Wall Units Research Articles

Scaling and mechanism of the propagation speed of the upstream turbulent front in pipe flow

The scaling and mechanism of the propagation speed of turbulent fronts in pipe flow with the Reynolds number has been a long-standing problem in the past decades. Here, we derive an explicit scaling law for the upstream front speed, which approaches a power-law scaling at high Reynolds numbers, and we explain the underlying mechanism. Our data show that the average wall distance of low-speed streaks at the tip of the upstream front, where transition occurs, appears to be constant in local wall units in the wide bulk-Reynolds-number range investigated, between 5000 and 60 000. By further assuming that the axial propagation of velocity fluctuations at the front tip, resulting from streak instabilities, is dominated by the advection of the local mean flow, the front speed can be derived as an explicit function of the Reynolds number. The derived formula agrees well with the speed measured by front tracking. Our finding reveals a relationship between the structure and speed of a front, which enables a close approximation to be obtained of the front speed based on a single velocity field without having to track the front over time.

Flow Field Analysis of a Turbulent Channel Controlled by Scalloped Riblets

Riblets are small protruding surfaces along the direction of the flow, and are one of the most well-known passive turbulent drag reduction methods. We investigated a scalloped riblet, the shape of which was constructed by smoothly connecting two third-order polynomials and was not as sharp in the tip as corresponding triangular riblets with the same height-width ratio. Numerical simulations were performed for turbulent channel flow with and without riblet control at an estimated optimum width of W+=20 and a height-width ratio of 0.5. A drag reduction rate of 5.77% was obtained, which is generally larger than the reported drag reduction rates of corresponding triangular riblets from the literature. Mean flow fields and second-order statistics of velocity, vorticity, and Liutex, a quantity introduced to represent vortices, were reported. It was found that streamwise vortices just above the riblet tips, which have a length scale of 200−300 in wall units, have an important influence on those statistics and thus the turbulence generation cycle and the drag reduction mechanism. Pre-multiplied energy spectra of streamwise velocity and the Liutex component were reported to reveal the length scales in the flow field. Instantaneous vortical flow fields visualized by the Liutex method were provided with emphases on the streamwise vortices just above riblet tips. It should be noted that the class of scalloped riblets is suitable for investigations on the influences of curvatures at the tip and the valley of the riblet in future.

Performance of all-air wall induction unit for displacement ventilation in a four-bed hospital ward during the cooling period

The existing induction air-conditioning system faces ventilation shortcomings and condensation risks. We developed an all-air wall induction air terminal unit (wall IU) that doesn't require reheating and eliminates water pipes to address condensation. Moreover, with modifications to its supply method, it can achieve displacement ventilation (DV) in an adiabatic environment. However, the real-world performance of this system needs to be further confirmed. This study examines wall IU performance during the cooling period. A full-scale experiment was conducted in a room furnished as a four-bed hospital ward with an outdoor climate chamber (OCC) to simulate summer conditions. IU's cooling and ventilation performance was evaluated by examining steady-state temperatures and tracer gas distributions. Moreover, CFD simulations to reproduce the experiment were performed to supplement the results. Patients' thermal comfort and infection risk while cohabiting with an infected individual were estimated using detailed temperature, airflow, and concentration data from an experiment-based CFD analysis. Furthermore, the DV stratification height under various perimeter heat loads and performance comparison between wall and ceiling-mounted induction units were also studied. In conclusion, the wall IU improved ventilation efficiency by achieving DV under cooling supply conditions. Satisfied thermal comfort is ensured based on ADPI and DR evaluation. DV's stratification height can be kept high enough under a 20 °C equivalent temperature difference when using a single glass window or more after thermal insulation enhancement. Compared with ceiling-mounted induction units, wall-mounted units can reduce infection risk by more than double while making cold drafts less noticeable.

Simulation of multiphase flow in pipes with simplified models of deposited beds

Turbulent particle-laden flows in pipes can result in particle deposition leading to the formation of solid beds. The presence of such beds modifies the flow field, resulting in secondary motions in the plane of the pipe cross-section, which in turn impact particle transport. In this work turbulent pipe flows with equal mass flow rates and solid beds of height Hb = 0 (full pipe), 0.5R (three-quarter pipe), and R (half pipe) are predicted using direct numerical simulation, with the beds represented simplistically as flat surfaces. The particulate phase is one-way coupled to the flow at a volume fraction of 10−3 and particle motion is solved for using a Lagrangian point-particle approach. The Reynolds numbers computed based on bulk velocity and equivalent pipe diameter for the full, ¾ and, ½ pipes are 5,300, 5,909 and 7,494, respectively. The same particle size is used in all the simulations and their respective Stokes numbers, based on the shear timescale, are 0.5, 1.2 and 1.9, respectively. The results for flows with beds show that the fluid flow exhibits secondary vortices and an increase in the mean streamwise vorticity caused by corners in the cross-sectional plane of the pipes, with their intensity near the upper curved wall increasing with Hb. However, the upper vortices remain relative weak compared to those in lower regions of the pipes. The increase in mean streamwise vorticity in the half pipe is larger than that in the three-quarter pipe near the upper curved wall, while similar near the flat pipe floor due to the resistance of the curved wall to secondary motions. The movement of the particles in the cross-sectional plane is consistent with that of the secondary flows, but with slightly lower velocities. In regions near the wall away from the pipe corners, particle concentration in the half pipe is lower than in the three-quarter pipe, most likely due to its thinner boundary layer. This is reversed for concentration maxima near the pipe corners because of the magnitude of the secondary flows. Finally, the secondary flow changes the deposition or resuspension rate of the particles, particularly near the pipe corners, but these are always less than equivalent rates in the full pipe flow, which is likely caused by the magnitude of the wall unit.

Self-similar, spatially localized structures in turbulent pipe flow from a data-driven wavelet decomposition

This study aims to extract and characterize structures in fully developed pipe flow at a friction Reynolds number of $\textit {Re}_\tau = 12\,400$ . To do so, we employ data-driven wavelet decomposition (DDWD) (Floryan & Graham, Proc. Natl Acad. Sci. USA, vol. 118, 2021, e2021299118), a method that combines features of proper orthogonal decomposition and wavelet analysis in order to extract energetic and spatially localized structures from data. We apply DDWD to streamwise velocity signals measured separately via a thermal anemometer at 40 wall-normal positions. The resulting localized velocity structures, which we interpret as being reflective of underlying eddies, are self-similar across streamwise extents of 40 wall units to one pipe radius, and across wall-normal positions from $y^+=350$ to $y/R=1$ . Notably, the structures are similar in shape to Meyer wavelets. Projections of the data onto the DDWD wavelet subspaces are found to be self-similar as well, but in Fourier space; the bounds of self-similarity are the same as before, except streamwise self-similarity starts at a larger length scale of $450$ wall units. The evidence of self-similarity provided in this study lends further support to Townsend's attached eddy hypothesis, although we note that the self-similar structures are detected beyond the log layer and extend to large length scales.

Tests on reinforced concrete U-shaped walls subjected to torsion and flexure

This data paper describes an experimental campaign on two reinforced concrete (RC) U-shaped walls, including details of the test units, materials, test setup, loading protocol, instrumentation, primary observations and response of each unit during testing, organization of the test data in the public open source research data repository (DOI: 10.14428/DVN/FDJ4EU), and examples of derived data. The two wall units represent the lower stories of a half-scale 6-story prototype core wall that was designed for medium-to-high ductility. The tests conducted aimed at assessing the flexural and torsional response and capacity of RC U-shaped walls. The first wall unit, UW1, was tested for axial-flexure and subjected to quasi-static reverse-cyclic bending about its weak axis until failure. Furthermore, UW1 was subjected to an additional overturning moment to increase the shear span. The second wall unit, UW2, was tested for axial-torsion and subjected to a quasi-static reverse-cyclic rotation about its vertical axis until failure. Both wall units were subjected to a constant axial load ratio of 5%. A range of conventional instrumentation was used to capture salient global and local measurements of the wall, including three-dimensional (3D) digital image correlation on multiple faces. High-definition fiber-optic sensors were also utilized to capture the entire strain profile of several longitudinal reinforcing bars in each wall unit.

Analysis of thermal solutions in challenging spots of masonry constructions

AbstractAutoclaved aerated concrete (AAC) is a construction material with very good thermal properties. In many countries, the majority of masonry structures are developed using a single wall from low‐density AAC. In some countries, like Poland, the most popular solution is an AAC construction wall with covering insulation material. In the house construction process, several challenging spots can be observed where thermal bridges can develop. One of them will be the place of installation of external window roller blinds on a reinforced concrete lintel or an aerated concrete lintel. In fact, there are several solutions to this problem, but there is no clear answer in the literature about which of those will be optimal in terms of reducing the loss of thermal energy, the cost of which is growing year by year. The research program includes tests of the thermal conductivity of a wall unit built of AAC of various density classes, with the most popular thermal insulation materials available on the market, such as: EPS polystyrene, XPS polystyrene, or aluminum mats.

In-plane measurements and computational fluid dynamics prediction of permeability for biocompatible NiTi gyroid scaffolds fabricated via laser powder bed fusion

Laser powder bed fusion (LPBF) is considered a promising technology for manufacturing porous, biomimetic, and patient-specific scaffolds for bone repair. Scaffold permeability is one of the key factors to be considered for acquiring the required mass-transport properties in bone tissue engineering. This study aims to reveal the relationship between the design parameters of gyroid-based porous structure and scaffold permeability. A set of gyroid samples was manufactured from intermetallic NiTi alloy. Nine configurations of porous structures were obtained by varying the main design parameters, namely wall thickness and unit cell size. The in-plane method was employed to measure the permeability coefficient for the gyroid structures. Computational fluid dynamics simulations of the porous structures were performed to predict the targeted properties in an implant at the design stage before LPBF manufacturing. The results of the simulations were validated with the obtained experimental results. Geometrical accuracy and surface morphology of the as-built samples were investigated with various techniques. Biocompatibility assessment of the gyroid scaffolds was performed with human cell culture experiments. 

Steady Bi-dimensional Crossflow Plasma Jets in Turbulent Channel Flows

In this study, the possibility of reducing the friction drag exerted by turbulent flows by means of wall-mounted plasma actuators is experimentally investigated. Two large plasma actuators (PAs) arrays were operated in a channel-flow facility. They were conceived to replicate, the flow control approach investigated by Mahfoze and Laizet (Int J Heat Fluid Flow 66:83–94, 2017) by means of numerical simulations. Namely, steady and relatively largely spaced (378 wall units) actuators were lain down such to induce stationary crossflow-directed fluid motions. Different actuation parameters (actuators’ configurations and supplied voltages) and flow Reynolds numbers were tested. Flow static pressure measurements were performed along with the actuators mechanical and electrical characterization. The resulting values of drag manipulation and actuation efficiency are reported. The tested flow actuation led to overall higher values of flow friction drag, whereas values overcoming the value of 30% of drag reduction were measured at the more downstream actuation positions. The discrepancy with the above reference is deemed to be mainly due to the finite flow actuation hereby considered. Nevertheless, a slightly different Reynolds number was here considered while the actuators effect was measured to be considerably weaker.

Thermal streak spacing in fully developed duct flow with different Reynolds and Prandtl numbers

The present study investigates the behaviour of thermal streaks on a heated foil which is cooled with turbulent flow in a square duct channel. Real-time infrared thermography is used to visualize and measure the spacing between the thermal streaks. A stainless-steel foil with a thickness of 25 microns is cooled by water. The experiments were performed in a range of Reynolds numbers from 5000 to 20000 and Prandtl numbers from 3 to 7. The mean temperature, root-mean-square of temperature and autocorrelation function have been calculated and used to measure the average thermal streak spacing and power spectra in the spanwise and streamwise directions. The root mean square temperature was 0.3 °C to 0.5 °C which corresponds to roughly 10% of the mean temperature difference between foil and water. The uncertainty in mean temperature difference and root mean square temperature was around 5% and 10%, respectively. The measured thermal streak spacing was 100 wall unit to 180 wall unit under the present experimental range. The uncertainty in measured thermal streak spacing was around 2.5%. The effects of Reynolds number, Prandtl number and heat flux on the thermal streak spacing and also on the statistics of the temperature field have been presented and discussed in this paper. A new correlation has been proposed to predict the dimensionless thermal streak spacing. The error in the prediction is estimated within ± 15 %.

Influence of Flow Configuration and Thermal Wall Boundary Conditions on Turbulence During Premixed Flame-Wall Interaction within Low Reynolds Number Boundary Layers

The influence of flow configuration on flame-wall interaction (FWI) of premixed flames within turbulent boundary layers has been investigated. Direct numerical simulations (DNS) of two different flow configurations for flames interacting with chemically inert isothermal and adiabatic walls in fully developed turbulent boundary layers have been performed. The first configuration is an oblique wall interaction (OWI) of a V-flame in a turbulent channel flow and the second configuration is a head-on interaction (HOI) of a planar flame in a turbulent boundary layer. These simulations are representative of stoichiometric methane-air mixture under atmospheric conditions and the non-reacting turbulence for these simulations corresponds to the friction velocity based Reynolds number of Re_{tau }=110. It is found that the mean wall shear stress, mean wall friction velocity and the mean velocity statistics are affected during FWI and the behaviour for these quantities varies under the different flow configurations as well as for the different thermal wall boundary conditions. The behaviour of the quenching distance and mean wall heat flux under isothermal wall conditions is found to be significantly different between the two flow configurations. The variation of the non-dimensional temperature in wall units for cases with isothermal walls suggests that the temperature in the log-layer region is significantly altered by the evolving wall heat flux in both flow configurations. Statistics of the mean Reynolds stresses and turbulence dissipation rate show that the flame significantly alters the behaviour of turbulence due to thermal expansion effects and flow configuration plays an important role.

Backflow structures in turbulent pipe flows at low to moderate Reynolds numbers

The statistical characteristics and the evolution of the backflow structures are investigated in wall-bounded flows at Reynolds numbers up to $Re_{\tau }=1000$ . The backflow is caused by the joining of large-scale high- and low-speed structures in the vicinity of the wall and is formed at the tail tip of the low-speed structure. The distribution density of the backflow structures and the percentage area of the backflow region on the wall both increase with the Reynolds number. The backflow structures have an average lifespan of 8 wall units which is found to be slightly longer in the pipe than the channel, and they are convected downstream at the average velocities of the buffer region of approximately 10 wall units, similar to Cardesa et al. (J. Fluid Mech., vol. 880, 2019, R3). The backflow structures occasionally split and merge, and can form detached from the wall. Evidence shows that the split, merged and wall-detached backflow structures are caused by the near-wall structures. The split backflow structures are on average, larger and more spanwise-elongated which are split due to the spanwise shearing of the near-wall streaks. A backflow structure is formed detached from the wall when the trailing end of its carrier low-speed structure ‘sits’ on the near-wall high-speed streaks. The wall-detached backflow structures tend to become wall-attached by approaching the wall when undergoing a similar life cycle to the normal backflow of growth and decay with spanwise elongation because the backflow region at the tail of the low-speed structure is continuously pressed down to the wall by the high-speed structure driven by persistent vortical structures in the buffer region.

Study of compressible turbulent plane Couette flows via direct numerical simulation

Compressible turbulent plane Couette flows are studied via direct numerical simulation for wall Reynolds numbers up to $Re_w=10\ 000$ and wall Mach numbers up to $M_w=5$ . Various turbulence statistics are compared with their incompressible counterparts at comparable semilocal Reynolds numbers $Re^*_{\tau,c}$ . The skin friction coefficient $C_f$ , which decreases with $Re_w$ , only weakly depends on $M_w$ . On the other hand, the thermodynamic properties (mean temperature, density and others) strongly vary with $M_w$ . Under proper scaling transformations, the mean velocity profiles for the compressible and incompressible cases collapse well and show a logarithmic region with the Kárman constant $\kappa =0.41$ . Compared with wall units, the semilocal units yield a better collapse for the profiles of the Reynolds stresses. While the wall-normal and spanwise Reynolds stress components slightly decrease in the near-wall region, the inner peak of the streamwise component notably increases with increasing $M_w$ – indicating that flow becomes more anisotropic when compressible. In addition, the near-wall turbulence production decreases as $M_w$ increases – due to rapid wall-normal changes of viscosity caused by viscous heating. The streamwise and spanwise energy spectra show that the length scale of near-wall coherent structures does not vary with $M_w$ in semilocal units. Consistent with those in incompressible flows, the superstructures (the large-scale streamwise rollers) with a typical spanwise scale of $\lambda _z/h\approx 1.5{\rm \pi}$ become stronger with increasing $Re_w$ . For the highest $Re_w$ studied, they contribute about $40\,\%$ of the Reynolds shear stress at the channel centre. Interestingly, flow visualization and correlation analysis show that the streamwise coherence of these structures degrades with increasing $M_w$ . In addition, at comparable $Re^*_{\tau,c}$ , the amplitude modulation of these structures on the near-wall small scales is quite similar between incompressible and compressible cases – but much stronger than that in plane Poiseuille flows.

Intermittency and critical mixing in internally heated stratified channel flow

Through direct numerical simulations we investigate the effects of spatiotemporal intermittency as a result of stable stratification in surface heated stratified open channel flow. By adapting the density inversion criterion method of Portwood et al. [J. Fluid Mech., vol. 807, 2016, R2] for our flow, we demonstrate that the flow may be robustly separated into regions of active turbulence for which $Re_B \gtrsim {O}(10)$ and surrounding quiescent fluid, where $Re_B$ is the buoyancy Reynolds number. The intermittency in the flow spontaneously manifests as a deformed horizontal interface between the upper quiescent and lower turbulent flow, characterised by vigorous mixing from ‘overturning’ shear instabilities. The resulting vertical intermittency profile is accurately predicted by a local Monin–Obukhov length normalised by viscous wall units $\varLambda ^+$ such that the flow displays intermittency within the parameter range of $2.5 \lesssim \varLambda ^+ \lesssim 260$ . By considering conditional averages of the ‘turbulent’ and ‘quiescent’ flow separately, we find the ‘turbulent’ flow within this region to be described by constant critical gradient Richardson and turbulent Froude numbers of $Ri_{g,c} \approx 0.2$ and $Fr_c \approx 0.3$ . We find that the turbulent flow continues to display a $\varGamma \sim Fr^{-1}$ relationship when $Fr < Fr_c$ , whereas the quiescent flow shows no correlation between $\varGamma$ and $Fr$ , where $\varGamma$ is the flux coefficient. Hence, we demonstrate directly that for our flow, the emergence of an asymptotic ‘saturated’ $\varGamma$ regime in the limit of a low ‘global’ $Fr$ occurs due to intermittency and increasing contributions to measurements of $\varGamma$ from the quiescent flow.

OC-017 Robotic abdominal wall program surgery: Initial experience of our first 101 cases

Abstract Aim Minimally invasive surgery in abdominal wall had spread exponentially in the last decade, specially with the robotic surgery appearance. In our hospital, the Abdominal Wall Unit started the robotic program in 2019. The experience of our team in laparoscopic surgery allowed us to introduce the robotic surgery in complex abdominal wall pathologies. The aim of this study is to describe our experience of the progressive evolution that we had in our first 101 cases Material and Methods Patients with ventral/incisional hernias with medium and high complexity with no loss of domain, parastomal and inguinal hernias. The surgeries were performed by X or Xi DaVinci robotic platform. Result 101 patients (115 surgeries) were considered in our study. All of them underwent robotic abdominal wall surgery. Women were 46 (45,5%) and men 55 (54,5%). The mean IBM was 34 (±3,7). The main hernia was in the midline: 53 (52,5%), the more frequent techniques were eTEP 32 (31,7%), RoboTAR (intrabdominal transversus abdominis muscle release) 19 (18,8%), bilateral inguinal TAPP 16 (15,9%) and Pauli 15 (14,9%). Also, we performed 11(10,9%) patients with our protocol with botulinum toxin called Robotox decreasing the TAR procedure. The RoboTAR had the most operating time mean: 315 min (260–350) while eTEP had 225 min (180–269). 4 (3,9%) patients had a recurrence with a follow-up of 26 months (3–37). Conclusion The implementation of a robotic program in an abdominal wall surgery unit is possible but it is necessary to standarize a training program to improve progressively the skills.

OC-018 Elective recurrent inguinal hernia repair. Value of an Abdominal Wall Surgery Unit

Abstract Purpose The aim of this study is to analyze the influence of an abdominal wall surgery unit in postoperative complications (within postoperative 90 days), hernia recurrence and chronic postoperative inguinal pain after elective recurrent inguinal hernia repair. Methods This retrospective cohort study included patients who underwent elective recurrent inguinal hernia repair between January 2010 and October 2021. The short- and long-term outcomes between the groups of patients operated on by the Abdominal Wall Surgery Unit and groups operated on by surgeons outside of the specialized abdominal wall group were compared. A logistic regression model was performed for hernia recurrence. Results A total of 250 patients underwent elective surgery for recurrent inguinal hernia during the study period. Patients in the abdominal wall surgery group were younger (P= <0.001) and had less comorbidities (P= <0.001). Complications were similar in both groups. Patients in the abdominal wall surgery group presented fewer recurrences (15% vs 3%; P= 0.001). In the multivariate analysis, surgery performed by the abdominal wall surgery unit was related to fewer recurrences (HR= 0.164; 95% CI= 0.44–0.613; P=0.007). Conclusions Recurrent inguinal hernia repair is associated with better results in terms of recurrence when it is performed by a specialized abdominal wall unit.

Метод определения коэффициента эффективной теплопроводности пористого материала на основе минимальной поверхности типа Schoen’s I-WP(R)

Currently, to develop the materials with predictable properties at the macrostructural level, triply periodic minimum surfaces (TPMS) are used. The production of materials with an ordered structure of TPMS has become available due to the active development of additive technologies. Such materials have high strength-weight ratio, which is important in many structural problems. The study of the thermophysical properties of materials is necessary for the further design of various types of thermal insulation, heat exchangers, etc. In this regard, the study of the thermophysical properties of materials with an ordered structure based on TPMS is a topical issue. The paper proposes to study the thermophysical properties of materials with an ordered structure of Schoen’s I-WP(R) TPMS. Using the ANSYS software package (Steady-State Thermal module), numerical simulation of the heat transfer process in the material under study is carried out. The study is carried out for a material common in additive technologies, that is PETG plastic. The article presents the results of a study of the thermophysical properties of a material with an ordered macrostructure based on Schoen’s I-WP(R) triply periodic minimal energy surfaces. Based on the simulation results, graphical and analytical dependences of the heat flux density and effective thermal conductivity of the material are obtained. The results of the heat flux density are obtained in different directions with variable geometric parameters of the structure (the thickness of the cell wall and the length of the edge of the cube in which the cell is inscribed). Using the ANSYS software package, numerical simulation of the heat transfer process in the material under study is performed. The calculation results show a linear dependence of the effective thermal conductivity of the homogenized material on the wall thickness of the unit cell. It is shown that the intensity of heat transfer depends not only on the wall thickness and unit cell size, but also on the direction of the heat flux. The obtained results of the study can be used to create materials with predictable thermal conductivity by changing the dimensions (cell wall thickness and cube edge length) and high strength-weight ratio. The production of the material is possible with the help of additive technologies.

잔향가변장치 적용에 따른 다목적 음악연습실에서 음환경 변화특성 평가

This study investigated the practical effects of the reverberation variable device on controlling reverberation time of the sound fields in a multi-purpose music practice room. The surface of the reverberation variable device can be changed to absorb or reflect the incident sounds by the integrated electronic control system. The developed devices was installed on the surfaces of a music practice room having a length and width of 13.5 m and 3.5 m, respectively, for practical verification. The reverberation variable devices in a total of 96 units were employed in the room with a surface area of 78 m2. Especially, 50 units and 46 units of device were installed on the walls and the ceiling, respectively. As a result, the maximum changed reverberation time by frequency bands was measured as 0.15 s at 125 Hz, 0.63 s at 250 Hz, 0.59 s at 500 Hz, 0.37 s at 1000 Hz, and 0.3 s at 2000 Hz. The reflection mode with all the reverberation variable devices in closed state showed the highest reverberation time of 1.29 s at 500 Hz. The sound absorption mode with all the reverberation variable devices in opened state showed the lowest reverberation time of 0.7 s at 500 Hz. According to the partial control, the reverberation time at 500 Hz was 0.89 s when half of the wall units were opened, and 0.81 s when the half of the ceiling units were opened. Additionally, the design application of the developed devices was examined in accordance with ISO 23591.

Reinforcement learning of control strategies for reducing skin friction drag in a fully developed turbulent channel flow

Reinforcement learning is applied to the development of control strategies in order to reduce skin friction drag in a fully developed turbulent channel flow at a low Reynolds number. Motivated by the so-called opposition control (Choi et al., J. Fluid Mech., vol. 253, 1993, pp. 509–543), in which a control input is applied so as to cancel the wall-normal velocity fluctuation on a detection plane at a certain distance from the wall, we consider wall blowing and suction as a control input, and its spatial distribution is determined by the instantaneous streamwise and wall-normal velocity fluctuations at distance 15 wall units above the wall. A deep neural network is used to express the nonlinear relationship between the sensing information and the control input, and it is trained so as to maximize the expected long-term reward, i.e. drag reduction. When only the wall-normal velocity fluctuation is measured and a linear network is used, the present framework reproduces successfully the optimal linear weight for the opposition control reported in a previous study (Chung & Talha, Phys. Fluids, vol. 23, 2011, 025102). In contrast, when a nonlinear network is used, more complex control strategies based on the instantaneous streamwise and wall-normal velocity fluctuations are obtained. Specifically, the obtained control strategies switch abruptly between strong wall blowing and suction for downwelling of a high-speed fluid towards the wall and upwelling of a low-speed fluid away from the wall, respectively. Extracting key features from the obtained policies allows us to develop novel control strategies leading to drag reduction rates as high as 37 %, which is higher than the 23 % achieved by the conventional opposition control at the same Reynolds number. Finding such an effective and nonlinear control policy is quite difficult by relying solely on human insights. The present results indicate that reinforcement learning can be a novel framework for the development of effective control strategies through systematic learning based on a large number of trials.

Experimental Study of Bearing Capacity of Steel Frame with Portal Steel Plate Shear Wall Member

The steel frame with portal steel plate shear wall unit presented here is a new steel structure by assembling the portal steel plate shear wall into the steel frame through bolt connection. For study of shear resistance and bearing capacity of the structure, it is necessary to conduct experimental research. Three frame specimens are made here, and two geometric shapes and the case of setting stiffeners and not setting stiffeners are considered respectively. The concentrated lateral load is applied in the test and displacements and strain were measured. The results show that the setting of the shear wall element has an obvious effect on improving the lateral stiffness of the steel frame, and the failure also shows obvious plastic deformation characteristics. It is also seen that the connection has no evident failure or deformation, which means the connection construction is effective.