCN model compared to the LSTM and GRU models. Among the models studied, the highest and lowest areas under the relative operator and cost curves of the TCN model were 0.92 and 0.08, respectively. The results of this study showed that, based on the TCN model, 45% of the area of the region was in the very high susceptibility class, containing 83% oflood events. Also, based on the LSTM model, 23% of the region contained 60% of the flood events, and based on the GRU model, 10% of the region contained 4% of the flood events. The flood hazard zonation maps obtained in this study can be the basis for planning and crisis management caused by flood events, reducing flood hazard, and managing it. Furthermore, based on the excellent performance of all three TCN, LSTM, and GRU models, the results of this research are necessary and essential for the development of future projects of various organizations operating in many developing countries. Conclusion and Suggestions After running all three TCN, LSTM, and GRU models, the AUC-ROC index results in both training and testing step showed excellent performance of the mentioned models. The maps prepared from flood hazard zonation using all three models indicated high accuracy in zone classification and appropriate distribution of flood points in high and very high class. The fit of the TCN model in predicting flood occurrence in the test phase was better compared to the other two models. The evaluation of classification accuracy and validation of the models showed that two TCN and LSTM models, having the appropriate thresholds for classification, are the priority of their application for zonation and predicting the hazard of flooding. Finally, in this study, the area under the value or cost curve index values for TCN, LSTM, and GRU models were obtained as 0.08, 0.10, and 0.11, respectively. Based on the importance of morphometric features in the zoning of flood hazard maps, it is suggested that future research should use these metrics along with other common metrics in the preparation of zoning maps. Furthermore, based on the findings of this study, the application of deep learning methods and their combinations is recommended. It is also suggested that novel hybrid approaches and suitable optimization algorithms be investigated and evaluated in other watersheds of Golestan Province. | ||
| کلیدواژهها [English] | ||
| Deep learning models, floods, hazard, prediction, Qarasu, zonation | ||
| مراجع | ||
|
Aboah A, Adu-Gyamfi Y, Gursoy SV, Merickel J, Rizzo M, Sharma A. 2023. Driver maneuver detection and analysis using time series segmentation and classification. Journal of Transportation Engineering. Part A, Systems. 149(3): 1–15. https://doi.org/10.1061/JTEPBS.TEENG-7312 GDNRWM- General Department of Natural Resources and Watershed Management of Golestan Province, Iran. 2018. Behera I, Jain N, Martha TR, Bose M. 2026. Optimizing sampling strategies for AI-based landslide susceptibility mapping: a case study of Idukki, Kerala. Nat Hazards. 122(1). https://doi.org/10.1007/s11069-025-07842-8 Bentivoglio R, Isufi E, Jonkman SN, Taormina R. 2022. Deep learning methods for flood mapping: a review of existing applications and future research directions. Hydrology and Earth System Sciences. 26(16): 4345–4378. https://doi.org/10.5194/hess Chen H, Lu T, Huang J, He X, Yu K. 2023. An improved VMD-LSTM model for time-varying GNSS time series prediction with temporally correlated noise. Remote Sensing. 15: 3694. https://doi.org/10.3390/ rs15143694 Chen J, Li Q, Wang H, Deng M. 2020. A machine learning ensemble approach based on random forest and radial basis function neural network for risk evaluation of regional flood disaster: a case study of the Yangtze River Delta, China. International Journal of Environmental Research Public Health. 17(1): 49. https://doi.org/10.3390/ijerph17010049 Chen W, Panahi M, Tsangaratos P, Shahabi H, Ilia I, Panahi S, Ahmad BB. 2019. Applying population-based evolutionary algorithms and a neuro-fuzzy system for modeling landslide susceptibility. Catena. 172: 212–231. https://doi.org/10.1016/j.catena.2018.08.025 Chen Y, He Y, Zhang LF. 2021. Prediction of InSAR deformation time-series using a long short-term memory neural network. International Journal of Remote Sensing. 42(18): 6921–6944. https://doi.org/10.1080/01431161.2021.1947540 Choubin B, Moradi E, Golshan M, Adamowski J, Sajedi-Hosseini F, Mosavi A. 2019. An ensemble prediction of flood susceptibility using multivariate discriminant analysis, classification and regression trees, and support vector machines. Science of the Total Environment. 651: 2087–2096. https://doi.org/10.1016/j.scitotenv.2018.10.064 Convertino M, Annis A, Nardi F. 2019. Information-theoretic portfolio decision model for optimal flood management. Environmental Modelling and Software 119: 258–274. https://doi.org/10.1016/j.envsoft.2019.06.013 Costache R, Zaharia L. 2017. Flash-flood potential assessment and mapping by integrating the weights-of-evidence and frequency ratio statistical methods in GIS environment–case study: Bâsca Chiojdului River catchment (Romania). Journal of Earth System Science. 126(4): 1–19. https://doi.org/10.1007/s12040-017-0828-9 Dalu MT, Shackleton CM, Dalu T. 2018. Influence of land cover, proximity to streams and household topographical location on flooding impact in informal settlements in the Eastern Cape, South Africa. International Journal of Disaster Risk Reduction 28: 481–490. https://doi.org/10.1016/j.ijdrr.2017.12.009 Dano UL, Balogun AL, Matori AN, Wan Yusouf K, Abubakar IR, Said Mohamed MA, Aina YA, Pradhan B. 2019. Flood susceptibility mapping using GIS-based analytic network process: A case study of Perlis, Malaysia. Water. 11(3): 615. https://doi.org/10.3390/w11030615 Dodangeh E, Choubin B, Eigdir AN, Nabipour N, Panahi M, Shamshirband S, Mosavi A. 2020a. Integrated machine learning methods with resampling algorithms for flood susceptibility prediction. Science of the Total Environment. 705: 135983. https://doi.org/10.1016/j.scitotenv.2019.135983 Dodangeh E, Panahi M, Rezaie F, Lee S, Bui DT, Lee CW, Pradhan B. 2020b. Novel hybrid intelligence models for flood-susceptibility prediction: Meta optimization of the GMDH and SVR models with the genetic algorithm and harmony search. Journal of Hydrology. 590: 125423. https://doi.org/10.1016/j.jhydrol.2020.125423 Drummond C, Holte RC. 2006. Cost curves: An improved method for visualizing classifier performance. Machine Learning. 65: 95–130. https://doi.org/10.1007/s10994-006-8199-5 Eslaminezhad SA, Eftekhari M, Akbari M. 2020. GIS-based flood risk zoning based on data-driven models. Journal of Hydraulic Structures. 6(4): 75–98. https://doi.org/10.22055/jhs.2021.36629.1163 Gebrehiwot A, Hashemi-Beni L, Thompson G, Kordjamshidi P, Langan T. 2019. Deep convolutional neural nNetwork for flood extent mapping using unmanned aerial vehicles data. Sensors. 19(7): 1486. https://doi.org/10.3390/s19071486 Islam ARMT, Talukdar S, Mahato S, Kundu S, Eibek KU, Pham QB, Linh NTT. 2021. Flood susceptibility modelling using advanced ensemble machine learning models. Geoscience Frontiers. 12(3): 101075. https://doi.org/10.1016/j.gsf.2020.09.006 Kaur S, Kaur A, Kumar M. 2026. Recent Advances in Computational Methods in Science and Technology: Volume 1 (1st ed.). CRC Chemical Rubber Company Press. https://doi.org/10.1201/9781003662839 Khosravi K, Shahabi H, Pham BT, Adamowski J, Shirzadi A, Pradhan B, Dou J, Ly HB, Gróf G, Ho HL. 2019. A comparative assessment of flood susceptibility modeling using Multi-Criteria Decision-Making Analysis and Machine Learning Methods. J. Hydrol Journal of Hydrology. 573: 311–323. https://doi.org/10.1016/j.jhydrol.2019.03.073 Kia MB, Pirasteh S, Pradhan B, Mahmud A R, Sulaiman WNA, Moradi A. 2012. An artificial neural network model for flood simulation using GIS: Johor River Basin, Malaysia. Environ. Earth Science. 67: 251–264. https://doi.org/10.1007/s12665-011-1504-z Kratzert F, Herrnegger M, Klotz D, Hochreiter S, Klambauer G. 2019. Neural hydrology–interpreting LSTMs in hydrology. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). 11700 Lecture Notes in Computer Science. pp. 347–362. https://doi.org/10.1007/978-3-030-28954-6 LeCun Y, Bengio Y, Hinton G. 2015. Deep learning. Nature. 521(7553): 436–444. https://doi.org/10.1038/nature14539 Li J, Fan C, Zhao K, Zhang Z, Duan P. 2025. Landslide displacement prediction using time series InSAR with combined LSTM and TCN. Application to the Xiao Andong landslide, Yunnan Province, China. Natural Hazards. 121(4): 3857–3884. https://doi.org/10.1007/s11069-024-06937-y Maier HR, Jain A, Dandy GC, Sudheer KP. 2010. Methods used for the development of neural networks for the prediction of water resource variables in river systems: Current status and future directions. Environmental Modelling and Software. 25: 891–909. https://doi.org/10.1016/j.envsoft.2010.02.003 Malekian A, Khozani AO, Ashoornejad Gh. 2012. Zoning of flooding potential of Akhtarabad watershed using fuzzy hierarchical analysis method. Natural Geography Research. 4(44):162–191. (In Persian). https://doi.org/10.22059/jphgr.2012.30246 Mosavi A, Ozturk P, Chau KW. 2018. Flood prediction using machine learning models: Literature Review. Water. 10(11): 1536. https://doi.org/10.3390/w10111536 Nachappa TG, Piralilou ST, Gholamnia K, Ghorbanzadeh O, Rahmati O, Blaschke T. 2020. Flood Susceptibility Mapping with Machine Learning, Multi-Criteria Decision Analysis and Ensemble Using Dempster Shafer Theory. Journal of Hydrology. 590: 125275. https://doi.org/10.1016/j.jhydrol.2020.125275 Nhu VH, Mohammadi A, Shahabi H, Ahmad BB, Al-Ansari N, Shirzadi A, Clague JJ, Jafari A, Chen W, Nguyen H. 2020. Landslide susceptibility mapping using machine learning algorithms and remote sensing data in a tropical environment. International Journal of Environmental Research and Public Health 17(14): 4933. https://doi.org/10.3390/ijerph17144933 Pham BT, Avand M, Janizadeh S, Phong T V, Al-Ansari N, Ho LS, Das S, Le HV, Amini A, Bozchaloei SK, Jafari F, Prakash, I. 2020. GIS based hybrid computational approaches for flash flood susceptibility assessment. Water. 12(3):683. https://doi.org/10.3390/w12030683 Pourghasemi HR, Moradi HR, Aghda SMF, Gokceoglu C, Pradhan B. 2014. GIS-based landslide susceptibility mapping with probabilistic likelihood ratio and spatial multi-criteria evaluation models (North of Tehran, Iran). Arabian Journal of Geosciences. 7: 1857–1878. https://doi.org/10.1007/s12517-012-0825-x Pourtaghi ZS, Pourghasemi HR. 2015. GIS-based groundwater spring potential assessment and mapping in the Birjand Township. Southern Khorasan Province, Iran. Hydrogeology Journal. 22: 643–662. https://doi.org/10.1007/s10040-013-1089-6 Pradhan B. 2009. Flood susceptible mapping and risk area delineation using logistic regression, GIS and remote sensing. Journal of Spatial Hydrology. 9(2): 1–18. https://scholarsarchive.byu.edu/josh/vol9/iss2/4 Rahmati O, Darabi H, Panahi M, Kalantari Z, Naghibi SA, Ferreira CS, Kornejady A, Karimidastenaei Z, Mohammadi F, Stefanidis S, Tien Bui D. 2020. Development of novel hybridized models for urban flood susceptibility mapping. Scientific Reports. Jul 31. 10(1): 12937. https://doi.org/10.1038/s41598-020-69703-7 Rahmati O, Haghizadeh A, Stefanidis S. 2016a. Assessing the accuracy of GIS-based analytical hierarchy process for watershed prioritization. Gorganrood River Basin, Iran. Water Resources Management. 30(3): 1131–1150. https://doi.org/10.1007/s11269-015-1215-4 Rahmati O, Pourghasemi HR, Zeinivand H. 2016b. Flood susceptibility mapping using frequency ratio and weights-of-evidence models in the Golastan Province, Iran. Geocarto International. 31(1): 42–70. https://doi.org/10.1080/10106049.2015.1041559 Rozalis S, Morin E, Yair Y, Price C. 2010. Flash flood prediction using an uncalibrated hydrological model and radar rainfall data in a Mediterranean watershed under changing hydrological conditions. Journal of Hydrology. 394: 245–255. https://doi.org/10.1016/j.jhydrol.2010.03.021 Sajedi-Hosseini F, Malekian A, Choubin B, Rahmati O, Cipullo S, Coulon F, Pradhan B. 2018. A novel machine learning-based approach for the risk assessment of nitrate groundwater contamination. Science of the Total Environment. 644: 954–962. https://doi.org/10.1016/j.scitotenv.2018.07.054 Saleh A, Tan ML, Nazri FM, Yaseen ZM, Zhang F. 2025. Advancing tropical flood susceptibility mapping with multi-head attention mechanism and deep learning models. Advances in Space Research. https://doi.org/10.1016/j.asr.2025.07.022 Shahabi H, Shirzadi A, Ghaderi K, Omidvar E, Al-Ansari N, Clague JJ, Geertsema M, Khosravi K, Amini A, Bahrami S, Rahmati O. 2020. Flood detection and susceptibility mapping using sentinel-1 remote sensing data and a machine learning approach: Hybrid intelligence of bagging ensemble based on k-nearest neighbor classifier. Remote Sensing. 12 (2): 266. https://doi.org/10.3390/rs12020266 Shirani K, Sadri MA. 2023. Prioritizing effective factors and zoning of flood events using probabilistic models and GIS in the Gorganrood watershed of Golestan province. Watershed Management Research. 36(1): 111–131. (In Persian). https://doi.10.22092/wmrj.2022.358648.1473 Shirani K, Pasandi M, Arabameri A. 2018. Landslide susceptibility assessment by dempster–shafer and index of entropy models, Sarkhoun basin, southwestern Iran. Natural Hazards. 93: 1379–1418. https://doi.org/10.1007/s11069-018-3356-2 Shirani K, Peyrowan HR, Naderi Samani R. 2024. Applied principles of zonation and prediction landslide susceptibility (Machine Learning and Programming in MATLAB). Soil Conservation and Watershed Management Institute. 259 p. (In Persian). Tehrany MS, Pradhan B, Jebur MN. 2014. Flood susceptibility mapping using a novel ensemble weights-of-evidence and support vector machine models in GIS. Journal of Hydrology. 512: 332–343. https://doi.org/10.1016/j.jhydrol.03.008 Vafakhah M, Loor SMH, Pourghasemi HR, Katebikord A. 2020. Comparing performance of random forest and adaptive neuro-fuzzy inference system data mining models for flood susceptibility mapping. Arabian Journal of Geosciences. 13: 417. https://doi.org/10.1007/s12517-020-05363-1 Wang Y, Fang Z, Wang M. 2020. Comparative study of landslide susceptibility mapping with different recurrent neural networks. Computers and Geosciences. 138: 104445. https://doi.org/10.1016/j.cageo.2020.104445 Zhang D, Yang J, Li F, Han S, Qin L. 2022. Landslide risk prediction model using an attention-based temporal convolutional network connected to a recurrent neural network. IEEE Access. 10: 37635–37645. https://doi.org/10.1109/ACCESS.2022.3165051 Zhang YG, Tang J, He ZY. 2021. A novel displacement prediction method using gated recurrent unit model with time series analysis in the Erdaohe landslide. Nat Hazards. 105: 783–813. https://doi.org/10.1007/s11069-020-04337-6 Zhao ZA, He Y, Yao S, Yang W, Wang W, Zhang L, Sun Q. 2022. A comparative study of different neural network models for landslide susceptibility mapping. Advances in Space Research. 70(2): 383–401. https://doi.org/10.1016/j.asr.2022.04.055 | ||
|
آمار تعداد مشاهده مقاله: 81 تعداد دریافت فایل اصل مقاله: 22 |
||