Research on the Hydraulic Characteristics of Active Ship Collision Avoidance Devices for Hydrodynamic High-Energy Beam Bridges under Relatively Optimum Deployment Conditions

Abstract

To address the limitations of existing bridge anti-ship collision devices, which cannot protect both ships and bridges, this study introduced a hydraulic high-energy beam for inland navigation safety. Using a bridge as the technical basis and a typical ship in a navigable river section as the research object, the reasonable deployment angle of the device was investigated and the optimal jet ratio of the device R (the ratio of the high-energy beam jet to the mainstream flow velocity) was clarified through combined numerical simulations and a generalized model test. The ship’s motion response state was subsequently validated when the device was reasonably deployed. The results showed that the reasonable deployment angles of the device were 0°, 15°, and 30°. R = 4 served as the optimal jet ratio. Furthermore, the peak value of turbulence intensity in the Y direction was noticeably smaller than in the other three groups, with a stable change. The coordinate error of the key positions in the numerical simulations and generalized model test of ship motion response was less than 10%, the maximum error of the transverse coordinate of the deflection position was −9.8% and the maximum error of the longitudinal coordinate was −7.0%. The maximum error of the transverse coordinate of the maximum deflection position was −6.8% and the maximum error of the longitudinal coordinate was 3.7%. The numerical simulations and generalized model tests of ship motion response fit well.

1. Introduction

Bridges play a critical role in land transport. However, piers in waterways serve as obstacles to navigable ships, endangering safe ship navigation [1,2,3,4]. This issue is amplified by the increasing tonnage and number of ships, highlighting the conflict between water-related piers. As a result, there are significant challenges to the safety of ship navigation in inland river bridge waters. Existing bridge collision avoidance systems can be divided into passive collision avoidance and intelligent early warning systems, according to the role of collision avoidance [5]. Existing passive collision avoidance devices mainly consist of collision fenders and energy dissipation collision avoidance setups [6]. In recent years, materials such as fiber-reinforced polymers [7,8,9,10,11], carbon-fiber-reinforced polymers [12,13], and rubber and concrete damping supplies [14,15,16,17] have gained traction due to their energy absorption capabilities and use in passive applications as anti-vessel collision devices. Existing intelligent early warning systems mainly consist of laser-ranging alarm systems, which serve as the main safety component for ship navigation in inland bridge waters. Existing intelligent warning systems include laser-ranging alarm systems, infrared surveillance, and acoustic alert systems. Some studies have integrated new algorithms, such as automatic target recognition and the tracking (ATR) framework based on sparse coding [18], as well as new algorithms based on image and motion-related information [19], into intelligent warning systems to improve warning accuracy. However, these intelligent warning systems only have an effect on ship yaw resulting from specific factors. If the intelligent warning system fails to warn a ship effectively, passive collision avoidance devices are still needed to ensure the safety of the bridge [20]. Once a ship bridge collision occurs, it affects the normal navigation of the waterway and can even cause major bridge damage and negative societal impacts. For example, on 28 February 2019, the Russian cargo ship Sigran collided with the Gwanganri Bridge in Busan, South Korea, after deviating from the navigation channel due to improper operation shown in Figure 1a. Similarly, on 6 April 2019, a bridge over the Mojo River in Pará, Brazil, partially collapsed after it was hit by a ship, resulting in approximately 200 m of bridge deck damage and two cars falling into the river, as shown in Figure 1b.

Current anti-collision devices lack true active collision prevention. In cases where collisions cannot be avoided, passive protective devices are still needed to ensure bridge safety, protect both the ship and the bridge, and prevent other issues. Therefore, an active anti-collision method for bridges based on hydrodynamic energetic beams was proposed in this work, using hydrodynamic energetic beams to form an intervention belt with pressure greater than on the backflow surface. The energetic beams were gradually curved along the direction of the flow and eventually aligned with the direction of the cross-flow [21]. Subsequently, an intervention belt was used to direct the bow of the yawing ship to realize active collision avoidance. An existing warning system and passive anti-ship collision device formed the Trinity Bridge anti-ship collision system. Consisting of the bridge abutment in the bridge section, the river flow conditions, and the channel level, the bridge channel was divided into a warning area, emergency area, intervention area, and bridge area. Three lines of defense were built to ensure the safety of navigation in the bridge area waters, as shown in Figure 2. A hydraulic high-energy beam device was installed in the intervention area, and high-accuracy detectors and a water flow interference linkage system were deployed on the bridge. When a yawing ship entered the emergency area and was still yawing after an invalid warning, the hydraulic high-energy beam device was deployed. The operation principle of the hydraulic high-energy beam device is shown in Figure 3. By monitoring the ship’s movement trajectory in the waters around the bridge area, a hydraulic high-energy beam device in the intervention zone was set on the side of the bow of the vessel to change the ship’s direction and ensure active collision avoidance. This reduced the risk of the ship hitting the collision avoidance devices and causing damage to the ship, enhancing the safety of the navigable sections of the river and the bridge [19].

In this study, a bridge in the Jialing River basin was taken as the engineering background, and a typical ship in a navigable river section was taken as the research object. To prove the effectiveness of the device [20], numerical simulations and generalized model tests were carried out using a Fluent overset grid [22,23] technology and RNG k-ε turbulence model [24,25] to explore the reasonable deployment angle of the device and clarify the optimal jet ratio of the device, R. Finally, by comparing the optimal jet ratio and the difference of the ship’s motion and response state at the reasonable deployment angle, we provided a basis for the research and development of safe, efficient, and green collision avoidance technology and equipment for ships and bridges.

2. Model Building

2.1. Engineering Background

A bridge in the Jialing River basin served as the engineering background. Building upon the original passive anti-ship collision device, a new approach was developed in this study to provide a ship deflection collision avoidance method based on hydrodynamic high-energy beam intervention in the flow field for ship collision prevention. The bridge had a span of 1458 m, with the greatest risk of ship collision occurring at the No. 6 main perennial pier (Mao 2021), as shown in Figure 4. Therefore, the primary focus of this work was the No. 6 main pier for anti-ship collision research.

2.2. Generalized Model Building

The test was carried out in the outdoor test hall of the Chongqing Bridge Navigation Safety and Collision Avoidance Engineering and Technology Research Center in a rectangular section flume 30 m × 2 m × 0.9 m in size, with a 2% drop in the bottom of the flume; The generalized model tests were designed according to the geometric similarity criterion, and the relevant physical scales are shown in

Table 1. Scale of Relevant Physical Quantities.

Name of Physical Quantity	Geometric Scale	Flow Velocity Scale	Water Flow Scale	Gravimetric Scale
Scale	1:100	1:10	1:100,000	1:1,000,000
Formula	$\mathsf{\lambda }$	${\mathsf{\lambda }}_{\mathrm{V}}={\mathsf{\lambda }}^{1/2}$	${\mathsf{\lambda }}_{\mathrm{F}}={\mathsf{\lambda }}^{2.5}$	${\mathsf{\lambda }}_{\mathrm{G}}={\mathsf{\lambda }}^3$

.

According to the hydrological conditions of the navigable river section and the dimensions of typical navigable ships [26], a 1000 t class ship with 1500 t of full-loaded displacement was selected as the research object, with dimensions of 64.5 m × 12 m × 5.4 m (length × width × depth). The relevant test equipment included the ship model, bridge pier and collision avoidance device model, hydraulic high-energy beam device, ultrasonic flow velocity meter (±1% Fs ± 2 mm/s), ultrasonic level meter (measuring range of 5 m, blind zone <0.3 m, ±0.25% FS), high-pressure pump, and high-definition video camera. The boat model mass was 1.5 kg, as shown in Figure 5a, with a flume flow rate of 0.23 m/s. A high-energy beam device was used with a diameter of 0.006 m and length of 0.01 m, with a PVC nozzle spacing arrangement of 0.01 m and diameters of 0.0075 m, with a PVC body, as shown in Figure 5b. The principle of operation for the suction pump consisted of downstream channel suction through the supercharger delivery to the nozzle spray, with a 370 W constant-pressure variable-frequency supercharger. The suction pump nozzle structure is shown in Figure 5b, and the overall water pump is shown in Figure 5c.

An ultrasonic level meter was used to monitor the test water depth; an ultrasonic tachometer monitored the water flow rate; and a high-definition camera was set up in the test area directly above the bridge pier to record video of the ship’s motion response. In the water inlet portion of the rectification wall and pressure wave plate, the control test water depth was set to greater than 0.3 m to eliminate the effect of shallow water on the test [27]. The test setup had several considerations, namely, the test abutments and the overall arrangement of the passive collision avoidance devices in the test area, which were positioned 2.8 m from the left side of the entrance. In addition, the vertical distance from the side wall was 0.5 m to eliminate the shore wall effect on the ship’s impact [28,29]. The central deployment of the high-energy beam device was set at a vertical distance of 0.3 m from the side wall. The test region was set at a horizontal distance from the left side of the entrance to prevent damage to the bridge abutment in the high-energy beam intervention zone in the mainstream flow direction. The net distance between the piers was less than 1.5 D for the dangerous navigation distance (D: width of the piers) [29], as the ship was more dangerous downstream [30]. This study positioned the ship along the channel boundary from the side wall 0.5 D (77 m) downstream of the most dangerous conditions as the ship’s initial conditions. When the ship was 0.5 m from the nozzle, the device was turned on. A schematic showing the model test setup is shown in Figure 6.

2.3. Numerical Modeling

Numerical simulations of the ship’s response under the action of the intervention zone were carried out using the Fluent 2020R2 and the overset technique. The mesh was updated by spring-based smoothing and local re-meshing, and double precision solving was used to ensure the accuracy of the results [31,32], as well as prevent the degradation of the mesh around the ship or the formation of negative volume due to ship displacements that were significantly greater than the mesh size during the simulation. RNG k-ε turbulence model was selected to ensure interpolation precisions, as well as reduce computer round-off errors. The response state of the ship’s motion was compared with the test results of the generalized model.

2.3.1. Computational Domain Layout and Boundary Conditions

A 600 m × 200 m computational domain simulation test area was established in Fluent-DM, and the overall layout of the test area was the same as the test. The piers and dynamic collision avoidance devices were set up on the left side of the test area at a distance of 280 m from the boundary of the left entrance, and the vertical distance to the side wall was 50 m. The center of the high-energy beam device had a vertical distance of 30 m from the side wall, and the horizontal distance to the left entrance of the test field was 30 m. The ship moved along the channel boundary 0.5 D (77 m) from the side wall and deployed the device when it was 50 m away from the nozzle. The detailed parameters of the numerical model ship are shown in

Table 2. Detailed parameters of the modeled ship.

Parameter	Value	Parameter	Value
Length (m)	64.5	${\mathrm{I}}_{\mathrm{Z}\mathrm{Z}}\left(\mathrm{kg}·{\mathrm{m}}^2\right)$	$5.14\times {10}^8$
Width (m)	12	Load displacement (kg)	$1500\times {10}^3$

, and the computational domain layout and naming are shown in Figure 7.

The computational setup involved specific boundary conditions. For the inlet conditions, we specified the left boundary of the computational domain as the velocity inlet (inlet), given that the water flow conditions for the X-axis were positive. For the exit conditions, we specified the right boundary of the computational domain as the free outlet (outlet). For the piers and collision avoidance devices, solid wall boundary conditions (pier) were implemented, while the boundaries on the two sides of the computational domain were set to solid wall boundary conditions (wall), and the node velocity was 0. For the hydrodynamic high-energy beam nozzle conditions, a computational domain of the hydrodynamic high-energy beam nozzle was specified at the velocity of inlet 2 (inlet2), given the water flow conditions for the Y-axis forward. In terms of the dynamic grid conditions, the ship boundary was defined as (ship), and the boundary of the double ship width area around the ship was set as the designated dynamic grid boundary (overset).

2.3.2. Reliability Verification and Meshing

The number of device nozzles was selected as 16 with a deployment angle of 15°. Since the mesh size at the waterline interface significantly affects the mesh quality, we perform the mesh delineation at the interface grid sizes of 0.012 m, 0.013 m, 0.014 m, and 0.015 m. The grid refinement was applied to the outside of the nozzle of the hydrodynamic high-energy beam device. The area surrounding the ship, which extended twice the width of the ship, was designated as the foreground grid. The grid transition rate was 0.2, and the grid growth rate was 1.2, resulting in a set of 20 layers. Total grids are 4.69 million, 5.59 million, 6.46 million, and 7.41 million, and the grid skewness is kept less than 0.8 for the verification of grid independence. The peaks of the stress ephemeral curves of the devices at different grid numbers are compared, as shown in

Table 3. Meshing independence verification.

	Model 1	Model 2	Model 3	Model 4
Waterline grid size (M)	0.012	0.013	0.014	0.015
Total grids (Million)	469	559	646	741
Device stress peak (KN)	69.07	62.58	61.96	62.30

.

When the total number of model grids is 4.69 million, the error of peak device stress is 10.37% compared to 5.59 million, and the total number of model grids is 5.59 million, the error of peak device stress is 1.0% compared to 6.46 million. Therefore, in this paper, we choose the full domain grid number of 5.59 million for the time independence validation, and establish five monitoring points at X/D = 15, Y/D = 2, 4, 6, 8, 10, and compare the V_X and V_Y of the five monitoring points at the moment of 30 s under the calculation of different time steps, and validate the temporal independence of the step size of 0.0125 s, 0.01 s, 0.0075 s, and 0.005 s as in

Table 4. Time-step irrelevance verification.

	Point 1	Point 2	Point 3	Point 4	Point 5
VX (0.0125 s)	1.255	1.570	1.299	2.283	1.986
VX (0.01 s)	1.209	1.457	1.204	2.127	1.834
VX (0.0075 s)	1.185	1.445	1.179	2.107	1.800
VX (0.005 s)	1.191	1.408	1.130	2.175	1.829
VY (0.0125 s)	1.732	1.596	1.612	1.574	1.008
VY (0.01 s)	1.618	1.485	1.517	1.463	0.946
VY (0.0075 s)	1.621	1.448	1.493	1.449	0.940
VY (0.005 s)	1.595	1.405	1.535	1.463	0.914

. The relative error of V_X is 7.01%, and the relative error of V_Y is 6.98% for the time steps 0.0125 s and 0.01 s, and the relative error of V_X is 1.56%, and the relative error of V_Y is 1.11% for the time steps 0.01 s and 0.0075 s, so in this paper, we chose 0.01 s as the time step for the numerical simulation. The typical working condition of the test area for local refinement of the test area is shown in Figure 8, and the overall meshing of the computational domain is shown in Figure 9.

2.3.3. The RNG k-ε Turbulence Model

The RNG k-ε model is derived using a statistical method known as reformulated group theory. It is formally similar to the standard k-ε model but includes the following improvements: The RNG model adds a term to its ε equation, which improves the accuracy of the high-speed flow. Due to the strong anisotropy of high-energy beams and the different turbulence models with different assumptions and derivations with their own scope of application, each model was suited to specific applications [33,34]. To obtain more accurate simulation results, in this work, we compared the realizable k-ε, standard k-ε, and RNG k-ε turbulence models because the standard k-ε turbulence model could not reflect this type of flow, and the realizable k-ε model produced unphysical turbulence viscosity in certain calculation processes. In this work, the RNG k-ε turbulence model was finally chosen [35], which more accurately simulated the issue of large curvature and anisotropy of the streamlines. The flow control equations and the RNG k-ε turbulence model are provided in the following equations.

Continuity equation

$$\frac{\mathsf{\partial }\mathsf{\rho }}{\mathsf{\partial }t}+\frac{\mathsf{\partial }\left(\mathsf{\rho }{u}_i\right)}{\mathsf{\partial }{x}_i}=0$$

where $t$ is the time in s; $u$ is the velocity of the water in m/s; $\mathsf{\rho }$ is the density of the water in kg/m³.

Momentum equation

$$\frac{\mathsf{\partial }\left(\mathsf{\rho }{u}_i\right)}{\mathsf{\partial }t}+\frac{\mathsf{\partial }\left(\mathsf{\rho }{u}_i{u}_j\right)}{\mathsf{\partial }{x}_j}=-\frac{\mathsf{\partial }p}{\mathsf{\partial }{x}_i}+\frac{\mathsf{\partial }}{\mathsf{\partial }{x}_j}\left[\left(\mu +{\mu }_i\right)\left(\frac{\mathsf{\partial }{\mu }_i}{\mathsf{\partial }{x}_j}+\frac{\mathsf{\partial }{\mu }_j}{\mathsf{\partial }{x}_i}\right)\right]$$

where $p$ is the pressure in Pa, $\mu$ is the molecular dynamics viscosity coefficient, $u_i$ is turbulent viscosity coefficient, ${\mu }_i=\mathsf{\rho }{C}_{\mu }{k}^2/\epsilon$, $C_{\mu }=0.085$.

The turbulence model RNG k-ε equations are shown below.

$$\frac{\mathsf{\partial }\left(\mathsf{\rho }k\right)}{\mathsf{\partial }t}+\frac{\mathsf{\partial }\left(\mathsf{\rho }{\mu }_{i}k\right)}{\mathsf{\partial }{x}_i}=\frac{\mathsf{\partial }}{\mathsf{\partial }{x}_i}\left[\left(\mu +\frac{{\mu }_t}{{\sigma }_k}\right)\frac{\mathsf{\partial }k}{\mathsf{\partial }{x}_i}\right]+G-\mathsf{\rho }\epsilon$$

where $k$ is turbulent kinetic energy, ${\sigma }_k=0.7179$

$$\frac{\mathsf{\partial }\left(\mathsf{\rho }\epsilon \right)}{\mathsf{\partial }t}+\frac{\mathsf{\partial }\left(\mathsf{\rho }{\mu }_i\epsilon \right)}{\mathsf{\partial }{x}_i}=\frac{\mathsf{\partial }}{\mathsf{\partial }{x}_i}\left[\left(\mu +\frac{{\mu }_t}{{\sigma }_{\epsilon }}\right)\frac{\mathsf{\partial }\epsilon }{\mathsf{\partial }{x}_i}\right]+C_1\frac{\epsilon }{k}G-C_2\mathsf{\rho }\frac{{\epsilon }^2}{k}$$

where $\epsilon$ is the turbulent dissipation rate, ${\sigma }_{\epsilon }=0.7179$, $G={\mu }_t\left(\frac{\mathsf{\partial }{u}_i}{\mathsf{\partial }{x}_j}+\frac{\mathsf{\partial }{u}_j}{\mathsf{\partial }{x}_i}\right)\frac{\mathsf{\partial }{u}_i}{\mathsf{\partial }{x}_j}$

$$C_1=1.42-\frac{\tilde{\eta }\left(1-\tilde{\eta }/{\tilde{\eta }}_0\right)}{1+\beta {\tilde{\eta }}^3}, S_{ij}=\frac{1}{2}\left(\frac{\mathsf{\partial }{u}_i}{\mathsf{\partial }{x}_j}+\frac{\mathsf{\partial }{u}_j}{\mathsf{\partial }{x}_i}\right), S=\sqrt{2S_{ij}{S}_{ij}}, \tilde{\eta }=Sk/\epsilon$$

$${\tilde{\eta }}_0=4.38, \beta =0.0154, C_2=1.68$$

The X and Y direction turbulence intensity equation

$${\mathrm{T}}_{\mathrm{U}}=\frac{\mathsf{\delta }}{{\mathrm{V}}_{\mathrm{X}}}=\frac{\sqrt{{{\mathrm{u}}^{\prime }}^2}}{{\mathrm{V}}_{\mathrm{X}}}$$

$${\mathrm{T}}_{\mathrm{U}}=\frac{\mathsf{\delta }}{{\mathrm{V}}_{\mathrm{Y}}}=\frac{\sqrt{{{\mathrm{u}}^{\prime }}^2}}{{\mathrm{V}}_{\mathrm{Y}}}$$

where: $\mathsf{\delta }$ is the root mean square of pulsating flow rate, u is the instantaneous velocity.

3. Influence Analysis of the Device Deployment Angle on the Ship Motion Response State and Device Stress

Due to the considerable time required to set up and adjust the generalized model test for each working condition, in this study, we first used numerical simulations. The primary aim was to compare and analyze the response state of the ship’s motion when the device was set up at different angles, along with the stress–time curve during the process of opening and closing the device. We also explored the reasonable setting angle and determined the optimal jet ratio through the turbulence intensity in the X and Y directions in the device’s downstream local waters.

3.1. Analysis of the Influence of Device Deployment Angle on the Ship Motion Response State

When the device deployment angle was less than −45°, it hindered the formation of the intervention belt due to insufficient nozzle mixing intensity. When the intervention belt deployment angle was more than 45°, the width of the flow surface exceeded the width of the bridge abutment flow surface. Finally, the number of nozzles was selected as 16, and the installation angles of the device were set to −30°, −15°, 0°, 15°, 30°, and 45° for the six conditions to screen the installation angle. The velocity maps for each condition are shown in Figure 10.

As shown in Figure 10, when the device was installed at −15°, 0°, 15°, and 30°, the ship’s motion response showed a better effect, and the intervention belt effect on the ship was produced by the bridge abutment with enough safety distance after the heading changed back to the right. When the device was installed at −15°, the ship’s motion response was less effective, and while the final heading did not go back to the right, sufficient distance was maintained to produce bridge abutment, thus achieving the purpose of active collision avoidance.

When the device was deployed at −30° and 45°, the response effect of ship movement was worse, and when the device was deployed at −30°, the intervention zone at the nozzle and the mainstream intense mixing of energy dissipation was larger due to the direction of the device generating the intervention zone with the velocity component against the mainstream flow. This resulted in incomplete morphological development. When the device was deployed at −30°, the ship collided with the boundary of the test area after the intervention zone. In addition, when the device was deployed at 45°, the ship almost collided with it.

3.2. Analysis of the Effect of the Angle of Arrangement of the Device on the Stress of the Device

The ship motion response state was better at −15°, 0°, 15°, and 30°. The device condition from open to close was monitored for a total of 75 s to achieve device stress monitoring, resulting in a device stress change over time curve, as shown in Figure 11.

According to the stress change curve of the device, when the device was installed at angles of 0°, 15°, and 30°, the ship’s motion response was better, and the stress peak at the rear side of the device was smaller. When the device was deployed at −15°, two stress peaks were observed, one positive and one negative, and both were large, indicating that due to the negative angle between the high-energy beam and the main stream at the nozzle, the energy dissipation was large as a result of intense mixing. Therefore, this angle was not suitable for deployment. The device set at 15° showed the smallest peak stress, with 30° showing a peak value of 44.42%, while −15° showed a peak value of 26.67%.

4. Analysis of the Effect of Jet Ratio on the Turbulence Intensity of the Local Flow Field in the Bridge Area

To achieve active anti-collision while minimizing changes in the local flow pattern within the bridge area, the intervention zone was designed to pivot away from the bow of the ship. The intervention zone with the strongest mixing with the main stream was selected with reference to existing studies, and jet ratios of R = 3, 3.5, 4, 4.5, and 5 were screened [36]. The turbulence strength was used to characterize the transverse and vertical turbulence characteristics of the local waters downstream of the device, which was set as the ratio of the root-mean-square (rms) value of the pulsating flow velocity to the corresponding mean flow velocity. This reflected the strong and weak degree of water body flow pulsation. The turbulence intensity was set as the ratio of the root mean square of the pulsating flow rate to the corresponding time-averaged flow rate, which indicated that the dispersion of the instantaneous flow rate was near its mean value. This signified the strength of the water body flow rate pulsation, where a higher turbulence intensity indicated that this part of the water body was subjected to more intense mixing [37]. The turbulence intensity was taken at 75 monitoring points as Figure 12, positioned at X/D = 3, 6, 9, 12, 15 and Y/D = −10, −8, −6, −4, −2, 0, 2, 4, 6, 8, 10, 12, 14, 16, 18, as shown in Figure 13a–j.

According to the above comparison diagrams, the turbulence intensity in the X and Y directions changed drastically and showed no obvious distribution pattern. In addition, the peak value of turbulence intensity in the Y direction was more dispersed in R = 3, 3.5, and 5, which indicated that the mixing of the high-energy beam with the main stream was drastic and had a larger range for these three jet ratios. Notably, the X-direction turbulence intensity was more concentrated, and the peak value was smaller. In addition, the Y-direction turbulence intensity peak value was small, and the degree of change was relatively smooth for the optimal R = 4 condition, which was used for subsequent investigation. At this point, the X-direction turbulence intensity was more concentrated, and the integrated maximum peak value was small. The maximum peak value occurred at X/D = 3 and Y/D = 12, and the peak size was 36.44. The degree of change in the Y-direction turbulence intensity was relatively smooth, with the maximum peak appearing at X/D = 3 and Y/D = 8, and the peak size was 6.21.

5. Comparison of Ship Motion Response States

5.1. Working Conditions

An optimal jet ratio of R = 4 was achieved for the numerical simulations, and the device was reasonably placed at angles of 0°, 15°, and 30°. Subsequently, the effect of the ship’s motion response was studied, and the working conditions are shown in

Table 5. Numerical simulation working conditions.

Number of Nozzles	Device Angle of Deployment
	0	15	30
8	T1	T2	T3
12	T4	T5	T6
16	T7	T8	T9

. The working conditions of the generalized model tests were the same as those for the numerical simulations shown in

Table 6. Generalized model test scenarios.

Number of Nozzles	Device Angle of Deployment
	0	15	30
8	T10	T11	T12
12	T13	T14	T15
16	T16	T17	T18

.

5.2. Comparison of Ship Motion Response States for Typical Working Conditions

Due to the pressure difference between the front and rear boundaries of the intervention zone, the main body of the intervention zone gradually changed its direction under the action of the main stream into a downstream penetration section parallel to the main stream. Three key moments occurred where a change in the ship’s motion response was observed through the action of the intervention zone. In Moment 1, the ship approached the intervention zone, as shown in Figure 14a. In Moment 2, the intervention zone acted to deflect the bow of the ship, leading to a change in its heading, as shown in Figure 14b. In Moment 3, the ship managed to establish a sufficient safety distance from the bridge abutment after the action of the intervention zone, and its heading returned to its original direction, as shown in Figure 14c.

5.3. Comparison of Ship Motion Response States

The background difference method was used to collect the motion response process of the ship yawing to the course correction by the intervention belt, and the data of each condition were post-processed to export the pictures. The pixel points of the bow and stern positions were subsequently read and transformed into the natural coordinate system, as shown in Figure 15a–d.

According to the error analysis graph, the numerical simulations and the generalized model test deflection position had a maximum error of −9.8% in the horizontal coordinates and −7.0% in the vertical coordinates. The maximum deflection position had a maximum error of −6.8% in the horizontal coordinates and −3.7% in the vertical coordinates, and all values were less than 10%.

6. Conclusions

(1) The reasonable deployment angles of the device were found to be 0°, 15°, and 30°, resulting in better ship motion response and more reasonable device stress. When the device was installed, the peak stress on the rear side was the smallest, and it was only 44.42% of the 30° installation and 26.67% of the −15° installation. One positive and one negative stress peak were observed at −15° installation, which consisted of the largest peaks;

(2) R = 4 was determined as the optimal jet ratio of the device. At this time, the X-direction turbulence intensity was more concentrated with a smaller peak, and the Y-direction turbulence intensity peak was small, with a relatively smooth degree of change. The maximum peak appeared at X/D = 3 and Y/D = 12, with a peak size of 36.44. The Y-direction turbulence intensity of the degree of change was relatively smooth, and the maximum peak appeared at X/D = 3 and Y/D = 8, with a peak size of 6.21. The maximum peak was 6.21 at X/D = 3 and Y/D = 8;

(3) Numerical simulations and a generalized model test of the ship motion response deflection position showed a maximum error of −9.8% in the horizontal coordinate and −7.0% in the vertical coordinate. The maximum deflection position had a maximum error of −6.8% in the horizontal coordinate and −3.7% in the vertical coordinate, with errors all less than 10%. For the same number of nozzles, the larger the spray angle, the more forward the horizontal coordinate of the ship’s deflection, with a maximum side shift position. In addition, the spray angle and number of nozzles could be adjusted according to the use of the device in actual application;

(4) The numerical simulation results and generalized model test ship motion response fit well, showing that the hydraulic high-energy beam device had a significant bridge anti-ship collision effect. Notably, it could quickly increase the distance between the ship and the piers to ensure the navigational safety of inland waterways.

7. Patents

A three-in-one method for preventing boat collisions on wading pier (CN114934481A). A bridge active anti-ship collision method (CN114973771A).