Fishing port fishing boat berthing safety monitoring system

Abstract

The invention relates to the technical field of fishing port safety monitoring, and discloses a fishing port fishing boat berthing safety monitoring system. The system comprises a fishing boat dynamic sensing module which collects dynamic position and attitude data of each fishing boat in a fishing port in real time; the berth state analysis module generates berth occupation state distribution information based on the data; the environmental parameter acquisition module acquires fishing port hydro-meteorological parameters in real time; the safety constraint initialization module outputs an initial safety monitoring constraint including an allowable yaw threshold value and a maximum allowable collision risk value in combination with a berth occupancy state and hydro-meteorological parameters; the parking risk assessment module executes risk quantitative analysis and generates risk level identification data by taking the initial safety monitoring constraint as a condition; a scheduling parameter optimization module optimizes berth resource scheduling according to the risk level identification data to obtain safety scheduling control parameters; and the alarm signal generation module converts an alarm dimension according to the safety scheduling control parameter and the dynamic position data updated in real time, and outputs a multi-stage acoustic alarm signal.

Description

Technical Field

[0001] This invention relates to the field of fishing port safety monitoring technology, specifically a fishing port fishing vessel berthing safety monitoring system. Background Technology

[0002] In the development of coastal fisheries, fishing ports, as the core places for fishing vessels to moor, replenish supplies, and rest, are directly related to the safety of fishermen's lives and property and the order of fishery production. With the expansion of the fishery scale, the number of fishing vessels moored in fishing ports has increased year by year, and the types of fishing vessels have become more diverse, with significant differences in tonnage, which has made the mooring environment in fishing ports increasingly complex.

[0003] In most fishing ports, the management of fishing vessel berthing still relies on traditional methods, primarily through manual patrols to record the position of fishing vessels and the status of their berths. Manual monitoring not only consumes a significant amount of manpower but is also limited by the frequency and scope of patrols, making it difficult to achieve real-time tracking of fishing vessel movements. When fishing vessels shift due to factors such as tides and wind, manual monitoring often lags behind, failing to detect potential collision risks in a timely manner.

[0004] In terms of berth status management, existing technologies lack a systematic analysis of berth occupancy. Berth allocation within fishing ports is typically fixed, failing to consider the differences in berthing needs among fishing vessels of varying tonnages and the impact of real-time environmental changes on berth suitability. This results in low berth resource utilization, with some areas experiencing overcrowding while others remain idle for extended periods. This irrational resource allocation further exacerbates berthing risks.

[0005] The impact of environmental factors on the safety of fishing vessel mooring is equally significant. Changes in hydrological and meteorological parameters such as wind speed, wind direction, tidal height, and current velocity directly alter the attitude of fishing vessels. However, existing monitoring methods mostly collect environmental parameters separately, failing to effectively correlate them with dynamic data of fishing vessels, making it difficult to formulate targeted safety constraints. When encountering sudden weather events, it is impossible to quickly assess the extent of the impact of environmental changes on the mooring safety of individual fishing vessels, hindering timely implementation of preventative measures.

[0006] Traditional risk assessment and alarm mechanisms rely heavily on experience-based judgments and lack quantitative analysis. The standards for defining the degree of vessel deviation and collision risk are vague, and alarm signals are singular and lack a tiered mechanism. When risks occur, managers struggle to accurately determine the risk level and scope of impact based on alarm information, leading to inefficient emergency response and increasing the likelihood of accidents. These problems pose significant challenges to the safe management of fishing vessels in fishing ports, necessitating a comprehensive, real-time, and intelligent monitoring system to address them. Summary of the Invention

[0007] The purpose of this invention is to provide a safety monitoring system for fishing vessels mooring in fishing ports, in order to solve the problems mentioned in the background art.

[0008] To achieve the above objectives, the present invention provides a safety monitoring system for fishing vessels mooring in fishing ports, the system comprising:

[0009] The fishing vessel dynamic perception module is used to collect dynamic position and attitude data of each fishing vessel in the target fishing port in real time.

[0010] The berth status analysis module is used to generate berth occupancy status distribution information based on the dynamic position data and attitude data.

[0011] The environmental parameter acquisition module is used to acquire the hydrological and meteorological parameters of the target fishing port in real time.

[0012] The safety constraint initialization module is used to output initial safety monitoring constraints based on the berth occupancy status distribution information and the hydro-meteorological parameters, wherein the initial safety monitoring constraints include an allowable yaw threshold and a maximum allowable collision risk value.

[0013] The mooring risk assessment module is used to perform quantitative analysis of fishing vessel mooring risks by using the initial safety monitoring constraints as assessment conditions, and to generate risk level identification data.

[0014] The scheduling parameter optimization module is used to perform berth resource scheduling optimization based on the risk level identification data to obtain safe scheduling control parameters.

[0015] The alarm signal generation module is used to perform alarm dimension conversion based on the safety scheduling control parameters and the real-time updated dynamic location data, and output multi-level acoustic alarm signals.

[0016] Preferably, the fishing vessel dynamic sensing module includes:

[0017] A grid partitioning unit is used to divide the port area of ​​the target fishing port into a dynamic grid cell array;

[0018] The trajectory tracking unit is used to locate the real-time grid coordinates of each fishing vessel in the dynamic grid cell array based on radar scan data and AIS positioning data.

[0019] The attitude sensing unit is used to acquire roll and pitch angle data of each fishing vessel through the shipboard tilt sensor.

[0020] The data fusion unit is used to fuse the real-time grid coordinates, roll angle data and pitch angle data into the dynamic position data and attitude data.

[0021] Preferably, the berthing risk assessment module performs quantitative analysis of fishing vessel berthing risks, including:

[0022] A dynamic feature extraction unit is used to calculate the real-time distance change rate between adjacent fishing vessels based on the dynamic position data.

[0023] The safety membership calculation unit is used to perform fuzzy classification on the real-time distance change rate using a predefined membership function to obtain the membership degree of the first type of collision risk and the membership degree of the second type of collision risk.

[0024] The constraint matching unit is used to match and calculate the membership degree of the first type of collision risk, the membership degree of the second type of collision risk, and the allowable yaw threshold.

[0025] The risk quantification unit is used to generate risk level identification data, which includes collision probability value and risk duration, based on the matching results.

[0026] Preferably, the berth status analysis module includes:

[0027] The berth adaptation algorithm unit is used to divide the standard berth area according to the wharf structure characteristics of the target fishing port;

[0028] The occupancy status identification unit is used to map the real-time grid coordinates to the standard berth area and output a berth occupancy rate distribution map;

[0029] The safety distance analysis unit is used to calculate the minimum distance between adjacent berths based on the berth occupancy rate distribution map.

[0030] The safety constraint initialization module updates the allowable yaw threshold based on the minimum spacing value.

[0031] Preferably, the safety constraint initialization module further includes:

[0032] The fluid dynamics simulation unit is used to perform fluid simulation of the berth area based on the flow velocity data, wind direction data, and wave height data in the hydro-meteorological parameters.

[0033] The constraint tuning unit is used to correct the maximum permissible collision risk value based on the simulation results.

[0034] The risk diffusion model unit is used to generate risk propagation time limit data based on the corrected maximum permissible collision risk value.

[0035] Preferably, the alarm signal generation module performs alarm dimension conversion including:

[0036] A spatial coding unit is used to convert the real-time grid coordinates into acoustic alarm spatial codes;

[0037] The risk mapping unit is used to map the collision probability value in the risk level identification data to a baseline alarm frequency.

[0038] A timeliness control unit is used to set the alarm duration based on the duration of the risk.

[0039] The signal synthesis unit is used to fuse the reference alarm frequency, alarm duration, and acoustic alarm spatial encoding to generate the multi-level acoustic alarm signal.

[0040] Preferably, the acoustic alarm spatial coding includes:

[0041] The vertical dimension encoding unit is used to convert the grid vertical coordinates into a gradient of sound wave frequency changes;

[0042] Lateral dimension encoding unit, used to convert grid horizontal coordinates into acoustic wave phase offset;

[0043] A risk intensity encoding unit is used to convert the collision probability value into a sound wave amplitude intensity value.

[0044] Preferably, the scheduling parameter optimization module performs berth resource scheduling optimization by including:

[0045] The scheduling rule base unit is used to store historical collision avoidance scheduling schemes and corresponding execution effect data;

[0046] The parameter evaluation unit is used to filter historical collision avoidance scheduling schemes that meet the time constraints using the risk propagation time limit data;

[0047] A multi-dimensional optimization unit is used to fine-tune the parameters of the selected historical collision avoidance scheduling schemes based on the risk level identification data.

[0048] The control parameter generation unit is used to output the safety scheduling control parameters, which include collision avoidance path, speed adjustment command and priority identifier.

[0049] Preferably, the system further includes:

[0050] The execution feedback module is used to drive the berth guidance device of the target fishing port according to the multi-level acoustic alarm signal;

[0051] The execution feedback module includes:

[0052] The instruction parsing unit is used to generate a berth guidance sequence based on the priority identifier in the safety scheduling control parameters;

[0053] The signal driving unit is used to send the multi-level acoustic alarm signal to the berth guidance device in the designated area according to the berth guidance sequence;

[0054] The status feedback unit is used to collect the execution status data of the berth guidance device in real time and feed it back to the scheduling parameter optimization module.

[0055] Preferably, the scheduling parameter optimization module further includes:

[0056] An adaptive correction unit is used to update the historical collision avoidance scheduling scheme based on the execution status data;

[0057] The timeliness recalculation unit is used to recalculate the risk propagation timeliness data based on the updated historical collision avoidance scheduling scheme;

[0058] The alarm signal generation module adjusts the alarm duration based on the recalculated risk propagation time limit data.

[0059] Compared with the prior art, the beneficial effects of the present invention are:

[0060] This fishing vessel berthing safety monitoring system in the fishing port demonstrates a multifaceted positive role in port safety management through the organic collaboration of multiple modules. The dynamic sensing module for fishing vessels breaks through the limitations of traditional manual monitoring. By utilizing real-time collected dynamic position and attitude data, it can continuously track subtle changes in each fishing vessel, allowing managers to grasp the dynamic status of the vessels in real time and avoid missing risk warning opportunities due to information lag.

[0061] The berth status analysis module generates berth occupancy distribution information based on dynamic data, clearly presenting the usage status of each berth within the fishing port, thus changing the previous chaotic situation of berth management. Managers can use this information to rationally plan berth usage, avoiding overcrowding of fishing boats or berth idling, ensuring fuller utilization of berth resources, and optimizing the spatial layout within the fishing port.

[0062] The environmental parameter acquisition module incorporates hydrological and meteorological parameters into the monitoring system, ensuring that safety management no longer overlooks the impact of the natural environment. Real-time data on wind speed, wind direction, and tides provides crucial references for formulating subsequent safety constraints, making safety monitoring more aligned with the actual environmental conditions of the fishing port and taking into account the varying berthing needs of fishing vessels under different environmental conditions.

[0063] The initial safety monitoring constraints output by the safety constraint initialization module, combining berth status and environmental parameters, are not fixed standards but rather reference values ​​that are dynamically adjusted based on actual conditions. The setting of permissible yaw thresholds and maximum permissible collision risk values ​​provides a concrete basis for risk assessment, avoiding the subjectivity of previous experience-based judgments and making safety management more scientific and targeted.

[0064] The mooring risk assessment module clearly distinguishes the degree of risk faced by different fishing vessels through quantitative analysis of the generated risk level labels, enabling managers to quickly identify high-risk targets and take corresponding prevention and control measures. This quantitative analysis method makes the risk assessment results more intuitive and accurate, facilitating precise policy implementation by managers.

[0065] The scheduling parameter optimization module optimizes berth resource allocation based on risk levels, enabling dynamic adjustments to berth resources while ensuring safety. By optimizing scheduling parameters, fishing vessels are positioned more rationally, reducing potential collision risks and improving the overall berthing safety of the fishing port.

[0066] The alarm signal generation module outputs multi-level acoustic alarm signals, overcoming the limitations of traditional single-level alarm signals. Different alarm levels correspond to different risk degrees, allowing managers to quickly assess the urgency of the situation and take appropriate countermeasures, improving emergency response efficiency and reducing the likelihood of accidents. The collaborative work of all modules forms a complete closed loop from data collection and analysis to decision-making and alarm activation, comprehensively enhancing the level of safety management for fishing vessels berthing in fishing ports. Attached Figure Description

[0067] Figure 1 This is a timing diagram of the fishing port and fishing vessel berthing safety monitoring system described in this invention;

[0068] Figure 2 A flowchart of the workflow for the fishing vessel dynamic perception module;

[0069] Figure 3 The flowchart shows the workflow of the berth status analysis module.

[0070] Figure 4 A flowchart for dimensional transformation of the alarm signal generation module. Detailed Implementation

[0071] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0072] Please see Figure 1 This invention provides a safety monitoring system for fishing vessels mooring in fishing ports, the system comprising:

[0073] By integrating multi-source data acquisition and intelligent analysis modules, the system achieves dynamic monitoring and proactive early warning of fishing vessel berthing safety. The system's dynamic sensing module collects real-time data on the dynamic positions of each fishing vessel within the port area, such as latitude and longitude coordinates, and attitude data, such as hull tilt angles. The berth status analysis module generates a berth occupancy distribution map based on this data, marking the real-time occupancy status of each berth and the distance between adjacent fishing vessels. The environmental parameter acquisition module simultaneously acquires hydrological and meteorological data, such as wind speed, wave height, and tidal current speed. The safety constraint initialization module, combining berth distribution and hydrological and meteorological parameters, dynamically calculates the safety margin for allowable yaw threshold vessel position deviation and the maximum allowable collision risk value (collision probability upper limit). The berthing risk assessment module, using this constraint as a condition, quantitatively analyzes the probability and duration of collision risks between fishing vessels, generating red / yellow / green risk level labels. The scheduling parameter optimization module optimizes berth resource scheduling based on the risk level, outputting safe scheduling control parameters including avoidance paths and speed commands. The alarm signal generation module fuses these parameters with real-time position data, generating a multi-level acoustic alarm signal with spatial orientation indication through acoustic coding technology.

[0074] Example 1: See Figure 2 The harbor waters are divided into a square grid array with sides of 10 meters. The grid origin is fixed at the base of the harbor control tower. Each grid cell has a unique row and column number, for example, from G-01-01 to G-50-80, covering the entire harbor area. Radar equipment scans the water surface once per second, generating raw distance and azimuth data upon detecting reflected signals from fishing vessels. The Automatic Identification System (AIS) sends a location message containing the vessel's name, length, and beam information every five seconds. The trajectory tracking unit receives these two types of signals, first converting the radar polar coordinates to Cartesian coordinates, and then aligning them with the AIS latitude and longitude data in a unified coordinate system. When the radar detects a fishing vessel at an azimuth of 30 degrees and a distance of 150 meters from the control tower, the AIS simultaneously reports the vessel's latitude and longitude as 36.05 degrees North and 120.32 degrees East. The system uses a spatial matching algorithm to confirm that the two are the same target, and then calculates that the vessel is located in the center area of ​​grid cell G-15-08 based on the grid coordinate system. In the event of a brief signal loss, a dynamic prediction algorithm is used to estimate the fishing vessel's position. This algorithm uses the trajectory of the previous five seconds to predict the most likely grid position for the next second.

[0075] A dual-axis tilt sensor installed in the middle of the fishing boat's deck collects hull attitude data at five-second intervals. When the fishing boat is affected by wind and waves, the sensor records the roll angle (left and right tilt) and pitch angle (forward and backward pitch). For example, in one instance, the collected roll angle was 3.2 degrees to the left and the pitch angle was 1.8 degrees forward. This raw data is transmitted to a shore-based receiving station via a wireless transmission module. After receiving the grid position information and attitude data, the data fusion unit integrates them into structured data packets. Each data packet contains a timestamp accurate to milliseconds, the target fishing boat identification code, the row and column number of the grid, the roll angle value, and the pitch angle value. These data packets are transmitted to a central processing server via a long-distance wireless communication network, forming a time-sorted dynamic database of fishing boats.

[0076] The berthing risk assessment module extracts and analyzes the position sequences of adjacent fishing vessels from the database. The dynamic feature extraction unit selects two fishing vessels berthed in adjacent berths, for example, fishing vessel A is located in grid G-15-08 and fishing vessel B is located in grid G-15-09. The system retrieves the position records of both vessels over a continuous 30-second period and calculates the straight-line distance between the centers of the two grids at each time point. When it is found that fishing vessel A moves two grid cells from the center point of G-15-08 towards fishing vessel B within ten seconds, the system calculates the rate of change of distance between the two vessels to be 0.4 meters per second. The safety membership calculation unit inputs this rate value into a preset risk classification rule, which sets that when the speed at which the two vessels approach each other exceeds 0.05 meters per second, the first type of collision risk judgment process is initiated. If the actual distance between the two vessels is less than five meters, the system automatically upgrades the risk judgment level. The constraint matching unit compares the above judgment results with the allowable yaw threshold issued by the safety constraint initialization module. When the actual yaw distance exceeds the threshold, a high-risk label is generated. The risk quantification unit continuously monitors the duration of the risk state. For example, when the two ships remain close together for more than 120 seconds, the collision probability value increases to more than 0.9, and the state is recorded as having lasted for 150 seconds in the risk level identification data.

[0077] Example 2: See Figure 3The berth status analysis module's berth adaptation algorithm unit imports digital map data of the fishing port. This map includes the precise outline of the wharf's concrete structure and underwater topographic contour lines. The algorithm identifies straight shoreline segments suitable for berthing and divides berths using a typical fishing boat length of 30 meters as the basic unit. For example, on the 300-meter-long eastern wharf, the system evenly divides it into ten standard berth slots, numbered P01 to P10. Each berth slot is equipped with a rectangular electronic fence, whose boundary line extends from three meters from the wharf shoreline to a water depth of five meters. After receiving the real-time grid coordinates reported by the fishing boat dynamic sensing module, the occupancy status identification unit determines the spatial relationship between these coordinates and the electronic fence boundary of the berth slot. When the center point coordinates of the fishing boat are sampled six times consecutively within the fence area of ​​a certain berth slot, the berth is determined to be occupied, and the fishing boat's identification code is recorded. When a fishing vessel's center point is detected in berth P03 at coordinates X=125.6 meters and Y=37.8 meters, while another vessel in berth P04 has a center point of X=156.2 meters and Y=38.1 meters, the system highlights these two adjacent berths in red on the berth occupancy distribution map. The safety distance analysis unit retrieves the hull dimension data of these two fishing vessels. The hull profile is calculated by extending a rectangular area outward from the center point, with half the length and half the width of the vessel. Calculations show that the outermost point on the starboard side of the fishing vessel in P03 is X=138.3 meters, and the outermost point on the port side of the fishing vessel in P04 is X=153.7 meters. Based on the Euclidean distance formula, the actual minimum distance is calculated as follows:

[0078]

[0079] in: This indicates the minimum spatial distance between two ships (unit: meters). This represents the longitudinal coordinate value of the first measurement point (i.e., the starboard side point of fishing vessel P03). This represents the longitudinal coordinate value of the second measurement point (i.e., the port side point of fishing vessel P04). In specific calculations, and These represent the longitudinal coordinates of the corresponding measurement points on the two ships. Since the measurement points are on a horizontal straight line, the difference in their longitudinal coordinates is less than 0.2 meters and can be ignored. Therefore, the calculated result is 15.4 meters. Because the system's preset safety standard is 5 meters, the current distance meets the safety requirements. When the distance approaches the critical value, the system will activate a three-measurement verification mechanism.

[0080] The environmental parameter acquisition module continuously collects hydrological and meteorological data. A flow velocity sensor, installed at a depth of three meters on the wharf pillar, provides flow velocity vectors twice per second. A wind vane and anemometer are installed atop a twenty-meter-high observation tower, and wave buoys transmit wave height periodic data every ten seconds. The fluid dynamics simulation unit of the safety constraint initialization module calls upon a three-dimensional harbor basin numerical model, which is pre-loaded with harbor basin bottom elevation point cloud data and shoreline structural material property coefficients. For example, when acquiring real-time data on a northeasterly wind speed of force 7, a wave height of 0.9 meters, and a flow velocity of 1.2 meters per second, the simulation program establishes a computational grid covering a 500m × 300m water area and performs fluid kinematics solutions at specific time steps. Simulation results show that a maximum lateral impact force of 150 Newtons is generated in the P03 to P05 berth area. This impact force parameter reflects the magnitude of the lateral thrust generated by the wind and waves on the fishing boat. The constraint optimization unit compares this measured physical quantity with a preset threshold: when the impact force exceeds 100 Newtons, the allowable collision risk value decreases linearly from the baseline value of 0.8. The specific optimization rule is that for every 10 Newtons increase in impact force, the maximum permissible collision risk value decreases by 0.02. Currently, the upper limit of the risk value corresponding to an impact force of 150 Newtons is adjusted to 0.7, meaning the system's tolerance for collision risk is reduced. The risk diffusion model unit establishes a network model based on the berth topology, with each berth considered a network node, and adjacent berths connected by edges. Based on the risk level distribution obtained from fluid simulation, when berth P03 is marked as a high-risk node, the system activates a propagation delay algorithm to calculate the time required for the risk to spread along the berth network. This algorithm considers two variables: environmental intensity and vessel density. For example, under Force 7 wind and wave conditions, the predicted time parameter for the high-risk state to propagate from P03 to adjacent P02 and P04 is 168 seconds.

[0081] The dynamic changes in berth occupancy status trigger a real-time safety constraint update mechanism. When the safety distance analysis unit detects that the minimum distance between adjacent berths is less than six meters, it automatically generates a correction command for the allowable yaw threshold. For example, during high tide, a fishing boat at berth P07 drifts towards berth P08 due to a loose anchor chain, reducing the minimum distance between the two boats to 3.5 meters. At this time, the system reads the real-time attitude data of these two fishing boats to confirm that they are drifting in a non-operational state, and immediately compresses the allowable yaw threshold from the default five meters to 80% of the new value, i.e., 2.8 meters. Simultaneously, the fluid dynamics simulation unit starts a rapid simulation mode for this local area, performing micro-fluid modeling only on the water area with a 50-meter radius centered on P07. Simulation calculations show that the lateral impact force is significantly enhanced when the drift direction is consistent with the tidal current direction. Based on this, the constraint optimization unit further reduces the maximum allowable collision risk value from 0.7 to 0.65. The risk diffusion model unit then adjusts the prediction parameters, setting that if the risk is not eliminated within 200 seconds, it will diffuse to the downstream berth. The berth status analysis module generates an updated berth occupancy distribution map every 30 seconds. This data, combined with the fluid impact force distribution map, forms the dynamic input source for the safety constraint initialization module. When hydrological parameters undergo a sudden change, such as a two-level increase in wind speed, the system automatically shortens the simulation time step to improve response speed and recalculates the risk propagation time limit parameters for all berths.

[0082] Example 3: See Figure 4 After receiving the real-time grid coordinates and risk level identification data of the fishing vessel, the alarm signal generation module initiates the alarm dimension conversion process. The spatial coding unit processes the grid location information of the target fishing vessel. For example, when the fishing vessel is located in grid G-15-08, the vertical dimension coding unit extracts the vertical coordinate value 15 and generates the sound wave frequency parameters through a linear transformation rule. This rule sets the base frequency to 2000 Hz, and increases the frequency by 20 Hz for each additional grid vertical coordinate unit. The specific calculation expression is as follows:

[0083]

[0084] in: Indicates carrier frequency (unit: Hertz). This indicates the vertical index of the grid where the fishing vessel is located. The horizontal dimension encoding unit extracts the horizontal coordinate value 08, setting a phase angle offset of 36 degrees for each grid unit; this fishing vessel generates a phase offset of 288 degrees. The risk intensity encoding unit reads the collision probability value p=0.95 and sets the sound wave amplitude to 95% of the maximum amplitude. The amplitude intensity reflects the urgency of the alarm; a larger value indicates stronger sound wave energy. The risk mapping unit determines the base alarm frequency based on the collision probability value. The system presets three frequency thresholds: 0.5 Hz low frequency for probability values ​​below 0.3, 2 Hz medium frequency for the 0.3 to 0.7 range, and 5 Hz high frequency for values ​​above 0.7. The current probability value of 0.95 triggers the 5 Hz base frequency. The timeliness control unit retrieves the risk duration data; this fishing vessel has been in a high-risk state for 150 seconds. The alarm duration calculation formula sets a base duration of 10 seconds, adding 10 seconds for every 60 seconds of overtime, thus generating an alarm duration of 30 seconds.

[0085] The acoustic timing control parameters ensure that the alarm signal is broadcast cyclically within a specified time period. The signal synthesis unit uses frequency modulation technology to integrate all parameters. The carrier frequency is 2240 Hz, generated by longitudinal encoding, and the phase angle is set to a 288-degree offset according to lateral encoding. The amplitude intensity is maintained at 95% of the risk intensity encoding output level. The reference alarm frequency of 5 Hz is used as the modulation signal, controlling the carrier amplitude to change regularly at 5 times per second. The final synthesized multi-level acoustic alarm signal is played through a distributed sonar array, with each berth area corresponding to a specific combination of frequency and phase acoustic characteristics. For example, the alarm signal of the fishing vessel located at G-15-08 is characterized by a 2240 Hz carrier frequency, a 288-degree phase angle delay at the start point of each cycle of the acoustic waveform, an amplitude intensity of 95% of the maximum value, and five amplitude fluctuations per second. Crew members can identify the location characteristics of the sound source by hearing; a high-frequency buzzing sound represents an emergency, and changes in the rhythm of the sound wave reflect the duration of the risk.

[0086] The directional sonar equipment deployed in the system is installed at the top of the dock light poles, with each pole covering three grid units. When an alarm signal is played, the equipment adjusts the sound wave emission angle according to the phase shift. For example, a 288-degree phase shift corresponds to the sound wavefront rotating to the northeast, making the sound pressure level in grid area G-15-08 15 dB higher than other areas. The alarm duration control system has a built-in interruption detection mechanism; if the risk level indicator is updated to a safe state within the 30-second alarm cycle, the signal output is immediately terminated. All acoustic parameters are refreshed every two seconds to ensure that the alarm signal remains synchronized with the real-time position and risk changes of the fishing vessels.

[0087] Example 4: The scheduling parameter optimization module retrieves a historical collision avoidance record dataset stored in the scheduling rule base. This dataset contains seven core fields: timestamp, initial risk value, berth area code, fishing vessel tonnage class, action type, total time, and risk reduction. The parameter evaluation unit uses the risk propagation time limit as a filter to retrieve effective solutions. For example, when the current propagation time limit is 200 seconds, the system filters out all historical records where the total time from the issuance of the instruction to the risk relief does not exceed 200 seconds. Data from record number HJ-2023-087 shows that in a high-risk scenario where two fishing vessels are 3.2 meters apart in area P09, the system sent a left-shift instruction requiring one of the fishing vessels to move 2.5 meters westward, with an execution time of 115 seconds, and the risk value decreased from 0.94 to 0.21. Records meeting the criteria enter the optimization process. The system establishes an optimization objective function while constraining the shortest path and the minimum collision probability, where the path length weight coefficient is set to 0.7 and the collision probability weight is 0.3. The multidimensional optimization unit uses a genetic algorithm for iterative calculation. The initial population size is set to one hundred solution vectors, each containing three variables: lateral offset, longitudinal offset, and execution speed. After fifty generations of iteration, the optimal solution is output: a lateral movement of -2.2 meters represents a westward displacement, a longitudinal displacement of 0.3 meters represents a fine-tuning of the docking direction, and an execution speed of 0.8 meters per second. When the speed adjustment command is converted into engine control parameters, the movement speed is converted into a propeller speed value according to the fishing vessel power performance comparison table: a 300-ton fishing vessel needs to maintain a propeller speed of 800 revolutions per minute to execute a displacement of 0.8 meters per second. The priority indicator calculation relies on a two-factor weighted formula, with the fishing vessel tonnage factor accounting for 0.6% and the current risk level factor accounting for 0.4%. When the risk value of a 500-ton fishing vessel at berth P09 is 0.9, the priority value reaches 0.82, while the priority of a 100-ton fishing vessel in the same area with a risk value of 0.9 is 0.66. The priority indicator controls the sorting logic of the command queue; all tasks with a priority exceeding 0.7 are marked for immediate execution (see Table 1).

[0088] Table 1: Example of historical collision avoidance scheduling scheme data structure.

[0089] Record ID Timestamp berth area Tonnage (tons) Execute action Time elapsed (seconds) Risk changes HJ-2023-056 202305061132 P07 280 Move 1.8 meters to the left 98 0.92→0.30 HJ-2023-087 202308211540 P09 350 Move 2.5 meters west 115 0.94→0.21 HJ-2023-102 202310170918 P11 180 Move 3.0 meters to the right 152 0.89→0.28 HJ-2023-155 202312050743 P05 420 Moved 1.5 meters eastward 83 0.96→0.25

[0090] After receiving the safety scheduling control parameters, the instruction parsing unit in the execution feedback module generates a berth guidance sequence in descending order of priority. When three sets of tasks exist simultaneously—P09 (priority 0.82), P05 (priority 0.78), and P11 (priority 0.73)—the instruction queue is arranged in the order of P09→P05→P11. The signal drive unit transmits instruction data packets to the berth guidance device via a 900 MHz wireless channel. The data packet structure includes the target grid number, acoustic coding parameters, and light color coding. After receiving the instruction, the acoustic and light guidance pile located in the P09 area activates the acoustic module to play a 2240 Hz sound wave signal with a phase angle of 252 degrees, synchronously driving the red LED to flash at a 5 Hz frequency. The status feedback unit monitors the device's operating status. The guidance pile current sensor continuously detects the motor's operating current. When the current value driving the servo motor exceeds 1500 mA for ten consecutive seconds, a mechanical blockage fault is determined in the device. At this point, the system generates status code F-03 and returns it to the scheduling parameter optimization module via the same wireless channel. The third digit of the status code, 3, represents the resistance over-limit fault type.

[0091] The actual execution data of the berth guidance device is recorded in the system operation log, including three key time points: the time of command issuance, the device response delay, and the displacement completion time. The displacement completion time is verified by the device's built-in ultrasonic ranging module: the task is considered complete when the monitored distance of the fishing vessel reaches 95% of the command value. For example, the HJ-2023-087 task record shows that the dispatch command was issued at 14:41:05, the target fishing vessel detected the audible and visual signal and started its engine at 14:41:08, and the ultrasonic ranging confirmed a movement of 2.37 meters at 14:42:55. The system automatically calculates key performance indicators. The net time for this task is 110 seconds from the issuance of the command to the achievement of the displacement target, with a 3-second device response delay included as an additional cost. All performance indicators are updated to the execution effect field of the historical collision avoidance record to provide data support for subsequent optimization iterations. When a fault status code is returned, the system activates a backup guidance strategy: suspending the audible and visual alarms in the area under the jurisdiction of the currently faulty device and automatically enabling the auxiliary prompting function of the adjacent berth guide piles. For example, when the P09 guide post malfunctions, the P08 post will play the same acoustically encoded prompt tone but with the amplitude reduced to 50% of the normal value, and the original red light will be changed to a slow yellow flashing mode.

[0092] Upon receiving execution status data, the scheduling parameter optimization module initiates a strategy self-checking mechanism. When it receives abnormal data indicating that the displacement completion time significantly exceeds historical records, it retrieves environmental monitoring records for the corresponding time period for correlation analysis. For example, if a task in area P11 takes 182 seconds, far exceeding the historical average, the system detects an abnormal northeasterly wind force of level 8 during that period and adds a wind speed compensation factor to the optimization model accordingly. When a fault status code is received, the module generates an equipment maintenance work order and automatically reduces the scheduling priority weight for that area: the priority calculation coefficient for that berth is reduced by 20% within 24 hours of the fault. All parameter adjustment records are written to the system change log, including four elements: adjustment time, adjusted parameter item, adjustment range, and associated event number. The scheduling rule base is updated using a structured query language. After a new successful solution is executed, it is immediately inserted into the database table. Every 50 new records inserted automatically trigger a data quality verification process to remove invalid records with missing execution effect fields or logical conflicts.

[0093] Example 5: The adaptive correction unit continuously monitors the status data stream returned by the execution feedback module. When it receives status code F-03, the system automatically activates the fault scenario recording function. This function captures a snapshot of environmental parameters at the moment of the fault, including three hydrological and meteorological data: wind speed, current speed, and wave height. It also records the target fishing vessel's real-time grid coordinates, onboard tilt sensor readings, and the original text of the dispatch instructions. These data, along with the fault code, constitute a new case record, which is appended to the historical collision avoidance scheme dataset in the dispatch rule base in a structured format. The case record contains sixteen fields. In addition to basic environmental parameters and fishing vessel status, it is specially marked with a fault type classification code and a timestamp. For example, in a fault record of a guide post in area P09, a complete set of parameters is stored, including a northeast wind of force 7, a current speed of 1.3 meters per second, a roll angle of 4.2 degrees, and the dispatch instruction "shift west by 2.5 meters". The newly added record is marked as a learning case identifier and stored separately from regular historical records.

[0094] The timeliness recalculation unit initiates a model update process every hour. This process retrieves all new records with learning case identifiers, extracts environmental parameters and fault features as training samples, and uses a state-action-reward framework for the reinforcement learning model. Berth area, environmental intensity, and vessel density are defined as state space elements, and scheduling instruction type is defined as action space elements. The model calculates reward values ​​by comparing the expected execution time with the actual execution time: a positive score is awarded when the actual execution time is more than 10% shorter than expected, and a negative score is penalized when it exceeds 15%. The Markov state transition matrix update process is as follows: the original matrix records the probability distribution of risk diffusion under different environmental conditions, for example, "the probability of risk P03 spreading to P04 in a level 7 wind is 0.85". The system adjusts this probability value based on new fault cases; if a new case shows that the actual diffusion speed is faster under level 7 wind conditions, the probability is increased to 0.92. The recalculation of risk propagation time limit is based on the updated state transition matrix, using a dynamic programming algorithm to solve for the worst-case propagation time limit. The calculation formula is:

[0095]

[0096] in: This indicates the recalculated risk propagation timeframe (in seconds). Represents the set of all berth states. This indicates the state of the k-th berth node. This represents the probability of transitioning from the current state to a higher-risk state. Indicates the time interval for single-step state transitions (default setting is 10 seconds). This indicates the maximum number of transition steps required for calculation. For example, the original model predicted that the risk transfer from P03 to P05 would take 200 seconds, while the new model outputs a result of 180 seconds.

[0097] The alarm signal generation module subscribes to risk propagation time limit updates in real time. When the time limit recalculation unit publishes new time limit parameters, the alarm duration calculation formula immediately switches to the new version. The original formula was "base 10 seconds plus an additional 10 seconds for every 60 seconds of duration," while the new formula is "base 10 seconds plus an additional 10 seconds for every 45 seconds of duration." Taking a fishing boat that has been in a high-risk state for 135 seconds as an example: the original formula calculated the alarm duration as a base value of 10 seconds plus two additional 10-second increments, totaling 30 seconds; the new formula calculates it as a base value of 10 seconds plus three additional 10-second increments, totaling 40 seconds. This switching process requires no manual intervention; the system automatically completes the parameter hot update through a publish-subscribe mechanism. When adjusting the duration, the alarm signal output system keeps the acoustic coding parameters unchanged, only extending the total signal playback time. If the risk state is resolved within the alarm period, the system immediately interrupts the signal output and generates a status resolution log.

[0098] The fault handling closed loop includes a reverse verification mechanism. When the guide pile is repaired and put back online, the system automatically issues a test command to verify the equipment status. The test command requires the guide pile to output three standard frequency acoustic signals sequentially, while simultaneously monitoring the motor's no-load current curve. If the test data meets the preset standard value range, the system marks the equipment status as recovered and removes the priority downgrading treatment for that area. All model update records are written to the version tracking database, including update trigger time, changed matrix parameters, and a list of affected berth areas. The system performs full historical data retraining monthly, merging new cases with the original dataset to regenerate the state transition matrix, ensuring effective integration of long-term learning effects. During retraining, the system continues to use the currently effective model for real-time risk calculation, and performs a smooth switch only after the new model has been validated.

[0099] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0100] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A safety monitoring system for fishing vessel berthing in a fishing port, characterized in that, include: The fishing vessel dynamic perception module is used to collect dynamic position and attitude data of each fishing vessel in the target fishing port in real time. The berth status analysis module is used to generate berth occupancy status distribution information based on the dynamic position data and attitude data. The environmental parameter acquisition module is used to acquire the hydrological and meteorological parameters of the target fishing port in real time. The safety constraint initialization module is used to output initial safety monitoring constraints based on the berth occupancy status distribution information and the hydro-meteorological parameters, wherein the initial safety monitoring constraints include an allowable yaw threshold and a maximum allowable collision risk value. The mooring risk assessment module is used to perform quantitative analysis of fishing vessel mooring risks by using the initial safety monitoring constraints as assessment conditions, and to generate risk level identification data. The scheduling parameter optimization module is used to perform berth resource scheduling optimization based on the risk level identification data to obtain safe scheduling control parameters. The alarm signal generation module is used to perform alarm dimension conversion based on the safety scheduling control parameters and the real-time updated dynamic location data, and output multi-level acoustic alarm signals.

2. The fishing port and fishing vessel berthing safety monitoring system according to claim 1, characterized in that, The fishing vessel dynamic sensing module includes: A grid partitioning unit is used to divide the port area of ​​the target fishing port into a dynamic grid cell array; The trajectory tracking unit is used to locate the real-time grid coordinates of each fishing vessel in the dynamic grid cell array based on radar scan data and AIS positioning data. The attitude sensing unit is used to acquire roll and pitch angle data of each fishing vessel through the shipboard tilt sensor. The data fusion unit is used to fuse the real-time grid coordinates, roll angle data and pitch angle data into the dynamic position data and attitude data.

3. The fishing port and fishing vessel berthing safety monitoring system according to claim 2, characterized in that, The mooring risk assessment module performs quantitative analysis of fishing vessel mooring risks, including: A dynamic feature extraction unit is used to calculate the real-time distance change rate between adjacent fishing vessels based on the dynamic position data. The safety membership calculation unit is used to perform fuzzy classification on the real-time distance change rate using a predefined membership function to obtain the membership degree of the first type of collision risk and the membership degree of the second type of collision risk. The constraint matching unit is used to match and calculate the membership degree of the first type of collision risk, the membership degree of the second type of collision risk, and the allowable yaw threshold. The risk quantification unit is used to generate risk level identification data, which includes collision probability value and risk duration, based on the matching results.

4. The fishing port and fishing vessel berthing safety monitoring system according to claim 3, characterized in that, The berth status analysis module includes: The berth adaptation algorithm unit is used to divide the standard berth area according to the wharf structure characteristics of the target fishing port; The occupancy status identification unit is used to map the real-time grid coordinates to the standard berth area and output a berth occupancy rate distribution map; The safety distance analysis unit is used to calculate the minimum distance between adjacent berths based on the berth occupancy rate distribution map. The safety constraint initialization module updates the allowable yaw threshold based on the minimum spacing value.

5. The fishing port and fishing vessel berthing safety monitoring system according to claim 4, characterized in that, The security constraint initialization module also includes: The fluid dynamics simulation unit is used to perform fluid simulation of the berth area based on the flow velocity data, wind direction data, and wave height data in the hydro-meteorological parameters. The constraint tuning unit is used to correct the maximum permissible collision risk value based on the simulation results. The risk diffusion model unit is used to generate risk propagation time limit data based on the corrected maximum permissible collision risk value.

6. The fishing port and fishing vessel berthing safety monitoring system according to claim 5, characterized in that, The alarm signal generation module performs alarm dimension conversion, including: A spatial coding unit is used to convert the real-time grid coordinates into acoustic alarm spatial codes; The risk mapping unit is used to map the collision probability value in the risk level identification data to a baseline alarm frequency. A timeliness control unit is used to set the alarm duration based on the duration of the risk. The signal synthesis unit is used to fuse the reference alarm frequency, alarm duration, and acoustic alarm spatial encoding to generate the multi-level acoustic alarm signal.

7. The fishing port and fishing vessel berthing safety monitoring system according to claim 6, characterized in that, The acoustic alarm spatial code includes: The vertical dimension encoding unit is used to convert the grid vertical coordinates into a gradient of sound wave frequency changes; Lateral dimension encoding unit, used to convert grid horizontal coordinates into acoustic wave phase offset; A risk intensity encoding unit is used to convert the collision probability value into a sound wave amplitude intensity value.

8. The fishing port and fishing vessel berthing safety monitoring system according to claim 7, characterized in that, The scheduling parameter optimization module performs berth resource scheduling optimization, including: The scheduling rule base unit is used to store historical collision avoidance scheduling schemes and corresponding execution effect data; The parameter evaluation unit is used to filter historical collision avoidance scheduling schemes that meet the time constraints using the risk propagation time limit data; A multi-dimensional optimization unit is used to fine-tune the parameters of the selected historical collision avoidance scheduling schemes based on the risk level identification data. The control parameter generation unit is used to output the safety scheduling control parameters, which include collision avoidance path, speed adjustment command and priority identifier.

9. The fishing port and fishing vessel berthing safety monitoring system according to claim 8, characterized in that, Also includes: The execution feedback module is used to drive the berth guidance device of the target fishing port according to the multi-level acoustic alarm signal; The execution feedback module includes: The instruction parsing unit is used to generate a berth guidance sequence based on the priority identifier in the safety scheduling control parameters; The signal driving unit is used to send the multi-level acoustic alarm signal to the berth guidance device in the designated area according to the berth guidance sequence; The status feedback unit is used to collect the execution status data of the berth guidance device in real time and feed it back to the scheduling parameter optimization module.

10. The fishing port and fishing vessel berthing safety monitoring system according to claim 9, characterized in that, The scheduling parameter optimization module also includes: An adaptive correction unit is used to update the historical collision avoidance scheduling scheme based on the execution status data; The timeliness recalculation unit is used to recalculate the risk propagation timeliness data based on the updated historical collision avoidance scheduling scheme; The alarm signal generation module adjusts the alarm duration based on the recalculated risk propagation time limit data.

Images