Lyapunov-based robust optimal control for time-delay systems with application in milling process

Arash Bahari Kordabad, Sébastien GrosInternational Journal of Dynamics and Control 2023 · DOI

TL;DR

This paper introduces a robust, optimal controller designed to eliminate "chatter"—destructive self-excited vibrations—in high-speed milling processes. By framing the milling dynamics as a time-delay system and using Lyapunov-Krasovskii stability theory, the authors developed a method to calculate a controller that maximizes the stable machining zone while minimizing control effort, outperforming existing "intelligent" controllers by over 3.5 times in error reduction.

What Problem Does This Paper Address?

In precision manufacturing, milling is a workhorse, but it has a "ghost in the machine" called chatter. Chatter is a self-excited vibration caused by the regeneration effect: as a tool cuts, it leaves a wavy surface; when the next tooth hits that wave, it vibrates, creating a feedback loop.

Mathematically, this is modeled as a Time-Delay System (TDS) because the current force depends on the state of the workpiece one tool-rotation ago. Controlling these systems is notoriously difficult because the delay can cause traditional controllers to become unstable. Furthermore, physical parameters like material stiffness, damping, and the "axial depth of cut" are often uncertain or change during the process. The challenge is to create a controller that is "robust" (works despite these uncertainties) and "optimal" (doesn't use excessive force/energy).

Methods

The authors propose a delay-independent robust optimal controller. Key technical components include:

  • Lyapunov-Krasovskii (L-K) Theorem: They use L-K functionals to derive stability conditions for the time-delay system. Unlike some methods that are "delay-dependent" (and thus sensitive to changes in spindle speed), this approach ensures stability regardless of the specific delay value.
  • Linear Matrix Inequalities (LMIs): The stability conditions are formulated as LMIs. This transforms a complex control problem into a convex optimization problem that can be solved efficiently.
  • Nonlinear Programming (NLP): To make the controller "optimal," they set the objective function to minimize the norm-2 of the controller gain. This ensures the system achieves stability with the minimum possible control effort, preventing "over-control" that could damage the actuators or the workpiece.
  • Semi-Discretization Method (SDM): This standard tool is used to generate Stability Lobe Diagrams (SLDs), which are maps showing the "safe" combinations of spindle speed and depth of cut.

Datasets

The paper does not use a large-scale public dataset. Instead, it utilizes high-fidelity physical parameters (mass, stiffness, damping, and cutting coefficients) derived from modal experimental procedures published in prior literature (specifically Moradi et al., 2013). These parameters represent a real-world 2-Degree-of-Freedom (2-DOF) milling setup, providing a realistic benchmark for their simulations.

Key Insights and Ideas

  • Bridging Linear and Nonlinear: While the controller is designed based on a linearized model of the milling process, the authors demonstrate it remains effective when applied to the full nonlinear dynamics (which include third-degree polynomial cutting forces).
  • Bifurcation Control: One of the most interesting findings is that the controller doesn't just suppress vibration; it changes the nature of the system's failure. It postpones "Hopf" and "Flip" bifurcations (the mathematical points where stability is lost) to much higher depths of cut, effectively doubling the productive capacity of the machine.
  • Efficiency over Complexity: Compared to Model Predictive Control (MPC), which requires solving an optimization problem at every single time step, this LMI-based approach solves the optimization once offline. This makes it much more feasible for real-time industrial implementation where microsecond response times are required.

Key Results

  • Stability Expansion: The Stability Lobe Diagrams show that the controlled system can operate at nearly double the axial depth of cut compared to an uncontrolled system without triggering chatter.
  • Performance Benchmark: The proposed controller was compared against a "Learning-based intelligent controller." The proposed method improved the Integral Absolute Error (IAE) index by a factor of 3.65, meaning it kept the tool significantly closer to its intended path with much less oscillation.
  • Vibration Reduction: In nonlinear simulations, the amplitude of limit cycles (sustained vibrations) was approximately halved, and settling times were significantly reduced.

Limitations and Future Work

  • Stochastic Disturbances: The current model is deterministic. The authors note that future work should incorporate external stochastic disturbances (random noise/vibrations) to better reflect a messy factory floor.
  • Dimensionality: The study focuses on a 2-DOF model. Extending this to more complex, high-dimensional dynamics (like 5-axis milling or flexible workpieces) remains a challenge.
  • Experimental Validation: While the simulations use experimentally-derived parameters, the authors suggest that the next logical step is a physical implementation on a CNC machine to verify the theoretical gains in a real-world environment.

Summary updated

Abstract

Abstract In this paper, we propose an optimal delay-independent robust controller based on the Lyapunov–Krakovskii theorem for the milling process as a time-delay system in the presence of varying axial depth of cut and different parametric uncertainty, such as stiffness, damping, etc. Milling is widely used in the manufacturing processes for the production of complex-shaped workpieces with high accuracy. However, chatter is a self-excited vibration that may cause adverse effects, such as tool damage, poor surface quality, excessive noise, etc. The dynamic model of the milling process is considered a two-dimensional time-delay system. A nonlinear programming with linear matrix inequality constraints is solved in order to obtain the controller gain where the objective function is corresponding to the norm-2 of controller gain, and the constraints guarantee robust stability. Using the semi-discretization method, stability lobes are shown in both uncontrolled and controlled plants to illustrate the improvement of the stable region via the proposed controller. Bifurcation phenomena have been improved with this controller by postponing the adverse effects to the higher values of the axial depth of cut, reducing the amplitude of limit cycles, and changing the type of bifurcation. Finally, we will compare the proposed controller with an intelligent controller in order to show the efficiency of the proposed method. It is shown that the proposed controller has improved the integral absolute error index by about 3.65 times compared with the intelligent controller.
BibTeX
@article{kordabad2023lyapunov,
  author = {Arash Bahari Kordabad and Sébastien Gros},
  title = {{Lyapunov-based robust optimal control for time-delay systems with application in milling process}},
  journal = {International Journal of Dynamics and Control},
  year = {2023},
  doi = {10.1007/s40435-023-01217-2},
}
Figure