The interaction is the problem.

We build computational models for systems whose parts cannot be understood in isolation. Every project begins with the physics of an interface, then moves through formulation, computation, verification and design.

An open braced structure on a pile cap, with embedded piles and qualitative waves in a dry-soil cutaway.
Program 01

Dynamic interaction

How do structures, foundations and soil exchange forces and motion?

Models that treat the structure, foundation, piles and soil as one coupled vibrating system.

Pile groups · flexible foundations · contact · IBEM-FEM
A vibration source, periodic embedded barriers and a receiver structure in one dry-soil cutaway.
Program 02

Vibration control

How can waves be redirected before they reach a sensitive structure?

Barriers, periodic systems and metamaterial concepts that redirect or attenuate vibration.

Ground waves · barriers · periodicity · vibration attenuation
An optimized branching structural support, a dashed design envelope and a downward load arrow.
Program 03

Optimization

How does the best design change when supports, interfaces and failure are modeled honestly?

Computational design methods that account for flexible supports, interfaces and failure modes.

Topology optimization · stress · buckling · flexible supports
A submerged barrier resting on the seabed, with conceptual wave paths near the water-soil interface.
Program 04

Offshore multiphysics

What new behavior appears when water, seabed and structures move together?

Coupled models of water, seabed and structures for vibration and marine-protection questions.

Scholte waves · PML · submerged barriers · fluid-soil coupling
An engineered root-like anchorage and a deformable organism sharing a dry-soil cutaway.
Program 05

Bio-inspired engineering

What can roots and burrowing organisms teach us about anchorage and movement through soil?

We study how roots and soil organisms interact with the ground, then translate those mechanics into new structural and robotic concepts.

Root anchorage · organism-soil interaction · soft excavation · biomimetic design

Research in evidence

Published findings and explicitly labelled preliminary results show how the programmes become concrete mechanical claims.

Graphical abstract with labels 1–12: a wind-turbine tower, elastic soil, fixed-base and coupled FEM–BEM models, pile-soil-pile interaction, and three continuous vertical frequency-response curves. The teal coupled-model curve reaches its first resonance before the petroleum-blue fixed-pile-head and navy fully fixed-base curves. The numbered legend and graph guide explain the labels and axes.
Redrafted diagramThe dynamic response of a piled wind-turbine tower depends on foundation flexibility and energy transmitted through the soil between piles. The coupled model represents these effects; simplified fixed-support models shift the predicted resonances.
View full-size figure

Adapted from the graphical abstract and Figure 6 in Oliveira, A.; Vasconcelos, A. C. A.; Labaki, J. (2023), Latin American Journal of Solids and Structures, 20(6), e491. DOI: 10.1590/1679-78257514. Original article published under CC BY.

Peer reviewed

Dynamic response of piled structures including pile-soil-pile interaction

The receiving structure changes the vibration field rather than merely observing it. Increasing separation reduces transferred vibration, while structural height shifts the coupled resonances. Pile spacing cannot be reduced to one simple monotonic rule.

Graphical abstract key

  1. Wind-turbine tower: an example of a structure whose dynamic response depends on its supporting soil.
  2. Elastic soil: carries energy away from the structure as propagating waves, producing radiation damping.
  3. Rigid-base approximation: neglects soil and foundation flexibility and may misrepresent the dynamic response.
  4. Coupled soil-pile-structure model: the proposed formulation includes the dynamic behavior of the supporting soil.
  5. Finite element method (FEM): represents the elastic superstructure, illustrated by the mesh on the tower base.
  6. Boundary element method (BEM): represents the surrounding soil and its interaction with the piles; the piles themselves use finite beam elements.
  7. Pile-soil-pile interaction: energy passes between piles through the intervening soil, represented by the teal wave paths.
  8. Vertical frequency-response function: vertical displacement at the tower response point per unit vertical force, compared for three support assumptions.
  9. Coupled model, teal curve: includes flexible piles, soil response and interaction between piles (Case 3 in the paper).
  10. Fixed pile-head supports, petroleum-blue curve: only the discrete pile-head locations are fixed; soil interaction is omitted (Case 2).
  11. Fully fixed base, navy curve: the entire bottom surface is fixed; soil interaction is omitted (Case 1).
  12. First-resonance shift: the fully fixed-base model overestimates the first vertical natural frequency compared with the coupled model in this example.
How to read the graph

Horizontal axis: normalized excitation frequency a₀ = ωd/cₛ, from 0 to 3. Here ω is angular excitation frequency, d is pile diameter, and cₛ is the soil shear-wave speed. Vertical grid lines are spaced by 0.5 in a₀.

Vertical axis: magnitude of vertical displacement per unit vertical force, |u_zz| = |U_zz/F_z|, in m/N, on a logarithmic scale. The displayed range is 10⁻¹³ at the bottom to 10⁻⁸ at the top; successive horizontal grid lines differ by a factor of ten.

The drawings are schematic, not to scale; the depicted pile count and mesh are illustrative. The continuous curves were digitized from the unobscured amplitude panel of the published Figure 6, not recalculated from raw simulation data. Their colors follow the SSI Lab palette.

Open journal article
Conceptual paired views of a finite numerical soil domain and an ideal reference half-space.
Conceptual illustrationConceptual introduction to the numerical verification problem.

Work in progress

Reference response reproduced in a finite COMSOL domain

After correcting the boundary treatment and calibrating the domain, the simulated vertical response follows the reference curve closely through the main resonance and its decay.

View original evidence figure
Comparison between Guilherme's COMSOL response and the reference paper for vertical excitation.
Original internal validation figure. COMSOL result in blue and reference solution in black, 2025.
Technical-linework interpretation of the Bag Sentável optimized skeleton.
Conceptual illustrationConceptual cover based on the visible idea and silhouette of the preliminary result.

Work in progress

A structural skeleton that preserves the cargo volume

The initial 3D optimization produced a connected load path between the seating surface and the supports while keeping the central cargo cavity free of structural material.

FAPESP project
View original evidence figure
Initial three-dimensional optimized skeleton for the Bag Sentável project.
Original BESO proof of concept. Stress, buckling and experimental validation remain future stages.
An expanding cylindrical soil cavity with qualitative stress-redistribution contours.
Conceptual illustrationQualitative conceptual cover for the expanding-cavity mechanics.

Work in progress

Expansion creates a moving soil-stress field

Each expanding body segment generates a localized stress concentration in the surrounding soil. The travelling pattern provides a mechanical bridge between prescribed body motion, anchorage and soil disturbance.

FAPESP IC project
View original evidence figure
Computed stress-proxy field around an expanding cylindrical soil cavity.
Original preliminary field at t = 5.00 s. Internal soil-interaction result, 2026.

We are preparing research codes, interactive tools and reproducible examples for public use.

Explore Resources

See who develops these questions and which partnerships make them possible.