Investigation of Seasonal Frozen Soil effect on Structural Seismic Behavior

1
Effects of Seasonally Frozen Soil on the Seismic
Behavior of Bridge Bent-Foundation-Soil System
Presented at the ASCE Structures Congress 2007
Feng Xiong Post Doctoral Researcher, UAA May
18, 2007
2
Acknowledgement

• Sponsors
• NSF and State of AK Funded EPSCoR Program
• Alaska Dept. of Transportation and Public
  Facilities
• U.S. Geological Surveys ANSS Program
• UAA
• Co-Workers
• Drs. Zhaohui (Joey) Yang and Uptal Dutta, UAA
• Mr. Elmer Marx, Mr. Richard Pratt from AK DOTPF
• Dr. Niren Biswas, UAF

3
Outline

• Background
• Strong-motion instrumentation of the Port Access
  Bridge Phase I Phase II
• Previous results on frozen soil effects
• Study of frozen soil effects on bridge
• Modeling of a soil-pile system
• Modeling of a bridge bent
• Conclusion and discussion

4
Background

• South-central Alaska is a cold region with high
  seismicity the effect of seasonal frost on
  bridge behaviour has not been a focus of past
  studies.
• Strong motion instrumentation of the Port Access
  Bridge
• Phase I Summer of 2004
• Phase II current scheduled to be completed in
  June, 2007

5
Strong-motion instrumentation Phase I
6
Strong-motion instrumentation Phase II
7
Previous results Seasonal Frost Effects

• 21 earthquakes and 449 train-induced vibrations
• The average frequency of first transverse mode is
  0.99 Hz for summer season, and 1.12 Hz for winter
  season, or 12 of change
• The average damping ratio of the first transverse
  mode is 1.54 and 1.27 in simmer and winter
  season.

8
Investigation of Seasonally Frozen Soil Impact on
the Bridge bend-foundation-soil System
1. Modeling of a soil-pile system 2. Modeling of
a bridge bend
9
Influence of Seasonally Frozen Soil on the
Soil-pile System

• Soil-pile system modeling
• Beam element piles (HP 12X53 Steel)
• Tetrahedral solid element soil and concrete
  cap
• No gap element between soil, cap and piles

• Elastic-plastic cyclic analysis
• Drucker-Prager material for unfrozen soil
• elastic material for pile and concrete
• elastic material for frozen soil

10
Results and Analysis Hysteretic behavior

• Unfrozen soil hysteretic loop demonstrates large
  energy dissipation capability frozen soil
  behaves like an elastic material.
• Damping ratios estimated from hysteretic loop
  are 7.7 and 0.3 for unfrozen and frozen soil,
  respectively.
• Main plastic strain develops in surface soil.
  This is the reason that no obvious yielding is
  observed in frozen soil.

The top layer
The second layer
0.07
0.285
11
Equivalent soil spring coefficient

• The spring coefficient is estimated from FE
  modeling results.
• The horizontal stiffness of frozen soil is about
  10 times larger than that of unfrozen soil,
  however the stiffness only slightly increases in
  vertical direction and 4 times in rocking
  direction.
• In unfrozen condition the results from elastic
  analysis and elastic-plastic analysis differ by
  15, and the difference decreases to about 1.0
  for frozen condition, implying that the soil
  yield is not prominent in the frozen condition.

12
Influence of frozen soil depth

• When frozen soil depth increases, soil stiffness
  also increases.
• Even when the frozen soil depth is only 0.5 m,
  the horizontal stiffness still increases 5.1
  times.
• The change of soil stiffness is only sensitive
  to upper 1.0-1.5 m frozen soils.

13
Influence of Seasonally Frozen Soil on the Bridge
Bent

• Bridge Bent modeling
• 6 bents (2A, 3A, 4A, 7, 13,17) were
  selected from total 23 bents.
• Bilinear kinematic hardening material for steel
  and bilinear isotropic hardening material for
  concrete
• Rigid beam element for cap beam
• Mass density of cap beam (representing
  superstructure mass) was adjusted to match the
  frequency identified in previous study.

14
Dynamic property results

• The results indicate a significant effect by
  frozen soil on structural frequency (3-25)
• The change of frequency due to frozen ground
  depends on the overall structural stiffness (H/L
  ratio) the smaller H/L, the larger the change in
  the frequency.
• The inconsistent change in dynamic properties
  caused by seasonally frozen soil would result in
  unfavorable responses in bridge.

15
Push-over Analyses of Typical Bents

• Bent 2A, 7, 13, 17
• Three boundary conditions unfrozen, frozen and
  fixed.
• Displacement controlled push-over analysis
• Failure condition tensile strain in steel tube
  reaching 0.03

16
Push-over Analysis Results

• Ultimate lateral displacement capacity decreases
  when ground freezing compared with unfrozen
  condition, 717 decrease has been found for
  different bents.
• Shear demand at pier base increases when ground
  freezing at yield displacement level 1652
  increase has been found.
• The increase of shear demand decreases with the
  plastic deformation.

17
Conclusion

• Significant impact in the stiffness of the
  soil-pile system due to the soil freezing is
  found. The stiffness in the horizontal direction
  could increase by about 10 times compared with
  unfrozen condition.
• Seasonally frozen soil could also affect the
  energy dissipation capacity of the soil-pile
  system.
• The stiffness of soil-pile system generally
  increases with the depth of frozen soil. However,
  the stiffness change is more sensitive to the
  freezing of top layer soils.
• FE results predicted in bridge bents agree well
  with the identified results. The influence on
  dynamic properties due to frozen soil increases
  with the overall stiffness. The maximum increase
  in frequency is found to be 25.
• Under frozen soil condition, the ultimate
  lateral displacement capacity decreases and the
  shear demand increases. As large as 52 of
  increase in shear demand is found at the bent
  yielding.

18
Thank you!