1 Introduction

In general, the maximum weight of deep-sea installations cannot exceed 1000 tons. However, with the continuous development of oil and gas exploitation in deepsea areas, the weight of deep-sea installations is continuously increasing, thereby leading not only to a higher operational risk but also a higher demand for crane capacity. To meet those demands, many methods have been proposed, such as sheave installation method (Stock et al.,2002), pendulous installation method (Roveri et al., 2005),pencil buoy method (Risoey et al., 2007), and subsea deployment system (Joensen and Paul, 2011). In practice,the rental fees for installation technology and equipment are high. Moreover, the increasing weight of installations challenges the crane capacity of vessels. Therefore, a new installation technology is needed to meet such challenges.The researchers proposed a buoyancy module auxiliary installation technology by loading a buoyancy module on a subsea production system to reduce the lifting weight and the required compensation ability for the heave compensator (Fig.1). To verify the feasibility of this method, a physical experiment was conducted; results indicate that this method is feasible in principle (Xu et al., 2016). The installation technology includes two processes: the SPS lowering-down process and buoyancy module retrieval process. An analysis of the responses of the hoisting rope,the SPS, and the floating-up velocity of the buoyancy module can help in implementing and improving the buoyancy module auxiliary installation technology.

滤波器原型的频率响应特性主要由半波长谐振器的长度和加载枝节线的长度决定。为了验证以上的分析,本文利用HFSS软件对滤波器的传输特性进行了计算机全波电磁仿真。如图4(a)所示,通过改变短路枝节线的长度L2,滤波器第1通带的中心频率可以获得较大范围的改变,与此同时,滤波器第2通带的中心频率保持不变,第3和第4通带的中心频率有微小的变化。图4(b)和图4(c)显示出L3和L4对第3通带的频率有明显的影响,而对第4通带的频率影响较小。从图4(d)可知,L5对第4通带的频率有明显的影响,而对其他频率几乎没有任何影响。

When SPS is installed on the sea, the ship, the hanger rope, and the system are affected by waves and currents.Continuous models of deep-sea remotely operated vehicle systems are established (Driscoll et al., 2000a; Lueck et al.,2000), which are based on frequency domain and time domain analysis. Discrete models of the coupled system are then established (Driscoll et al., 2000b; Zhu et al.,2009a). The discrete model is more widely used than the continuous model because it can calculate the response system in a larger scale. However, the disadvantage of the discrete model is that it requires large computing resources. The harmonics motion and tension transfer functions between the ship and the cage are determined (Zhu et al., 2009b), and an expression of tension along the tether can be derived based on the functions. For the deep-sea oil and gas equipment retrieval process through deep water, wave area, and sea level phases, coupled motion mathematical models of the crane vessel-cable-equipment systems are established (Zhou et al., 2015). Results suggest that, during the deep-sea oil and gas equipment retrieval process, the coupled system is relatively safe in the deep-water phase and more dangerous when the equipment crosses through the wave area and sea level phases,thereby preventing the cable from rupturing as a result of the sudden load.

After the installation, the buoyancy modules will be released for recycling. Knowing when and where the buoyancy modules will rush out from the water in advance is not only helpful in searching for them effectively and improving the recovery but also can avoid collision between the buoyancy modules and the recycling ship.Numerous researchers have studied the floating-up problem (Huang et al., 2013; Kang et al., 2012; Sun et al.,2015). Han and Tao (2001) proposed that frictional resistance and shape resistance should be separate to calculate flow resistance in the deep water. Gong et al. (2008) calculated the resistance coefficient and the maximum speed of auto-returned sampler based on ANSYS software. Yan et al. (2011) presented a method of forecasting the maximum steady speed of the full-scale buoy floating freely in the water on the basis of force analysis. In this study, we simulate the SPS lowering-down process and the buoyancy retrieval process with a new method, which is based on computational fluid dynamics (CFD) software FLOW-3D. According to the simulated results, the responses of hoisting rope and SPS, as well as the velocity law of buoyancy module during the floating-up period, are obtained. These responses are crucial to the implementation and improvement of the buoyancy module auxiliary installation technology.

Fig.1 Equipment diagram of the installation method with buoyancy aid: 1, hanging bracket; 2, winch; 3, heave compensator; 4, hanger rope; 5, installation; 6, buoyancy.

2 Mathematical Model

Numerous simulations of moving boundary problems in ocean engineering have been performed. To overcome the difficulties caused by large amplitude movement in the simulation of moving boundary problems and to calculate the force accurately, arbitrary Lagrange-Euler (ALE)method is introduced to solve the problem. Noh (1963)and Trulio (1966) first proposed an ALE method in a finite difference scheme. This method was further generalized by Hirt et al. (1974). Ramaswamy and Kawahara(1987) developed the method to track large motion on a free surface, and Floryan and Rasmussen (1989) used the ALE method to solve moving boundary problems successfully. ALE method is a body-fitted dynamic grid technology, with the grids moving together with the moving objects. The CFD software FLUENT adopts the bodyfitted dynamic grid method only to deal with moving boundary problems. However, the body-fitted dynamic grids are difficult to generate when the model is complicated and the movement amplitude is large. Moreover, the computational cost is high and guaranteeing the quality of the grid is difficult. The fractional area/volume obstacle representation (FAVOR) method adopted in FLOW-3D is an immersed boundary method (Jing et al., 2014), which has led to the successful development of a general moving object capability, the principle of which permits the modeling of any type of rigid body motion (six degrees of freedom, fixed axis, and fixed point) on a fixed mesh(Bhinder et al., 2009). In the FAVOR method, the motion of the simulation model is unrelated to the background grids. Thus, no matter how large the motion amplitude of the model is, the grids do not need to be generated again.Therefore, compared with the body-fitted dynamic grid method, the FAVOR method not only generates grids more easily but also reduces calculation more effectively(Liang et al., 2015; Ding et al., 2015a). Furthermore,compared with the traditional immersed boundary method,the FAVOR method can describe the boundary more accurately, thereby simulating water flow in a high Reynolds number precisely (Ferdos and Dargahi, 2016). This ability compensates for the inadequacy of the body-fitted dynamic grid method. Thus, FLOW-3D has been extensively used to model various problems in hydraulic engineering (Dargahi, 2010; Griffith et al., 2007; Ho et al.,2006; Johnson and Savage, 2006). Hence, the SPS low-ering- down process and buoyancy retrieval process are simulated in the physical experiment by using FLOW-3D.

2.1 Governing Equation

The fluid is assumed to be an incompressible viscous fluid, and the governing equations of FLOW-3D that are used to establish the numerical flume are continuity equation and Navier-Stokes equation, which are as follows(Wang et al., 2012):

where u, v, and ω denote the velocity components in the x,y, and z directions, respectively; Ax, Ay, Az and VF are flowable area fractions in the x, y, and z directions, and the volume fraction, respectively, which are related to the FAVOR grid technology in FLOW-3D; ρ is the density of fluid; p stands for the pressure; Gx, Gy and Gz are gravitational accelerations in the x, y and z directions, respectively; and fx, fy and fz are viscous force accelerations in the x, y and z directions, respectively.

Fig.6 shows two remarkable reductions in the tension curve caused by flotage, which correspond to the SPS water entry process and the buoyancy module water entry process, respectively. However, fluctuation barely occurs in the rope tension decreasing process, which is remarkably different from the result of wave condition. Except for the moment when the SPS and the buoyancy module enter the water, the current has little influence on rope tension.

where F represents the fluid volume function.

2.2 Turbulence Model

To simulate the wave breaking condition accurately when the waves and SPS interacts, the RNG k-ε model is chosen as the turbulence model, which is expressed as(Ding et al., 2015b)

where κT denotes the turbulence kinetic energy; PT is the production of κT caused by shear effect; εT is the dissipation rateare the diffusion terms; and CDISI and CDISI2 are the model parameters.

Lagrangian volume-of-fluidadvection method is applied to capture the change in free wave surface, in which empty cells are given a value of 0, and full cells receive a value of 1 (Shahrokhi et al., 2013). Central difference and upwind schemes are adopted in the discretization of the convective term with nonconservation type, and the discretized equations for velocity-pressure terms are solved by using the generalized minimum residual method.

3 Numerical Flume

3.1 Boundary Conditions and Initial Conditions

A three-dimensional numerical wave flume is established to analyze the experimental results in detail. The front of the flume is set as the wave boundary (wave),which can generate regular waves in different wave heights and periods. Velocity can also be added to the fluid in the wave boundary to simulate the joint action of waves and currents. The end of the flume is arranged as an outflow boundary and allows fluid backflow, which has a significant effect on eliminating waves, especially for linear small-amplitude waves. The wall and bottom of the flume are set as no-slip wall boundaries.

3.2 Mesh Generation

The size of the numerical flume is 15 m×0.6 m×1 m,which is the same as the physical experiment tank. The meshing method is as follows (Fig.2): In the x and y directions, the grids are refined around the structure and become gradient meshes toward the sides, with one wavelength containing 50–100 grids. In the z direction, the grids are refined at the free surface, and the grid size is approximately 1/10 wave height. The size of the other grids gradually increases toward both sides.

Fig.2 Mesh of the numerical wave flume in the x–z directions and the y direction.

4 Simulation of the SPS Lowering-Down Process

4.1 Basic Assumption of the Simulation Model

Study on the variation of direct shearing test of pile-soil interface on excess pore water pressure in clayey soil

4.2 Establishment of Model Experiment

The numerical model of the SPS is established according to the experimental model. In the physical experiment,the equipment includes buoyancy models, a weight, tension sensor, data collection system, wave generator, and sling equipment. The buoyancy module is made of hollow buoyant materials and is filled with sand to control density; its weight is similar to that of the SPS. The tension sensor hangs over the buoyancy module and weighed using ropes to measure the tension changes. The other end of the rope crosses through the crown block and is controlled by operators. When waves become stable, the operator releases the rope slowly, and the systems enter the water gradually. During the system lowering-down process, we can obtain the tension changes from the data collection system connected to the tension sensor. Detailed information about the experiment procedures are given by Xu et al. (2016).

The numerical models are set in FLOW-3D, including the buoyancy model and the SPS model (Fig.3). The parameters of both models are detailed in Table 1. The motion form of the two objects is set as coupled motion in the GOM model, from which we can obtain the objects’moving speed and displacement at every moment. The SPS and buoyancy model are connected with a rope, whose attribute can be set in the spring model. All the objects are lowered at a speed of 0.03 m s−1. Depending on the GOM and spring models, we can obtain the SPS disturbance and tension changes after calculation. When all the models are established, we run the simulation under three conditions, namely, wave action, current action, and wavecurrent coupled action (Table 2).

请论文通信作者或第一作者在《南方能源建设》、《广东电力》、《浙江电力》、《内蒙古电力技术》网站或邮箱(如下)注册和投稿，投稿时选择“综合能源系统规划与运行”栏目或备注。作为专题稿件，各刊会加速审稿，缩短周期；不收取任何版面费和审稿费，发表后根据文章质量以及基金支持情况支付较高稿费。

今天早上我在使用微波炉时遇到一件怪事，启动微波炉加热食物时，微波炉中有火花冒出，关掉微波炉重新启动，还是有火花冒出。怎么回事呢？

Fig.3 SPS lowering-down model.

Table 1 Parameters of the buoyancy and SPS model

Model Density(kg m−3) Size (m) Speed (m s−1)Weight(N)Buoyancy module 372 0.1×0.1×0.047 0.03 1.7 SPS 7990 R=0.0375,H=0.05 0.03 17.3

Table 2 Running conditions

dition Wave height (m) Period (s) Velocity (m s−1)e 0.03 0.8 –rent – – 0.1 e-current pled 0.03 0.8 0.1

4.3 Result Analysis

随着业务规模和市场环境的快速变化，国际大石油公司对天然气业务正从注重上游向打造一体化全产业价值链转变。在上游领域，国际大石油公司继续在澳大利亚、中亚、北美和非洲等地区收购或开发一批深水和LNG项目，呈现项目大型化和集中化趋势；在下游领域，积极推进天然气化工和交通燃料的商业化应用，加快发展以北美等地区廉价的天然气资源为原料的化工业务。

According to the hanger rope tension curve shown in Fig.4, four viewpoints can be obtained: 1) Good agreement is found between the simulation results and the experimental data, thereby suggesting the simulation accuracy of FLOW-3D; 2) Two marked reductions are observed in the tension curve caused by flotage, which correspond to the SPS water entry process and the buoyancy module water entry process, respectively; 3) Some fluctuations occurred in the rope tension decreasing process,and the number of fluctuations is approximately the same as the waveform numbers when the SPS and buoyancy module underwent the water entry processes; 4) The final rope tension is reduced to 60 percent compared with the no-water entry process, which indicates that buoyancy module reduces a load of rope effectively. In sum, the aid of buoyancy module can meet the demand of load reduction, thereby validating the feasibility of the method.However, under wave condition, the range of tension in the buoyancy module water entry process is twice that of the SPS water entry process. According to Tang et al.(2008), the rope is much easier to crack when it is tightened and loosened alternately. Thus, the hanger ropes should be added and the SPS should be installed during good weather to reduce the discontinuous load of the rope.

此外，南充市旅游景点交通通达性呈现一定的规律特征：城市景区通达性优于乡镇景区通达性，平地景区通达性普遍优于山丘景区通达性，5A景区通达性明显优于4A及其以下等级景区通达性，原有景区通达性优于新建景区通达性。区域经济发展水平以及景点的知名度对旅游景点的整体交通网络可达性指数影响较大。

Fig.4 Tension curve of the hanger rope along water depth under wave condition. (a) experimental result; (b) simulation result.

Fig.5 Disturbance curve along water depth under wave condition: (a) and (b) are the horizontal and vertical disturbances of SPS, respectively; (c) and (d) are the horizontal and vertical disturbances of buoyancy module, respectively.

4.3.2 Under current condition

The fluid transport equation in FLOW-3D is expressed as follows (Wang et al., 2012):

The disturbance of the SPS and the buoyancy module under the current condition is distinguished from that under wave condition. Two remarkable fluctuations appear in the disturbance curve (Fig.7), whether in the horizontal or vertical direction; these fluctuations correspond to the SPS water entry process and the buoyancy module water entry processes, respectively. The material of the buoyancy module is lighter than that of SPS; thus,the second remarkable fluctuation is larger than the first one. As mentioned previously, the structures swing symmetrically from the left to the right under wave condition.However, the structures swing in the flow direction only under the current condition. As the water depth increases,the swing amplitude stabilizes with a certain offset distance rather than zero, thereby enlarging the rope tension to a certain degree. The offset distance is so small that the induced tension is not obvious under the given circumstance. Moreover, the current action has little effect on the structures in the vertical direction except for the moment when they enter the water.

Fig.6 Tension curve of the hanger rope along water depth under current condition. (a) experimental result; (b) simulation result.

Fig.7 Disturbance curve along water depth under the current condition: (a) and (b) are the horizontal and vertical disturbances of SPS, respectively; (c) and (d) are the horizontal and vertical disturbances of buoyancy module, respectively.

The above shows that current action mainly causes an offset distance in the horizontal direction and barely influences the rope tension. Moreover, the offset distance increases because of the presence of the buoyancy module. Therefore, guidelines are needed in the installation process to control the offset caused by the current.

现阶段，“跨界”已成为当下盛行的元素。媒介传播也需要尝试跨界相融，国家领导人于2017年元旦通过《人民日报》的“两微一端”给全国人民拜年的形式吸引不少读者。[5]加之各大自媒体上能够进行分享，微信与QQ等社交网站上处处可见。从最开始的纸媒到如今的互联网媒体，虽然《人民日报》与时俱进，但没有VR技术作为支撑，这样的传播奇迹是很难实现的。因此，当VR技术与新闻相融时也可将各种媒介载体进行跨界相融。

4.3.3 Under wave-current coupled condition

The chosen current load is smaller than the wave load in the wave-current coupled condition. Thus, the tension(Fig.8) and the disturbance (Fig.9) are close to the results of the wave condition (Figs.4 and 5). The analysis and comparison show that the tension and disturbance under the wave-current coupled condition are not equal to the sum of the tension and disturbance under the wave and current condition separately but are slightly less than the sum. This finding indicates that current has an inhibiting effect on wave action to a certain degree.

4.3.1 Under wave condition

5.4.3 Influence of water depth

Fig.8 Tension curve of the hanger rope along water depth under wave-current coupled condition. (a) experimental result;(b) simulation result.

Fig.9 Disturbance curve along water depth under wave-current coupled condition: (a) and (b) are the horizontal and vertical disturbances of SPS, respectively; (c) and (d) are the horizontal and vertical disturbances of buoyancy module, respectively.

5 Simulation of the Buoyancy Module Retrieval Process

After installation, the buoyancy module should be released for recycling (Fig.10(a)). The light material of the buoyancy module causes the module to float after being released. However, after a period of rising, the buoyancy module will rush out from the water with a certain velocity and pop-up distance, thereby potentially posing a risk to the recycling ship and other structures on the sea.With knowledge of how the velocity of buoyancy module changes during the floating-up period, some measures can be taken to perfect the technology. Numerical models are established to solve this problem.

Fig.10 Schematic diagram of the buoyancy module retrieval process: 1, hanging bracket; 2, winch; 3, heave compensator; 4, hanger rope; 5, installation; 6, buoyancy module; 7, drag parachute.

5.1 Basic Assumption of the Simulation Model

The buoyancy module is released in the deep sea. Two assumptions are made based on the deep-sea conditions.First, wave energy mainly focuses on the sea surface and decreases rapidly along the water depth. As a result, wave load has little influence on the vertical motion of buoyancy module around the seafloor, and wave action can be ignored during the floating-up period. Second, during the floating-up period, the speed of the buoyancy module is much faster than the current speed. Thus, the buoyancy module is mainly affected by water resistance in the vertical direction. Therefore, current action is ignored for the vertical motion of the buoyancy module. Guo et al. (2015)made a similar assumption in their study on the mooring vertical movement model.

5.2 Buoyancy Module Floating-up Model

The buoyancy module and the numerical flume are established for a physical experiment. The size of the buoyancy module is 0.1 m×0.1 m×0.047 m, of which the density is 372 kg m−3. The numerical flume is 1 m long, 1 m wide, and 2 m high (Fig.11). To save calculation time,gradient mesh is adopted, and the minimum size is 0.01 m.The total number of the grids is 1344800. The wall and bottom of the flume are set as no-slip wall boundaries.The initial water level is 1.8 m, and the initial time step is 0.005 s.

Fig.11 Numerical flume.

Fig.12 Floating-up vertical velocity of buoyancy module.

5.3 Analysis of Computing Results

After simulation calculation, a comparison is performed between the simulative floating-up process and experimental process at the same moment, and the comparison results are shown in Fig.13. The displacement of the buoyancy module in the simulative result agrees well with that in the experimental result, thereby verifying the correctness of the floating-up simulation model.

The floating-up vertical velocity of the buoyancy module is shown in Fig.12. The floating-up velocity increases rapidly at the initial stage and reaches a relatively stable speed in a short time. At that moment, the velocity is approximately 0.6 m s−1. Afterwards, the velocity of the buoyancy module speeds up gradually for a while and reaches a balancing speed eventually. The balancing speed is approximately 0.7 m s−1. At the balancing speed, the buoyancy module floats up continually and emerges from the surface. Notably, a pop-up distance exists when the buoyancy module emerges from the water and then falls on the surface, vibrating until it stabilizes.

手性药物领域在1997年于伦敦举办的国际药物成分大会上发生了根本性地改革，对于已经开发上市的消旋体药物或者非对映异构体的混合物而言，可以将其转换为单一的活性对映体进行开发，这种方式被称之为手性转换（chiral switch）［10‐11］。该政策无疑扩大了手性药物的专利保护范围并延长了药物的专利寿命。值得注意的是，给定药物的手性转换并不一定意味着外消旋体以前已经上市而手性转换的基本标准是药物手性状态的改变。手性转换的优势包括：（1）通过提高药效，降低毒性以及更好的选择性来改善药物的治疗指数；（2）药物起效更快；（3）降低药物‐药物间的相互作用；（4）降低患者药物的暴露剂量。

From the perspective of mechanical equilibrium, the floating-up buoyancy module is affected by gravity, flotage,and drag force, without considering the effect of accelerated speed. When the buoyancy module rises at a balancing speed, the balance motion equation of buoyancy can be expressed as

where σ and V represent the mass density and volume of the buoyancy module, respectively; ρ is the mass density of sea water; and FD is the drag force, which is composed of frictional drag Ff and pressure drag Fp. Pressure drag is decided by the shape of the structure. Thus, it is also called form drag according to Li and He (2006) and is expressed as follows:

where Cf and Cp represent the coefficient of frictional drag and pressure drag, respectively; Af is the shear stress area; and Ap is the meeting area of the flows.At the initial floating-up stage, the velocity of the buoyancy module is small. The flotage is much larger than resistance, which results in a great floating-up acceleration. As a result, the floating-up velocity increases rapidly in a short time. According to Eqs. (6–8), resistance is proportional to the floating-up velocity and increases rapidly along with velocity. As the velocity increases, resistance is close to the sum of gravity and flotage. Thus, the floating-up acceleration decreases and the floating-up velocity gradually stabilizes, tending to the balancing speed. Therefore, the balancing speed of the buoyancy module is the result of gravity, flotage, and resistance.

Overall, after the module is released in the deep sea,the velocity of the buoyancy module cannot increase all throughout the process. Through a short period of floating-up, the buoyancy module reaches a balancing speed and floats up continually at this speed until it emerges from the surface.

However, a pop-up distance exists when the buoyancy module rushes out from the water, thereby potentially posing a risk to the recycling ship and other structures on the sea. Thus, measures should be taken to reduce the rushing-out velocity.

Fig.13 Comparison of the experimental and simulative floating-up processes at the same moment.

5.4 Factors That Influence the Floating-up Velocity

As shown in Eqs. (6–8), the floating-up velocity is related to the mass density, shape, and drag coefficient of the buoyancy module. The drag coefficient can be obtained from an empirical formula. However, the relationship between the floating-up velocity and the mass density, shape, and drag coefficient of the buoyancy module still needs to be discussed.

Three assumptions are made for the following research.First, sway, surge, and heave of crane ship do not affect the hanger rope system, which means that crane ship remains still on the sea. Second, the hanger rope has no mass, so that a load of wave and flow that lays on it can be ignored. Finally, the SPS is lowered down at a constant speed.

According to wave theory, the wave energy mainly focuses on the water surface. Therefore, as the water depth increases, the wave energy and the structure disturbance are reduced, as shown in Fig.5. However, whether in the horizontal or vertical direction, two remarkable disturbances are caused by wave action when these two structures enter the water. The two structures swing to the left and the right symmetrically in the horizontal direction under wave condition. Moreover, given that the buoyancy module is made of light materials, the largest disturbance occurs in the buoyancy module water entry process. The disturbance of the buoyancy module actively affects the SPS disturbance. In addition, the vertical disturbance is smaller than the horizontal disturbance because of gravity.The fluctuations in the rope tension are also related to disturbances in both the horizontal and vertical directions.

5.4.1 Influence of mass density

The floating-up velocity is related to the mass density of the buoyancy module under the condition that the shape and volume of buoyancy module, as well as the flow field, do not change. The mass density of the buoyancy module is set as σ2=420 kg m−3, σ3=62 kg m−3, respectively, which are the same with the experimental models, and then a comparison is performed with the result with the density of σ1=372 kg m−3. Results are shown in Fig.14, and two viewpoints can be acquired. First, the change of mass density has little influence on the initial acceleration process but affects the final balancing speed markedly. Second, the final balancing speed is inversely proportional to the mass density. Therefore, under the condition that the shape and volume of the buoyancy module, as well as the flow field, do not change, a large mass density corresponds to a small final balancing speed and thereby less danger during the installation period.However, as a buoyancy auxiliary installation facility, a buoyancy module with a large mass density provides less flotage and is less helpful in the installation process. As a consequence, the mass density of the buoyancy module is decided considering both demand and safety. If the required density is too small, some measures should be taken to decrease the floating-up velocity.

高中物理必修知识是整个高中阶段物理知识学习的重点和难点，只有做好这一阶段的学习才能为选修课的学习打下基础.教师在课程完成后，首先回顾教学过程，分组讨论阶段学生是否充分发挥主观能动性，积极思考动手实践，其次思考课程时间分配上，有没有留白过多实践而造成效率低下，最后总结打点计时器的实验过程有没有达到引出加速度概念的目的.通过课后作业的回收情况综合判断本节课的学习成果.

Fig.14 Floating-up velocity of different buoyancy module densities.

5.4.2 Influence of shape

2.维生素B1治疗。轻症患者：给予口服维生素B1。重症（如心型、脑型）及消化道功能紊乱者：对症治疗，如纠正心力衰竭、抗惊厥治疗等。应注意静脉注射忌用葡萄糖溶液稀释，以免因血中丙酮酸增加而加重病情。肾上腺皮质激素、ACTH、过量的烟酸和叶酸均妨碍维生素B1的利用，均应避免。

Under the condition that the density and volume of the buoyancy module do not change, that is, the same flotage is provided, the final balancing speed of the buoyancy module is greatly related to the shape. The sizes of different buoyancy module shapes are shown in Table 3. In general, frictional drag is the main factor that affects final balancing speed. To ensure the comparability of the cases,the meeting areas of flowing for the buoyancy modules with different shapes are approximately equal to each other.

Table 3 Sizes of different-shaped buoyancy modules

Shape Density(kg m−3) Size (m) Meeting area of flowing (m2)Square 372 0.1×0.1×0.047 1×10−4 Cylinder 372 R=0.056 h=0.047 1×10−4 Sphere 372 R=0.048 0.73×10−4

As shown in Fig.15, the final balancing speeds of the square and cylindrical buoyancy modules are the same.However, the balancing speed of the spherical buoyancy module is much larger than that of the other two buoyancy modules. Therefore, square and cylinder are more suitable shapes for the buoyancy module. The cylindrical buoyancy module is uniformly stressed; thus, it remains stable more easily than the square buoyancy module during the floating-up period. Therefore, a cylinder is the most suitable shape for the buoyancy module.

Fig.15 Floating-up velocity of different buoyancy module shapes.

Overall, the aid of the buoyancy module can ensure load reduction, thereby proving that this method is feasible. However, after analyzing the results, some points still require attention. First, under wave condition, the range of tension in the buoyancy module water entry process is twice that of the SPS water entry process. To prevent ropes from rupturing in the lowering-down process, a number of hanger ropes should be added. The SPS should be installed during good weather to reduce the discontinuous load of the rope. Second, given that the buoyancy module is made from a light material, the largest disturbance occurs during the buoyancy module water entry process. During the installation process, an offset is caused by the current action. Therefore, guidelines are needed in the installation process to control the installation offset. Both two suggestions are valuable to the implementation and improvement of the buoyancy module auxiliary installation technology.

阅读是人类认识世界和获取知识的基本方式，是孩子学习和成长的重要途径。国际21世纪教育委员会指出，要适应未来社会的发展，教育必须围绕四种基本学习能力来重新设计和重新组织。这四种能力分别是—学会认知、学会做事、学会共同生活和学会生存，其中认知的主要方式就要通过阅读来实现。今天，对于以知识为基础来建构经济体系的国家和民族而言，有多少公民具备快速阅读的能力，是关系国家未来竞争力的重要指标。阅读能力的高低直接影响着一个国家和民族的未来，世界各发达国家都把培养儿童阅读能力作为国家教育改革的重点。语文是基础教育领域的核心课程，阅读教学又是语文教学的核心。因此，语文教学改革必须从阅读开始。

Buoyancy module auxiliary installation is mainly used in deep-sea conditions. Thus, whether the water depth has an influence on balancing speed is an issue that engineering technicians are concerned with. To verify the relationship between water depth and floating-up velocity, we add water depth up to 3.8 m. A comparison of buoyancy module velocities at different water depths is shown in Fig.16. Good agreement is found between the two conditions, thereby indicating that water depth has no influence on floating-up velocity. Therefore, no matter how deep the buoyancy modules are released in the water, the motion states are the same before the balancing speed is reached, and they all reach the balancing state in a short time with the same balancing speed.

Fig.16 Comparison of buoyancy module velocity in different water depths.

6 Model of Buoyancy Module with Drag Parachute

As previously mentioned, after a period of rising, the buoyancy module will emerge from the water with a certain velocity and pop-up distance, thereby potentially posing a risk to the recycling ship and other structures on the sea. To reduce the rush-out velocity and avoid unknown dangers, we add a drag parachute to the buoyancy module to increase its resistance (Fig.10(b)). Experiment and simulation are performed to verify the effect of drag parachute.

20世纪60、70年代对于美国福利发展而言，具有分水岭的意义。这一阶段，美国政府福利政策呈现出前所未有的自由化倾向，公共福利规模发生鲜有的爆炸式增长，黑人经济环境的恶化与政治影响力的扩大，是理解这次福利发展的关键［5］62，67。

Restricted by experimental conditions, three assumptions are made to perform the following research. First, the drag parachute is assumed to be a rigid structure. Second,the drag parachute was already stretched in the initial state, and the stretched shape is a semi-sphere. Third, the density of the drag parachute is 1000 kg m−3; thus, the pull of the parachute equals the flow resistance only.

The module-parachute model is shown as Fig.17 and Table 4. The diameter of the parachute is 12 cm. The floating-up process becomes slightly complicated because of the drag parachute (Figs.18 and 19). The process can be divided into the acceleration, deceleration, and uniform stages. A comparison between the experimental and the simulative results is shown in Fig.20. Good agreement is observed between the experimental and the simulative displacements.

Fig.17 Model of buoyancy module with drag parachute.

Table 4 Parameters of the buoyancy module and drag parachute

Model Density (kg m−3) Size (m)Buoyancy module 372 0.1×0.1×0.047 Drag parachute 1000 R=0.06

Fig.18 Floating-up vertical velocity of buoyancy module with drag parachute.

Fig.19 Comparison of buoyancy velocity and buoyancy module with drag parachute velocity.

Similar to the single buoyancy module, the floating-up process of the module-parachute model is also the result of adaptation between floating-up velocity and drag force.However, a rope connects the buoyancy module with the parachute, thereby reducing the sensitivity of speed transportation. During the early stage, the velocity reaches a large value, and the velocity curve is not smooth. As a result of the large velocity, the drag force is greater than the sum of gravity and flotage, and then the speed slows down until the drag force equals the sum of gravity and flotage. Finally, the module-parachute floats up at the balancing speed.

The scale of the experimental model is 1:100. In accordance with the scale, the simulation results of the natural model are calculated in Table 5. At a 180 m water depth, the final speed of the buoyancy module is 7.2 m s−1,and the pop-up distance of the mass center is 4.1 m. With the aid of the drag parachute, the final speed of the buoy-ancy module is 3.3 m s−1, and the pop-up distance of the mass center is reduced to 1.37 m, which means that the bottom of the buoyancy module does not rush out from the water. The parachute can hold water when the buoyancy module rushes out. As a result, the pop-up distance can be small even though the buoyancy module has a high speed. These findings indicate that the drag parachute can reduce both the rush-out speed and the pop-up distance effectively, which means it can be taken into account in engineering application.

Fig.20 Comparison of experimental and simulative floating-up processes at the same moment.

Table 5 Comparison of experimental and natural models

Situation Model Final speed(m s−1) Time (s) Mass center Bottom Pop-up distance (m)Experimental size Module-parachute 0.32 3.47 0.011 −0.013 Buoyancy module 7.20 20.1 4.10 1.75 Natural size Module-parachute 3.30 33.9 1.37 −0.98 Buoyancy module 0.70 2.05 0.038 0.015

7 Conclusions

In this paper, we establish two simulation models based on FLOW-3D: the SPS lowering-down model and the buoyancy module retrieval model. On the basis of calculations and summarization, the following results were obtained:

1) Loading a buoyancy module on the SPS can reduce the lifting weight effectively, thereby verifying that the method is feasible.

2) When the buoyancy module enters the water under wave condition, the amplitude of tension fluctuation is twice that when the SPS enters the water. Therefore, the ability of hanger ropes to bear discontinuous loads should be strengthened to avoid ruptures during construction.

3) Under current condition, the displacement of SPS becomes three times larger because of the existence of the buoyancy module. Thus, guidelines are needed to control the displacement.

4) After the module is released, the velocity of the buoyancy module increases to a great speed rapidly and then reaches a balancing speed gradually. The buoyancy module floats up at the balancing speed and rushes out from the water with a pop-up distance.

5) In deep water, the floating-up velocity of the buoyancy module is related to its mass density and shape, and is not related to water depth.

工程结构试验是工程专业的必修课，包括科学研究和生产评估。其中，科学研究是研究和开发新材料，新结构和新技术的重要手段。生产测试用于施工质量评估，结构可靠性测试，剩余寿命估算和灾害结构事故识别。该课程具有较强的实用性，在工程人才培养中发挥着重要作用。然而，在传统的教学过程中，实践环节往往是在考试指南的基础上进行的，这减少了学生的参与。同时，测试环节的独立性太强，测试内容和使用的知识不一致，这可能会导致学生认知的局限。特别是，生产测试和科学研究测试的作用往往被忽视，理论与现实之间的联系也较少。

6) A drag parachute can reduce the floating-up velocity and pop-up distance effectively, and it should be taken into account in engineering practice.

Acknowledgements

The study was partially supported by the National Natural Science Foundation of China (Nos. 51479183 and 5177 9236).

References

Bhinder, M. A., Rahmati, M. T., and Aggids, G. A., 2009. A joint numerical and experimental study of a surging point absorbing wave energy converter (WRASPA). Asme International Conference on Ocean, Honolulu, Hawaii, 869-875.

Dargahi, B., 2010. Flow characteristics of bottom outlets with moving gates. Journal of Hydraulic Research, 48 (4): 476-482.

Ding, X. T., Xu, D. L., and Zhu, F., 2015a. Numerical simulation of hydraulic characteristic of swirling device based on FLOW-3D software. International Conference on Modelling,Identification and Control, Sousse, 1-4.

Ding, Y., Ma, R., and Li, N., 2015b. A simulation model for three-dimensional coupled wave-current flumes. Engineering Mechanics, 32 (10): 68-74.

Driscoll, F. R., Lueck, R. G., and Nahon, M., 2000a. Development and validation of a lumped-mass dynamics model of a deep-sea ROV system. Applied Ocean Research, 22 (3): 169-182.

Driscoll, F. R., Lueck, R. G., and Nahon, M., 2000b. The motion of a deep-sea remotely operated vehicle system: Part 2: Analytical model. Ocean Engineering, 27 (1): 57-76.

Ferdos, F., and Dargahi, B., 2016. A study of turbulent flow in large-scale porous media at high Reynolds numbers. Part I:numerical validation. Journal of Hydraulic Research, 54 (6):663-677..

Floryan, J. M., and Rasmussen, H., 1989. Numerical methods for viscous flows with moving boundaries. Applied Mechanics Reviews, 42 (12): 323-341.

Gong, J. D., Yu, J. Q., and Zhang, P. P., 2008. Numerical calculation of the resistance coefficient of an auto-returned seawater sampler and its method verification. Marine Sciences, 7(3): 169-178 (in Chinese with English abstract).

Griffith, R. A., Rutherford, J. H., Alavi, A., Moore, D., and Groeneveld, J., 2007. Stability review of the Wanapum spillway using CFD analysis. Bulletin, Canadian Dam Association, Fall 2007, 16-26.

Guo, Y. Q., Li, X. L., and Xiong, X. J., 2015. Mathematical model and result validation of mooring vertical movement in the course of floating-up. Coastal Engineering, 2: 11-23 (in Chinese with English abstract).

Han, G., and Tao, J. H., 2001. A simplified method of studying the kinematic properties of a new-type probe for measuring seawater temperature, salinity and depth and its experimental verification. Journal of Hydrodynamics, 4 (16): 467-471.

Hirt, C. W., Amsden, A. A., and Cook, J. L., 1974. An arbitrary Lagrangian-Eulerian computing method for all flow speeds.Journal of Computational Physics, 14 (3): 227-253.

Ho, D. K. H., Cooper, B. W., Riddette, K. M., and Donohoo, S.M., 2006. Application of numerical modelling to spillways in Australia. In: Dams and Reservoirs, Societies and Environment in the 21st Century. London, 951-959.

Huang, X. B., Huang, X. L., and Dong, Y. F., 2013. Hydrodynamic performance analysis of surfacing motion of submarine escape capsule. Journal of Naval University of Engineering,25 (3): 78-81 (in Chinese with English abstract).

Jing, S. Y., Wang, L., Du, H., and Wei, G., 2014. Applications of FLOW-3D in numerical simulation of fluid-structure interaction. In: National Conference on Hydrodynamics. Qingdao,423-428 (in Chinese with English abstract).

Joensen, A., and Paul, D., 2011. A low tech, low risk system for the installation of large structures in deep water. In: SPE Offshore Europe Oil and Gas Conference and Exhibition. Aberdeen, 1-8.

Johnson, M. C., and Savage, B. M., 2006. Physical and numerical comparison of flow over ogee spillway in the presence of tailwater. Journal of Hydraulic Engineering, 132 (12): 1353-1357.

Kang, Z., Jia, L. S., Sun, L. P., and Liang, W. Z., 2012. Design and analysis of typical buoyancy tank riser tensioner systems.Journal of Marine Science and Application, 11 (A03): 351-360.

Li, Y. Z., and He, W. Z, 2006. Engineering Fluid Mechanics.Tsinghua University Press, Beijing, 249-250.

Liang, B., Li, D., and Pan, P., 2015. Numerical study of local scour of pipeline under combined wave and current conditions. International Society of Offshore and Polar Engineers,1 (15): 316.

Lueck, R. G., Driscoll, F. R., and Nahon, M., 2000. A wavelet for predicting the time-domain response of vertically tethered systems. Ocean Engineering, 27 (12): 1441-1453.

Noh, W. F., 1963. CEL: A time-dependent, two-space dimensional, coupled Eulerian-Lagrangian code. Methods in Computational Physics, 3: 117.

Ramaswamy, B., and Kawahara, M., 1987. Arbitrary Lagrangian-Eulerian finite element method for the analysis of fluid flow with free surface. International Journal for Numerical Methods in Fluids, 7 (10): 1053-1075.

Risoey, T., Mork, H., Johnsgard, H., and Gramnaes, J., 2007.The pencil buoy method–A subsurface transportation and installation method. Proceedings of Offshore Technology Conference. Houston, Texas, 1-7.

Roveri, F. E., Cerqueira, M. B., and Machado, R. D., 2005. The utilization of the pendulous motion for deploying subsea hardware in ultra-deep water. Proceedings of 17th Conference on Deep Offshore Technology (DOT2005), Vitoria, 1- 11.

Shahrokhi, M., Rostami, F., Said, M. A. M., and Syafalni, S.,2013. Numerical modeling of baffle location effects on the flow pattern of primary sedimentation tanks. Applied Mathematical Modelling, 37 (6): 4486-4496.

Stock, P. F., de Cerqueira, M. B., and Roveri, F. E., 2002. A new method for deploying subsea hardware in deep water. Proceedings of the 2002 Deep Offshore Technology, New Or-leans, 1-8.

Sun, P. N., Li, Y. B., and Ming, F. R., 2015. Numerical simulation on the motion characteristics of freely rising bubbles using smoothed particle hydrodynamics method. Acta Physica Sinica, 17: 207-221 (in Chinese with English abstract).

Tang, Y. G., Zhang, S. X., Zhang, H. Y., and Zhang, R. Y., 2008.Snap tension induced by slack-taut in a mooring system.Journal of Vibration and Shock, 27 (4): 70-72 (in Chinese with English abstract).

Trulio, J. G., 1966. Theroy and Structure of the AFTON Codes.Technical report no. AFWL-TR-66-19, Air Force Weapons Laboratory, California.

Wang, Y. H., Bao, Z. J., and Wang, B., 2012. Three-dimensional numerical simulation of flow in stilling basin based on Flow-3D. Engineering Journal of Wuhan University, 45 (4): 454-458(in Chinese with English abstract).

Xu, S. S., Chu, X. J., Yu, L. J., Li, C. Y., and Dong, S., 2016.Experimental study on installation method of subsea heavy equipment in deep sea by buoyancy aid. Transactions ofOceanology and Limnology, 2016 (1): 44-51 (in Chinese with English abstract).

Yan, W. J., Liu, W. B., Liu, Y. H., and Lao, J. F., 2011. A convertible method of the model experiment about the buoy floating freely. Ship and Ocean Engineering, 40 (5): 172-173(in Chinese with English abstract).

Zhou, Q. J., Yu, J. X., Du, Z. F., and Liang, J., 2015. Research on movement response of deep sea equipment retrieving process. The Ocean Engineering, 33 (3): 51-58 (in Chinese with English abstract).

Zhu, K. Q., Liu, Y. L., and Jiang, K. Q., 2009a. Fast calculation method for the natural frequency & amplitude of surface ship critical motion of deep sea vertically tethered system. The Ocean Engineering, 27 (2): 41-45 (in Chinese with English abstract).

Zhu, K. Q., Liu, Y. L., Wei, Y., Zhang, F., and Jiang, K. D.,2009b. Dynamic response of underwater vertically suspend systems connected to the surface ship. Navigation of China,32 (1): 21-23.