15870

14170

كارشناسي ارشد

مهندسي و مديريت منابع آب

اصفهان : دانشگاه صنعتي اصفهان

1399

داوزاده، 74ص.‌ :مصور، جدول، نمودار

حميد رضا صفوي، محمد حسين گل محمدي

آزاده احمدي، رامتين معيني

1399/07/27

كتابنامه

عمران

مهندسي عمران

1399/07/27

2641422

7 چكيده فصل اول مقدمه 1 1 مقدمه وكليات 2 1 2 اهداف تحقيق 3 1 3 نوآوري تحقيق 4 1 4 روش و مراحل تحقيق 4 1 5 فرضيات و محدوديت 5 1 6 محتواي فصول پايان نامه 6 فصل دوم پيشينه پژوهش 2 1 مقدمه 7 2 2 پديده سيالب و تحليل آن 8 2 3 تجزيه و تحليل فرواني سيالب چند متغيره 8 2 4 تحليل فراواني سيالب با استفاده از توابع مفصل دو متغيره 01 2 4 1 تحليل فراواني وقايع حدي هيدرولوژيكي با استفاده از توابع مفصل ارشميدسي 01 2 4 2 تحليل فراواني سيالب با استفاده از توابع بيضوي 41 2 5 تحليل فراواني سيالب با استفاده از توابع مفصل چند متغيره 51 2 5 1 تحليل فراواني سيالب با استفاده از توابع مفصل ارشميدسي 51 2 5 2 تحليل فراواني سيالب با استفاده از توابع مفصل بيضوي 61 2 6 جمع بندي 61 فصل سوم مواد و روشها 3 1 مقدمه 81 3 2 استخراج پارامترهاي دبي اوج حجم و مدت يك واقعه سيل 91 3 3 مشخصههاي توزيعهاي آماري 02 3 4 تعيين ضرايب همبستگي متغيرها 12 3 4 1 ضريب كندال تاو 12 3 4 2 ضريب اسپيرمن 22 3 4 3 ضريب همبستگي پيرسون 22 r هفت

Trivariate Flood Frequency Analysis Using the Copula Functions Mohamad haytham klaho Eng h klaho@cv iut ac ir Civil Engineering Department Isfahan University of Technology Isfahan 84156 83111 IranDegree MSc Language FarsiSupervisors DR Hamid R Safavi Dr Mohammad Hossein GolmohammadiAbstractThroughout history severe floods have caused great human and financial losses in various countriesaround the world The occurrence of devastating floods has intensified in recent decades due toclimate change as well as human intervention in catchments and urban development Floods areoften caused by a combination of meteorological physiographic and human interventions Therefore the study and analysis of flood frequency in hydrology and water resources managementis one of particular importance In this context univariate frequency analysis of flood peak is notsufficient to predict future floods based on classical distribution functions Therefore multivariateanalysis of flood phenomenon including flood peak volume and duration of flood is of interest tohydrologists today For multivariate analysis of floods using classical possible functions normal Gamble etc there is a serious reason for assuming the same functions for all variables and theexistence of correlations between variables also limits the application of possible classicalmultivariate functions in flood analysis Therefore due to these limitations in the application ofclassical multivariate probabilistic functions the use of copula functions is recommended Althoughbivariate analysis of floods is also of particular importance more comprehensive analyzes aremainly performed in the form of three variables In this research the analysis and modeling of thedependence structure of two and three variable flood limit values using copula functions ofArchimedean and Elliptical families has been considered Copula functions of Archimedean familyincluding Clayton Frank Gamble Ali Mikhail Haq and Joe functions and the Elliptical familyincluded Gaussian and T student functions For this purpose 55 years of annual flood data recordedat the Dez Dam hydrometric station located on the Karun River were used The results showed thatbased on goodness of fit criteria and Q Q plot graph Gaussian Elliptic copula function of twovariables between flood peak and flood volume with parameter 0 32 and between peak dischargeand flood continuation time with parameter 0 35 and Archimedean copula function of Claytonbetween Flood volume and continuity with parameter 1 76 are selected as the best copula functions According to other results of this research in order to select the appropriate copula functions with abivariate coupling and then connect it to the third variable as C2 C1 Q V F D 2 the Ali Mikhail Haq copula function with parameter 0 43 And for the state C2 C1 Q D F V 2 andalso C2 C1 V D F Q 2 the Gaussian copula function with parameters of 0 87 and 0 02 respectively is the most suitable functions Simultaneous analysis of trivariate was performed andshowed that according to the criteria of goodness of fit including maximum likelihood Akaike Root Mean Square Error and Nash Sutcliffe the values of these criteria were 165 58 33 12 0 365 0 916 respectively the Clayton copula with a parameter of 0 31 has the best performance in thiscase After selecting the best joint function the common and conditional cumulative distributionfunctions of two and three variables were analyzed according to their importance in the design andmanagement of hydraulic structures as well as the periodic and periodic return or and conditional two and three variable floods that play an important role in flood risk analysis werecalculated and it was shown that calculating the two variable conditional return period for short term periods is more reliable than the three variable conditional return period Also the three variable conditional return period for long term periods is more reliable than the two variableconditional return period Keywords Floo

حميد رضا صفوي، محمد حسين گل محمدي

آزاده احمدي، رامتين معيني