##
Toprak dolgu barajlara ve istinat duvarlarına gelen deprem etkisi

Toprak dolgu barajlara ve istinat duvarlarına gelen deprem etkisi

##### Dosyalar

##### Tarih

1993

##### Yazarlar

Damar, Nice

##### Süreli Yayın başlığı

##### Süreli Yayın ISSN

##### Cilt Başlığı

##### Yayınevi

Fen Bilimleri Enstitüsü

##### Özet

İstinat duvarları ve toprak dolgu barajlar depremler s ırasında ve sonrasında çeşitli etkilere maruz kalırlar. Deprem bu yapılarda kaymalar, çatlaklar, gerilme yığılmaları seklinde olumsuz sonuçlar doğurur. Tez esas olarak 2 ana bölümden oluşmuştur. İlk bölümde toprak dolgu barajlara gelen deprem etkisi ele alınmıştır. Depremden barajın etkilenmesine sebep olan parametreler teker teker ele alınıp, açıklanmıştır. Sonra depreme karsı yapılan tahkikler sıralanmıştır. Tezin ikinci bölümünde istinat duvarlarına gelen deprem yükü ele alınmıştır. Ankrajlı ve ağırlık istinat duvarları ayrı ayrı ele alınmıştır, örnek bölümünde de toprak dolgu barajlara ve ağırlık istinat duvarlarına gelen deprem yükü birer örnekle sunulmuştur. İnşaat mühendisliğinde son yıllardaki teknik gelişmeler, özellikle bilgisayarların gelişimi bu tip problemleri sonlu eleman metodu ile çözümünü sağlamıştır.

Several factors a-Ffect behaviours of an earthfill dam during earthquakes. These are hydrodynamic effects, damping of an earthquake motion, geometry of valleys and dams, compaction type, strenght and material properties of embankment dam, acceleration of an earthquake, period of an eartquake, duration of an eartquake and magnitude of an earthquake. If slopes of an earthfill dam are not perpendicular to water surface, the hydrodynamic effect is negligible. According to laboratory tests and field tests, parameters effecting shear modulus and damping factors of a dam are strain amplitude, confining pressure, void ratio and number of cycles of a loading. If cohesive soils are not considered, parameter effecting shear modulus and damping ratio is degree of saturation. Geometry of valleys and dams is an important factor that is affecting earthquake induced behaviour of a dam. By considering the ratio og height to lenght of a dam, we can determine whether an earthquake motion is two dimensional or three dimensional. In these models amplitude, period and frequency differ according to dimension of a model. Compaction type is also a vital factor. Rolled embankments are more resistant to earthquake motion when compared with hydraulic fill dams. Materials of a dam can be clay, mixed clay or sand. In earhquake conditions clayey soils are preferred. Foundation is the most vulnerable zone of an earhfill dam and we must be very cautious. Bedrock is preferred in foundation section. Clayey embankments resting on bedrock foundations resist earthquakes with a magnitude in the order of 8.25 and accelerations in the order between 0.35g and O.SOg. Earthfill dams generally resist moderate earthquake motions. Period of any eartquake is not confined with duration of an eartquake, but it continues afterwards, damages occur afterwards. Magnitude and duration of an earthquake is directly proportional. If one increases, the IX other also increases. As results of an earthquake, in earth-fill dams failures. slides, settlements, cracks and deformations occur. Also liquefaction is a result of an earthquake. Because of vibrations and increase in pore pressures decreases in strenght of soil accrue. At the latest point, failure occurs. If earthquake induced shear stress exceed an embankment's or a foundation's shear strenght, serious slides take place. Permanent deformations occur. Newmark (1957) has defined failure. Failure occurs in a well defined surface. This is simulated by sliding block model. Material behaves elastically before failure, and behaves plastically after failure. In loose sands and silts, pore pressure accrues during earthquake and liquefaction occurs. Liquefaction is related with soil properties, enviromental factors and earthauake's properties. If cohensionless soil is saturated, liquefaction mostly take place. Pore pressure increases and strenght of material declines. Earthquake induced deformations are closely related with earthquake's amplitude, accelaration and frequency content. Duration of a vibration is also vital. The most important results of earthquake are settlements and cracks. Settlements usually occur at crest and cracks usually occur at body of an eartfill dam. Horizantal vibrations induce tension stress and they cause longitudinal cracks. Settlements also cause cracks. Settlements and cracks are interrelated. In order to determine situation of a dam after an earthquake several analysis methods are developed. Tere are five methods. Classical stability method (slice method), advanced stability method, total stress and effective stress method, shear deformation theory and finally finite element method. Slice method is the most widespread method. Slope is divided into slices on a sliding arc and sismic coefficients are determined, finally stability is checked. This method considers only sliding during earthquakes and because of this X- deficiency it is not used vitally in recent years. Acceleration is considered constant throughout a slice. In recent years -finite element method is widespread. Finite element method divides a dam into small members which are connected at nodal points and stability is analized according to all members. In this method there are deficiencies at boundary conditions, but it is the most reliable method. In the second part of my thesis retaining walls are described. Retaining walls are divided into two parts as anchored retaining walls and gravity retaining walls. During earthquake in anchored retaining walls licşjıefaction and ruptures take place. 'Sismic stability analysis is developed by Mononobe-Okabe (1955). Sismic stability of anchored retaining walls is related with embedment depth. This depth must be increased. - Gravity retaining walls are dimensioned according to displacements, total failure analysis or overturning analysis is considered. Anchored retaining walls are analized according to -Forces developing in the system. Analysis are carried out in order to determine whether wire will rupture or not. During earthquakes in gravity retaining walls inertial forces are very important, but in anchored retaining walls, they are trivial. Achored retaining walls are flexible and gravity retaining walls are rigid. In bulkheads, sheet piles and quay walls, great perroanant deformations occur as a result of strong seismic force. The stiffness of the anchorage, its slope and spacing distances are important parametres. The flexibility of the wall and the number of anchorages also effect seismic reaction. During earthquakes the anchored wall and backfill move together, but the soil above and below the wall move seperately. The model is nonlineer. The soil is elastoplastic, homogen and isotropic hardening material. The greatest bending occur at the middle of the span. Anchorage is modelled as a spring in soil.. The design of sheet pile walls, anchored bulkheads and siffiilar anchored retaining wall Is, is made according to active and passive soil pressures. The cause of results induced by an earthquake is estimated as liquefaction of the soil and anchorage rupture. Anchorages are generally the XI main cause of damage during eartquakes. In order to prevent the removal of the anchorage -from the active zone the anchorage rod must be made long. The magnitude and distribution of the horizantal soil pressure is effected by the properties of the backfill, the bending of the retaining structure with respect to soil and the properties of the soil between the structure and the soil. Even with respect to static loads soil and earth interaction is very complicated. Classical soil pressure theory is prepared by Coulomb. He has 2 main assumptions: (1) The soil is unflexible and plastic; and (2) active and passive wedges slide on plane rupture surfaces. During earthquake retaining walls are subject to additional inertial forces. The solution of this problem according to Coulomb is made by Mononobe, Okabe and Matsua. According to them, during earthquakes, the active soil pressure increases and passive soil pressure decreases. With the increase in the magnitude of an earthquake tepric sliding plane approaches to horizantal state. As a result of this, the factor of safety decreases. If the wall moves outwards, it moves with a constant velocity. There are three kinds of displacement. These are: CI) only sliding or horizantal displacement; (2) the rotation of the top; (3) the rotation of the bottom edge. The first state necessitates the existence of moment and for this reason it is unprobable. Second state happens when the anchorage is situated outside of the active sliding zone. According to field observations the third state happens when the anchorage is swept from the active zone or when it yields prematurely. Field observations has proved that the third state is point of interest with sliding under static design conditions. The most desired yield state is the rotation of the anchorage. In this state, the anchorage is outside of active sliding zone and during seismic loading it provides neutralizing force. If there are inertial forces exceeding the pressure thrust in the soil, during rotation failure occurs. Horizantal active thrust is affected by the dynamic coefficient of active soil pressure, the unit weight of the soil, the height of the active part, internal friction angle of the active side and the friction angle between the soil XII and the wall at the active side. Horizantal passive thrust is affected by the dynamic coefficient of the passive soil pressure. With the increase in ground accelaration of the active zone the dynamic coefficient of the active soil pressure cfecreases and the dynamic coefficient of passive soil pressure increases. When the factor of safety of the retaining wall appproaches 1, failure occurs. When the accelaration angle is greater than internal friction angle the equations can not be solved. Soil acts like a viscous liquid. The factor of safety with respect to failure is the ratio of moments.that are preventing the overturning of the structure and that are overturning the structure. During earthquake additional forces accrue. These are Ğ/nasnic increases. These additional forces affect the overturning moment and the other moments. The stability of the anchored retaining wall greatly depend on the passive resistance of the soil and the embedment depth. At the starting moment of the movement the centre of the passive pressure is 1/3D below the ground level. In a frictionless medium it is 0.2D below the ground level. Under static loads the increase in the embedment depth cause the increase of the factor of safety. On the other hand, some observers claim that increasing the embedment depth has no effect on the seismic stability. The displacements in anchored retaining walls with the increase of embedment depth has no priority over the retaining walls &«hich hasn't increased embedment depth. There is a claim that increasing the embedment depth has no influence on the displacements above the ground level. In order to ensure equilibrium observers prefer to increase the depth of the pressure wedge or decrease the passive res i s tence. The -Factor of safety of the passive ground pressure coefficient raust be 2. In this instance there is a plane sliding surface. In a dense cohesionless fill, at the beginning of the event, with vibration stiffness is ensured. Then step by step sliding plane occur. When sliding surfaces are completed, ther is a decline in the force. When the sliding surfaces are completed, the internal friction angle is not at its maximum value, but at its residual value. The embedment depth of the anchorage is found according to the equilibrium of the moments. This depth must be increased 20 per cent. XIII Increasing number of anchorages in an anchored retaining wall has positive effect on sismic stability. Increasing nwber of anchorages in an anchored retaining wall decreases displacement at deep end of a wall, whereas displacement at top of a wall increases. In anchored retaining wall displacements increase proportionally with accelerations. If soil doesn't liquify a one anchored retaining wall suffers a small and a moderate earthquake without serious damage. Ffeximum amplitude of a wall occurs at an unsupported portion of a wall. If amplitude increases, acceleration also icrease. Stiffer anchorages carry more load and suffer less displacement. The more inclination an anchorage has, the mars load it carries. But stiffness of an anchorage is effective only in a restricted area, only in vicinity of an anchorage. Stiffer anchorages tolerate bigger accelerations. Iclination of an anchorage is effective only between anchorage levels. Gravity retaining walls slide onwards or bend during eartquakes. At one end of a foundation, if bearing capacity declines, settlements occur. Caution must be taken in order to decrease back pressure. Cement grouting is one solution. If a foundation is sandy, one must also be cautious. Liquefaction or e decline in bearing capacity takes place. We use several national standards in computing sismic stability of gravity retaining walls. The most widespread procedure is Mononobe-Okabe (1955) formula. In this formula backfill is considered as dry, cohensionless and homogenous. Internal friction angle is assumed to be constant. Sliding line is considered flat. During eartquakes, base acceleration and inertial forces save opposite sides. If one is inwards, other is outwards. In Richards and Elms (1992) model, a procedure that is taking amplitude into account is used.' Computation of sismic behaviour of reinforced soil gravity retaining wall is important. It is more economical when compared with others. Sismic stability of reinforced soil gravity retaining wall is analized by 2 procedures. In my thesis, also Turkish, Japanese and Indian standards are introduced.

Several factors a-Ffect behaviours of an earthfill dam during earthquakes. These are hydrodynamic effects, damping of an earthquake motion, geometry of valleys and dams, compaction type, strenght and material properties of embankment dam, acceleration of an earthquake, period of an eartquake, duration of an eartquake and magnitude of an earthquake. If slopes of an earthfill dam are not perpendicular to water surface, the hydrodynamic effect is negligible. According to laboratory tests and field tests, parameters effecting shear modulus and damping factors of a dam are strain amplitude, confining pressure, void ratio and number of cycles of a loading. If cohesive soils are not considered, parameter effecting shear modulus and damping ratio is degree of saturation. Geometry of valleys and dams is an important factor that is affecting earthquake induced behaviour of a dam. By considering the ratio og height to lenght of a dam, we can determine whether an earthquake motion is two dimensional or three dimensional. In these models amplitude, period and frequency differ according to dimension of a model. Compaction type is also a vital factor. Rolled embankments are more resistant to earthquake motion when compared with hydraulic fill dams. Materials of a dam can be clay, mixed clay or sand. In earhquake conditions clayey soils are preferred. Foundation is the most vulnerable zone of an earhfill dam and we must be very cautious. Bedrock is preferred in foundation section. Clayey embankments resting on bedrock foundations resist earthquakes with a magnitude in the order of 8.25 and accelerations in the order between 0.35g and O.SOg. Earthfill dams generally resist moderate earthquake motions. Period of any eartquake is not confined with duration of an eartquake, but it continues afterwards, damages occur afterwards. Magnitude and duration of an earthquake is directly proportional. If one increases, the IX other also increases. As results of an earthquake, in earth-fill dams failures. slides, settlements, cracks and deformations occur. Also liquefaction is a result of an earthquake. Because of vibrations and increase in pore pressures decreases in strenght of soil accrue. At the latest point, failure occurs. If earthquake induced shear stress exceed an embankment's or a foundation's shear strenght, serious slides take place. Permanent deformations occur. Newmark (1957) has defined failure. Failure occurs in a well defined surface. This is simulated by sliding block model. Material behaves elastically before failure, and behaves plastically after failure. In loose sands and silts, pore pressure accrues during earthquake and liquefaction occurs. Liquefaction is related with soil properties, enviromental factors and earthauake's properties. If cohensionless soil is saturated, liquefaction mostly take place. Pore pressure increases and strenght of material declines. Earthquake induced deformations are closely related with earthquake's amplitude, accelaration and frequency content. Duration of a vibration is also vital. The most important results of earthquake are settlements and cracks. Settlements usually occur at crest and cracks usually occur at body of an eartfill dam. Horizantal vibrations induce tension stress and they cause longitudinal cracks. Settlements also cause cracks. Settlements and cracks are interrelated. In order to determine situation of a dam after an earthquake several analysis methods are developed. Tere are five methods. Classical stability method (slice method), advanced stability method, total stress and effective stress method, shear deformation theory and finally finite element method. Slice method is the most widespread method. Slope is divided into slices on a sliding arc and sismic coefficients are determined, finally stability is checked. This method considers only sliding during earthquakes and because of this X- deficiency it is not used vitally in recent years. Acceleration is considered constant throughout a slice. In recent years -finite element method is widespread. Finite element method divides a dam into small members which are connected at nodal points and stability is analized according to all members. In this method there are deficiencies at boundary conditions, but it is the most reliable method. In the second part of my thesis retaining walls are described. Retaining walls are divided into two parts as anchored retaining walls and gravity retaining walls. During earthquake in anchored retaining walls licşjıefaction and ruptures take place. 'Sismic stability analysis is developed by Mononobe-Okabe (1955). Sismic stability of anchored retaining walls is related with embedment depth. This depth must be increased. - Gravity retaining walls are dimensioned according to displacements, total failure analysis or overturning analysis is considered. Anchored retaining walls are analized according to -Forces developing in the system. Analysis are carried out in order to determine whether wire will rupture or not. During earthquakes in gravity retaining walls inertial forces are very important, but in anchored retaining walls, they are trivial. Achored retaining walls are flexible and gravity retaining walls are rigid. In bulkheads, sheet piles and quay walls, great perroanant deformations occur as a result of strong seismic force. The stiffness of the anchorage, its slope and spacing distances are important parametres. The flexibility of the wall and the number of anchorages also effect seismic reaction. During earthquakes the anchored wall and backfill move together, but the soil above and below the wall move seperately. The model is nonlineer. The soil is elastoplastic, homogen and isotropic hardening material. The greatest bending occur at the middle of the span. Anchorage is modelled as a spring in soil.. The design of sheet pile walls, anchored bulkheads and siffiilar anchored retaining wall Is, is made according to active and passive soil pressures. The cause of results induced by an earthquake is estimated as liquefaction of the soil and anchorage rupture. Anchorages are generally the XI main cause of damage during eartquakes. In order to prevent the removal of the anchorage -from the active zone the anchorage rod must be made long. The magnitude and distribution of the horizantal soil pressure is effected by the properties of the backfill, the bending of the retaining structure with respect to soil and the properties of the soil between the structure and the soil. Even with respect to static loads soil and earth interaction is very complicated. Classical soil pressure theory is prepared by Coulomb. He has 2 main assumptions: (1) The soil is unflexible and plastic; and (2) active and passive wedges slide on plane rupture surfaces. During earthquake retaining walls are subject to additional inertial forces. The solution of this problem according to Coulomb is made by Mononobe, Okabe and Matsua. According to them, during earthquakes, the active soil pressure increases and passive soil pressure decreases. With the increase in the magnitude of an earthquake tepric sliding plane approaches to horizantal state. As a result of this, the factor of safety decreases. If the wall moves outwards, it moves with a constant velocity. There are three kinds of displacement. These are: CI) only sliding or horizantal displacement; (2) the rotation of the top; (3) the rotation of the bottom edge. The first state necessitates the existence of moment and for this reason it is unprobable. Second state happens when the anchorage is situated outside of the active sliding zone. According to field observations the third state happens when the anchorage is swept from the active zone or when it yields prematurely. Field observations has proved that the third state is point of interest with sliding under static design conditions. The most desired yield state is the rotation of the anchorage. In this state, the anchorage is outside of active sliding zone and during seismic loading it provides neutralizing force. If there are inertial forces exceeding the pressure thrust in the soil, during rotation failure occurs. Horizantal active thrust is affected by the dynamic coefficient of active soil pressure, the unit weight of the soil, the height of the active part, internal friction angle of the active side and the friction angle between the soil XII and the wall at the active side. Horizantal passive thrust is affected by the dynamic coefficient of the passive soil pressure. With the increase in ground accelaration of the active zone the dynamic coefficient of the active soil pressure cfecreases and the dynamic coefficient of passive soil pressure increases. When the factor of safety of the retaining wall appproaches 1, failure occurs. When the accelaration angle is greater than internal friction angle the equations can not be solved. Soil acts like a viscous liquid. The factor of safety with respect to failure is the ratio of moments.that are preventing the overturning of the structure and that are overturning the structure. During earthquake additional forces accrue. These are Ğ/nasnic increases. These additional forces affect the overturning moment and the other moments. The stability of the anchored retaining wall greatly depend on the passive resistance of the soil and the embedment depth. At the starting moment of the movement the centre of the passive pressure is 1/3D below the ground level. In a frictionless medium it is 0.2D below the ground level. Under static loads the increase in the embedment depth cause the increase of the factor of safety. On the other hand, some observers claim that increasing the embedment depth has no effect on the seismic stability. The displacements in anchored retaining walls with the increase of embedment depth has no priority over the retaining walls &«hich hasn't increased embedment depth. There is a claim that increasing the embedment depth has no influence on the displacements above the ground level. In order to ensure equilibrium observers prefer to increase the depth of the pressure wedge or decrease the passive res i s tence. The -Factor of safety of the passive ground pressure coefficient raust be 2. In this instance there is a plane sliding surface. In a dense cohesionless fill, at the beginning of the event, with vibration stiffness is ensured. Then step by step sliding plane occur. When sliding surfaces are completed, ther is a decline in the force. When the sliding surfaces are completed, the internal friction angle is not at its maximum value, but at its residual value. The embedment depth of the anchorage is found according to the equilibrium of the moments. This depth must be increased 20 per cent. XIII Increasing number of anchorages in an anchored retaining wall has positive effect on sismic stability. Increasing nwber of anchorages in an anchored retaining wall decreases displacement at deep end of a wall, whereas displacement at top of a wall increases. In anchored retaining wall displacements increase proportionally with accelerations. If soil doesn't liquify a one anchored retaining wall suffers a small and a moderate earthquake without serious damage. Ffeximum amplitude of a wall occurs at an unsupported portion of a wall. If amplitude increases, acceleration also icrease. Stiffer anchorages carry more load and suffer less displacement. The more inclination an anchorage has, the mars load it carries. But stiffness of an anchorage is effective only in a restricted area, only in vicinity of an anchorage. Stiffer anchorages tolerate bigger accelerations. Iclination of an anchorage is effective only between anchorage levels. Gravity retaining walls slide onwards or bend during eartquakes. At one end of a foundation, if bearing capacity declines, settlements occur. Caution must be taken in order to decrease back pressure. Cement grouting is one solution. If a foundation is sandy, one must also be cautious. Liquefaction or e decline in bearing capacity takes place. We use several national standards in computing sismic stability of gravity retaining walls. The most widespread procedure is Mononobe-Okabe (1955) formula. In this formula backfill is considered as dry, cohensionless and homogenous. Internal friction angle is assumed to be constant. Sliding line is considered flat. During eartquakes, base acceleration and inertial forces save opposite sides. If one is inwards, other is outwards. In Richards and Elms (1992) model, a procedure that is taking amplitude into account is used.' Computation of sismic behaviour of reinforced soil gravity retaining wall is important. It is more economical when compared with others. Sismic stability of reinforced soil gravity retaining wall is analized by 2 procedures. In my thesis, also Turkish, Japanese and Indian standards are introduced.

##### Açıklama

Tez (Yüksek Lisans) -- İstanbul Teknik Üniversitesi, Fen Bilimleri Enstitüsü, 1993

##### Anahtar kelimeler

Barajlar,
Deprem,
Jeoteknik,
Dams,
Earthquake,
Geotechnics