Computations of wind flow over changes in surface roughness

The equations of motion for two-dimensional mean wind flow over single changes of surface roughness are solved numerically, using a scheme which has been adapted from computations of wind flow over hills. A simple eddy-viscosity/mixing-length closure is used which adequately models the mean flow quantities.Results from a number of computations of flow over different roughness changes are presented, and empirical formulae found which describe the growth of the new boundary layers. Downwind surface shear stress variation is also found to follow very closely a semi-empirical non-dimensional form based on the predictions of Townsend's theory. Results for flows over double roughness changes (with roughness far upwind and downwind being equal) are also obtained, and regions of influence for such flows are examined.Finally, empirical formulae are given for the interpolation of velocities and stresses within the non-equilibrium regions of the flow.     

Calculation of wind-driven cross ventilation in buildings with large openings

A simplified macroscopic method is commonly used for wind-driven ventilation analysis of buildings with small openings. Consequently, it is reasonable to question if and under what conditions will this method provide accurate results in predicting ventilation flow rates in buildings with large openings. We investigate a single-zone cubic building with two equal large openings using a computational fluid dynamics approach. We analyzed the driving forces and the ventilation flow rates due to wind as a function of the geometry, size and relative location of the two openings. The ventilation flow rates are found to be affected by both wind flows around and through the building when the two openings are relatively large. The simplified macroscopic method can provide reasonable engineering accuracy (i.e., less than 10% error) when the porosity of the building envelope does not exceed a critical value. This critical value is not a constant; instead it depends significantly on the degree of alignment between the wind direction and the character of the dominant stream tube associated with the flow through the room. We found that the simplified macroscopic method fails to provide acceptable accuracy when this stream tube is truly dominant and parallel to the wind direction. The effects of wall thickness and aspect ratio of openings are also investigated.     

Problems with the Static Regain method

Many attempts have been made to devise a simple yet accurate algorithm for duct design. The Static Regain Method, which is one attempt, has gained wide popularity due to its simplicity and its seemingly sound theoretical basis. However, a paper by authors Tsal and Behls challenged many long held ideas concerning this method. In this paper some of their arguments are critically reviewed.The Static Regain method equates the static pressure increase across a junction with the pressure drop in the succeeding section. It uses the Bernoulli equation to compute the static pressure regain across a junction. However, Tsal and Behls argued that the Bernoulli equation may not be used for this purpose since it is only valid for streamline flow. Based on the energy equation, Tsal and Behls derived an expression for the static pressure regain across a junction. In this paper it is proposed that the derivation was over-simplified since the energy equation was applied between two junction nodes. Instead it should have been applied to an appropriate control volume.One of the popular applications of the Static Regain method is to create equal static pressures at successive junctions along a main duct. It is claimed that this will result in equal branch flows for identical branches. To disprove this claim Tsal and Behls considered the change in total pressure in a system and showed that irrespective of the static pressures equal flows cannot be attained since total pressure balancing is violated. Once again it is suggested in this paper that their proof is based on an over-simplified view of the problem. It is shown that under certain circumstances it is indeed possible to attain equal flows in identical taps although equal static pressures do not necessarily have anything to do with it.Based on the balancing requirement, a method is proposed purely to demonstrate that when equal static pressures are created at take-off points; it does not follow that equal flows in identical branches will be the result. A realistic duct design problem was solved by using both this method and the Static Regain method. In general only the proposed method succeeded in designing a system that ensures exact total pressure balancing at the design airflows. A duct section is considered here to be a length of duct which has no change of cross sectional area, shape, flow or friction coefficient.     

Flow autocorrelation: a dyadic approach

The paper proposes and investigates a new index of flow autocorrelation, based upon a generalization of Moran’s I, and made of two ingredients. The first one consists of a family of spatial weights matrix, the exchange matrix, possessing a freely adjustable parameter interpretable as the age of the network, and controlling for the distance decay range. The second one is a matrix of chi-square dissimilarities between outgoing or incoming flows. Flows have to be adjusted, that is their diagonal part must first be calibrated from their off-diagonal part, thanks to a new iterative procedure procedure aimed at making flows as independent as possible. Commuter flows in Western Switzerland as well as migration flows in Western US illustrate the statistical testing of flow autocorrelation, as well as the computation, mapping and interpretation of local indicators of flow autocorrelation. We prove the present dyadic formalism to be equivalent to the “origin-based” tetradic formalism found in alternative studies of flow autocorrelation.     

Flow mechanisms and flow capacity in idealized long-street city models

It is an open question whether a street network of a city has a certain flow capacity characterizing the flow that can pass through the street network. It s our hypothesis that at least the simple street network has a certain flow capacity. With the purpose of exploring this we studied numerically and experimentally the flow capacity in some idealized long-street models continuously lined with buildings and exposed to a parallel approaching wind. The height of all the models is the same (H = 69 mm). Three groups of models were studied: models with the same uniform street width (W = H) but different lengths (L = 21.7H, 43.5H, 72.5H); models with the same length (L = 43.5H) but different uniform width (W = H, 2H, 4H); and models with a change of width at half distance, L/2. In the last of the three cases, the width of the upstream half was always the same (W1 = H), but there was a wider (W2 = 1.25H, 1.5H, 2H) or narrower (W2 = 0.75H, 0.5H) downstream half. We normalized flow rates by a reference flow rate equal to the flow rate through an opening far upstream with the same area as the windward entry. The normalized flow rate through the windward entry was about 1.0 in all cases. For a sufficiently long-street models, a flow balance is established, creating a fully developed region with a constant horizontal flow (flow capacity) and zero vertical mean velocity. The street length does not affect the flow capacity but as expected the width of the street affects the flow capacity.     

The horizontal flow of buoyant gases toward an opening

Well-established equations exist for calculating mass flow rates of thermally buoyant gas layers out of openings. The present work presents an alternative method of deriving such an expression in two-dimensional flows for non-uniform approach velocities induced by the layer's buoyancy. The method makes a number of assumptions, leading to the application of the total energy equation at a virtual vena contracta outside the opening.

The resulting mass flow formula is quantitatively equal to the established formula for a wide exit and C_d = 0.6, i.e., for a deep downstand, and so agrees with published data as well as does the existing formula. The two formulae diverge for different values of C_d and for gas flows much hotter than 300 °C.

Since the new formula gives more pessimistic (i.e., larger) mass flow rates for wide openings with no downstand (and hence larger extract fans or vent areas in the smoke reservoir “downstream” of those openings), it is suggested that it should be used by designers of smoke ventilation systems in such circumstances.

It is also deduced that, for “realistic” buoyant-layer flows, the mass weighted average is typically 0.73 θ_max for θ_max 300 °C.     

Recursive transport flow dynamics with time dependent a priori informationRecursive transport flow dynamics with time dependent a priori information

Two properties of a dynamic network flow model based on a slow process of structural adjustment inspired by principles used in models utilising a priori information is analysed. Initially it is proved that each time period is characterised by monotonically non-increasing transportation costs among flows. Secondly, we analyse the recursive sequence of connected periods. This dynamic sequence is shown to converge to the linear programming solution connected with cost minimisation. Evidently those properties have to be taken into consideration when this class of network flow models is used in forecasting of future transport flows in the process of infrastructure planning. Received: November 1996 / Accepted: November 1997     

Stress-dependent shear wave splitting and permeability in fractured porous rock

It is well known that shear wave propagates slower across than parallel to a fracture, and as a result, a travelling shear wave splits into two directions when it encounters a fracture. Shear wave splitting and permeability of porous rock core samples having single fracture were experimentally investigated using a high-pressure triaxial cell, which can measure seismic shear wave velocities in two directions mutually perpendicular to the sample axis in addition to the longitudinal compressive wave velocity. A single fracture was created in the samples using a modified Brazilian split test device, where the cylindrical sample edges were loaded on two diametrically opposite lines by sharp guillotines along the sample length. Based on tilt tests and fracture surface profilometry, the method of artificially induced tensile fracture in the sample was found to create repeatable fracture surfaces and morphologies. Seismic velocities of the fractured samples were determined under different levels of stress confinement and fracture shear displacement or mismatch. The effective confining stress was varied from 0.5 MPa to 55 MPa, while the fractures were mismatched by 0 mm, 0.45 mm and 1 mm. The degree of matching of the fracture surfaces in the core samples was evaluated using the joint matching coefficient (JMC). Shear wave splitting, as measured by the difference in the magnitudes of shear wave velocities parallel (V_S1) and perpendicular (V_S2) to the fracture, is found to be insensitive to the degree of mismatching of the fracture joint surfaces at 2 MPa, and decreased and approached zero as the effective stress was increased. Simple models for the stress- and JMC-dependent shear wave splitting and fractured rock permeability were developed based on the experimental observations. The effects of the joint wall compressive strength (JCS), JMC and stress on the stress dependency of joint aperture were discussed in terms of hydro-mechanical response. Finally, a useful relationship between fractured rock permeability and shear wave splitting was found after normalization by using JMC.     

Indoor air flow analysis based on lattice Boltzmann methods

The modeling of convective flows based on a 3D lattice Boltzmann approach for low Mach number flows with variable density combined with a large eddy turbulence model is presented. The ability to handle non-Boussinesq density variation problems is depicted for two-dimensional Rayleigh–Bénard convection at a Rayleigh number Ra=800,000.A complex three-dimensional example shows the status of our work with respect to turbulent flow in and around a building, so far without consideration of the energy equation in the full scale 3D case. Integrated within a CAD environment, the spatial geometric model, based on an IFC building product data model, is discretized using a hierarchic data structure. Results are presented for a Reynolds number Re=75,000 computed on a high-performance parallel vector computer.State-of-the-art visualization techniques integrate the simulation results and the CAD model into a virtual reality environment. The VR environment allows also for an interactive analysis of thermal comfort criteria, being demonstrated for an indoor air flow simulation of an open-plan office.     

VENTCF2: an algorithm and associated FORTRAN 77 subroutine for calculating flow through a horizontal ceiling/floor vent in a zone-type compartment fire model

An algorithm and associated FORTRAN 77 subroutine, called VENTCF2, for calculating the effects on two-layer compartment fire environments of the quasi-steady flow through a circular, shallow (i.e. small ratio of depth-to-diameter), horizontal vent connecting two spaces is presented. The two spaces can be either two inside rooms of a multi-room facility or one inside room and the outside ambient environment local to the vent. The flow is determined by consideration of standard orifice-type flows driven by cross-vent pressure differences and, when appropriate, the combined pressure- and buoyancy-driven flows which occur when the density configuration across the vent is unstable, i.e. a relatively cool, dense gas in the upper space overlays a less dense gas in the lower space. The algorithm calculates rates of flow exchange between the two spaces based on previously reported model equations. Characteristics of geometry and the instantaneous environments of the two spaces are assumed to be known and specified as inputs. Outputs calculated are the rates and properties of the vent flow at the elevation of the vent as it enters the top space from the bottom space and/or as it enters the bottom space from the top space. Rates of mass, enthalpy and products of combustion extracted by the vent flows from upper and lower layers of inside room environments and from outside ambient spaces are determined explicitly. VENTCF2 is an advanced version of the algorithm /subroutine VENTCF in that it includes an improved theoretical and experimental basis. The subroutine is completely modular and it is suitable for general use in two-layer, multi-room, zone-type fire model computer codes. It has been tested numerically over a wide range of input variables and the results of some of these tests are described.     

Effect of rotational oscillation upon fluid forces about a circular cylinder

The present work aims at a computational study of the effect of various parameters that influence fluid-force behavior when a circular cylinder is subjected to rotational oscillations. A numerical simulation is conducted using the unsteady form of Reynolds-averaged Navier-Stokes equations combined with the k-ε model of turbulence. The study is carried out to examine the influence of various flow parameters, such as oscillation non-dimensional frequencies (0.1-2), rotational non-dimensional amplitudes (0.25-3) and Reynolds numbers (2000-30,000). Special attention is focused on the resonance condition at lower frequency and the subsequent drag reduction at higher frequency. It is found that the peak value of fluid forces on the circular cylinder increases abruptly when the forcing frequency is closer to the vortex shedding frequency. This is followed by a drag reduction at higher forcing frequencies, which becomes very strong at a non-dimensional oscillation frequency around 1 and at rotational amplitudes larger than 1. The former phenomenon is found to intensify with an increase in Reynolds numbers, but the latter almost preserves its strength in the range of Reynolds numbers considered here.     

Recent developments in pedestrian flow theory and research in Russia

Predtechenskii and Milinskii's seminal work [Predtechenskii VM, Milinskii AI. Planning for foot traffic flow in buildings. Revised and updated edition. Moscow: Stoiizdat; 1969] in relation to pedestrian flows is well known. However, analysis of the experimental results and observations obtained from this series of experimental studies revealed the inherent statistical non-homogeneity of pedestrian flow speeds [Kholshevnikov VV. The study of human flows and methodology of evacuation standardisation. Moscow: MIFS; 1999]. As such, the results of these individual experiments cannot be integrated to produce a valid general expression V=f(D) for each type of pedestrian flow path, where V is the flow velocity and D is the flow density. This paper presents further pedestrian flow research conducted in Russia post 1969.     

Wind conditions and ventilation in high-rise long street models

We regarded high-rise cities as obstacles and channels to wind. We first studied wind conditions and ventilations in idealized high-rise long street models experimentally and numerically with a constant street width (W = 30 mm), variable street heights (H = 2 W, 2.5W, 3W, 4W), variable street lengths (L = 47.4W, 79W, 333W, 667W) and a parallel approaching wind. The flow rates penetrating into windward entries are a little larger than the reference flow rate in the far upstream free flow through the same area with windward entries in all models. The stream-wise velocity decreases along the street as some air leaves upwardly across street roofs. Near the leeward entry, there is a downward flow which brings some air into the street and results in an accelerating process. In the neighborhood scale long streets (L = 47.4W and 79W), wind in taller streets is stronger and the ventilation is better than a lower one. For the city scale long streets (L = 333W and 667W), a constant flow region exists where the vertical velocity is zero and the stream-wise velocity remains constant. In such regions, turbulent fluctuations across the street roof are more important to air exchange than vertical mean flows. In a taller street, the process to establish the constant flow conditions is longer and the normalized balanced horizontal flow rate is smaller than those in a lower street. In the city scale long streets, the turbulence exchange rate can be 5–10 times greater than the mean flow rate.     

Investigation of flow properties in natural streams using the entropy concept

This paper examines the discharge and velocity distributions in natural open channel flows using the entropy theory. Flow measurements were carried out at four different cross‐sections in central Turkey. The mean and maximum velocities at these stations exhibited a linear distribution and the entropy parameter was calculated to be M=1.31. Using this value, discharges for all flow conditions were calculated as a function of the measured maximum velocities (u_max). It was observed that the u_max/H and z_max/H ratios remained relatively constant when 0.2≤y/T≤0.8, especially for the wider channels. Using these constant values for each station, u_max and z_max could be determined solely as a function of the water depth H. Although the calculated velocities were higher than those measured at some verticals, the entropy‐based approach presents an attractive alternative to the traditional flow‐measurement techniques for the determination of flow properties because of its simplicity and quick application.     

Cathedral Mountain debris flows,Canada

Historic debris flow activity along the north side of Cathedral Mountain in the southern Rocky Mountains of British Columbia, began in 1925 and has increased in frequency up to 1985. A typical debris flow event involves approximately 100,000 m³ of material. Debris flow velocities and discharges above the head of the fan crossed by the Trans-Canada Highway and the C.P.R. mainline are 5.5 m/sec and 210 m³/sec. Most of the large debris flow events are associated with jökulhlaups from Cathedral Glacier. Jökulhlaup discharges of at least 10,000 and perhaps as much as 24,000 m³ of water mobilize these debris flows. Part of the water may have come from a small ephemeral lake on the south side of the glacier. The balance must have been stored within the glacier. The onset and acceleration of debris flow activity was apparently induced by the recession of Cathedral Glacier. Source areas of debris flow sediments have retreated upslope since initiation of debris flow activity. C.P.R. began pumping meltwater from the glacier in 1985 and no jökulhlaups or significant debris flows have occurred since. This preventive measure should either eliminate jökulhlaups or reduce their magnitudes should they occur. Without jökulhlaups, debris flow hazard in the area should be reduced both in frequency and in magnitude.     

CFD modelling of natural displacement ventilation in an enclosure connected to an atrium

Computational fluid dynamics (CFD) is used to investigate buoyancy-driven natural ventilation flows in a single-storey space connected to an atrium. The atrium is taller than the ventilated space and is warmed by heat gains inside the single-storey space which produce a column of warm air in the atrium and drive a ventilation flow. CFD simulations were carried out with and without ventilation openings at the bottom of the atrium, and results were compared with predictions of analytical models and small-scale experiments. The influence of key CFD modelling issues, such as boundary conditions, solution controls, and mesh dependency were investigated. The airflow patterns, temperature distribution and ventilation flow rates predicted by the CFD model agreed favourably with the analytical models and the experiments. The work demonstrates the capability of CFD for predicting buoyancy-driven displacement natural ventilation flows in simple connected spaces.     

Mechanism of Bearing Capacity of Spread Footings Reinforced with Micropiles

The Hyogoken-Nambu Earthquake in 1995 caused extensive damages to the foundations of bridges. Ever since, methods to improve the bearing capacity of existing foundations have become an important aspect of foundation engineering in Japan. Micropiles are considered to provide promising solutions. The mechanism which enhances the bearing capacity of surface footings reinforced with micropiles is the subject of investigation in this study. As an initial phase, model tests were conducted to understand the load-displacement behavior of surface footings with and without micropiles on loose, medium dense, and dense layers of sands. Salient factors which influence the behavior of the footings were selected and their influence on bearing capacity was examined through a comprehensive series of model tests. Notable improvements in the bearing capacity of surface footings reinforced with vertical micropile groups were observed in the case of dense sand which is dilative during shear. To assess quantitatively the degree of improvement in the bearing capacity of surface footings reinforced with micropiles, an index R called “Network Effect Index” was introduced in this study. The index R of unity means that the bearing capacity of footings reinforced with micropiles is simply equal to the summation of the individual value of the surface footing and that of the micropile group. An index R of more than two is achieved in this study where surface footings reinforced with a group of vertical micropiles bear on a dense layer of dilative sand. By contrast, with loose and medium dense sand, which are contractive in nature, the index R is found to be less than unity.     

Long-term trends in hydro-climatology of a major Scottish mountain river

The River Dee, in North East Scotland, is a mountainous river strongly influenced by patterns of snow accumulation and melt from the Cairngorm Mountains. Analysis of this river's flow record from 1929-2004, the longest in Scotland, supports anecdotal evidence that river extreme flows are increasing. There was no detectable change in the overall annual flow patterns. However, an analysis of seasonal data suggested a shift towards increased flows in spring (March-May) and decreased flows in summer (June-August) over the 75 years of the record. Flows in spring exceeded 29 m³ s^− 1 for 50% of the time over the earliest part of the record (1930 to 1954), whereas in the last 25 years of the record (1979 to 2004) 50% of the flows exceeded 35 m³ s^− 1. Precipitation is increasing in the spring and decreasing in July and August. If these trends continue they have important implications for water management in the Dee, with a potential increase in flood risk in spring and the increased possibility of drought in summer. Combined with this increase in flows the river appears to be more responsive to precipitation events in the catchment. In large heterogeneous catchments with a marginal alpine/high latitude climate it is difficult to assess the amount of precipitation falling as snow and its relative accumulation and ablation dynamics on daily to seasonal time scales. Changes in the temporal pattern of coherence between flow and precipitation are thought to be linked to changing snow patterns in the upland part of the catchment. A decreased amount of precipitation occurring as snow has led to higher coherence. We also show that in responsive systems it is important to record river flows at an hourly rather than daily time step in order to characterise peak flow events.     

Experimental and numerical studies of flows through and within high-rise building arrays and their link to ventilation strategy

Urban ventilation implies that wind from rural areas may supply relatively clean air into urban canopies and distribute rural air within them to help air exchange and pollutant dilution. This paper experimentally and numerically studied such flows through high-rise square building arrays as the approaching rural wind is parallel to the main streets. The street aspect ratio (building height/street width, H/W) is from 2 to 5.3 and the building area (or packing) density (λ_p) is 0.25 or 0.4. Wind speed is found to decrease quickly through high-rise building arrays. For neighbourhood-scale building arrays (1-2 km at full scale), the velocity may stop decreasing near leeward street entries due to vertical downward mixing induced by the wake. Strong shear layer exists near canopy roof levels producing three-dimensional (3D) vortexes in the secondary streets and considerable air exchanges across the boundaries with their surroundings. Building height variations may destroy or deviate 3D canyon vortexes and induced downward mean flow in front of taller buildings and upward flow behind taller buildings. With a power-law approaching wind profile, taller building arrays capture more rural air and experience a stronger wind within the urban canopy if the total street length is effectively limited. Wider streets (or smaller λ_p), and suitable arrangements of building height variations may be good choices to improve the ventilation in high-rise urban areas.     

Ventilation system with spiral recuperator

A new kind of longitudinal flow spiral recuperator for the heat recovery in ventilation systems of buildings was studied experimentally and analytically.The aim of this work is to analyze the possibility of using air handling units with new type of spiral recuperator in order to recover heat in ventilation systems of buildings. For the reason that the air flows are parallel to the symmetry axis of the longitudinal flow spiral recuperator, in this unit pressure drops are smaller than in commonly known spiral exchangers. Because of the counter flow, a greater value of thermal effectiveness ? is reached for the same value of the number of transfer units in comparison to cross-flow recuperators.The exploitation of the new type of spiral recuperator in winter periods brings significant savings. The results obtained from computations are very encouraging for a widespread use of these devices.