Geri Dön

Tehlike analiz yöntemlerinin İstanbul, İzmir, Dinar bölgeleri için karşıkaştırılması

Başlık çevirisi mevcut değil.

  1. Tez No: 55647
  2. Yazar: VOLKAN OKUR
  3. Danışmanlar: DOÇ.DR. HÜSEYİN YILDIRIM
  4. Tez Türü: Yüksek Lisans
  5. Konular: İnşaat Mühendisliği, Civil Engineering
  6. Anahtar Kelimeler: Belirtilmemiş.
  7. Yıl: 1996
  8. Dil: Türkçe
  9. Üniversite: İstanbul Teknik Üniversitesi
  10. Enstitü: Fen Bilimleri Enstitüsü
  11. Ana Bilim Dalı: Belirtilmemiş.
  12. Bilim Dalı: Belirtilmemiş.
  13. Sayfa Sayısı: 93

Özet

ÖZET Önemli sorunlardan birisi depremlerin zaman, yer ve büyüklük gibi özelliklerin önceden saptanamamasıdır. Depremler zaman yer ve şiddet bakımından rassallık gösterdikleri için depremle ilgili paremetrelerin hesabı için istatistiksel yöntemler gereklidir. Bu çalışmada ilk olarak sismik tehlike analizinde kullanılan istatistiksel yöntemlere yer verilmiştir. Bu yöntemlerden en çok kullanılanları Gutenberg-Richter yöntemi, Gumbel yöntemi, Poisson yöntemi ve Weibull yöntemi ayrıntılı olarak açıklanmıştır. Açıklanan yöntemlerin esaslarından hareketle istanbul bölgesi, İzmir bölgesi ve Dinar bölgesi sismotektonik yöreleri için tarihsel ve aletsel dönem sismik verileri ayrı ayrı değerlendirilerek gelecekteki ihtimali depremler için sismik tehlike analizleri yapılmıştır. Gözlem dönemleri ile aşılma olasılıkları arasındaki ilişki araştırılmış, belirli bir dairesel alan ele alınmış ve bu alan küçültülerek sismik tehlike analizine etkisi araştalmışur. Son bölümde ise çeşitli araştırmacılar tarafından önerilen azalım ilişkileri kullanılarak maksimum yer ivme değerleri hesaplanmıştır. XI

Özet (Çeviri)

COMPARISON OF SEISMIC HAZARD ANALYSIS METHODS FOR REGIONS OF ISTANBUL, İZMİR AND DİNAR SUMMARY Earthquakes are one of the biggest natural disasters which have been a continuing source of destruction throughout the history of mankind. With the high seismicity (nearly %90 of the country is tectonically active) Turkey has had lots of severe earthquakes in her history. Earthquakes are differ from other natural disasters with its randomness in time, place and magnitude as well. At present man can design structures which can stand any level of dynamic or static loading. On the other hand designing of these structures brings too much cost for such safety. So the problem is what is the optimum solution?. Recently some studies about forecasting the earthquakes are being made but still there is not a convenient method as earthquakes do not occur at the same level of magnitude and frequency all over the world. In certain areas like plate borders, large earthquakes occur more frequently than other areas. Since earthquakes and their effects are random phenomena it must be considered probabilistic behavior of earthquakes. Probability analyses have been widely used in recent years by many researchers in estimation of future earthquakes. This probabilistic approach has introduced a new decision tool in engineering interest, seismic risk analysis. Seismic risk analysis involves the adverse effects of future earthquakes on society might suffer and also estimating the probability of these events in future time period (EEIR, 1989). First probability models in estimating future earthquakes were laid by Cornell, 1968. The essence of his approach was to establish the cumulative probability of exceedance of some measures from seismic sources like line sources or point sources or area sources (Skipp, 1995). In general, the goal is to identify the faults that are believed to be tectonically active and include in the analyses. If specific faults have not been identified, the common way is to model earthquakes either as a single point of energy release or as a rupture on a fault with a random location. Once source zones have been identified, seismicity parameters are used to quantify the distribution of future earthquakes in each zone (EEIR, 1989). There other models were laid by other researchers like Algermissen and Perkins. They produced a map estimating the maximum ground acceleration expected from earthquakes throughout the United States. Their technique was based on Cornell in common. Another researcher Mortgat introduced a Bayesian niodel xufor seismic hazard mapping. The program is aimed at the hazard mapping of large regions. A Poissonian model is used for the earthquake recurrence but the magnitude-frequency distribution is taken to follow a Bernoulli model rather than the standard logarithmic Gutenberg-Richter relation. The disadvantage of this program is its complex and long calculations for real engineering hazard problems. Another program was developed by WoodwardrClyde for application in active tectonic regions. In this program below a certain magnitude recurrence times are modeled as a Poissonian process, above this magnitude a Semi-Markov model is used (Skipp, 1995). The other methods that were laid in seismic hazard analysis and also carried in this study as follows: Gutenberg-Richter Process Gumbel (Theory of extremes) Poisson Distribution Analysis Weibull Distribution Analysis Richter and Gutenberg described a method for estimating the magnitudes of the future earthquakes. The method is required all the past earthquakes almost any size and any number. After listing the earthquakes occurred in a region, by using the least square regression approach a correlation is found corresponding to the equation Log N = a - bM in which N is the total number of earthquakes with a magnitude equal to or greater than M and a and b are the regression coefficients. This equation corresponds to a seismic history period of T2 years, which is called the study period. If another new study period of Ti years (design period) is considered, the relationship between the two sets of equations will obtain from Nı/N2=Tı/T2 In this equation, the assumption is that the number of earthquakes greater than a particular magnitude is proportional to the time period. It can be understood that greater the time period, larger is the number of earthquakes. Taking the logarithms and arranging the above equations we get the return period Td for any maximum magnitude as follows Log Td = Log T2 - (a - bMd) X1HAnnual probability of exceedance of magnitude Md is R=l/Td Earthquake catalogues contains two types of information. First one is historical period information that's before 1900's and in terms of Modified Mercalli scale and second one is the instrumental period information that is after 1900's (after the use of seismographs). The catalogues contain historical data of largest earthquakes and generally no data about the smaller earthquakes. Therefore more accurate results are needed if only the largest earthquakes of the past are available. Gumbel in 1958 proposed a method of yearly maxima which includes only the maximum magnitude for each year. This type of probability analysis may be useful and effective as the lackness of historical data and the deficiencies of the relevant earthquake catalogues and also provides a convenient method to obtain estimates of the frequencies of occurrence of events on the extreme of a statistical distribution and to estimate recurrence times for these events. It is assumed that there is no upper bound to the magnitude of an earthquake (Knopoff, Kagan, 1977). The Gumbel distribution function is given by as follows G(M) = exp(-aexp(-j3M)) in which M is the earthquake magnitude and a,P are the regression coefficients and G(M) is the cumulative relative frequency of occurrence of the magnitude M. The probability of exceedance however is defined by R(M) = Pr(M)=l-G(M) R(M) is the probability of occurrence with in any one year, of a particular earthquake with magnitude M or higher (Tezcan, 1994). Another model is Poisson process and based on the following assumptions. 1) An event can occur at random and at any time or any point or any point in space. 2) The occurrence of an event in a given time (or space) interval is independent of that in any other nonoverlapping intervals. 3) The probability of occurrence of an event in a small interval At is proportional to At, and can be given by vAt, where v is the mean rate of occurrence of the event (assumed to be constant); and the probability of two or more occurrences in At is negligible (of higher orders of At). XIVOn the basis of these assumptions, the number of occurrences of an event in t is given by the Poisson distribution; that is, if Xtis the number of occurrences in time interval t, then P(Xt= x) = (vt)x.exp(-vt)/x! x = 0, 1,2, where v is the mean occurrence rate that means the average number of occurrences of the event per unit time interval (Ang, Tang, 1975). The probability of exceedance can be defined as follows P(M, T) = 1 - exp(-N(M).T) where T is time period and N(M) is a function value related to any magnitude and can be determined from the magnitude-frequency relationship. Weibull distribution analysis was applied by Hagiwara (1974), to seismic hazard analysis, and based on the assumption that if crustal strain increases linearly with time, so it is reasonable to think that a probability law governs the crustal rupture time. Earthquake occurrence can be regarded as a probabilistic phenomenon which can be treated by statistics of ultimate strain of the earth's crust. Weibull density function is defined as follows, fT(t) = n.y.trl.exp(-\x.ty) and the distribution function is, FT(t) = 1- expC-ji.f) where u, and y are regression coefficients. By using u. and y values, the probability of exceedance can be defined as follows, P = 1- exp(-u..f) In this study the models named above are applied respectively for forecasting earthquake magnitudes of the next 50 years, by using the past N years' data in selected regions, in Turkey. Istanbul, Izmir and Dinar seismotectonic regions are XVselected as they are the most tectonically active regions. In the past these regions were shaked by some severe earthquakes. The historical data goes back nearly 0 A.D. for three regions but the lackness of the data set the study of the historical period between 1800-1900 for these regions. The instrumental data period varies between 1901-1994 for Istanbul, 1900-1986 for Izmir and 1914-1994 for Dinar. For a design period 50 and 100 years maximum magnitudes for each regions were computed using the methods named above. For the province of Istanbul the maximum magnitude in 50 and 100 years are Mmax=6.6 and Mmax=7.1 respectively. For the province of Izmir the maximum magnitude in 50 and 100 years are Mmax =7.0 and Mmax=7.4 respectively. And finally for the province of Dinar the maximum magnitude in 50 and 100 years are Mmax=6.8 and Mmax=7.3 respectively. All of these magnitudes are the maximums chosen from the methods named above. The study period including the instrumental earthquakes were decreased in order of ten years and a relationship was investigated between the study period and the probability of exceedance. In the province of Istanbul when the study period were decreased probability of exceedances were increased. But in the regions of Izmir and Dinar probability of exceedances were decreased. In author's opinion it is the result of using fore shock and after shock datas for the regions of Izmir and Dinar as the number of these datas were so much and effected the analysis. Finally the effect of a distance from an origin point to probability of exceedance was investigated. From an origin point an area was chosen with a radius of R By decreasing R so the area the seismic hazard analysis models were applied for the regions named above. For the three regions from an origin point with radiuses of 150, 125, 100, 75 km. circles were chosen. As the radiuses were decreased probability of exceedances were decreased too. Attenuation relationships given by Ambrasseys and Bommer (1992) and Ambrasseys (1995) were applied for these three regions by taking the yearly probability of exceedances %15 for normal buildings, %0.5 for important buildings and %0.05 for very important buildings as well. As the probability of occurrences were nearly the same for the three regions, peak ground accelerations were found nearly same too. XVI

Benzer Tezler

  1. Tez No

    505163

    Türkiye'deki çelik demiryolu köprülerinin deprem etkisi altında olasılıksal yöntemler yardımı ile incelenmesi

    Investigation of seismic behaviour of steel railway bridges in Turkey with probabilistic aproaches

    MEHMET FATİH YILMAZ

    Doktora

    Türkçe

    Türkçe

    2018

    Deprem Mühendisliğiİstanbul Teknik Üniversitesi

    İnşaat Mühendisliği Ana Bilim Dalı

    DR. ÖĞR. ÜYESİ BARLAS ÖZDEN ÇAĞLAYAN

  2. Tez No

    75454

    Testing and evaluation of Konaklar Railway Bridge

    Konaklar Demiryolu Köprüsünün değerlendirmesi ve rafine bir bilgisayar modelinin hazırlanması

    CANES AVCI

    Yüksek Lisans

    İngilizce

    İngilizce

    1998

    İnşaat Mühendisliğiİstanbul Teknik Üniversitesi

    İnşaat Mühendisliği Ana Bilim Dalı

    PROF. DR. ERDOĞAN UZGİDER

  3. Tez No

    356077

    Coğrafi bilgi sistemleri (CBS) ve analitik hiyerarşi yöntemi (AHY) ile üretilen deprem tehlike haritalarının duyarlılık analizi

    Sensitivity analysis of earthquake hazard maps produced by geographic information systems (GİS) and analytic hierarchy process (AHP)

    MUSTAFA ALİ GÖKKAYA

    Yüksek Lisans

    Türkçe

    Türkçe

    2014

    Jeodezi ve Fotogrametriİstanbul Teknik Üniversitesi

    Geomatik Mühendisliği Ana Bilim Dalı

    DOÇ. DR. TURAN ERDEN

  4. Tez No

    232542

    Yaygın yöntemlerle yapılan yapısal analiz sonuçlarının kademeli yükleme etkisi gözönüne alınarak yapılan analiz sonuçlarıyla karşılaştırılması

    The comparision of the construction analyses results of the conventional methods to the results of analyses considering sequential loading effect

    YUSUF ZİYA ALP

    Yüksek Lisans

    Türkçe

    Türkçe

    2007

    İnşaat Mühendisliğiİstanbul Üniversitesi

    İnşaat Mühendisliği Ana Bilim Dalı

    PROF. DR. NAMIK KEMAL ÖZTORUN

  5. Tez No

    627700

    İş sağlığı ve güvenliği risk değerlendirme süreci için yeni bir yaklaşım: Tersane işletmelerinde uygulama

    New approach to risk asssessment process in health and safety at work: Practice in shipyard enterprises

    GÜFTE CANER AKIN

    Doktora

    Türkçe

    Türkçe

    2020

    Gemi Mühendisliğiİstanbul Aydın Üniversitesi

    İş Sağlığı ve Güvenliği Ana Bilim Dalı

    PROF. DR. HASAN ALPAY HEPERKAN

Geri Dön