Pumping water and oil through the world’s pipelines consumes roughly 10% of global electric power, most of it wasted overcoming the drag of turbulent flow. This project aims to kill that turbulence entirely. The problem is that turbulence is stubbornly persistent. While researchers have occasionally forced a turbulent pipe flow back into a smooth, energy-efficient laminar state in the lab, they have never understood *why* those control strategies worked or when they would fail. Without that theoretical foundation, no one has been able to scale the approach up to real pipelines, power station cooling systems, or industrial ductwork. This project will use machine learning and dynamical systems theory to build a mathematical explanation of how turbulence can be forced to die. That understanding will then be used to design new control strategies that suppress turbulence in flows of industrial scale. If it succeeds, the impact is direct and large. Vanishing frictional drag in pipelines would massively cut carbon emissions from pumping systems, helping meet net-zero targets. The same fundamental physics applies to flows over aircraft wings and turbine blades, so the work could also improve efficiency in aviation and power generation. The research is fundamentally curiosity-driven—it seeks a universal theory of why wall-bounded turbulence exists—but its practical payoff is unusually clear.
View original technical description
An enormous amount of fluids — from water to oil and natural gas — is transported across the globe through pipes and ducts. In the United Kingdom alone there is over 215,277 miles of water pipelines, enough to travel the circumference of the world 8 times. Most often these flows are turbulent, and the associated frictional losses are much larger than those of laminar flows, making it a far less energy-efficient way of transporting fluids. According to estimates, around 10% of the global electric power consumption is spent by pumping systems to overcome frictional drag in pipelines, including not only large-scale oil/gas pipelines, but also domestic networks. Fighting against climate change, the most desirable, yet challenging, outcome that a flow control method could achieve is to completely extinguish turbulence, hence zeroing the associated frictional losses. Even when relaminarisation has been achieved in lab experiments and numerical simulations, it was not possible to explain on a theoretical basis why the control strategy worked, and under which conditions. This lack of understanding has so far prevented up-scaling of relaminarising control strategies for deployment and implementation in practical engineering systems. To overcome these limitations, this project will exploit recent advances in machine-learning and data-driven methods to unravel the physical mechanisms underlying the phenomenon of forced relaminarisation and to provide a mathematical description of its dynamics through the theory of dynamical systems. The understanding gained in this way will be leveraged to develop novel control strategies, based on the same principle, to completely suppress turbulence in pipeline flows of industrial interest. Such vision has important societal and economic implications because vanishing turbulence will massively curb carbon emissions, thus leading to improved air quality and contributing to meeting the net-zero-by-2050 target. This achievement is also of great fundamental interest as it would provide a better understanding of the universal mechanisms sustaining wall-bounded turbulence. As this knowledge applies not only to pipelines but to many other flows of industrial relevance (e.g. flows over the wing of an aeroplane or a turbine blade), it will enable us to control and improve e?iciency of these systems in a variety of engineering and technological applications.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know