A method and system for self-assessment of a ship's navigation capability after damage

Abstract

This invention relates to the field of navigation technology, and more particularly to a method and system for self-assessing the navigation capability of a damaged ship. The method includes: during normal navigation, collecting data from all core navigational support sensors of the ship and establishing a sensor data correlation benchmark model; after a collision, collecting sensor data from all navigational support equipment in real time, and performing single-sensor correlation anomaly scoring, stability anomaly scoring, redundancy anomaly scoring, and physical rationality anomaly scoring; summing the correlation anomaly scores, stability anomaly scores, redundancy anomaly scores, and physical rationality anomaly scores to obtain a total single-sensor anomaly score; and assessing the ship's navigation capability after damage based on the total anomaly score of all sensors. This invention can automatically diagnose and assess the operational status of navigational support equipment based on the output parameters of navigational support sensors, providing a basis and guidance for ship mission execution and crew maintenance.

Description

Technical Field

[0001] This invention relates to the field of radio navigation technology, and in particular to a method and system for self-assessing the navigation capability of a ship after it has been damaged. Background Technology

[0002] Due to the complex maritime environment, ship collisions and other accidents occur frequently, and collisions are also a means of countermeasures. Collisions can cause multi-dimensional damage to ship navigational safety sensors, thereby affecting the ship's navigational safety capabilities. Navigational safety sensors are core equipment for ship navigation, environmental monitoring, and navigational status perception. Damage to these sensors can lead to the ship becoming "blind," "deaf," and "disoriented," making it unable to locate itself, perceive its environment, or assess its own status, thus resulting in a fundamental loss of the ship's capabilities.

[0003] Currently, the assessment of maritime safety capabilities after a ship collision relies heavily on manual inspection, which suffers from low efficiency, poor accuracy, and an inability to quickly provide a basis for emergency decision-making. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method and system for self-assessing the navigation capability of a ship after damage. After a ship collision, the system can automatically diagnose and assess the operating status of navigation support equipment based on the output parameters of navigation support sensors, providing a basis and guidance for ship missions and crew maintenance, and supporting emergency decision-making.

[0005] This invention is achieved through the following technical solution: A method for self-assessing the navigation capability of a ship after damage, comprising the following steps: S1: During normal navigation, collect data from all core navigational support sensors of the ship, preprocess the data, establish a sensor data correlation benchmark model including a physical constraint model and a redundancy verification model, and calculate the Pearson correlation coefficient and deviation threshold between the data of all core navigational support sensors of the ship. S2: After a ship collision, in high-speed acquisition mode, sensor data of all navigation support equipment after the ship is damaged are collected in real time, and the sensor data of all navigation support equipment after the ship is damaged are preprocessed. S3: Calculate the Pearson correlation coefficient matrix of the sensor data of the navigation support equipment after the ship is damaged based on the preprocessed sensor data of the navigation support equipment after the ship is damaged, and calculate the change in the correlation coefficient of the same type of physical quantity. Based on the change in the correlation coefficient and the Pearson correlation coefficient matrix of the sensor data of the navigation support equipment after the ship is damaged, perform a single sensor correlation anomaly score. S4: Compare the mean, standard deviation and deviation threshold of the real-time deviation sequence of the sensor data of the navigation support equipment after the ship is damaged, and score the stability anomaly of a single sensor based on the comparison results. S5: Conduct redundancy verification and testing on the navigation support equipment after the ship is damaged, and score the redundancy anomaly of a single sensor based on the redundancy verification and testing results; S6: Detect the physical rationality of sensor data after ship damage based on a physical constraint model, and score the physical rationality anomaly of a single sensor based on the detection results; S7: Summing the correlation anomaly score, stability anomaly score, redundancy anomaly score, and physical rationality anomaly score of a single sensor to obtain the total anomaly score of a single sensor. The sensor fault level is determined based on the total anomaly score of the single sensor, and the navigation capability of the ship after damage is assessed based on the fault levels of all sensors.

[0006] The optimized data collected in step S1 from all core maritime safety sensors includes: position, speed, and heading data from the GNSS receiver; position, speed, heading, angular velocity, and acceleration data from the inertial navigation system; speed data from the log; heading data from the compass; wind speed and direction data from the wind sensor; dynamic data of surrounding vessels from the automatic identification system; and target detection data from the radar.

[0007] Furthermore, the preprocessing of sensor data in steps S1 and S2 is as follows: a precise time protocol is used to synchronize the time of all core maritime safety sensors of the ship; outliers in the data are removed using the 3σ criterion; missing data is filled in using linear interpolation; and the timestamps and coordinate systems of all data are unified.

[0008] The optimized timestamp accuracy for collecting sensor data from maritime support equipment in steps S1 and S2 is ≤1s.

[0009] Furthermore, the construction process of the physical constraint model in step S1 is as follows: S111: Based on the ship's speed, heading, and time interval, derive the ship's theoretical displacement according to equation (1) and constrain the position change: (1); in: Indicates that the ship is The theoretical displacement along the heading direction within a given time period. Indicates the ship's speed relative to the water. Indicates the data sampling time interval. Indicates the ship's true heading angle. Indicates that the ship is The theoretical displacement perpendicular to the heading direction within a given time period; S112: Velocity constraints are applied based on the vector composition relationship (2) between the ship's land speed, water speed, and ocean current speed. (2); in: Represents the ship's velocity vector relative to the ground. This represents the ship's velocity vector relative to the water. Represents the ocean current velocity vector; S113: Combine the maximum thrust of the ship's main engine and the structural characteristics of the hull to set the longitudinal acceleration threshold and the lateral acceleration threshold of the ship. Under normal navigation conditions, acceleration is constrained by equation (3): (3); in: Indicates the longitudinal acceleration of the ship. Indicates the longitudinal acceleration threshold of the ship. Indicates the lateral acceleration of a ship. Indicates the threshold value for lateral acceleration of a ship; S114: Set the ship's heading angular velocity threshold based on the ship's steering performance. Under normal steering conditions, the angular velocity is constrained by equation (4): (4); in: Indicates the angular velocity of the ship's heading. This indicates the threshold value for the ship's heading angular velocity.

[0010] Furthermore, in step S1, the Pearson correlation coefficient matrix of the sensor data of the navigation support equipment is calculated according to equation (5): (5); in: Indicates the first The and the first Pearson correlation coefficient matrix of sensor data, Indicates the number of samples. Indicates the first The sensor at the first The measured value at time, Indicates the first The sensor at the first The measured value at time.

[0011] Furthermore, in step S3, the change in the correlation coefficient of the same type of physical quantities is calculated according to equation (6): (6); in: This indicates the change in the correlation coefficient. Indicates the first [day] after the ship suffered damage. The and the first The Pearson correlation coefficient matrix of sensor data.

[0012] In the optimized step S5, when performing redundancy verification testing on the navigation support equipment after ship damage, a 95% confidence interval is used to determine the redundancy verification threshold. ,in: Indicates the first The and the first The mean of the data deviation sequence of each sensor, Indicates the first The and the first The standard deviation of sensor data.

[0013] In the optimized step S6, the physical rationality of the sensor data after the ship is damaged is detected based on the physical constraint model, including position change detection, velocity logic detection, and acceleration or angular velocity detection. The physical rationality anomaly score is the sum of the anomaly scores of the three detections: position change detection, velocity logic detection, and acceleration or angular velocity detection.

[0014] A self-assessment system for navigation capability after ship damage, used to execute a self-assessment method for navigation capability after ship damage as described in any of the above, comprising a sensor data correlation benchmark model construction module, a sensor data acquisition and preprocessing module for all navigation support equipment after ship damage, a correlation anomaly scoring module, a stability anomaly scoring module, a redundancy anomaly scoring module, a physical rationality anomaly scoring module, and a navigation capability assessment module. The sensor data correlation benchmark model construction module is used to collect data from all core navigational support sensors of the ship during normal navigation, preprocess the data, establish a sensor data correlation benchmark model including a physical constraint model and a redundancy verification model, and calculate the Pearson correlation coefficient between the data from all core navigational support sensors of the ship. The sensor data acquisition and preprocessing module for all navigation support equipment after ship damage is used to acquire sensor data of all navigation support equipment after ship collision in real time in high-speed acquisition mode, and to preprocess the sensor data of all navigation support equipment after ship damage. The correlation anomaly scoring module calculates the Pearson correlation coefficient matrix of the sensor data of the navigation support equipment after the ship is damaged based on the preprocessed sensor data of the navigation support equipment after the ship is damaged, and calculates the change in the correlation coefficient of the same type of physical quantity. Based on the change in the correlation coefficient and the Pearson correlation coefficient matrix of the sensor data of the navigation support equipment after the ship is damaged, a single sensor correlation anomaly score is performed. The stability anomaly scoring module is used to compare the mean, standard deviation and deviation threshold of the real-time deviation sequence of sensor data of the navigation support equipment after the ship is damaged, and to score the stability anomaly of a single sensor based on the comparison results. The redundancy anomaly scoring module is used to perform redundancy verification and detection on the navigation support equipment after the ship is damaged, and to score the redundancy anomaly of a single sensor based on the redundancy verification and detection results. The physical rationality anomaly scoring module detects the physical rationality of sensor data after ship damage based on a physical constraint model, and scores the physical rationality anomalies of a single sensor based on the detection results. The navigation capability assessment module is used to sum the correlation anomaly score, stability anomaly score, redundancy anomaly score, and physical rationality anomaly score of a single sensor to obtain a total anomaly score for a single sensor. The sensor fault level is determined based on the total anomaly score of the single sensor, and the navigation capability of the ship after damage is assessed based on the fault levels of all sensors.

[0015] Beneficial effects of the invention: The present invention provides a method and system for self-assessing the navigation capability of a ship after damage, which has the following advantages: 1. A preliminary assessment can be given in about 10 seconds after the collision, and a full quantitative assessment can be completed within 30 seconds, with a rapid response.

[0016] 2. By integrating multiple technical means such as statistical correlation, stability, physical rationality, and redundancy verification, it can effectively detect hidden faults that cannot be detected manually, and significantly reduce the false judgment rate.

[0017] 3. No crew members need to enter dangerous equipment compartments; the assessment is completed entirely through data analysis, thus avoiding secondary disasters that could harm personnel after a collision.

[0018] 4. No hardware modifications are required to the existing ship navigation system; the function can be achieved simply by adding a software module. This significantly reduces ship downtime, provides clear guidance for crew maintenance, and has good economic benefits and promotional value. It is applicable to all types of civilian and military vessels. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the overall process of the present invention. Detailed Implementation

[0020] A self-assessment method for a ship's navigation capability after damage is illustrated in Figure 1, and includes the following steps: S1: During normal navigation, collect data from all core navigational support sensors of the ship, preprocess the data, establish a sensor data correlation benchmark model including a physical constraint model and a redundancy verification model, and calculate the Pearson correlation coefficient and deviation threshold between the data of all core navigational support sensors of the ship. Specifically, the collected data from all core maritime safety sensors include, but are not limited to: position, speed, and heading data from GNSS receivers; position, speed, heading, angular velocity, and acceleration data from inertial navigation systems; speed data from logs; heading data from compasses; wind speed and direction data from wind sensors; dynamic data of surrounding vessels from the Automatic Identification System (AIS); and target detection data from radar.

[0021] The following methods can be used to preprocess sensor data: use a precise time protocol to synchronize the time of all core navigational support sensors on the ship; use the 3σ criterion to remove outliers in the data (data exceeding the mean ± 3 times the standard deviation can be removed); use linear interpolation to fill in missing data; and unify the timestamps (accurate to the millisecond level) and coordinate systems of all data (the China Geodetic Coordinate System 2000 can be used to ensure data consistency).

[0022] When collecting sensor data from maritime support equipment, the timestamp accuracy should be ≤1s to avoid data analysis errors caused by time deviations.

[0023] Furthermore, the construction process of the physical constraint model in step S1 is as follows: S111: Based on the ship's speed, heading, and time interval, derive the ship's theoretical displacement according to equation (1) and constrain the position change: (1); in: Indicates that the ship is The theoretical displacement along the heading direction within a given time period. Indicates the ship's speed relative to the water. The value can be measured by a speedometer. Indicates the data sampling time interval. Indicates the ship's true heading angle. It can be measured by an inertial navigation system or a compass. Indicates that the ship is The theoretical displacement perpendicular to the heading direction within a given time period; S112: Velocity constraints are applied based on the vector composition relationship (2) between the ship's land speed, water speed, and ocean current speed. (2); in: The velocity vector of a ship relative to the ground can be obtained through GNSS measurements. The velocity vector of a ship relative to water can be measured using a speed log. This represents the ocean current velocity vector, which can be provided by hydrological data and used to verify the reasonableness of the velocity data.

[0024] S113: Combine the maximum thrust of the ship's main engine and the structural characteristics of the hull to set the longitudinal acceleration threshold and the lateral acceleration threshold of the ship. Under normal navigation conditions, acceleration is constrained by equation (3): (3); in: Indicates the longitudinal acceleration of the ship. Indicates the longitudinal acceleration threshold of the ship. Indicates the lateral acceleration of a ship. Indicates the threshold value for lateral acceleration of a ship; If the longitudinal acceleration and lateral acceleration of the ship exceed the corresponding thresholds, the data is considered abnormal.

[0025] S114: Set the ship's heading angular velocity threshold based on the ship's steering performance. Under normal steering conditions, the angular velocity is constrained by equation (4): (4); in: Indicates the angular velocity of the ship's heading. This indicates the threshold value for the ship's heading angular velocity.

[0026] If the ship's angular velocity exceeds the threshold, it is considered an abnormal data point.

[0027] Furthermore, in step S1, the Pearson correlation coefficient matrix of the sensor data of the navigation support equipment is calculated according to equation (5): (5); in: Indicates the first The and the first Pearson correlation coefficient matrix of sensor data, Indicates the number of samples. Indicates the first The sensor at the first The measured value at time, Indicates the first The sensor at the first The measured value at time.

[0028] Pearson correlation coefficient matrix Used to characterize the degree of inherent correlation between data from different sensors The value range is [-1, 1]. The closer the absolute value is to 1, the stronger the correlation between the two.

[0029] The established sensor data correlation benchmark model can serve as a reference for assessing the navigation capability of a ship after it has been damaged.

[0030] Redundancy check models are used to identify redundant navigation equipment on ships, which typically include multiple GNSS receivers, multiple inertial navigation systems, multiple compasses, and multiple logs (which are redundant with each other).

[0031] S2: After a ship collision, in high-speed acquisition mode, sensor data of all navigation support equipment after the ship is damaged are collected in real time, and the sensor data of all navigation support equipment after the ship is damaged are preprocessed. Under normal conditions, the sampling frequency is generally 1Hz, while in high-speed acquisition mode, the sampling frequency can reach 10Hz. The acquisition time range covers 1 minute before the collision until the end of the evaluation, ensuring that the changing trend of sensor data before and after the collision is captured.

[0032] The data types collected are completely consistent with those in step S1, including measurement data from all maritime safety sensors, ensuring comparability with the baseline model data related to the sensor data.

[0033] The method for preprocessing sensor data from all navigational support equipment after a ship is damaged is the same as the method for preprocessing all core navigational support sensor data collected during normal navigation.

[0034] S3: Calculate the Pearson correlation coefficient matrix of the sensor data of the navigation support equipment after the ship is damaged based on the preprocessed sensor data of the navigation support equipment after the ship is damaged, and calculate the change in the correlation coefficient of the same type of physical quantity. Based on the change in the correlation coefficient and the Pearson correlation coefficient matrix of the sensor data of the navigation support equipment after the ship is damaged, perform a single sensor correlation anomaly score. The method for calculating the Pearson correlation coefficient matrix of the sensor data of the navigation support equipment after the ship is damaged is the same as the method for calculating the Pearson correlation coefficient matrix of the sensor data of the navigation support equipment in step S1.

[0035] Specifically, the change in the correlation coefficient of the same type of physical quantities can be calculated according to equation (6): (6); in: This indicates the change in the correlation coefficient. Indicates the first [day] after the ship suffered damage. The and the first The Pearson correlation coefficient matrix of sensor data.

[0036] Specifically, for significantly correlated sensor pairs, i.e. ( ), calculate the correlation coefficient matrix of real-time sensor data after the collision, compare it with the baseline matrix, and calculate the change in correlation coefficient. Calculate the single-sensor correlation anomaly score according to Table 1. If there is no significantly correlated sensor, the single-sensor correlation anomaly score is 0.

[0037] Table 1

[0038] S4: Compare the mean, standard deviation and deviation threshold of the real-time deviation sequence of the sensor data of the navigation support equipment after the ship is damaged, and score the stability anomaly of a single sensor based on the comparison results. Specifically, for significantly correlated sensor pairs ( ),like If so, it is determined to be an abnormal mean deviation (sensor parameter drift); If so, it is determined to be a decrease in stability (loose sensor components or poor contact).

[0039] Scoring rules: 1 point is added for each anomaly, up to a maximum of 3 points. If there is no significantly associated sensor, the single sensor stability anomaly score is 0 points.

[0040] S5: Conduct redundancy verification and testing on the navigation support equipment after the ship is damaged, and score the redundancy anomaly of a single sensor based on the redundancy verification and testing results; Specifically, when conducting redundancy verification testing on maritime safety equipment after ship damage, a 95% confidence interval is used to determine the redundancy verification threshold. ,in: Indicates the first The and the first The mean of the data deviation sequence of each sensor, Indicates the first The and the first The standard deviation of sensor data.

[0041] During redundancy verification, if the measurement deviation exceeds the redundancy verification threshold range, the redundant equipment is deemed to be malfunctioning.

[0042] Scoring rules: 1 point is added for each anomaly, up to a maximum of 3 points; if there is no redundant equipment, this item will receive 0 points.

[0043] S6: Detect the physical rationality of sensor data after ship damage based on a physical constraint model, and score the physical rationality anomaly of a single sensor based on the detection results; Specifically, the physical rationality detection of sensor data after ship damage is based on a physical constraint model, including position change detection, velocity logic detection, and acceleration or angular velocity detection. The physical rationality anomaly score of a single sensor is the sum of the anomaly scores of the three detections: position change detection, velocity logic detection, and acceleration or angular velocity detection.

[0044] When performing position change detection, the actual position change of the ship (measured by GNSS and inertial navigation system) can be compared with the theoretical position change (calculated by the position constraint formula in step S1). If the difference exceeds twice the reference standard deviation, it is determined that the position data is contradictory. When performing speed logic testing, the vector composition relationship between the ground speed, water speed, and ocean current speed is verified. If the vector difference exceeds 0.5 m / s (the allowable error for ship speed measurement), the speed data is determined to be contradictory. When detecting acceleration or angular velocity, it can be verified whether the ship's acceleration and angular velocity are within the set safety threshold range. If they exceed the threshold, the data is judged to be abnormal.

[0045] Scoring rules: 1 point is added for each instance of physical logic contradiction or data anomaly, up to a maximum of 3 points.

[0046] S7: Summing the correlation anomaly score, stability anomaly score, redundancy anomaly score, and physical rationality anomaly score of a single sensor to obtain the total anomaly score of a single sensor. The sensor fault level is determined based on the total anomaly score of the single sensor, and the navigation capability of the ship after damage is assessed based on the fault levels of all sensors.

[0047] Based on the analysis results of the above four items, the total anomaly score for a single sensor is summarized, and the specific total anomaly scores for single sensor data are shown in Table 2: Table 2

[0048] Table 3 shows the assessment of a ship's overall navigation capability based on the total anomaly score from a single sensor. Table 3

[0049] By employing the aforementioned method to self-assess the navigation capabilities of damaged vessels, a preliminary assessment can be provided approximately 10 seconds after a collision, with a comprehensive quantitative assessment completed within 30 seconds. This rapid response is hundreds of times more efficient than traditional manual inspection, providing crucial time for emergency evacuation and personnel avoidance. Furthermore, the assessment method integrates statistical correlation, stability, physical plausibility, and redundancy verification techniques, effectively detecting hidden faults that are difficult to detect manually (such as loose components or parameter drift), significantly reducing the false positive rate. During the assessment, crew members do not need to enter hazardous equipment compartments; the assessment is completed entirely through data analysis, avoiding secondary disasters and injuries to personnel after a collision. Moreover, no hardware modifications to the existing ship navigation system are required; only software modules need to be added to achieve the functionality, significantly shortening ship downtime and providing clear guidance for crew maintenance. This approach demonstrates excellent economic viability and widespread applicability to all types of vessels.

[0050] A self-assessment system for navigation capability after ship damage, used to execute a self-assessment method for navigation capability after ship damage as described in any of the above, comprising a sensor data correlation benchmark model construction module, a sensor data acquisition and preprocessing module for all navigation support equipment after ship damage, a correlation anomaly scoring module, a stability anomaly scoring module, a redundancy anomaly scoring module, a physical rationality anomaly scoring module, and a navigation capability assessment module. The sensor data correlation benchmark model construction module is used to collect data from all core navigational support sensors of the ship during normal navigation, preprocess the data, establish a sensor data correlation benchmark model including a physical constraint model and a redundancy verification model, and calculate the Pearson correlation coefficient between the data from all core navigational support sensors of the ship. The sensor data acquisition and preprocessing module for all navigation support equipment after ship damage is used to acquire sensor data of all navigation support equipment after ship collision in real time in high-speed acquisition mode, and to preprocess the sensor data of all navigation support equipment after ship damage. The correlation anomaly scoring module calculates the Pearson correlation coefficient matrix of the sensor data of the navigation support equipment after the ship is damaged based on the preprocessed sensor data of the navigation support equipment after the ship is damaged, and calculates the change in the correlation coefficient of the same type of physical quantity. Based on the change in the correlation coefficient and the Pearson correlation coefficient matrix of the sensor data of the navigation support equipment after the ship is damaged, a single sensor correlation anomaly score is performed. The stability anomaly scoring module is used to compare the mean, standard deviation and deviation threshold of the real-time deviation sequence of sensor data of the navigation support equipment after the ship is damaged, and to score the stability anomaly of a single sensor based on the comparison results. The redundancy anomaly scoring module is used to perform redundancy verification and detection on the navigation support equipment after the ship is damaged, and to score the redundancy anomaly of a single sensor based on the redundancy verification and detection results. The physical rationality anomaly scoring module detects the physical rationality of sensor data after ship damage based on a physical constraint model, and scores the physical rationality anomalies of a single sensor based on the detection results. The navigation capability assessment module is used to sum the correlation anomaly score, stability anomaly score, redundancy anomaly score, and physical rationality anomaly score of a single sensor to obtain a total anomaly score for a single sensor. The sensor fault level is determined based on the total anomaly score of the single sensor, and the navigation capability of the ship after damage is assessed based on the fault levels of all sensors.

[0051] In summary, the present invention provides a self-assessment method and system for navigation capability after ship damage. After a ship collision, it can automatically, quickly, and accurately assess the availability of navigation support equipment and navigation support capability based on the output parameters of navigation support sensors. This solves the problems of existing navigation support capability assessment methods that rely on manual labor, are inefficient, and have poor accuracy. It provides a basis and guidance for ship missions and crew maintenance, and provides reliable support for emergency decision-making.

[0052] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for self-assessing the navigation capability of a ship after damage, characterized in that: Includes the following steps: S1: During normal navigation, collect data from all core navigational support sensors of the ship, preprocess the data, establish a sensor data correlation benchmark model including physical constraint model and redundancy verification model, and calculate the Pearson correlation coefficient matrix and deviation threshold between the data of all core navigational support sensors of the ship. S2: After a ship collision, in high-speed acquisition mode, sensor data of all navigation support equipment after the ship is damaged are collected in real time, and the sensor data of all navigation support equipment after the ship is damaged are preprocessed. S3: Calculate the Pearson correlation coefficient matrix of the sensor data of the navigation support equipment after the ship is damaged based on the preprocessed sensor data of the navigation support equipment after the ship is damaged, and calculate the change in the correlation coefficient of the same type of physical quantity. Based on the change in the correlation coefficient and the Pearson correlation coefficient matrix of the sensor data of the navigation support equipment after the ship is damaged, perform a single sensor correlation anomaly score. S4: Compare the mean, standard deviation and deviation threshold of the real-time deviation sequence of the sensor data of the navigation support equipment after the ship is damaged, and score the stability anomaly of a single sensor based on the comparison results. S5: Conduct redundancy verification and testing on the navigation support equipment after the ship is damaged, and score the redundancy anomaly of a single sensor based on the redundancy verification and testing results; S6: Detect the physical rationality of sensor data after ship damage based on a physical constraint model, and score the physical rationality anomaly of a single sensor based on the detection results; S7: Summing the correlation anomaly score, stability anomaly score, redundancy anomaly score, and physical rationality anomaly score of a single sensor to obtain the total anomaly score of a single sensor. The sensor fault level is determined based on the total anomaly score of the single sensor, and the navigation capability of the ship after damage is assessed based on the fault levels of all sensors.

2. The self-assessment method for navigation capability after ship damage according to claim 1, characterized in that: The data collected in step S1 from all core maritime safety sensors of the ship include: position, speed, and heading data from the GNSS receiver; position, speed, heading, angular velocity, and acceleration data from the inertial navigation system; speed data from the log; heading data from the compass; wind speed and direction data from the wind sensor; dynamic data of surrounding ships from the automatic identification system; and target detection data from the radar.

3. The self-assessment method for navigation capability after ship damage according to claim 1, characterized in that: The preprocessing of sensor data in steps S1 and S2 is as follows: a precise time protocol is used to synchronize the time of all core maritime safety sensors on the ship; outliers in the data are removed using the 3σ criterion; missing data is filled in using linear interpolation; and the timestamps and coordinate systems of all data are unified.

4. The self-assessment method for navigation capability after ship damage according to claim 1, characterized in that: The timestamp accuracy when collecting sensor data from the maritime support equipment in steps S1 and S2 is ≤1s.

5. The self-assessment method for navigation capability after ship damage according to claim 1, characterized in that: The process of constructing the physical constraint model in step S1 is as follows: S111: Based on the ship's speed, heading, and time interval, derive the ship's theoretical displacement according to equation (1) and constrain the position change: （1）； in: Indicates that the ship is The theoretical displacement along the heading direction within a given time period. Indicates the ship's speed relative to the water. Indicates the data sampling time interval. Indicates the ship's true heading angle. Indicates that the ship is The theoretical displacement perpendicular to the heading direction within a given time period; S112: Velocity constraints are applied based on the vector composition relationship (2) between the ship's land speed, water speed, and ocean current speed. （2）； in: Represents the ship's velocity vector relative to the ground. This represents the ship's velocity vector relative to the water. Represents the ocean current velocity vector; S113: Combine the maximum thrust of the ship's main engine and the structural characteristics of the hull to set the longitudinal acceleration threshold and the lateral acceleration threshold of the ship. Under normal navigation conditions, acceleration is constrained by equation (3): （3）； in: Indicates the longitudinal acceleration of the ship. Indicates the longitudinal acceleration threshold of the ship. Indicates the lateral acceleration of a ship. Indicates the threshold value for lateral acceleration of a ship; S114: Set the ship's heading angular velocity threshold based on the ship's steering performance. Under normal steering conditions, the angular velocity is constrained by equation (4): （4）； in: Indicates the angular velocity of the ship's course. This indicates the threshold value for the ship's heading angular velocity.

6. The self-assessment method for navigation capability after ship damage according to claim 1, characterized in that: In step S1, the Pearson correlation coefficient matrix of the sensor data of the navigation support equipment is calculated according to equation (5): （5）； in: Indicates the first The and the first Pearson correlation coefficient matrix of sensor data, Indicates the number of samples. Indicates the first The sensor at the first The measured value at time, Indicates the first The sensor at the first The measured value at time.

7. The self-assessment method for navigation capability after ship damage according to claim 6, characterized in that: In step S3, the change in the correlation coefficient of the same type of physical quantities is calculated according to equation (6): （6）； in: This indicates the change in the correlation coefficient. Indicates the first [day] after the ship suffered damage. The and the first The Pearson correlation coefficient matrix of sensor data.

8. The self-assessment method for navigation capability after ship damage according to claim 1, characterized in that: In step S5, when performing redundancy verification testing on the navigation support equipment after ship damage, a 95% confidence interval is used to determine the redundancy verification threshold. ,in: Indicates the first The and the first The mean of the data deviation sequence of each sensor, Indicates the first The and the first The standard deviation of sensor data.

9. The self-assessment method for navigation capability after ship damage according to claim 1, characterized in that: In step S6, the physical rationality of the sensor data after the ship is damaged is detected based on the physical constraint model, including position change detection, velocity logic detection, and acceleration or angular velocity detection. The physical rationality anomaly score is the sum of the anomaly scores of the three detections: position change detection, velocity logic detection, and acceleration or angular velocity detection.

10. A self-assessment system for navigation capability after ship damage, characterized in that: The method for performing a self-assessment of navigation capability after a ship is damaged, as described in any one of claims 1 to 9, includes a sensor data correlation benchmark model construction module, a sensor data acquisition and preprocessing module for all navigation support equipment after the ship is damaged, a correlation anomaly scoring module, a stability anomaly scoring module, a redundancy anomaly scoring module, a physical rationality anomaly scoring module, and a navigation capability assessment module. The sensor data correlation benchmark model construction module is used to collect data from all core navigational support sensors of the ship during normal navigation, preprocess the data, establish a sensor data correlation benchmark model including a physical constraint model and a redundancy verification model, and calculate the Pearson correlation coefficient between the data from all core navigational support sensors of the ship. The sensor data acquisition and preprocessing module for all navigation support equipment after ship damage is used to acquire sensor data of all navigation support equipment after ship collision in real time in high-speed acquisition mode, and to preprocess the sensor data of all navigation support equipment after ship damage. The correlation anomaly scoring module calculates the Pearson correlation coefficient matrix of the sensor data of the navigation support equipment after the ship is damaged based on the preprocessed sensor data of the navigation support equipment after the ship is damaged, and calculates the change in the correlation coefficient of the same type of physical quantity. Based on the change in the correlation coefficient and the Pearson correlation coefficient matrix of the sensor data of the navigation support equipment after the ship is damaged, a single sensor correlation anomaly score is performed. The stability anomaly scoring module is used to compare the mean, standard deviation and deviation threshold of the real-time deviation sequence of sensor data of the navigation support equipment after the ship is damaged, and to score the stability anomaly of a single sensor based on the comparison results. The redundancy anomaly scoring module is used to perform redundancy verification and detection on the navigation support equipment after the ship is damaged, and to score the redundancy anomaly of a single sensor based on the redundancy verification and detection results. The physical rationality anomaly scoring module detects the physical rationality of sensor data after ship damage based on a physical constraint model, and scores the physical rationality anomalies of a single sensor based on the detection results. The navigation capability assessment module is used to sum the correlation anomaly score, stability anomaly score, redundancy anomaly score, and physical rationality anomaly score of a single sensor to obtain a total anomaly score for a single sensor. The sensor fault level is determined based on the total anomaly score of the single sensor, and the navigation capability of the ship after damage is assessed based on the fault levels of all sensors.

Images