Robust active vibration control by the receptance method

Abstract

In recent years, the pursuit of lighter and more energy-efficient structures has fuelled research and development efforts focused on creating innovative, flexible, and lightweight structures. While these advancements promise benefits, such as enhanced fuel efficiency, they often encounter challenges in the form of unwanted vibrations, particularly unstable oscillations known as resonance. Resonance occurs when external loads excite the structure near its natural frequencies, raising safety concerns such as mechanical deterioration, fatigue, and structural failure.

Conventional methods for suppressing structural vibrations involve passive modifications, such as physically altering the structural stiffness or adding supplementary damping properties, like dashpots. However, these modifications usually lead to additional weight penalties for the structure. Fortunately, with technological advancements, active vibration control (AVC) has emerged as a promising solution. AVC strategies deploy strategically placed actuators and sensors on the structure to dynamically adjust the damping levels and natural frequencies, thereby mitigating vibrations.

While the theory of AVC is well-established, practical implementations on various structures in real-world applications remain rare. This scarcity is primarily due to their reliance on highly accurate models of the structures, which are challenging to obtain due to the complexities associated with achieving precise numerical or analytical models. This thesis explores an experimental-based approach to AVC known as the Receptance Method, eliminating the need for numerical modelling and mitigating issues related to modelling accuracies.

However, owing to the experimental nature of the Receptance Method, the controller is sensitive to the measurement process. Therefore, this thesis aims to investigate and design a control system that can maintain robust performance in the presence of uncertainties. The formulation proposed in this thesis builds upon the Receptance Method, addressing the practical limitations associated with uncertain system parameters. It is demonstrated that when the proposed formulation is applied in conjunction with the Receptance Method, the controller maintains a predefined level of performance in the presence of uncertainties.

Furthermore, to facilitate the experimental implementation of the proposed method, this thesis presents an experimental procedure for quantifying uncertainties in a system using only a single measured nominal receptance dataset. This technique eases the implementation of the proposed robust controller in experimental settings. Experiment validations have demonstrated improvements in system performance and have proven effective in preventing the propagation of uncertainties in its poles. Finally, a series of numerical analyses on a continuous system are conducted to investigate the implications of employing suboptimally measured datasets in designing a receptance controller.

More information

Identifiers

ORCID for Neil Ferguson: orcid.org/0000-0001-5955-7477

Catalogue record

Export record