Conducting seismic risk analysis by Monte Carlo simulation method without using reduction relations

Number of pages: 124 File Format: Not Specified File Code: 29396
Year: Not Specified University Degree: Not Specified Category: Civil Engineering
Tags/Keywords: earthquake - Geophysics
  • Part of the Content
  • Contents & Resources
  • Summary of Conducting seismic risk analysis by Monte Carlo simulation method without using reduction relations

    Master's Thesis

    Geophysics-Seismology

    1392

    Abstract

    Estimating earthquake risk to calculate a certain level of damage and vulnerability in the design of structures is one of the most important topics and principles of earthquake engineering, but there are uncertainties in determining the characteristic parameters of earthquakes, attenuation relationships, predicting the time of future earthquakes, etc. The distance of the spring to the site of earthquakes and other things has made it complicated and new methods have been presented.

    Two deterministic and probabilistic methods for estimating seismic risk have been presented, the first method is based on one or more specific scenarios that usually show the worst result. And it does not give any information about the probability of the future earthquake, its location and time frame. The second method, which is the fundamental probabilistic model in earthquake risk analysis, consists of possible combinations of size and distance to the building, which causes the ground motion parameter (a) to increase from values ??(a0). The goal is to obtain the annual probability P( ). This earthquake risk analysis method was presented by Cornell (1960).   Although this method is very useful in risk analysis in earthquake engineering, it has disadvantages, such as that in this method we have to define the geometry of the seismic zone and determine the seismic parameters, which are accompanied by many uncertainties.

    In this thesis, a new probabilistic method called the Monte Carlo simulation method is used to estimate the seismic risk. This method uses multiple random sampling of an earthquake catalog to build a synthetic catalog. In this method,  It is possible to calculate the values ??of strong ground motion (PGA, PSA) using reduction relations and without using reduction relations and direct use of accelerometers recorded in the study area.   Since in this simulation method, there is no need to determine the geometry of the seismic springs and the characteristic parameters of the earthquake, the uncertainties related to these two are controlled. The uncertainty caused by the use of reduction relations is related to the inherent dispersion of data and also the difference in the functional form of different relations. Using the Monte Carlo simulation method, it is possible to use the real probability distribution of the data based on the observed data. Finally, the results of the conventional probabilistic method and the simulation method are compared either by using the attenuation relations directly or by using the scaled observed data. Keywords: seismic risk estimation, attenuation relations, uncertainty, Monte simulation. Carlo

    Introduction

    Many construction projects have been carried out during the past decades, and many investments have been made for the development of infrastructure industries. Many of these construction projects, including power plants, dams, oil and petrochemical facilities, are very sensitive and expensive.

      Iran's plateau is affected by active tectonic forces and is known as an earthquake-prone region of the world. The occurrence of destructive earthquakes such as the peak of 2002 with a magnitude of 4.6  Bam 2003 with a size of 3/6, Kajor 2004 with a size of 3/6 and Silakhor 2006 with a size of 1/6  The proof is for this claim.

    In the design of civil structures, one of the most effective forces is the force caused by the earthquake, and in many cases, due to the special characteristics of such structures, the force of the earthquake is the governing force in the design. Therefore, the demand for seismic risk analysis has increased significantly. Seismic risk analysis  It provides the occurrence rate of strong earth movement in the future and plays an important role in reducing the seismic risk.

    However, the analysis of seismic risk by conventional methods is probabilistic with many uncertainties, which has a great impact on the level of seismic design of structures. One of the biggest sources of uncertainty is the use of reduction relationships. The uncertainty caused by the use of reduction relations is related to the inherent dispersion of data and also the difference in the functional form of different relations.

    According to these explanations, the purpose of this research is to provide an efficient and engineering method to determine the seismic risk in which the uncertainties in the calculations are reduced.In this thesis, the Monte Carlo simulation method is used to determine the seismic risk, and the methods, approaches, problems and review of the technical literature in this field are discussed. Using the Monte Carlo simulation method, it is possible to use the real probabilistic distribution of data based on the observed data. Chapter One

     

     

     

    Overview of conventional risk analysis methods

      and Monte Carlo simulation method-1-Introduction

    Estimation of seismic risk to calculate the level Determining the damage and vulnerability in the design of structures is one of the most important topics and principles of earthquake engineering, but there are uncertainties in determining the characteristic parameters of an earthquake, reduction relationships, predicting the time of future earthquakes, mainly,  The distance from the source to the site of earthquakes and other cases has caused its complexity and the presentation of new methods.

    The purpose of seismic hazard estimation is to rationally evaluate the parameters of the powerful earth movement (maximum acceleration, maximum speed, intensity, etc.) in a certain period of time, which is usually the useful life of the structure. The economic consequences resulting from it are of particular importance and also the results of risk analysis can be important in government policy making and identification of high risk areas in Iran.

    At the beginning of this chapter, a review of the conventional methods of seismic risk analysis is given, in the rest of the chapter, the uncertainties [1] in the conventional methods are introduced and how to deal with them, and at the end of the chapter, a review of the new Monte Carlo simulation method is given.

     

    1-2-Overview of conventional seismic risk analysis methods

    Earthquake risk is a property of earthquake that causes damage or damage. Currently, there are two general methods for estimating seismic risk in an area. In the first method, which is known as deterministic method [2], by using reduction relations, earthquakes and deterministic characteristics of the desired area, such as the distance to the fault, etc., to determine the maximum acceleration of the site  becomes This method usually provides worst-case results and reports results deterministically without considering probabilities. The second method of seismic risk analysis is the probabilistic method [3], in which the seismic risk of a specific location is determined using the theory of total probability [4]. In this section, the determination method will be explained first, and then the principles of the earthquake risk analysis method will be examined. 1-2-1-Overview of the determination method in seismic hazard estimation The determination method is the first and simplest method used in earthquake risk studies. This is a very simple method to estimate the maximum believable acceleration for a site. which is not dependent on time unlike the probabilistic method. A certain method is very necessary for sensitive sites. The different stages of conducting a risk analysis in a deterministic way are:

    Identifying seismic sources related to the investigated location

    Determining the required parameters in the reduction relationship for each seismic source

    Choosing the appropriate reduction relationship[5]

    Determining the maximum acceleration rate of the desired design or the desired design range[6]

    This process is shown in Figure (1-1). In the following, each of these steps has been examined separately.

    -2-1-1-Identification of seismic springs

    One of the most important parts of seismic hazard analysis in each area is the identification of seismic springs. This step is common in both deterministic and probabilistic methods, in other words, the determination of seismic sources is the first step of conducting mandatory studies.

    In determining seismic sources, it is necessary to use all geological, seismic, and geophysical information. An important point in determining seismic springs is the uncertainty in determining the geometry of these springs. In some cases, by changing the geometry of the seismic source, the results of the analysis will have 100% changes.

     

    1-2-1-2-Determining the required parameters in the reduction relationship for each seismic source

    In the determination method, depending on the reduction relationship used, it is necessary to specify the parameters of the seismic source, the parameters of the movement path and the parameters of the desired building conditions. The parameters of seismic springs usually include the seismic magnitude and the distance between the source and the site according to the reduction ratio used.

  • Contents & References of Conducting seismic risk analysis by Monte Carlo simulation method without using reduction relations

    Abstract A

    Introduction B

    Chapter 1 Overview of conventional methods of risk analysis and Monte Carlo simulation method 1

    1-1-Introduction 1

    1-2-Overview of conventional methods of seismic risk analysis 1

    1-2-1-Overview of the determination method in seismic risk estimation 2

    1-2-1-1-Identification of seismic springs 3

    1-2-1-2-Determining the required parameters in the reduction relationship for each seismic source 3

    1-2-1-3-Choosing the appropriate reduction relationship 4

    1-2-1-4-Determining the maximum acceleration of the design or the desired design range 4

    1-2-2-Overview of the possible method in seismic risk estimation 4

    1-2-2-1-Identification of the seismic sources of the site 5

    1-2-2-2-Obtaining the distance distribution relationship for each source,

    Determining the return relationship and determining seismicity parameters 7

    A- Distance distribution for each source 7

    B-Determining the earthquake return relationship for each source 9

    C- Determining seismicity parameters in earthquake return models 14

    1-2-2-3-Choosing the reduction relationship 19

    1-2-2-4-Determining the risk curve 20

    1-2-3- Introducing different uncertainties in seismic risk analysis and how to deal with it 21

    1-2-4-Logic tree method 22

    1-3-Overview of Monte Carlo simulation method 23

    1-3-1-The nature of Monte Carlo method 24

    1-3-2-Application of Monte Carlo in seismic risk analysis 25

    1-3-3-Risk determination theory Seismic and seismic risk curve determination methods 26

    1-3-4-Monte Carlo algorithms 28

    1-3-4-1-Monte Carlo algorithm Abel and Kafka (1999) 28

    A-Interpolation of the observed values ??of strong ground motion instead of

    direct use of reduction relations 31

    1-3-4-2- Monte Carlo Algorithm Han and Choi (2007) 34

    1-3-5-Advantages and disadvantages of Monte Carlo risk analysis method with other risk analysis methods 39

    Chapter 2 reduction relationships and their uncertainty 42

    2-1-Introduction 42

    2-2-Methods for calculating strong earth movement 42

    2-2-1-Random or semi-theoretical method 42

    2-2-2-Experimental method 43

    2-3-Uncertainty of reduction relationships 46

    2-3-1-Examination of parameters affecting reduction relationships 46

    2-3-1-1-Parameters affecting attenuation relationships caused by strong earth movement 47

    2-3-1-2-Uncertainty caused by the regression model and method to determine

    coefficients of attenuation relationships 47

    Chapter 3, investigation of attenuation relationships in Iran using LH and LLH statistical tests 50

    3-1- Introduction 50

    3-2-Introduction of statistical tests for ranking and selection of reduction relationships 50

    3-2-1- Test of residuals distribution 51

    3-2-1-1- Test of variance and tendency to the center (median and mean) 52

    3-2-2- Test of the set of observed data values ??52

    3-2-2-1-LH test 52

    3-2-2-2-LLH test 54

    A - Introduction of method 54

    B - Description of method 54

    3-3-Introduction and examination of reduction relations 56

    3-4- Catalog of strong ground motion related to the double Ahar-Varzghan earthquake 61

    3-5-LH and LLH 67 statistical test results

    3-5-1-LH 67 test

    3-5-2-LLH 69 test

    Chapter 4 of Tabriz city seismic data collection and Iran accelerometer catalog 71

    4-1-Introduction 71

    4-2- Seismic characteristics of the area and preparation of catalog 71

    4-2-1- Determination of seismicity parameters 72

    4-2-1-1- Determination of coefficients and 73

    4-3- Catalog of observed accelerograms 75

    Chapter 5, possible risk analysis using Monte Carlo method from available accelerometer data

    and comparing the results of conventional method and Monte Carlo simulation 81

    5-1-Introduction 81

    5-2-Case study of advantages and disadvantages of Monte Carlo method with conventional methods of risk analysis 81

    5-2-1-No need to define the geometry of the seismic zone 81

    5-2-2-No The need to determine seismic parameters 84

    5-2-3- No need to choose a return relationship 84

    5-2-4- No need for distance distribution function 85

    5-2-5- No need for a reduction relationship 86

    5-3- Comparison of possible risk analysis for different introduced relationships 86

    5-3-1-Comparison of the risk curve of all three risk analysis methods for the Akar-Boomer reduction relationship 86

    5-3-2- Comparison of the risk curve of all three risk analysis methods for the Ambrasis reduction relationship 88

    5-3-3- Comparison of the risk curve of all three risk analysis methods for the Campbell-Bourgnia reduction relationship 89

    5-3-4- Comparison of the risk curve of all three risk analysis methods For Ghasemi-Sinaiyan mitigation relationship 90 5-3-5- Comparison of the risk curve of all three risk analysis methods for the saffron mitigation relationship 91 5-3-6- Comparison of the risk curve using the conventional method of possible risk analysis for different mitigation relationships 93 5-3-7- Comparison of the risk curve using the Monte Carlo simulation method using mitigation relationships 94 5-3-8- Comparison of the risk curve by the Monte Carlo simulation method using existing accelerometer data 95 5-3-9- Comparison of the risk curve of all three risk analysis methods using the weight of reduction relationships 96 References and sources list 101 Appendix 103 English abstract 106

Conducting seismic risk analysis by Monte Carlo simulation method without using reduction relations