Flood simulation and visualization framework using position-based fluids
Başlık çevirisi mevcut değil.
- Tez No: 622294
- Danışmanlar: PROF. DR. DANIŞMAN YOK
- Tez Türü: Doktora
- Konular: Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrol, Computer Engineering and Computer Science and Control
- Anahtar Kelimeler: Belirtilmemiş.
- Yıl: 2019
- Dil: İngilizce
- Üniversite: Texas A&m Unıversıty
- Enstitü: Yurtdışı Enstitü
- Ana Bilim Dalı: Belirtilmemiş.
- Bilim Dalı: Belirtilmemiş.
- Sayfa Sayısı: Belirtilmemiş.
Özet
Assessing flood resilience of urban environments has been an ongoing work as flood modeling techniques and various ways of visualizing flood help communities discover weaknesses and become more resilient against disasters. From 3D urban environment reconstruction to flood models that account for urban environments, tests are conducted to gain more insight regarding local dynamics that affect water flow. Recent work has been contributing to this field, by proposing methods to ease the preliminary evaluation yet efficiency is a problem, especially for real-time particle-based simulations. This work proposes a framework to generate the city model from OpenStreetMap (OSM) data, incorporate simulation parameters in the form of friction coefficients and storm drain information and then using mechanics such as precipitation and soil absorption to simulate and visualize flooding in 3D. As an efficient visual aid, a flexible depth mapping engine, which can transition between regular and adaptive grids is provided. To process the city information, OSM data alongside Satellite Radar Topography Mission (SRTM) elevation information is extracted. Upon efficient construction of a simulation-ready virtual environment, position-based fluids are utilized to supply an efficient method of 3D simulation as several shortcomings of other prominent approaches are alleviated through position-based dynamics that this method introduces. Localized rainfall mechanics with spatial distribution differentiator and soil absorption mechanics are embedded to accommodate particle addition and deletion from the simulation environment. Data collected from OpenStreetMap is utilized to generate digital city models of selected areas. Using 25km2 of sections from each city from different regions of the world, the features are visualized, and city models are generated. Preliminary evaluation showed that all 6 areas are reconstructed and visualized properly for further evaluation. Moreover, it is concluded that, the 3D urban environment reconstruction module built for this work can accomplish parsing and rendering areas of varying extent, adding and removing features, extracting simulation parameters through feature-level friction parameters and storm drain information extraction. Upon generation of such a city model, it can be passed as an input to the flood visualization framework for viewing and simulating. A limitation is that larger sections of cities may not utilize real-time viewing experience due to performance issues caused by large number of features being rendered. To supplement 3D particle-based flood simulations, a depth mapping engine is provided, which can efficiently track particle clusters. Considering several environmental parameters and simulation performance, the mapping engine has the power to choose a representation between regular and adaptive grids. Through utility of Jenks natural breaks for classification purposes, appropriate coloring is provided that adds context to the environments being investigated. An experiment was conducted with the position-based flood visualization framework to assess its visual fidelity to the river flooding event of 2000, in Bewdley, UK. Utilizing SRTM with 1 arc-second resolution for preparation of the 3D digital model of the river and surroundings, the flooding was represented with approximately 1 million particles. The results were satisfactory as the flood extent was represented properly. Furthermore, comparison with an existing model investigating the same event showed that the visual fidelity of FloodSim was superior to the other method. Consequently, the framework can be a useful tool for practitioners who have sufficient of information about the study area and would like to visualize flooding using a particle-based approach for real-time particle tracking, flood path analysis and incorporating precipitation to their models.
Özet (Çeviri)
ı
Benzer Tezler
- Yapı bilgi modellemesi (5D) ile maliyet yönetiminin avantaj ve dezavantajlarının tespiti
Determining advantages and disadvantages of 5D BIM cost management
FARUK GÜLERSES
Yüksek Lisans
Türkçe
2018
İnşaat Mühendisliğiİstanbul Teknik Üniversitesiİnşaat Mühendisliği Ana Bilim Dalı
DOÇ. DR. ESİN ERGEN PEHLEVAN
- Modeling and control of a new unmanned aerial vehicle (SUAVI) with tilt-wing mechanism
Döner-kanat mekanizmalı yeni bir insansız hava aracının (SUAVİ) modellemesi ve kontrolü
KAAN TAHA ÖNER
Yüksek Lisans
İngilizce
2009
Mühendislik BilimleriSabancı ÜniversitesiMekatronik Mühendisliği Ana Bilim Dalı
DOÇ. DR. MUSTAFA ÜNEL
- Web based GIS optimazation for coverage in wireless sensor network
Başlık çevirisi yok
MUSTAFA SHAKIR MAHDI AL-SALİH
Yüksek Lisans
İngilizce
2022
Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve KontrolAltınbaş ÜniversitesiBilişim Teknolojileri Ana Bilim Dalı
DR. ÖĞR. ÜYESİ OĞUZ KARAN
- Optimizasyon tabanlı bir su kaynakları sistemi modelleme aracının geliştirilmesi ve çok amaçlı sektörel tahsise uygulanması
Development of an optimization-based modeling tool for water resources systems and application to multi-objective sectoral allocation
SALİM YAYKIRAN
Doktora
Türkçe
2025
Çevre Mühendisliğiİstanbul Teknik ÜniversitesiÇevre Mühendisliği Ana Bilim Dalı
DOÇ. DR. ALPASLAN EKDAL
- Comparatıve analysıs of spatıal experıence ın physıcal and vırtual envıronments: The case of anadolu hısarı
Fiziksel ve sanal ortamlarda mekansal deneyimin karşılaştırmalı analizi: Anadolu hisarı örneği
BUKET TÜMBUL
Yüksek Lisans
İngilizce
2026
Mimarlıkİstanbul Teknik ÜniversitesiMimarlık Ana Bilim Dalı
DOÇ. DR. AHMET GÜN