Projects

1. Numerical investigation of hydraulic forces induced by drillstring whirling and rotation dynamics

Comprehensive computational fluid dynamics (CFD) simulations were conducted with the finite volume method (FVM) to estimate the hydraulic forces on drillstrings under diverse conditions. Coupled lateral-torsional movements of a drill pipe were reconstructed and approximated as the moving boundaries of the CFD model. With the hybrid/dynamic mesh, the proposed CFD model can effectively calculate the real-time annulus fluid velocity field and hydraulic forces with improved accuracy. The simulations span a wide range of drilling scenarios, exploring different drillstring whirling and rotational frequencies, with varying flow rates of Newtonian and non-Newtonian fluids in 2D and 3D domains.

2. Coupled managed pressure and temperature drilling in geothermal and HPHT wells

This paper introduces an integrated MPD-MTD control framework based on improved reduced drift-flux model (RDFM) that incorporates temperature dynamics, interface mass transfer, and a new lumped pressure dynamics model to describe geothermal and HPHT drilling. The proposed MPD-MTD control strategy utilizes MPD choke adjustments, flow rate modulation, and mud cooling to simultaneously regulate downhole pressure and temperature.

3. FEM

Numerical investigation of linear advection equation with SUPG method. Report

• Solved the linear advection equation with adaptive mesh refinement using deal.II and FEniCS on a Linux system.
• Conducted numerical experiments in 1D 2D and 3D domains.
• Utilized parallel computing with multiple processors accessing shared memory techniques to assemble the global matrix and accelerate computation.
• Compare the results and convergence rates from SUPG, Standard Galerkin, and Upwind methods.

4. Shallow water wave simulation

Numerical modeling of water wave evolution in shallow and variable-depth terrain using Finite Volume Method in Fortran. Report

• Developed a Fortran program to numerically solve fully nonlinear Boussinesq Wave Model from scratch.
• Applied the Monotonic Upstream-centered Scheme for Conservation Law (MUSCL) for spatial discretizing and utilized the third-order Strong Stability-Preserving (SSP) Runge-Kutta scheme for time integration.

5. Fluid model using Lattice Boltzmann Method (LBM)

Numerical investigation of a fluid domain activated by a waving plate using Lattice Boltzmann Method (LBM)

6. Image reconstruction using convex optimization

Delved into the realm of dictionary learning models and optimization algorithms to address the challenges of grayscale and color image processing. Our primary focus was on two key aspects: dictionary learning and descent methods. Report

7. Controller Design

Control System Design for an Underwater Vehicle

• Developed a Simulink model for simulating an underwater vehicle, incorporating a 4-DOF dynamic model of the ROV, along with a propulsion system model and a voltage allocation module.
• Designed a Proportional Integral Differential (PID) control system to correct deviations between the robot’s actual depth and heading, achieving 18.50% and 31.57% overshoot, with settling time of 5.13s and 12.84s for depth and heading, respectively.

8. Wind Speed Prediction

Wind Speed Prediction with Neural Networks and Signal Decomposition Techniques Report

• Conducted short-term wind speed forecasting using a Long Short-Term Memory (LSTM) neural network based on historical wind speed data.
• Employed signal processing techniques to preprocess raw wind speed data for improved model accuracy.