PIV粒子测速技术在污水排海管道内流研究中的应用

吴 玮

（苏州科技学院，江苏苏州 215003）

0 Introduction

Saline intrusion into marine sewage outfalls will greatly decrease the efficiency of sewage disposal， especially in some early－constructed structures without special safe－guard measures.During the periods of low sewage discharges，for example，at the initial operation stage of the outfall or at the abrupt interruption of pumping due to some reasons，seawater would intrude into the outfall due to the density difference of seawater and sewage.Saline wedge would block some of the risers and seawater would circulate in the outfall.Under this condition，sediment would deposit and marine organisms would grow on the inner walls of the outfall pipe，that would result in blockage and decrease the efficiency of the structure［1－5］.

For many years researchers had paid more attention to the study of buoyant jet of sewage flow［3－6］.Saline intrusion in long sea outfalls was not investigated until 1970s.The theoretic analysis was limited due to the complex mechanisms of this flow.So for a long time，experimental observation was the main means for the study and engineering design［7－8］.From 1980s，numerical simulation was investigated.In recent years，many researchers have made significant contributions towards this goal［12－17］.

For the experiment，of course，spot observation was impractical for the most of the engineerings.So in most cases，the laboratory experiment was used to optimize the structural design.In addition，the validation of the numerical model and the discussion of the parameters of the numerical model should resort to the laboratory experiment.

For the laboratory experiment，measuring technique was an important factor besides the similarity principle and scale effect.Due to the limitations of the measuring technique，the early experiments were major at qualitative observation pipes.The physical parameters measured in the experiment were critical discharges for saline intrusion and purging，geometric parameters of saline wedge in the outfall pipes.In the early experiment studies some macroscopic results such as the critical conditions of saline intrusion and purging，the effect of structural design［18－26］were achieved.However，the internal flow details could not be realized comprehensively.In 2002，the researchers in the University of Belfast measured the internal flow with Laser Doppler Velocimetry（LDV）and set up a two－dimensional numerical model［14－15］.The LDV was used to measure the single－point velocity but not flow field.The multi－points velocities could not be measured synchronously.In order to understand the internal flow field the experiment should be repeated for many times to achieve velocities at different points.

In this paper，the Particle Image Velocimetry（PIV）was used to measure the internal velocityfields of the outfall pipes.The PIV technique was not interfering with the flow field and could measure flow field synchronously.

1 Experimental study

In order to present the flow characteristics in the outfall at different inflowing sewage discharges，three series of velocity data were measured with PIV system.The development of flow patterns at different inflowing sewage discharges and the mechanisms of saline intrusion and purging were analyzed.

1.1 Design of the experimental system

An experimental system was established in the laboratory.For the experiment a simplified outfall including a main pipe and three risers were made of Perspex with equivalent roughness height of 0.001mm.The main pipe was 5mlong and had a circular cross section with radius of 0.075m.The risers had a circular cross section with radius of 0.025m.The distance of adjacent risers was 0.45m.Each riser had a height of 0.4m，which was the distance between the center of the main pipe and the exit of the riser.

The receiving saline water was modeled with an overflowing flume to keep the water level steady.Two pieces of partition plates were set in the Plexiglas tank between the risers to construct the independent receiving water region for each riser.

For the region size of the receiving water there were three points that should be considered.Firstly，the distance of the receiving water surface with the exit of the risers should be large enough that the saline intrusion could take place.Secondly，the receiving water surface should keep steady by overflowing.Besides，the range of the receiving water should be large enough to make sure that the boundary of the flume would not affect the discharge flow of the risers.In this experiment，the surface of the receiving water was 0.275mabove the exit of the risers.

The experimental system was shown in Fig.1.

Fig.1 Schematic diagram for the experimental system（Unit：mm）图1 实验装置示意图（单位：mm）

1.2 Measuring system

The velocity－field was measured with PIV system made by TSI Corporation.The basic principle of the PIV technique was shown in Fig.2.The tracking particles were released into the flow field.And then the measured section was illuminated with laser sheet.In this experiment，the laser illumination was installed at the right side of the outfall system.The laser sheet was generated by the laser illumination and projected into the centre－line cross section of the outfall system.The particle images of the measured section at different time were photographed by the camera.The velocities were calculated through correlation analysis of the two pieces of particle images.

Fig.2 Basic principle of PIV图2 PIV量测原理图

The laser illumination was double－pulse Nd：YAG system with the maximum laser intensity of 200mJ and the maximum frequency of 15Hz.The image－forming system was 28mm FL D／2.8Nikon variable focal camera lens.The format of image was 10bit.The effective measurement scope of single camera was 350mm×260mm.The time interval between two images of single camera could be adjusted according to the flow velocity of the measured range.The images were dealt with FFT cross－correlation algorithm with inquiry window of 64×64pixels.

1.3 Points for attention

Firstly，the cross sections of the main pipe and the risers were circular.The particle images would be deformed due to the refraction by different mediums inside and outside of the circular pipe（liquid and air）.To eliminate the effect of this refraction，a rectangular glass water tank full of clear water was designed.The measured section of the outfall pipe was put into this water tank and submerged into the water.

During the process of saline intrusion or purging，there would be two kinds of liquids（saline water and clear water）in the outfall pipe.The different refraction of these two kinds of liquids would have some effect on the particle images.However，this effect was not obvious and was ignored in this experiment.

Secondly，the big air bubbles staying in the outfall pipe would scatter the laser.This would greatly decrease the quality of the particle images.So in the experiment，measures should be taken to exhaust the air bubbles from the measured scope of the outfall pipe.

Thirdly，to avoid the effect of the background light，the experiment should be taken in the darkroom.

1.4 Experimental conditions

The aim of the experiment was to build up profiles of saline wedge formation and saline intrusion within the simplified model outfall.The initial condition during these tests was a fully intruded outfall thus the system was full of saline water.And then the sewage discharged into the outfall.Monitoring was carried out continuously from this initial state，through the transient stage until a steady state was achieved.During this process，the development of the flow field within the outfall was measured.

The sewage was modeled with clear water with density of 998kg／m3.The receiving saline water was modeled with sodium chloride solution with density of 1021kg／m3.The relative density difference between the two fluids was 2.3%.

The range of flow discharge in the experiment was selected according to the following two points.Firstly，the flow pattern should keep turbulent.In the experimental flow，the Reynolds number was 1800.Preliminary tests showed that the internal flow in the outfall would be turbulent conspicuously when the Reynolds number was 1800.Secondly，the range of the flow discharge should be sure that saline intrusion would occur at lower discharge and saline purging would occur at higher discharge.The flow discharges in the experiment were shown in Table.1.

Table.1 Experimental conditions表1 实验工况表

2 Experimental results

Measurements of velocities were taken at the centre－line cross section of the outfall.The measured range in the experiment was shown in Fig.3.

Fig.3 Measured range图3 量测区域示意图

Fig.4 Particle image in the experiment图4 实验中的粒子图像照片

The flow characteristics in the outfall with different inflowing sewage discharges were analyzed.With the discharges of 0.2×10－3m3／s and 0.6× 10－3m3／s，the flow patterns in the outfall were similar.Saline intrusion occurred in the outfall.With the discharge of 0.72×10－3m3／s，the flow pattern turned into saline purging.

The variation of the flow patterns could be expressed by the flow velocities in the risers.For the convenience of experession，the coordinate system was established as shown in Fig.5.The measured sections located in the centre－line was xz plane in the coordinate system，as shown in Fig.5.The coordinate x0was the distance from the measuring point to the left wall of the corresponding riser.The variable v was the velocity in the direction z.The variable u was the velocity in the direction x.The variable D was the radius of the main pipe.The positive value of v indicated discharging from the riser and the negative indicated flowing backward into the riser.The coordinate t represented the time of the flow developing from the initial.The velocities of the three risers were expressed by the velocities in the section 1－1，2－2，3－3as shown in Fig.5.The velocities in the main pipe were expressed by the velocities in the section 4－4，5－5and 6－6as shown in Fig.5.

Fig.5 Coordinate system and section－positions（Unit：mm）图5 坐标系及断面位置示意图（单位：mm）

The measured flow fields at the inflowing sewage discharges of 0.6×10－3m3／s and 0.72×10－3m3／s were described in the following.

2.1 The velocity－fields（Q0＝0.6×10－3 m3／s）

With the inflowing sewage discharge of 0.6× 10－3m3／s，one of the representative flow field measured in the experiment was shown in Fig.6.

In the initial stage of the flow，after the flow had persisted 30seconds in the experiment，the saline was extruded by the inflowing sewage and discharged from the risers that were indicated by the positive velocities in the risers.For convenience of analysis，as shown in Fig.6，the velocities of the three risers were expressed by the velocities in the section 1－1，2－2，3－3as shown in Fig.5.The average velocities in three risers were close.The velocities in the main pipe were expressed by velocities in the section 4－4，5－5，and 6－6as shown in Fig.5.The velocities distribution in the main pipe indicated that no stratification occurred in the outfall pipe.That is to say，the sewage wedge had not reached the measuring section yet.Under this condition，the discharging of saline from the risers was due to the extruding by the inflowing sewage.

Fig.6 Measured velocities in the risers and main pipe（Q0＝0.6×10－3 m3／s）图6 竖管及主排污管流速分布（Q0＝0.6×10－3 m3／s）

In the experiment，with the time passing 60s after the flow initiated，the flow developed into the second stage.The front of the inflowing sewage reached the riser 1and the discharge in the riser 1 began to increase rapidly.In the mean time，the discharge in the riser 2and riser 3decreased evidently.This could be seen from Fig.6.The velocities distribution in the section 4－4indicated that stratification occurred.The sewage wedge reached the 4－4section.

The above second stage had not lasted for a long time in the experiment.After a little while，with the time passing 90sfrom the initial，the flow developed into the third stage，the front of the sewage reached the riser 2and then the velocities in the riser 2began to increase.The velocities in the riser 3continued to decrease and soon flowed inversely，which indicated the receiving saline water intruded into the riser 3.

The above flow pattern in the third stage lasted until the end of the experiment.The experimental results indicated that about 300slater the velocities through the three risers tended to remain steady，indicating that the flow pattern tended to be steady.The velocities distribution in the main pipe indicated the patterns of the sewage wedge clearly.The stratification of the two fluids was present in the experiment.At the interface of the stratification under the riser 1，the vortex flow due to the entrainment could be seen.The saline fluid in the lower layer of the stratification was entrained by the upper sewage layer and discharged from the riser 1.In the main pipe under the riser 3，the lower layer fluid of the stratification flowed backward to the upstream.This indicated that the receiving saline water intruded into the riser 3and flowed backward to the upstream then discharged from the riser 2or riser 1.The saline circulated in the outfall.

2.2 The velocity－fields（Q0＝0.72×10－3 m3／s）

With the sewage discharge flow of 0.72×10－3m3／s，the development of the flow patterns in the outfall system was different with that in the discharge of 0.6×10－3m3／s.The velocities measured in the experiment were shown in Fig.7.

In the initial stage of the flow，after the flow had persisted 30seconds in the experiment，the saline was extruded by the inflowing sewage and dis－charged from the three risers that were indicated by the positive velocities in the risers.As shown in Fig.7，the average velocities in three risers were closed.The velocities distribution in the main pipe indicated that no stratification occurred in the main pipe.

Fig.7 Measured velocities in the risers and main pipe（Q0＝0.72×10－3 m3／s）图7 竖管及主排污管流速分布（Q0＝0.72×10－3 m3／s）

With the time passing 60safter the flow initiated the flow developed into the second stage.The stratification of the velocities in the main pipe indicated that the front of the inflowing sewage reached the section 4－4and section 5－5.The velocities in the riser 1and riser 2increased evidently.The velocities in the riser 3was evidently lower than that in the riser 1and riser 2，however，the values of the velocities still kept positive indicating that the saline still discharged from the riser 3.

The flow pattern in the second stage developed continuously.When the time passed 90s，for the riser 3，the velocities in the part of the riser’s section was negative indicating saline intrusion through part of the section in the riser 3.The reason was that the momentum of the discharge flow in the riser 3was too low to resist the saline.With the time passing 360s，the velocities in the riser 3were positive indicating that intruded saline was extruded from the riser 3by the sewage.In the experimental condition，the extruding of the saline from the riser 3was not intense and the flow pattern in the riser 3 was not steady.The probable reason was that the riser 3might be in the critical condition of saline intrusion and purging.If the inflowing sewage discharge increased，this extruding flow would be more intense and saline purging would be evident.

At the interface of the stratification under the riser 1，the vortex flow due to the entrainment was more intense than that under the riser 2and riser 3.The reason for this was that the entrainment was appreciably related to the momentum of the inflowing sewage.When the momentum of the inflowing sewage decreased to a low value，the entrainment was not evident and under this condition，the saline’s extrusion from the outfall depended on the interfacial turbulent mixing.

3 Conclusions and Discussions

With the technique of PIV，the internal flow fields of the outfall system during saline intrusion and purging in the simplified experimental model were measured.Compared with the previous singlepoint measuring technique such as LDV，the velocity field could be achieved synchronously and without interference.The development of the flow patterns at different inflowing sewage discharges could be clarified by the velocity－fields.

（1）For an outfall system with certain structure，the inflowing sewage discharge was a decisive factor that affected the internal flow pattern in the outfall system.When the inflowing discharge was low，saline circulation would occur in the outfall system.If the practical engineering was under this low discharge working condition，sediment would tend to deposit in the main pipe and result in block－age of the pipe.When the inflowing sewage discharge was high enough，saline circulation in the system could be avoided.So，in the practical engineering，it was very necessary to select a reasonable operation mode for the control of the process of the inflowing discharge.This was particularly important for those early structures without special safeguard measures.

（2）According to the velocity－fields of the internal flow in the outfall system，the flow patterns of saline intrusion and purging developed in several different stages.The development of the flow indicated that the internal flow in the outfall system during saline intrusion and purging was three－dimensional，unsteady，density stratified and with intensely entrainment and interfacial turbulent mixing.These flow characteristics gave some important suggestions for the numerical model.

（3）The measured velocity－fields indicated that the mechanisms to extrude the intruding saline from the system were related to the momentum of the inflowing sewage.When the inflowing discharge was large enough，the saline was entrained by the inflowing sewage and discharged from the system.The entrainment at the interface of the stratification of the sewage and saline was the main driving mechanism to extrude the saline from the system.The time scale for this extruding process was relatively short.When the inflowing discharge was low，some risers in the upstream were purged by the entrainment aforementioned.However，for the other risers in the downstream，the saline was extruded by the interfacial turbulent mixing.Besides，even if the discharge was high enough the saline wedge would still remain in the bottom of the main pipe of the outfall even after the risers had been purged.This remained saline wedge would be extruded only by the interfacial turbulent mixing.The time scale for this extruding process was relatively long.

For the first time，the technique of Particle Image Velocimetry（PIV）was used to measure the flow field of saline intrusion and purging inside the outfall system.The following points should be discussed in the further study.

（1）For the technique of PIV，the performance of the tracking particles following with the fluid was an important factor that affected the accuracy of the measurement.This performance was decided by the density of the tracking particles.The density of the tracking particles was closer to that of the fluid the better performance could be achieved.In this experiment，the tracking particles with the same density were released into the two fluids with different densities.This would have some effect on the accuracy of the measurement.So，the performance of the tracking particles following with the fluid for this density－stratified flow should be discussed in the further study.

（2）In the experiment，the different refractive index of the saline water and clear water would have some effect on the accuracy of the measurement.This should be discussed in the further study too.

Acknowledgment：

This research was supported by Laboratory of Engineering Hydraulics of Hohai University.Professor Yan Zhongmin supervised the author patiently.

［1］ WILKINSON D L.Purging of saline wedges from ocean outfalls［J］.Journal of Hydraulic Engineering，1984，110（12）：1915－1829.

［2］ WILKINSON D L.Seawater circulation in sewage outfall tunnels［J］.Journal of Hydraulic Engineering，1985，111（5）：864－858.

［3］ PHILIP Roberts.Under and underused the case for marine outfalls in waste water disposal［R］.Water 21，2010：22－26.

［4］ ECHAVARRI－ERASUN B，JUANES J A，PUENTE A，et al.Coastal outfalls，a sustainable alternative for improving water quality in north－east Atlantic estraries［J］.Journal of Environmental Monitoring，2010：1737－1746.

［5］ ZHAO Lin，CHEN Zhi，LEE Kenneth.Modelling the dispersion of wastewater discharges from offshore outfalls，a review［J］.Environment Research，2011，19：107－120.

［6］ HUAI Wen－xin，LI Wei.Prediction of characteristics for vertical round negative buoyant jets in homogenous ambient［J］.Journal of Hydrodynamics，Ser.B，2001：103－110.

［7］ ZHANG Yan，WANG Dao－zeng，FAN Jing－yu.Experimental investigations on diffusion characteristics of high concentration jet flow in near region［J］.Journal of Hydrodynamics，Ser.B，2001：117－121.

［8］ HAN Hui－ling，ZHANG Hong－min，LIANG Su－tao.Mean behavior of three dimensional line buoyant jets in cross flows［J］.Journal of Hydrodynamics，Ser.B，2003：32－36.

［9］ FAN Jing－yu，WANG Dao－zeng，ZHANG Yan.Three dimensional mean and turbulence characteristics of an impinging density jet in a confined cross flow in near field［J］.Journal of Hydrodynamics，Ser.B，2004，16（6）：737－742.

［10］MORT R B.The effect of wave action on long sea outfalls［D］.University of Liverpool，1989.

［11］BURROWS R，DAVIES P A.Studies of saltwater purging from a model sea diffuser［J］.Water Maritime ＆Energy，1996，118：77－87.

［12］GUO Zhen－ren，SHARP J J.Numerical model for sea outfall hydraulics［J］.Journal of Hydraulic Engineering，1996，122（2）：82－89.

［13］DOYLE B M，MACKINNON P A，HAMILL G A.Development of a numerical model to simulate wave induced flow patterns in a long sea outfall［C］.3rd International Conference on Advances in Fluid Mechanics，Montreal，Canada，2000：33－42.

［14］SHANNON N R.Development and validation of a twodimensional CFD model of the saline intrusion in a long sea outfall［D］.Queen’s University Belfast，2000.

［15］SHANNON N R，MACKINNON P A，HAMILL G A.Modeling of saline intrusion in a long sea outfall with two risers［C］.1st International Conference on Computational Methods Inmultiphase Flow，Orlando，USA，2001：531－541.

［16］WU Wei，YAN Zhon－min，WU Long－hua.Numerical simulation of saline intrusion and purging in long sea outfalls with three－dimensional model［J］.China Ocean Engineering，2006，20（2）：279－288.

［17］WU Wei，YAN Zhong－min.Evaluation of three－dimensional numerical model for saline intrusion and purging in sewage outfalls［J］.Journal of Hydrodynamics，2007，19（1）：48－53.

［18］DAVIES P A，CHARLTON J A，BETHUNE G H M.A laboratory study of primary saline intrusion in a circular pipe［J］.Journal of Hydraulic Research，1988，26（1）：33－47.

［19］BURROWS R，ALI K H M，DAVIES P A，et al.Studies of saltwater purging from a model sea outfall diffuser［J］.Journal of Water，Maritime and Energy，1996，118：77－87.

［20］BETHUNE G H M.A laboratory investigation of saline intrusion with particular reference to its effect on the performance of long sea outfalls［D］.University of Dundee，U.K.，1986.

［21］BURROWS R，ALI K H M，SPENCE K，et al.Experimental observation of salt purging in outfalls with soffit－connected risers［C］／／Proceedings Second International Conference on Hydrodynamics，Hong Kong，1996，12：901－906.

［22］BURROWS R，ALI K H M，SPENCE K，et al.Experimental observations of salt purging in a model sea outfall with eight soffit－connected risers［J］.Water Science＆Technology，1998，38（10）：269－275.

［23］BURROWS R，ALI K H M，DAVIES P A，et al.Hydraulic performance of outfalls under unsteady flow［R］.Dept.Civil Engineering University of Liverpool，final report on SERC project GR／G／21841.

［24］WOSE A E，BURROWS R，ALI K H M，et al.Purging studies on a model sea outfall diffuser，in water pollution III：modeling，measuring and prediction［J］.Computation Mechanics Publications：277－286.

［25］CHARLTON J A，DAVIES P A，BETHUNE G H M.Seawater intrusion and purging in multiport sea outfalls［C］／／Proceedings Institution of Civil Engineers，Part II，1987，83：263－274.

［26］BURROWS R，ALI K H M，WOSE A E.Laboratory studies of saline intrusion，salt wedge formation and sediment deposition in long－sea outfalls［C］／／Proceedings of the International Symposium on Environmental hydraulics，Hong Kong，1991：269－273.