Neural Networks Tame Turbulence: Breakthrough in Controlling Vortex-Induced Vibrations

Researchers Soha Ilbeigi, Ashkan Bagherzadeh, and Alireza Sharifi from the University of Tehran have developed a novel approach to controlling vortex-induced vibrations (VIVs) in cylindrical structures, a common issue in various engineering applications, including those relevant to the energy sector. Their work was published in the journal Mechanical Systems and Signal Processing.

Vortex-induced vibrations occur when fluid flows past a cylindrical structure, creating vortices that can cause the structure to oscillate. These vibrations can lead to fatigue and failure in structures such as marine risers, tall buildings, and renewable energy systems like wind turbines and offshore platforms. The researchers aimed to address this challenge by developing a model-based active control strategy integrated with a neural network (NN) to suppress VIVs, even in the presence of system uncertainties.

The proposed method employs a closed-loop control system that uses feedback from the system’s dynamic state to generate adaptive control commands. This allows the system to respond to changing flow conditions and nonlinearities. The researchers conducted a controllability analysis to assess the efficiency of their control strategy in mitigating VIVs. They implemented two control approaches: simple learning and composite learning. Both strategies significantly enhanced vibration suppression, achieving up to 99% reduction in vibrations despite uncertainties in the system.

The practical applications of this research for the energy sector are substantial. For instance, in offshore wind farms, the suppression of VIVs can enhance the efficiency, stability, and lifespan of wind turbines and other structures. Similarly, in the oil and gas industry, controlling VIVs in marine risers can prevent fatigue and failure, ensuring the safe and efficient extraction of resources. The researchers’ findings demonstrate the potential of their method to improve the performance and durability of structures subject to VIVs, making it a valuable tool for engineers and operators in the energy industry.

In summary, the researchers from the University of Tehran have developed an innovative approach to controlling vortex-induced vibrations in cylindrical structures, with significant implications for the energy sector. Their method, which combines model-based active control with neural networks, offers a promising solution to a longstanding challenge in engineering, paving the way for more efficient, stable, and durable structures in various applications.

This article is based on research available at arXiv.

Scroll to Top
×