Progress Reports: June 2006 December 2005
NUMERICAL MODELING OF TORPEDO ANCHORS
OBJECTIVE: This Research Project is aimed at development and application of numerical models for the analysis of installation, behavior and strength of torpedo anchors. The Project will utilize and extend earlier developments at OTRC on suction caissons. Studies will be conducted to establish the relative significance of factors affecting the performance of torpedo anchors and variants, thus contributing toward improved design of these deep-water foundations.APPROACH: The computational techniques to be developed in the course of this work will be based on the finite-element method and will take into account the nonlinear behavior of clayey and sandy soils where torpedo anchors may be installed. Procedures used in earlier work on suction caissons will require modifications to reflect the dynamics of torpedo entry and subsequent trajectory into the soil, the three-dimensional features stemming from off-vertical penetration and the presence of flukes, and rate-dependent aspects of soil behavior.
DEPLOYMENT OF RESULTS: By virtue of ease of installation, torpedo anchors appear to be promising foundations for deep-water applications. The proposed work will explore the mechanics underlying the performance of these anchors and will thereby seek to contribute to design improvements. Results of the work will be disseminated through presentations, dissertations, theses, reports and conference and journal articles.
ANTICIPATED NUMBER OF PHASES: 3
PROJECT PLAN FOR PHASE 1 (2005 -2006):
Scope and Plan: The literature on torpedo anchors or piles, also known as “rocket” anchors, and several variants installed by release from some depth, “free” fall through the seawater and penetration into the seabed, will be reviewed in order to fully understand the current state of the art in the design and practice of such anchoring systems. Alongside the literature review, a database of available experimental evidence and measured-performance records will be compiled. Our activities will then focus on the development and demonstration of numerical models that can be used effectively in simulating the installation and subsequent behavior of torpedo anchors. The following Tasks are envisioned during Phase 1 of the Project:
Task 1: A thorough examination of the literature will be conducted in this Task. Preliminary search suggests that current torpedo-pile technologies, although reportedly already in use, are not documented in widely available comprehensive reports, or easily retrievable journal articles. The most likely sources of information appear to be relatively few conference articles and progress reports of limited circulation. Therefore, we expect that our review of the literature will involve not only traditional information mining but, in addition, personal communications with engineers and other industry representatives.
Task 2: An up-to-date database of available measurements and experience related to installation, behavior and performance of torpedo anchors will be compiled in this Task. We are already familiar with other academic research efforts with regard to small-scale rocket-anchor model testing. Other important records of tests, especially full-scale, are only partially described in easily accessible conference articles or reports. Success in this Task will require, as in Task 1, communications, exchanges and discussions of ideas and, possibly, collaborations with engineers and researchers active in using and further developing torpedo piles.
Task 3: Numerical models of torpedo piles based on the finite-element method and built on our earlier extensive modeling experience with suction caissons will be developed in this Task. Various levels of sophistication and detail will be explored until reasonable balance between model fidelity and model practical effectiveness can be achieved. Nonlinear behavior of clayey and sandy soils where torpedo anchors may be installed will be taken into account, including consideration of rate-dependent effects of soil behavior, especially during installation. The models will be capable of simulating penetration into the seabed at arbitrary angle of torpedo-pile incidence. However, it is expected that, during Phase 1 of the Project, attention will be focused on the relatively more straightforward case of vertical incidence. As in our previous work on suction caissons, the computational procedure that will be formulated in this Project will provide information not only on the state of stress in the soil immediately after installation but, as well, on the dissipation of the pore-water pressure field and the accompanying changes in soil behavior and strength during, relatively long-term, reconsolidation of the soil-pile system, and estimates of the pull-out capacities of torpedo piles at various times after installation.
Task 4: Simulations of vertical torpedo-anchor installation will be pursued in this Task. The objectives will be twofold, to demonstrate the effectiveness of the numerical models in this relatively simple case, from the viewpoints of modeling and computational effort, and to obtain a baseline that can be used later in this Project in exploring off-vertical incidence, an issue of concern in torpedo-pile performance.
Task 5: A Report of our work and progress during Phase 1 will be prepared in this Task. It is expected that parts of the Report will be ready relatively early and will be made available as soon as possible thereafter, upon completion of the literature review and compilation of the test and performance database.
Anticipated Results: The knowledge available on torpedo anchors will be compiled. The procedure for nonlinear, dynamic finite-element analysis of vertical torpedo anchor penetration into the soil will be described along with examples of simulations.
PROJECT PLAN FOR PHASE 2 (2006 - 2007):
Scope and Plan: Off-vertical torpedo entry, penetration and extraction: finite-element modeling; preliminary computations and comparisons with available laboratory test data on torpedo anchor models.
Anticipated Results: The procedure for nonlinear, dynamic finite-element analysis of torpedo anchor response will be described and preliminary computational results will be reported.
PROJECT PLAN FOR PHASE 3 (2007 - 2008):
Scope and Plan: Parametric studies of torpedo anchor installation and holding axial and lateral capacities; effects of entry angle, speed, rate-dependence of soil behavior, flukes and other features.
Anticipated Results: The parametric studies along with the computational tools will be documented.
PRINCIPAL INVESTIGATOR (S) & OTHERS INVOLVED IN PROJECT:
PI: John L. Tassoulas
Other: 1 Ph.D. Student
DATE: May 2006
Project Title: Numerical Modeling of Torpedo Anchors
MMS Project: 557 TO Number: 39344
PI: John Tassoulas
COTR: S. Buffington
Estimated Completion Date: August 31, 2006
Project Description:
Aimed at development and application of numerical models for the analysis of installation, behavior and strength of torpedo anchors, this Research Project will utilize and extend earlier developments at OTRC on suction caissons in clayey soils. By virtue of ease of installation, torpedo anchors have shown promise in deep-water applications. Studies will be conducted to establish the relative significance of factors affecting the performance of torpedo anchors and variants, thus contributing toward improved design of these foundations. The computational techniques to be developed in the course of this work will recognize the interplay between nonlinear soil deformation and pore-water flow. Procedures used in earlier work on suction caissons will require modifications to reflect the dynamics of torpedo entry and subsequent trajectory into the soil, the three-dimensional features stemming from off-vertical penetration and the presence of flukes, and rate-dependent aspects of soil behavior.Progress:
We have completed our review of the literature on torpedo anchors and compilation of experimental data available in the public domain. Laboratory model, centrifuge and field test results have been identified for use in calibrating, evaluating and verifying the computational procedures to be developed in the Project. Furthermore, progress has been made toward synthesis of the initial version of the numerical model that will be used in simulations of vertical torpedo-anchor installation and performance. We have explored the possibility of using FLUENT, a computational-fluid-dynamics tool, in simulations of the torpedo-anchor installation process. The results are very promising. In FLUENT, the water-soil domain is represented as a system of fluid layers with variable viscosity and other properties. The treatment of the interface between the falling torpedo anchor and the water-soil layers as a moving boundary is facilitated by means of various meshing options. We have applied our FLUENT-based numerical procedure to the centrifuge model tests reported by O’Loughlin et al. (2004). These tests were performed on 1:200 scale model anchors with and without flukes in kaolin clay. Anchors with different impact velocities were tested by changing the release height. The length of the anchor is 0.075 m, the diameter 0.006 m and the mass 0.01675 kg. O’Loughlin et al. (2004) reported 0.180, 0.192 and 0.218 m penetration depths for impact velocities of 14.5, 19, 24 m/s respectively. The soil absolute viscosity was calibrated at 1 N•s/m2 on the basis of the test with 24 m/s impact velocity and the same value of viscosity (constant during penetration in the present computations) was used to predict the penetration depth for the two other tests. The computations produced predictions of the embedment depth about 4.0% higher than the measured values. Fig. 1 shows the calculated velocity profiles vs. depth for different impact velocities. The profiles confirm the finding of O’Loughlin et al. (2004) with regard to initially increasing velocity of the anchor in shallow soil after impact and entry into the soil. Fig. 2 shows the axial velocity of the soil around the anchor (with 24 m/s impact velocity) providing a visual representation of the added mass of the soil in dynamic analysis. Shown in Fig. 3 is the radial velocity of the soil suggesting that the tip of the moving anchor displaces soil outwards and then penetrates. Also, it can be seen that soil also moves to close the gap produced on top of the anchor while an animation of the penetration process shows that water is trapped at the top of the anchor.References:
FLUENT, Online User’s Guide, Fluent, Inc., www.fluent.com
O’Loughlin, C. D., Randolph, M. F., Richardson, M. (2004). “Experimental and Theoretical Studies of Deep Penetrating Anchors,” Paper No. OTC 16841, Proceedings, Offshore Technology Conference, Houston, Texas.
Figure 2: Axial velocity of soil around the anchor.
Figure 3: Radial velocity of soil around the anchor.
DATE: December 2005
Project Title: Numerical Modeling of Torpedo Anchors
MMS Project: 557 TO Number: 39344
PI: John Tassoulas
COTR: S. Buffington
Estimated Completion Date: August 31, 2006
Project Description: Aimed at development and application of numerical models for the analysis of installation, behavior and strength of torpedo anchors, this Research Project will utilize and extend earlier developments at OTRC on suction caissons in clayey soils. By virtue of ease of installation, torpedo anchors have shown promise in deep-water applications. Studies will be conducted to establish the relative significance of factors affecting the performance of torpedo anchors and variants, thus contributing toward improved design of these foundations. The computational techniques to be developed in the course of this work will recognize the interplay between nonlinear soil deformation and pore-water flow. Procedures used in earlier work on suction caissons will require modifications to reflect the dynamics of torpedo entry and subsequent trajectory into the soil, the three-dimensional features stemming from off-vertical penetration and the presence of flukes, and rate-dependent aspects of soil behavior.
Progress: A review of the literature on torpedo anchors is nearing completion alongside a compilation of experimental data available in the public domain. Laboratory model, centrifuge and field test results have been identified for use in calibrating, evaluating and verifying the computational procedures to be developed in the Project. Furthermore, progress has been made toward synthesis of the initial version of the numerical model that will be used in simulations of vertical torpedo-anchor installation and performance later this year.
