| dc.description.abstract | This thesis investigates a novel state-constrained control technique, namely ‘Barrier Avoidance Control’, that can be applied to a high-order system to address state barriers of complex shape, and then applies this control method to a downhole drilling system to avoid drilling at undesired operating conditions.
Drilling is one of the most critical processes in the exploration and production of shale oil & gas, enhanced geothermal energy, and minerals. Control of the drilling process is challenging due to the underactuated feature of a long drill string, large uncertainty in the downhole environments, and the nonlinear and nonsmooth nature of the bit-rock interaction that can cause severe vibrations. These harmful vibrations, such as stick-slip and bit bouncing, will be largely exacerbated if not properly controlled, since the oscillation will be transported by the wave propagation and reflection along the drill string. In addition, the tendencies of using directional/horizontal wells with deeper and more complex drilling environments further complicate the drilling control. The coupling of vibration modes in the directional/horizontal drilling can make certain regions in the state space, which correspond to particular working conditions, undesirable in the drilling operation. Thus, drilling process needs to avoid these undesired regimes, so as not to experience slow drilling rate, significant vibration, safety problems, and drilling system failure. State barrier avoidance is not only necessary for the steady state, but is also critical during control transients.
In this thesis, we will introduce the state barrier avoidance control theories to keep the system states away from the undesired state regime in both steady state and transients. Since existing methods on state barrier avoidance control cannot be directly applied to directional drilling applications, therefore, a novel barrier avoidance method is proposed, which can be applied to high-order systems and address state barriers with complex shapes. This new control method is then applied to directional drilling, whose effectiveness is demonstrated through both simulation and experimental results.
Here, the logic flow of this thesis is organized as follows: 1) Firstly, one of the most commonly used state-constrained control methods, namely the integral barrier Lyapunov function based control, is applied to the drilling control. The dynamic model contains a lumped-parameter drill string model and a bit-rock interaction governed by a delay-differential equation. A customized model transformation is employed to enable the barrier control design. However, it is found that this method can be applied to a system described by a lower-order model such as the vertical drilling scenario, but it is hard to be implemented to the higher-order model that is suitable for directional drilling. 2) Because of this, we propose a novel barrier avoidance control scheme, where a diffeomorphic transformation projects the constrained region in the original space into a radially large region in the new space, and converts the state-constrained control problem into a non-constrained problem. The method can offer large flexibility in the barrier avoidance control design, and thus is more promising to address control of a higher-order control system. 3) As the empirical results of the field tests show the desired operating zone of the drilling inherently behaves in a complex shape, the corresponding state constraint can be found in a complex barrier region. However, none of the existing studies on barrier avoidance control investigates the complex barrier case. Therefore, a new method is proposed to address the complex state constraints based on the transformation-based barrier avoidance framework. The method is then applied to the directional drilling control, and its effectiveness is evaluated through comprehensive simulation results. 4) Finally, we validate the state-constrained drilling control in an experimental study in a hardware-in-the-loop framework, which contains a lab-scale drill rig with a real size polycrystalline diamond compact bit, and drill string simulation in a real-time environment. The experiment setup captures severe vibration modes through the contact of the drill bit to the rock sample. The barrier avoidance control is next applied to this lab-scale drill rig testbed. The testing results validate the effectiveness and robustness of the proposed state-constrained controller, providing proof of implementing this barrier avoidance control technique to full-scale drill rigs in the field. | |
