Experimental Study of Hydrodynamic and Self-Buried Behavior of Submarine Pipeline with Perpendicular Spoilers

The spoiler is a kind of device to disturb current and promote burying. At present, all submarine pipeline spoilers at home and abroad are parallel spoilers, that is, the plane of the spoiler is parallel to the vertical plane of the pipeline axis. According to the results of indoor experiments, when the pipeline with the forward spoiler is installed perpendicular to the direction of water flow, the spoiler will accelerate the seabed erosion and cause the pipeline to endure downward pressure, which will eventually cause the pipeline self-buried to form a protection. However, when the pipeline direction is consistent with the flow direction, the self-buried behavior and protective effect is vanished. By aiming at the defect that the forward spoiler cannot be self-buried when the direction of the pipeline and the flow are basically parallel, the spoiler burying aid device perpendicular to the pipeline axis has been innovatively developed, and the hydrodynamic changes and sediment erosion characteristics near the pipeline after the installation of the device were studied based on the experiment. Results reveal that although the perpendicular spoiler cannot generate downforce, it can greatly increase the turbulent kinetic energy of the flow and the rate of sediment erosion. The larger the angle between the pipeline axis and the spoiler plane is, the larger the increase in turbulent energy will be. The increase in turbulent energy near the bed surface can reach up about 70% when the angle is 90°, while serious sediment erosion mainly occurs along both sides of the pipeline with a distance of about 2–4 times the pipe diameter. In the future, we can further explore the influence of the perpendicular spoiler size and installation position on the pipeline downforce and the effect of burying promotion. At the same time, field tests on the perpendicular spoiler burying aid device currently developed will conduct to observe the actual effect of perpendicular spoiler promoting pipeline scouring and burying, and improve submarine pipeline safety protection technology.


Introduction
In the exploitation of offshore oil and gas, submarine pipelines are widely used to transport fluid or gas medium, which has the advantages of low cost, large transportation capacity, continuous transportation and high efficiency. However, due to the harsh marine environment, submarine pipelines are directly affected by waves, tidal currents and other factors, which make the laying difficulty and operation risk of submarine pipeline relatively large (Xu, 2009;Zhang et al., 2013). In order to protect the pipeline from the complex submarine environment, it is particularly important to bury it stably under the seabed. Therefore, many scholars have conducted research on the protection technology of submarine pipeline sinking, self-buried, and vibration prevention (Jiang et al., 2018).
One of the protection technologies is to install spoiler above the pipeline. It was first proposed by Huisbergen (1984). By installing a device similar to fish fins in parallel above the pipeline, when the pipeline axis is perpendicular to the water flow, the turbulent kinetic energy of the water near the pipeline can be increased, the seabed erosion can be accelerated and the pipeline will endure downforce, which will lead to the pipeline to be automatically buried, and finally the pipeline can be buried into the seabed with the depth about twice the pipe diameter. As a new type of pipeline burying technology, it is an economical and effective pipeline protection technology due to its self-burial effect (Yang et al., 2016). The technology was first applied in the North Sea of the United Kingdom and achieved good economic and environmental benefits (Shan et al., 2015). It was used in the laying of submarine pipeline in Hangzhou Bay for the first time in China (Yang et al., 2016;Zhu et al., 2019).
Many related researches have been done on spoilers at home and abroad. Han et al. (2010) compared and analyzed the pressure distribution on the surface of submarine pipelines with flexible and rigid spoilers and without spoilers, the pressure difference between the two sides of the pipeline and the erosion at the bottom under the action of unidirectional flow and reciprocating flow. Dong et al. (2014) analyzed the changes in the flow field and vibration characteristics of the laminar flow area around the pipeline when the gap ratio between the pipeline and the seabed wall was different, and evaluated the vibration suppression effect of the horizontal spoiler on the submarine suspension pipeline. Shan et al. (2015) used FLUENT software to simulate the submarine pipeline with and without spoiler, analyzed the data of velocity and the shape of the scour pit below the pipeline, and compared the effects of spoiler usage on the pipeline self-burial. Zhang (2016) simulated and analyzed the flow field, velocity distribution, pressure distribution, lift/drag coefficient and bed shear force distribution around the pipe with and without spoiler. Yang and Shi (2018) studied the variation law of seabed scour depth under the action of unidirectional constant flow below submarine pipeline without spoiler, with spoiler installed and with different clearance ratio. Chiew (1992) studied the extreme scour depth of pipes with spoilers at different attack angles through experiments, and the scour development law on the back flow side of the pipe. Zhao and Wang (2009) studied the bed shear stress with or without spoiler by numerical simulation, and found that the height of spoiler had no significant effect on the bed shear stress. Yang et al. (2012) studied the influence of flow velocity and spoiler on the maximum scouring depth below the pipe through experiments, and found that both the Reynolds number and the pipe height have a great influence on the scouring depth, and proposed the corresponding scouring depth prediction formula. Zhu et al. (2013) simulated the erosion of the pipe with spoiler under the condition of constant flow, and found that the height of the spoiler and the gap between the pipe and the bed have great influence on the erosion. Öner (2016) simulated the two-dimensional turbulent flow field of a pipe with a spoiler under a constant flow condition, and found that installing a spoiler would cause a large separation zone downstream of the pipe, resulting in an increase in the drag force of the pipe and a decrease in the upward force.
Although scholars have done a lot of researches on spoilers, their works are mainly about parallel spoilers (that is, the spoiler plane is parallel to the pipe axis), and rarely involve the influence of the change of the angle between the pipeline and the water flow on the self-buried effect, and there is no relevant research report about the installation of perpendicular spoiler when the pipeline and water flow direction are consistent. From the view of the actual effect, the self-buried effect of the parallel spoiler is better only when the angle between the pipeline and the water flow direction is large. For example, after parallel spoilers were used for the Hangzhou Bay submarine pipelines, the self-buried ratio has reached more than 90%; the self-buried area is mainly concentrated in the pipeline route KP4−KP42 area, but the pipeline in the south bank of Andong Shoal KP42−KP47 area has been exposed for a long time. The pipelines in the self-buried area and the exposed area are both equipped with parallel spoilers. The main difference between them is that the angle between the pipeline and the water flow direction is varied. The angle between pipeline and flow direction in self-burial area is large (more than 50°), while pipeline in the exposed area is basically parallel to flow direction. Therefore, it is urgent to study the reasons of failure of parallel spoiler in the south bank beach and solve the exposed problem.
In view of this, this study intends to investigate the selfburial mechanism of parallel spoilers through indoor experiments, and the reasons for their self-burial failure under the condition of pipeline parallel to the flow direction, and to clarify the internal mechanism of long-term exposure of Andong Shoal pipeline on the south bank of Hangzhou Bay. In order to solve the problem of pipeline exposure on the south bank, aiming at the defect that the parallel spoiler cannot be buried automatically when the water flow direction is basically parallel to that of the pipeline route, a new perpendicular spoiler burying aid device installed perpendicular to the pipeline axis is developed. The effectiveness of the device is verified by the hydrodynamic and sediment erosion tests. The research results can improve the safety protection technology of submarine pipelines at downstream and provide safety protection solutions for the exposed problems of Hangzhou Bay pipelines.

Test equipment and instruments
The dimension of the test flume is 35 m×1.0 m×0.5 m (length×height×width). For the convenience of observation, a 3-m long glass side wall was installed in the middle of the water tank. And a rectangular pit of 0.5 m×0.5 m×0.2 m (length×height×width) was set in the middle of the glass side wall area to fill the test sand sample. The bottom slope of the fixed-slope flume was 1/500. The specific layout is shown in Fig. 1. The water tank was equipped with two 500 m 3 /h water pumps. At the water depth of 0.4 m, the maximum flow rate of about 1.3 m/s could be formed. The inlet flow of the water tank was measured by a thin-walled rectangular weir, and the outlet water depth was controlled by a flap tailgate.
The current velocity was measured by Ultrasonic Doppler Velocimeter ADV (Vectrino) produced by Sontek company. It can measure the three-dimensional flow velocity of a single point in real time (the combined velocity can be calculated), the sampling frequency was 25 Hz, the current ve-locity resolution was ±1 mm/s, and the measurement error was 1% of the current velocity.
Turbulent kinetic energy k can be calculated after flow velocity measurement: where u, v, and w are the turbulence intensity in flow direction, span direction and vertical direction, respectively. When the velocity and turbulent kinetic energy were measured, the relative deviation of the parameters between different types of spoiler (horizontal spoiler and perpendicular spoiler) could be compared: where y is the distance from the bed surface, and is the combined velocity of the three-dimensional current velocity at point y, subscripts v and h denoting the vertical and horizontal direction, respectively. At this time, D R is the relative deviation of the current velocity. When is replaced by turbulent kinetic energy k, D R is the relative deviation of turbulent kinetic energy.
The circumferential pressure of the pipeline was expressed by the piezometric head. The pressure taps (#1−#12, the circumferential angle interval being 30°) were circumferentially arranged on the middle section of the pipeline, and were led out of the water tank along the inside of the pipeline with a hose, and connect to the reading board, as shown in Fig. 2. In order to improve the accuracy and efficiency of the test, the high-resolution camera was used to capture the water level of the piezometer tube and to store it. In order to reduce the test error, a group photos of piezometric tube water level were taken every 2−3 minutes, and no less than 10 pictures were taken under each test condition. After the test, the water level of piezometric tube was read and calculated by self-developed software with an accuracy of ± 0.1 mm.
After the pipe circumferential pressure was measured, the total pipe circumferential pressure could be calculated (indicated by pressure head) by: where p i is the pressure at the i-th pressure tap, when =0°w ith the couterclockwise direction of , , , each hole was spaced by 30°, and =330°. Regarding the direction of m and n: (1) when the pipeline axis direction was perpendicular to the water flow direction, m was pointing upstream and n was vertical upward; (2) when the pipeline axis direction was parallel to the water flow direction, m was pointing to the right bank and n was vertical upward.  3.1 Parallel spoiler structure The "parallel direction" of the parallel spoiler means that the plane of the spoiler is parallel to the vertical plane where the pipeline axis is located. At present, spoilers of submarine pipelines at home and abroad are all parallel. The installation diagram and photo of the parallel spoiler are shown in Fig. 3. Fig. 3a shows the installation diagram of parallel spoiler in general, the height of the spoiler is generally 0.15−0.25 times the pipe diameter, the length of a single spoiler of the Hangzhou Bay submarine pipeline is 3.8 m, and the span between the spoilers is generally 0.3−0.6 m. In Fig. 3b, the outer diameter of the test pipe was D=2.6 cm, the parallel spoiler was welded to the pipe with a 1-mm thick iron sheet, the height was 4 mm (0.15D), the length was 3.95 cm (1.5D), and the span between spoilers was 0.6 cm (0.23D).

Self-burial mechanism of pipeline perpendicular to flow
This section briefly studies and explains the self-burial mechanism of pipelines with parallel spoilers. The comparison test of the pipeline with and without parallel spoiler was carried out. The test conditions are shown in Table 1.
According to the calculation in the table, the Reynolds number of the cylinder was 3.9×10 4 , which was located in the subcritical region. The circumferential pressure distribution of the pipeline is shown in Fig. 4a. Under condition of the pipeline without spoiler, the drag force F m around the pipeline pointed to the downstream direction, while the lift force F n was vertically upward, and the overall pressure was diagonally upward to the left. However, when the parallel spoiler was installed, the drag force F m decreased, and F n changed from lift force to downforce, and the resultant pressure force around the pipe pointed to the bottom left. After the pipeline was installed, the column turbulence was generated, which made it easy to produce turbulent coherent structure behind the pipeline, which caused scouring and transport of sediment. Because the working condition of this experiment was different from that of traditional measurement of sediment incipient motion (constant uniform shear flow on flat bed surface of open channel), it was de-   Fig. 4. Measurement results of water/sediment parameters of the pipeline with parallel spoilers. FAN Hong-xia et al. China Ocean Eng., 2021, Vol. 35, No. 2, P. 250-261 cided to adopt "pipeline initial velocity" to reflect the influence of pipeline (or installation of spoiler) on bed sediment erosion. The starting flow velocity of the pipeline is defined as the average flow velocity of the section when the bed surface sediment (small, medium, large) critical scour occurs under the condition of placing the pipeline on the flat bed surface (or installing the spoiler).
The initial velocity tests of pipeline under two kinds of bed conditions were carried out (Zhu et al., 2017), which were: iron pipes on the bed surface and iron pipes with spoiler installed on the bed surface. The outer diameter of the pipe placed on the bed was 2.6 cm, and the spoiler was welded to the pipe with a 1-mm thick iron sheet, the height was 4 mm (pipe diameter 15%), the length was 3.95 cm (the original body was 3.8 m), and the interval was 0.6 cm (the original Body 0.3−0.6 m). Before the test, the sediment sample was placed in a rectangular groove, and the bed surface was scraped with a scraping ruler, so that the mud surface was flat with the bottom of the tank, kept the bed surface smooth. After the bed surface is made, put the pipe aside. In order to avoid the pipe sinking, a thin cotton thread was used to lift the pipe in the middle of the pipe, and the two ends of the cotton thread were fixed on the side of the tank. During the experiment, water was slowly filled at the end of the flume. After reaching a certain water depth, the pump was turned on and the inlet flow was slowly increased to observe the critical scouring condition of the bed surface and measure the current velocity. In this study, the critical situation of local sediment initiation of the pipeline was divided into three types: a small amount of erosion, medium erosion and a large amount of erosion, which was similar to the traditional definition of sediment initiation state.
The results of pipeline initial velocity are shown in Table 2 and Fig. 4b, in which only the initial velocity of pipeline under medium critical scour is given in Fig. 4b. It can be seen from the chart that when the pipeline is installed with a parallel spoiler, the local turbulence of the fluid around the pipeline increases, making the sediment under the pipeline easier to hang and be washed away, and the initial current velocity of the pipeline is largely reduced by about 20%. Therefore, the spoiler can greatly reduce the initial current velocity of the pipeline and accelerate the seabed erosion.
From the above results, it can be seen that the main reason for the flushing and self-buying of pipelines with parallel spoilers is that the spoilers cause the seabed erosion acceleration and generate downforce on the pipeline. Fig. 5 illustrates the entire scouring and self-burial process of a sub- Note: U ie1 , U ie2 and U ie3 in the table refer to the initial velocity of pipeline under mild, medium and severe critical scouring conditions respectively; The superscript * indicates the initial current velocity of the pipeline calculated by the formula combined with the experiment results (Zhu et al., 2017), where m=1/6, U ie and U if are the initial current velocity of the pipe under the tank test and natural conditions, respectively, and h e and h f are the water depth under the tank test and natural conditions, respectively. The result of the final rushing velocity of the pipeline at a water depth of 1 m is the average of the calculation results of the three working conditions (10, 20, 31 cm). The rushing velocity of the pipeline at other water depths can be analogized. Fig. 5. Self-burying mechanism of the pipeline with parallel spoilers. marine pipeline under the action of tidal current: at first, a pipeline with a spoiler is placed flat on the seabed (Fig. 5a); when the tidal current direction is from left to right (Fig. 5b), under the long-term scouring of the unidirectional tidal current, the seabed on the downstream side of the pipeline will be eroded, part of the eroded material will move down, and other will be transported to the vicinity of the pipeline by the backflow from the downstream side of the pipeline; when the tidal current direction is from right to left (Fig. 5c), the sediment deposited on the upstream side of the pipeline (downstream side in the last tidal current) is washed away, and the seabed at the downstream side of the pipeline is eroded; when the tidal current flushed several times (Fig. 5d). A large part of the sediment near the pipeline has been washed away, and the seabed is undercut, and the erosion is severer near the pipeline, while the erosion is milder at the place far away from the pipeline; when there is a gap between the pipeline and the seabed, downward pressure is generated, and the pipeline is lowered and pressed tightly on the seabed (Fig. 5e). Figs. 5f−5i show another process of scouring and pushing down. In this process, the pipeline position is continuously lower than the mean sea bed elevation (Fig, 5j) and finally be backfilled to self-buried (Figs. 5k and 5l).

Failure reason of self-buried mechanism of pipeline paraller to flow
The axis of the pipeline in the Andong Beach in the southern bank of Hangzhou Bay is basically parallel to the current, and the angle between the axial direction of the pipeline and the direction of the tidal current is within 15°. The pipeline in the section of Andong Beach has been mainly exposed for a long time. The test results in May 2016 showed that the exposed area has expanded and the exposed area has moved to south, but the total exposed length has been improved, as shown in Fig. 6.
In this section, through experiments, when the pipeline with parallel spoiler is installed parallel to the flow direction, it is the reason for the failure of the pipeline self-buried effect. Table 3 shows the experimental conditions of the two comparative conditions. Fig. 7a shows the circumferential pressure distribution of the pipeline. When the pipeline axis is parallel to the water flow direction, the installation of the parallel spoiler will hardly affect the pressure distribution around the pipe. Under the conditions of pipeline without spoiler, the drag force F m around the pipeline points to the left bank, while the lift force F n is vertically upward, and the overall pressure is oblique to the left bank; after the pipe is installed with a spoiler, the direction of the drag force F m and lift F n does not change, only the values are somewhat changed.
Figs. 7b and 7c show the measurement results of the flow velocity and turbulent kinetic energy around the pipe.    When the spoiler was installed, the current velocity at the 1D position on the left and right sides of the pipeline did not change, and the turbulent kinetic energy hardly changed.
The parallel spoiler device has the best self-burial effect only when the pipeline is perpendicular to the flow direction. As the angle between the these two decreases, the selfburial effect gradually weakens. When the flow direction is parallel to the pipeline axis, it almost loses its self-burial effect. Although the sediment on the south bank of Andong Beach is easier to be scoured, the turbulent scouring effect is weak, and the surrounding hydrodynamic forces cannot generate downforce on the pipeline. Therefore, the pipeline is difficult to be buried automatically, which has caused the pipeline to be exposed since its operation.

innovation of perpendicular spoiler
Owing to the vast sea area at the site and the changing direction of ocean currents, the direction of the pipeline cannot be kept perpendicularly to the current direction all the time. In order to solve the defect that the parallel spoiler cannot make the pipeline self-buried when the water flow direction is basically parallel to the pipeline route, a new perpendicular spoiler burying aid device installed perpendicularly to the pipeline axis has been developed, as shown in Fig. 8. The word "vertical" of the perpendicular spoiler means that the plane of the spoiler is perpendicular to the vertical plane of pipeline axis. Fig. 8a shows the structure of the perpendicular spoiler. In order to facilitate the installation of the perpendicular spoiler in the pipeline, a semi-elliptical interface is opened on the rectangular plate; the long and short axis of the elliptical interface are determined by the installation angle of the spoiler, for example in Fig. 7, the installation angle of the spoiler (the angle between the √ 2D plane of the spoiler and the pipe axis) is 45°, and the long and short axes of the ellipse are and D, respectively. The ellipses with other installation angles can be inferred according to the law of the tangent plane of the cylinder. In addition, a wedge is glued to the back flow side of the spoiler. In addition, a wedge is attached to the back flow side of the spoiler, which can support the spoiler, prevent it from toppling under its own weight, and is easy to bond with the pipe. Fig. 8b shows the test photos of the perpendicular spoiler. The thickness of the perpendicular spoiler is 2 mm and the installation angle is 45°. Refer to the horizontal spoiler parameters of Hangzhou Bay (the height of the spoiler is 0.15D−0.25D, the length is 5D−15D, and the span between spoilers is 0.5D−2.5D). The vertical height of the perpendicular spoiler is 3.75 cm (0.25D), and the spacing between the spoilers is 15 cm (1D). The wedge is 1.3 cm long, 0.5 cm high, and has a sharp angle of 30°.
When the current direction is parallel to the pipeline, the perpendicular spoiler has a larger flow blocking area than the parallel spoiler, which will cause great interference to the incoming flow. When the water flows head-on to the perpendicular spoiler (the surface without wedge), the perpendicular spoiler has a certain inclination angle, which can lift the water flow to a certain extent and divide it to the two sides of the pipeline, resulting in the increase of turbulent kinetic energy of water on both sides of the pipeline, which promotes seabed erosion. When the water flow is reversed, because the perpendicular spoiler has the effect of pressing the water flow down to the bed surface, it will accelerate the seabed erosion, thus, under the action of the reciprocating flow, the perpendicular spoiler will play a role in the seabed erosion enhancement.

Hydrodynamic experiments analysis
In order to study the disturbing effect of perpendicular spoiler, hydrodynamic experiments with four installation angles were carried out. The test layout diagram is shown in Fig. 9, and the experiment conditions are shown in Table 4. The installation angle of spoiler refers to the angle between the plane of spoiler and the axis of pipeline. The relative positions of five test holes were specially set (Fig. 9d) to measure the pressure change between the two perpendicular spoilers.
The circumferential pressure distribution of the pipeline under various working conditions is shown in Fig. 10. It can be seen from the figure that, compared with the parallel spoiler, the perpendicular spoiler does not significantly change the circumferential pressure distribution of the pipeline.  The total circumferential pressure on the pipeline was further calculated, and the results are shown in Table 5. It can be seen that when the perpendicular spoiler is installed in the pipeline, the drag force F m at each section of the pipeline under all working conditions points to the left, while the lift force F n is vertically upward. The perpendicular spoiler did not result in the formation of downforce. Although there are certain differences in the total pipeline pressure among various working conditions, the overall difference is not significant. Fig. 11 shows the measurement results of current velocity of a pipeline with a perpendicular spoiler. Compared with the parallel spoiler, due to the existence of a large flow blocking area above the pipeline, which causes the strength of the main flow to be weakened above the pipeline axis, and the flow velocity at the top of the spoiler is reduced by about 40%−50% (see Figs. 11a and 11b), but the closer to the water surface, the smaller the deviation; the smaller the spoiler angle, the more obvious the effect of weakening the main flow above the pipeline axis. In addition, since the spoiler divides the main flow to both sides of the pipeline, the flow velocity on both sides increases to a certain extent (the absolute increase is 2.0−3.8 cm/s), and the relative increase is about 5%−10% (see Figs. 11c and 11d), there is no obvious linear relationship among the degree of deviation, water depth and spoiler angle. Fig. 12 shows the measurement results of turbulent kinetic energy for a pipeline with a perpendicular spoiler. Compared with the parallel spoiler, after the perpendicular spoiler is installed on the pipeline, the turbulent kinetic energy   258 FAN Hong-xia et al. China Ocean Eng., 2021, Vol. 35, No. 2, P. 250-261 above the pipeline axis increases rapidly (Fig. 12a). Generally, the closer to the spoiler, the larger the deviation, and the maximum deviation can reach 900% (Fig. 12b), the smaller the spoiler angle, the larger the deviation. At 1D on the left side of the pipe, compared with the parallel spoiler, the turbulent kinetic energy under the condition of the perpendicular spoiler largely increases (Fig. 12c). However, taking the height of pipe diameter as the boundary, the smaller the spoiler angle above it, the larger the deviation, and the larger the spoiler angle below it, the larger the deviation (Fig. 12d). The maximum increase of turbulent kinetic energy can reach 70%, and the maximum value can reach 25 cm 2 /s 2 .

Experimental analysis of sediment erosion
The hydrodynamic condition of Hangzhou Bay is dominated by tidal currents. In order to reduce the difficulty of the test, only the scouring in the direction of single tide was simulated, which was simplified as a constant flow test. The sediment erosion experiment of the pipeline with perpendicular spoiler under constant flow conditions was carried out. The conditions of experiment are shown in Table 6. The pipe in the model has an outer diameter of 2.65 cm and a total length of 18.5 cm. The perpendicular spoiler (angle of 60°) was made of a 2-mm thick plastic sheet, with a height of 0.66 mm (0.25D) and an interval of 2.65 cm (1D). Before the test, the sediment sample was placed in a rectangular groove, the bed surface was scraped flat, and then the pipeline was laid. In order to avoid the pipe from rolling over, the two ends of the pipe were slightly tightened with steel wire ropes so that they could only move up and down vertically. The other end of the wire rope was fixed on the fixed screw at the bottom of the water tank.
For the prototype sand test, the incipient velocity of a large amount of sediment was about 100 cm/s. In order to obtain better scour results, we increased the scour current velocity to 150 cm/s, and the scour time lasted for 1.5 hours. The before and after bed morphology is shown in Fig. 13. Since the perpendicular spoiler will greatly increase the turbulent kinetic energy on both sides of the pipe, the sediment on the bed surface is easier to move. It can be seen from Fig. 13b that the scouring depth area of 2−3 cm on the bed surface is mainly in the range of 2−4 times the pipe diameter on the left and right sides of the pipe, and the further to downstream, the larger the scouring area.
For the charcoal sand experiment (the bulk density of 0.32 g/cm 3 ), the incipient velocity for a large amount of sands was about 26 cm/s. In order to enhance the scour result, we increased the scour current velocity to 53 cm/s. The scour time lasted for 1 hour. The bed morphology after scouring is shown in Fig. 13c. Although the final scouring form of charcoal sand is not exactly the same as the prototype sand, the main erosion characteristics are basically the same. The erosion depth of 4−5 cm on the bed surface is mainly concentrated in the range of 2−4 times the pipe diameter on both sides of the pipeline, and the further to down- stream, the larger the scouring area. From the two kinds of sand erosion experiments, it can be seen that the installation of perpendicular spoiler will lead to significant erosion on both sides of the pipeline, which can promote the pipeline being scoured and buried. The experiment results prove the effectiveness of the perpendicular spoiler in accelerating pipeline erosion. In the future, further field experiments can be conducted in Hangzhou Bay to measure the actual efficiency of self-buried acceleration.
It should be noted that, compared with the pipeline model, although the size of the sand-buried cavity in this experiment is not very large, the pipeline has 6D, 7.5D, 9D, and 9D lengths of paved sand on the front, rear, left, and right sides of the pipeline, and the depth of paved sand is 7.5D. Wang et al. (2016) reported that the impact range of the cylinder in the upstream direction is 1.3D; the pipeline is regarded as a combination of two semi-cylindrical spur dikes. Ying and Jiao (2004) reported that the horizontal impact range of the cylinder is 1.5D, regarding the pipeline as a submerged dam. Zhu (2001) reported that the influence range of the return zone behind the cylinder is 5D−7D. Compared with the size of the test area in this study and the influence range of the cylinder, it can be seen that although the size of the test area is small, it has no effect on the test results. In addition, the scouring test in this paper does not last long, and the scouring depth of the bed surface in the test area is in the range of 0−5 cm, about 2D. Compared with the 7.5D sand paving distance, it has sufficient burial depth.

Conclusions
In this study, through indoor tank experiments, the characteristics of water and sediment movement near submarine pipelines with spoilers were studied. The mechanism of pipeline self-burial caused by parallel spoiler and the failure reason when the pipeline was parallel to the flow direction were analyzed. Aiming at the defect that when the water flow direction is basically parallel to the pipeline route, it is not conducive to self-buying, a new perpendicular spoiler burying aid device installed perpendicular to the pipeline axis was innovatively developed , and experimental research has proved that it can promote the self-buying of the pipeline. The main conclusions are as follows.
(1) When the pipeline axis is perpendicular to the flow direction, the installation of parallel spoilers in the pipeline will cause the pipeline to endure downforce, increase the turbulent kinetic energy on the downstream side of the pipeline, and reduce the initial scour velocity of the pipeline. Under the reciprocating action of tidal currents, the sand around the pipeline continuously scours, and the pipeline continues to press down. In the process of continuous scour and depression, the position of the pipeline is gradually lower than the mean seabed elevation, and finally backfilled to be self-buried. When the pipeline axis is parallel to the flow direction, the parallel spoiler will not increase turbulence and form downforce, resulting in the disappearance of the self-buried effect.
(2) A perpendicular spoiler is developed to accelerate the velocity of pipeline erosion and self-burying when the pipeline is parallel to the flow direction. After the perpendicular spoiler is installed on the pipeline, although it cannot lead to the formation of downforce, it can significantly increase the turbulent kinetic energy of the water and the rate of sediment erosion. The maximum increase of turbulent kinetic energy near the bed surface can reach 70%, while the sediment erosion is mainly concentrated in the range of about 2−4 times the pipe diameter on both sides of the pipeline. All experimental data in this study have been uploaded to Baidu Netdisk for readers to download and use. The download link is as follows: https://pan.baidu.com/s/ 1hZWZmKryB1wzBC4U2NCaXg, and extraction code is p979.