2212003 - 3D Soil Simulation Models in OpenSees

Project Title/ID Number	3D Soil Simulation Models in OpenSees—2212003
Start/End Dates	10/1/03—9/30/04
Project Leader	Boris Jeremic (UCD/F)
Team Members	Qing Liu (UCD/GS), Matthias Preisig (UCD/GS), Zhao Cheng (UCD/GS), Ian Tucker (UCD/US)

F=faculty; GS=graduate student; US=undergraduate student; PD=post-doc; I=industrial collaborator; O=other

Click on images to enlarge in a new window

1. Project Goals/Objectives:

Continuing development of the database of models. This work follows an early idea by Professor Elgamal where he envisioned development of standard set of models for clays, sands. This idea is extended to include standard models for concrete, steel, and also to standard cross sections (as used by Caltrans for example) for piles, columns, and beams. In addition to that, a comprehensive database of models will be developed using XML standard so that users can discover available models and use them in their simulations. For example, a XML based database will feature model description, capabilities (comparison with available lab tests), a set of validation examples, and implementation description. It is envisioned that such a database, after reaching certain level of maturity, can be used by practicing engineers (for example Caltrans engineers) for their bridge design and maintenance simulation needs. In addition to the above work an initial work on incorporating sensitivity analysis (with respect to system parameters for soil models) will be started (discussed this with Joel a month or so ago).
Continuing development of models and analysis for dynamic (seismic) and kinematic (lateral spreading) soil-structure interaction. Development of a series of models with different sophistication, and demonstration of their capabilities and analysis of levels of model sophistication for given site conditions (should a full 3D, fully coupled, soil-structure analysis be performed or is the simple p-y spring approach enough, or we can go even simpler by connecting a structure to a set of linear, perhaps nonlinear, springs to mimic foundation-soil system). In particular, different models will be developed in conjunction with Humboldt Bay and I-880 bridges. The idea is to assess the influence of Soil-Structure-Interaction on the seismic response of complete bridge systems and make recommendations to practicing engineers. The influence of different types of soils (sands, clays, liquefiable, layers...), seismic input (duration, fling effects, directivity...), type of foundation systems (modern pile groups, old pile groups with retrofit, battered piles...) will be addressed and the level of sophistication of modeling during simulations will be recommended (for example if you have a pile group in liquefiable soils it may not be recommended to use fully fixed bridge foundation system.

We note that during last year we have performed a number of static and dynamic analysis of single pile and pile group foundation systems, in order to gain confidence in our numerical modeling, In addition to that a detailed 3D models for foundation systems for Humboldt Bay and I-880 bridges are being built and made ready for various simulations. A fully coupled formulation and implementation (upU) has been finished and partially tested. This formulation is undergoing more verifications and validation and will be used in near future simulations of pile foundations. We are also working on large deformation issues, as they are present in cases of kinematic loading of soils on pile groups.

2. Role of this project in supporting PEER’s mission (vision):

This project forms one of the basic elements of PEER's mission in that it creates computational tools that are used to assess the performance of the engineering systems.

3. Methodology Employed:

Methodologies used are based on rational mechanics and computer sciences and are implemented into a versatile computer platform. Current work is also focusing on making those implemented formulations and algorithms more accessible to practicing engineers by creating a set of specialized graphical modules that link directly to OpenSees but hide the details from practicing engineers.

4. Brief Description of past year’s accomplishments (Year 6) & more detail on expected Year 7 accomplishments:

During last year, we have developed a basis for large deformation analysis of inelastic solids (including soils) and have also verified our coupled formulation (u-p-U). These two formulations are being merged right now in order to be able to analyze lateral spreading kinematic loading cases on structural foundations during liquefaction.

In addition to that we have incorporated the Domain Reduction Method into OpenSees and enhanced it to be used with inelastic soil models,

For example, figure below show results of vertical wave propagation in soft and stiff soils. The plotted lines are the time histories of displacements from the depth of 30 meters to the surface. The seismic amplification in soft soil (clay) is apparent!

Figure 1. Seismic wave propagation resulting from the same earthquake acting on a stiff and soft soil site

The methodology allows for 3D wave propagation analysis with inelastic effects (plasticity) and choice of free field (like the one above) or full interaction with inelastic structures.

5. Other Similar Work Being Conducted Within and Outside PEER and How This Project Differs:

This project is part of larger effort centered around the OpenSees development and the analysis of bridge structures.

6. Plans for Year 8 if project is expected to be continued:

Continuing development of computational geo-mechanics formulations and implementations. These will include continuing development on a number of fronts, related to the core PEER mission. There are two main thrusts, related to the:

Dynamic Soil-Foundation-Structure (SFS) interactions (during the seismic loading), and
Kinematic SFS interaction (effects of lateral spreading on structural foundations)

Specific developments can be organized as follows:

Mechanics (formulation, implementation, verification, validation and application) of porous, coupled media (large deformations, full coupling of solid and fluid) and application to lateral spreading cases).
Stochastic inelasticity of soils (and other solids and structures), using stochastic differential equations, trace propagation of uncertainties within simulation process in closed form (and then numerically after being discretized into FEM models)
Seismic wave propagation and Soil-Foundation-Structure interaction using Domain Reduction Method, free field versus near field motions, influence of those on simulation efforts.

In addition to that, implementation testing will continue and be expanded.

7. Describe any actual instances where you are aware your results have been used in industry:

There is a lot of interest from Caltrans, USGS and some private companies in using our developments. I am not sure if the programs and results have actually been used yet but I suspect that in near future they will.

8. Expected Milestones & Deliverables:

We plan to continue to publish (on the web) a series of our models for testbed bridge foundations. In addition to that, a database of models will soon be submitted to the CVS repository. In addition to that we are preparing a detailed report on work performed in last year and will submit it to PEER by the end of November.