Prediction of the drag reduction effect of pulsating pipe flow based on machine learning

Kobayashi, Wataru; Shimura, Takaaki; Mitsuishi, Akihiko; Iwamoto, Kaoru; Murata, Akira

doi:10.1016/j.ijheatfluidflow.2021.108783

Cited by 17 publications

(10 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A similar control strategy was recently studied experimentally by Kobayashi et al 19 , who employed machine learning to devise and test different waveforms. In our numerical study, however, we employ a simpler and more realistic waveform, and aim at verifying the robustness of the approach with respect to the investigation method and the control parameters, establishing that the advantage is not limited to reductions in drag, but extends to savings in energy.…”

Section: The Unsteady Pumpingmentioning

confidence: 99%

“…This constitutes a hybrid form of control, in which no additional devices are required (as in passive techniques), and the active pumping phase is employed to increase the overall efficiency of the system (as in active techniques). Building upon recent progresses made in understanding the transient nature of turbulent flows and the effects of a time-varying pumping 16,17 , our approach exploits an unsteady power delivery, inspired by the work of Iwamoto, Sasou and Kawamura 18 and Kobayashi et al 19 , designed to move the flow smartly back and forth between the turbulent regime and a quasi-laminar one. In the following, we describe in detail our innovative forcing technique, and discuss the results of the demanding numerical simulations we performed.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Saving energy in turbulent flows with unsteady pumping

Rota

Monti

Rosti

et al. 2023

Sci Rep

View full text Add to dashboard Cite

Viscous dissipation causes significant energy losses in fluid flows; in ducts, laminar flows provide the minimum resistance to the motion, whereas turbulence substantially increases the friction at the wall and the consequent energy requirements for pumping. Great effort is currently being devoted to find new strategies to reduce the energy losses induced by turbulence. Here we propose a simple and novel drag-reduction technique which achieves substantial energy savings in internal flows. Our approach consists in driving the flow with a temporally intermittent pumping, unlike the common practice of a constant pumping. We alternate “pump on” phases where the flow accelerates, and “pump off” phases where the flow decays freely. The flow cyclically enters a quasi-laminar state during the acceleration, and transitions to a more classic turbulent state during the deceleration. Our numerical results demonstrate that important energy savings can be achieved by simply modulating the power injection into the system over time. The physical understanding of this process can help the industry in reducing the waste of energy, creating economical benefits and preserving the environment by reducing harmful emissions.

show abstract

Section: The Unsteady Pumpingmentioning

confidence: 99%

mentioning

confidence: 99%

Saving energy in turbulent flows with unsteady pumping

Rota

Monti

Rosti

et al. 2023

Sci Rep

View full text Add to dashboard Cite

show abstract

“…The point corresponding to the canonical turbulent channel flow at Re τ = 180 (or Re b ≈ 2800) is also plotted; by definition, it sits on the turbulent curve. For comparison, we include results obtained by Iwamoto et al (2007) and the more recent ones presented by Kobayashi et al (2021). Two points reported by Frohnapfel et al (2012) and corresponding to opposition control (Choi, Moin & Kim 1994) are highlighted as well.…”

Section: Control Performancementioning

confidence: 99%

“…The colour of the filled symbols encodes the forcing period, and their shape refers to the value of the duty cycle. The results fromIwamoto et al (2007),Kobayashi et al (2021) and opposition control data reported byFrohnapfel et al (2012) are represented as blue, red and yellow crosses, respectively. The black dot on the reference line corresponds to Re τ = 180.…”

mentioning

confidence: 99%

On–off pumping for drag reduction in a turbulent channel flow

et al. 2023

View full text Add to dashboard Cite

We show that the energy required by a turbulent flow to displace a given amount of fluid through a straight duct in a given time interval can be reduced by modulating in time the pumping power. The control strategy is hybrid: it is passive, as it requires neither a control system nor control energy, but it manipulates how pumping energy is delivered to the system (as in active techniques) to increase the pumping efficiency. Our control employs a temporally periodic pumping pattern, where a short and intense acceleration (in which the pumping system is on) followed by a longer deceleration (in which the pumping system is off) makes the flow alternately visit a quasi-laminar and a turbulent state. The computational study is for a plane channel flow, and employs direct numerical simulations, which present specific computational challenges, for example the highly varying instantaneous value of the Reynolds number, and the importance of discretisation effects. Particular care is devoted to a meaningful definition of drag reduction in the present context. The ability of the forcing to yield significant savings is demonstrated. Since only a small portion of the parameter space is investigated, the best performance of the control technique remains to be assessed.

show abstract

“…We perform this sectional estimation capitalizing on a convolutional neural network (LeCun et al, 1998). The CNN has recently been identified as one of the promising tools for data-driven fluid flow analyses including state estimation (Fukami et al, 2019a(Fukami et al, , 2021aKobayashi et al, 2021), flow control (Lee et al, 1997;Rabault et al, 2019), reduced-order modeling (Murata et al, 2020;Hasegawa et al, 2020b,a;Fukami et al, 2020c;Kim and Lee, 2020;Nakamura et al, 2021b;Maulik et al, 2021), and turbulence modeling (Fukami et al, 2019b;Duraisamy et al, 2019;Lapeyre et al, 2019;Pawar et al, 2020;Thuerey et al, 2020;Font et al, 2021), thanks to a filter operation inside the CNN (Fukami et al, 2020a).…”

Section: Convolutional Neural Network-based State Estimator For Fluid...mentioning

confidence: 99%

Robust training approach of neural networks for fluid flow state estimations

Nakamura,

Fukagata

2021

Preprint

View full text Add to dashboard Cite

State estimation from limited sensor measurements is ubiquitously found as a common challenge in a broad range of fields including mechanics, astronomy, and geophysics. Fluid mechanics is no exception -state estimation of fluid flows is particularly important for flow control and processing of experimental data. However, strong nonlinearities and spatio-temporal high degrees of freedom of fluid flows cause difficulties in reasonable estimations. To handle these issues, neural networks (NNs) have recently been applied to the fluid flow estimation instead of conventional linear methods. The present study focuses on the capability of NNs to various fluid flow estimation problems from a practical viewpoint regarding robust training. Three types of unsteady laminar and turbulent flows are considered for the present demonstration: 1. square cylinder wake, 2. turbulent channel flow, and 3. laminar to turbulent transitional boundary layer. We utilize a convolutional neural network (CNN) to estimate velocity fields from sectional sensor measurements. To assess the practicability of the CNN models, physical quantities required for the input and robustness against lack of sensors are investigated. We also examine the effectiveness of several considerable approaches for model training to gain more robustness against the lack of sensors. The knowledge acquired through the present study in terms of effective training approaches can be transferred towards practical machine learning in fluid flow modeling.

show abstract

Prediction of the drag reduction effect of pulsating pipe flow based on machine learning

Cited by 17 publications

References 18 publications

Saving energy in turbulent flows with unsteady pumping

Saving energy in turbulent flows with unsteady pumping

On–off pumping for drag reduction in a turbulent channel flow

Robust training approach of neural networks for fluid flow state estimations

Contact Info

Product

Resources

About