Loading...

Invesigating the Damage Effect of Earthquake Induced landslides on Buried Pipelines by Physical Modeling

Farahi Jahromi, Hadi | 2018

430 Viewed
  1. Type of Document: Ph.D. Dissertation
  2. Language: Farsi
  3. Document No: 51464 (09)
  4. University: Sharif University of Technology
  5. Department: Civil Engineering
  6. Advisor(s): Jafarzadeh, Fardin
  7. Abstract:
  8. This thesis intends to evaluate the effect of landslides on dynamic response of soil slopes and buried pipelines. The core findings of the research are itemized as below: According to the experiments, dynamic slope response which moderately develops at deeper depths at toe and lower section, displays a clear downtrend at upper section and crest. Also, the pipe response depends on pipe route and slope displacement pattern. The pipe deformations have positive correlation with depth at upper section and crest, but show opposite behavior at toe and lower section. Moreover, the horizontal strains impact on total strains in buried pipes reduces at greater depths.The pipe strain analysis shows that pipe route classification depends on sliding direction-pipe axis intersection angle. As for zero intersection angle, the slope crest has imposed maximum deformations to buried pipes. By increasing the intersection angle to 30o and 70o, the maximum deformation zone has moved to upper slope part and mid-slope section respectively. This zone has moved toward lower section at 90o intersection angle. The slope toe proves the safest route in all intersection angles.Pipeline angle with landslide direction has proved an important parameter in pipe response. The test results display that pipe strains become the highest at 90o angle. Decreasing the intersection angle will lower the pipe deformations, consequently the minimum strains occur at zero intersection angle. Positioning a Styrofoam sheet in pipe trench will decrease buried pipe deformations at lower section of a slope face and toe. The tests results which show limited effect of this approach on lowering pipe strains at upper section of a slope face, proves it useless at crest.Various approaches have been developed by researchers to predict earthquake induced landslide displacements. Five statistical measures are applied to compare the measured and predicted displacements and evaluate the precision of the predictive approaches. By combining the outcomes and using a ranking procedure, the approaches are scored. Consequently, the approaches of Fotopoulou and Pitilakis (2015) and Hsieh and Lee (2011) score the highest. Additionally, a correcting coefficient is assigned to each approach to convert the predicted displacements to observed displacements. An exponential equation is then developed to correlate the correcting coefficient to Ky/PGA ratio for each approach.The numerical modelings are divided into three categories. The first category intends to verify the numerical models with the results of laboratory scale physical tests. Various parameters such as shear wave velocity, meshing size, rayleigh damping coeffcients and soil dilation angle have been selected and inputted to the ABAQUS F.E. program. The numerical results have been then compard with laboratory results and optimum values have been determined. The results display that layering the soil in a slope according to shear strains has a positive impact on predicting the slope displacments similar to the obsereved laboratory responses. The second category of numerical models investigates the effect of various parameters such as slope width, slope length, dilation angle, the spring and dashpot coefficients, damping value, elastic modulus and soil density in dynamic slope response. These two numerical model categoreis intend to verify the numerical modelings to predict the actual slope response as recorded in the earthquake laboratory. The third category investigates the effect of several parameters on slope and pipeline responses in a full scale model. The effect of the slope width to slope height ratio and the trench geometry on an embedded pipe deformations have been investigated. Moreover the natural frequalcies of a slope and a pipeline which have been derived from empricall formulas have been compared with numerical calcluations. The effect of internal pressure on dynamic pipeline response has been also modeled. The models also study the effect of various trench routes in a slope on pipeline deformations. Finally, the burial depth impact on an embedded pipe deformation in the middle part of a slope has been investigated
  9. Keywords:
  10. Buried Pipline ; Landslide ; Shaking Table ; Dynamic Loading ; Seismic Retrofit ; Physical Modeling ; Numerical Modeling

 Digital Object List

 Bookmark

No TOC