High-speed stern bearing wear detection system and method

Abstract

The invention discloses a high-speed stern bearing wear detection system and method. The high-speed stern bearing wear detection system comprises a sensor module, a data acquisition module, a data processing module and an early warning module, the sensor module is used for collecting size data, temperature data and iron ion concentration data; the data acquisition module is electrically connected with the sensor module and is used for receiving data acquired by the sensor module and transmitting the data to the data processing module; the data processing module is electrically connected with the data acquisition module and is used for analyzing and processing the received data so as to judge the wear degree of the tail bearing; the early warning module is electrically connected with the data processing module and used for sending out a corresponding early warning signal according to the abrasion degree judged by the data processing module. According to the high-speed stern bearing wear detection system and method provided by the invention, real-time monitoring of the wear degree of the high-speed stern bearing is realized, and the accuracy and timeliness of detection are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of ship machinery condition monitoring and intelligent diagnosis, and in particular to a high-speed stern bearing wear detection system and method. Background Art

[0002] In a ship's propulsion system, the tail bearing is a core component supporting the propeller shaft system, and its operating status directly affects the ship's navigation safety and power efficiency. High-speed ship stern bearings are subjected to multiple forces such as alternating loads, seawater corrosion, and mud and sand abrasive particles for a long time, and the working environment is particularly harsh. With the increasing trend of larger and faster ships, the load intensity per unit area of ​​the tail bearing has increased by more than 60% compared to before, placing higher demands on wear detection technology. The industry currently generally uses auxiliary methods such as vibration spectrum analysis and lubricant metal content detection, combined with new monitoring methods such as acoustic emission technology and fiber optic sensing. An assessment system covering mechanical performance degradation and surface morphology changes has been established, providing data support for bearing life prediction.

[0003] However, existing tail bearing wear monitoring technologies still face specific challenges in practical application. Traditional detection methods mainly rely on disassembly and measurement during annual dry docking. Although accurate radial clearance data can be obtained, the detection cycle conflicts with the ship's operating plan, and it is difficult to promptly detect sudden abnormal wear that occurs between dry dockings. Although the online monitoring systems developed in recent years can achieve operational status tracking, the fusion and analysis of multi-source heterogeneous data is still insufficient. In particular, the coupling mechanism of electrochemical corrosion and mechanical wear caused by changes in seawater salinity has not yet been effectively separated by a feature separation model, which may lead to misjudgment of the degree of wear. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a high-speed stern bearing wear detection system and method to achieve high-speed stern bearing detection accuracy.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: A high-speed stern bearing wear detection system includes: a sensor module, a data acquisition module, a data processing module and an early warning module; The sensor module is used to collect size data, temperature data and iron ion concentration data; The data acquisition module is electrically connected to the sensor module and is used to receive data collected by the sensor module and transmit it to the data processing module; The data processing module is electrically connected to the data acquisition module and is used to analyze and process the received data to determine the wear degree of the tail bearing; The early warning module is electrically connected to the data processing module and is used to issue a corresponding early warning signal according to the degree of wear determined by the data processing module.

[0006] In order to solve the above technical problems, another technical solution adopted by the present invention is: A method for detecting wear of a high-speed stern bearing comprises the following steps: S1. Measure the change in the spacing between the sealing rings and the change in the longitudinal distance to the tail bearing using an ultrasonic sensor, and collect the cooling water temperature and iron ion concentration using a temperature sensor and an iron ion sensor, respectively; S2. performing temperature compensation correction on the distance variation based on the cooling water temperature and deducting the background value from the iron ion concentration to obtain a corrected concentration; S3. Constructing a wear quantification model based on the longitudinal distance change, the corrected spacing change, and the corrected concentration, wherein the model includes at least a temperature index term, an iron ion term, and a cross term, and each term is provided with a weight coefficient; S4. Using a recursive algorithm with a forgetting factor to update the weight coefficient, using an iron ion concentration sample to calibrate and verify the error of the weight coefficient, and determining the weight coefficient; S5. Detecting the wear condition of the tail bearing based on the wear quantification model after determining the weight coefficient.

[0007] The beneficial effects of the present invention are: providing a high-speed ship stern bearing wear detection system and method, realizing real-time monitoring of the wear degree of high-speed ship stern bearings, improving the accuracy and timeliness of detection, and being able to effectively avoid safety hazards caused by sudden abnormal wear, thereby ensuring the navigation safety and power efficiency of the ship; by fusing multi-source heterogeneous data, including spacing data, temperature data and iron ion concentration data, and adopting advanced algorithm models for comprehensive analysis and processing, the system can more comprehensively evaluate the wear status of the stern bearing, reduce misjudgment caused by single data, improve the accuracy and reliability of diagnosis, and also has intelligent early warning and alarm functions, which can timely issue early warning signals according to changes in the degree of wear, reminding relevant personnel to take measures to deal with it, thereby avoiding serious failures and downtime losses caused by increased wear. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 A system module framework diagram according to an embodiment of the present invention; Figure 2 Schematic diagram of the structure of a stern bearing in an embodiment of the present invention; Figure 3 is a flow chart of a method in an embodiment of the present invention; Description of labels: 1. First sealing ring; 2. Second sealing ring; 3. Tail bearing; DETAILED DESCRIPTION To illustrate the technical content, achieved objectives and effects of the present invention in detail, the following description is given in conjunction with the embodiments and accompanying drawings.

[0009] Reference Figure 1 As shown, the high-speed stern bearing wear detection system includes: a sensor module, a data acquisition module, a data processing module and an early warning module; The sensor module is used to collect size data, temperature data and iron ion concentration data; The data acquisition module is electrically connected to the sensor module and is used to receive data collected by the sensor module and transmit it to the data processing module; The data processing module is electrically connected to the data acquisition module and is used to analyze and process the received data to determine the wear degree of the tail bearing; The early warning module is electrically connected to the data processing module and is used to issue a corresponding early warning signal according to the degree of wear determined by the data processing module.

[0010] From the above description, it can be seen that a high-speed stern bearing wear detection system is provided, which realizes real-time monitoring of the wear degree of high-speed stern bearings, improves the accuracy and timeliness of detection, can effectively avoid safety hazards caused by sudden abnormal wear, and ensure the navigation safety and power efficiency of the ship; by fusing multi-source heterogeneous data, including spacing data, temperature data and iron ion concentration data, and using advanced algorithm models for comprehensive analysis and processing, the system can more comprehensively evaluate the wear status of the stern bearing, reduce misjudgment caused by single data, improve the accuracy and reliability of diagnosis, and also has intelligent early warning and alarm functions, which can issue early warning signals in time according to changes in the degree of wear, reminding relevant personnel to take measures to deal with it, thereby avoiding serious failures and downtime losses caused by increased wear.

[0011] The sensor module includes: ultrasonic sensor unit, temperature sensor unit and iron ion monitoring sensor unit; refer to Figure 2 As shown, the ultrasonic sensor unit is arranged between the first sealing ring 1 and the second sealing ring 2 of the tail bearing, and is used to measure the distance data X between the first sealing ring 1 and the second sealing ring 2; The ultrasonic sensor unit is also arranged below the tail bearing 3 and is used to measure the distance data Y between the tail bearing 3 and the first sealing ring; The temperature sensing unit is used to monitor the temperature data of the tail bearing cooling water in real time; The iron ion monitoring sensor unit is used to monitor the concentration data of iron ions in cooling water in real time; The data processing module receives the spacing data, the temperature data and the iron ion concentration data, and processes the data collected by the ultrasonic sensor unit using ultrasonic imaging technology to form a 2D image of the internal sealing ring of the tail bearing.

[0012] As a core component of a ship's propulsion system, the operating status of the stern bearing is directly related to the ship's navigation safety and power efficiency. Traditional detection methods suffer from long cycle times and low accuracy, making it difficult to detect wear and take timely action. The detection system described in this invention, however, enables real-time monitoring of stern bearing wear and provides intelligent early warnings, effectively preventing safety hazards caused by wear and improving ship reliability and safety. Furthermore, this system helps optimize maintenance plans, reduce maintenance costs, and improve ship operational efficiency.

[0013] The high-speed stern bearing wear detection system comprehensively and accurately monitors stern bearing wear. Through the collaborative operation of ultrasonic, temperature, and iron ion monitoring sensors, the system collects critical data in real time and uses this data to create a 2D image of the stern bearing's internal seal, accurately capturing distance measurements at multiple locations. This not only improves wear detection accuracy but also provides robust data support for subsequent fault analysis and repair decisions.

[0014] When determining the wear degree of the tail bearing, the data processing module assesses the wear degree based on the changes in the spacing data between the first and second sealing rings, as well as the changes in the spacing data from the first sealing ring to the bottom of the tail bearing, obtained from the 2D image, combined with the temperature data and the iron ion concentration data. During actual use, the weighting of each parameter is continuously corrected based on the actual detection data. The specific method and steps for assessing the wear degree are as follows: Based on the sensor module, the size data, temperature data and iron ion concentration data are obtained and normalized. Assign weight coefficients to various data and calculate the comprehensive quantitative value of the tail bearing wear degree. The calculation formula is: ; Where, is the comprehensive quantitative value of the tail bearing wear degree, is the distance change between the first sealing ring and the second sealing ring, is the distance change from the first sealing ring to the ultrasonic sensor below the tail bearing, is the temperature value detected by the temperature sensor, The iron ion concentration detected by the iron ion monitoring sensor. is the temperature change, is the baseline value of iron ion concentration, 、 、 、 、 is the weight coefficient, is the material thermal expansion correction coefficient, is the nonlinear adjustment coefficient affected by temperature.

[0015] Stern bearing wear is a complex process influenced by multiple factors. Analysis based on a single factor often fails to fully reflect the true extent of wear, easily leading to misjudgments or omissions. By constructing a comprehensive quantitative model that incorporates multiple influencing factors, we can more accurately capture the changing patterns of stern bearing wear, providing a scientific basis for timely maintenance measures. Furthermore, the impact of wear factors may vary with increasing vessel operation time and changing operating conditions, necessitating continuous parameter weighting adjustments to maintain the accuracy and effectiveness of the evaluation model.

[0016] By introducing weighting and correction factors, wear assessment becomes more accurate and comprehensive. Normalization ensures effective comparison and comprehensive analysis of data from different dimensions, improving the reliability of the assessment results. Furthermore, by continuously adjusting the weighting of each parameter based on actual test data, the model dynamically adapts to the wear characteristics of the tail bearing under different operating conditions, making the assessment model more realistic.

[0017] The method for continuously correcting the proportion of each parameter based on actual detection data is as follows: An initial parameter weight model based on historical data was established. After deployment, sensors were used to continuously collect seal displacement, temperature, and iron ion concentration data, and correlated these with actual wear. After pre-processing the collected data by filtering outliers, correcting temperature compensation, and normalizing dimensions, a sliding window mechanism is used to retain valid data for a preset time period. The preprocessed data is input into the recursive least squares algorithm with a forgetting factor to iteratively update the weight coefficients, while constraining the single parameter adjustment amplitude to not exceed the set threshold; Regularly compare the prediction error with historical data through offline verification. When the error exceeds the limit, the weight parameters are rolled back and manual verification is triggered. An independent injection calibration experiment is performed for the iron ion concentration parameter to eliminate environmental interference. The parameter update frequency is dynamically adjusted according to the stability of the operating conditions, and real-time high-frequency updates are initiated when a sudden change in temperature or iron ion concentration is detected. At the same time, a manual intervention interface is set to force a weight reset or impose empirical constraints.

[0018] The early warning module is provided with a normal wear degree range. When the wear degree Q value calculated by the data processing module deviates from the normal wear degree range, the early warning module sends a wear alarm signal.

[0019] The wear process of the stern bearing is influenced by a variety of complex factors, the intensity and mode of action of which may vary over time. By establishing an initial weight model based on historical data and adjusting it with feedback from real-time data, the system can gradually learn the wear assessment model that best reflects actual conditions. Furthermore, regular offline verification and manual calibration mechanisms further ensure the accuracy and robustness of the model. The early warning module is designed to meet the needs of safety management, ensuring that swift action can be taken when the degree of wear exceeds the normal range to ensure the safe operation of the ship. By continuously adjusting the weighting of various parameters, the system can adapt to the changing wear characteristics of the stern bearing under different operating conditions, reducing misjudgments due to changes in the environment or operating conditions. Furthermore, the early warning module's configuration enables timely response to abnormal changes in the degree of wear, providing maintenance personnel with valuable reaction time and effectively preventing serious failures caused by wear.

[0020] refer to Figure 3 As shown, the high-speed stern bearing wear detection method includes the following steps: S1. Measure the change in the spacing between the sealing rings and the change in the longitudinal distance to the tail bearing using an ultrasonic sensor, and collect the cooling water temperature and iron ion concentration using a temperature sensor and an iron ion sensor, respectively; S2. performing temperature compensation correction on the distance variation based on the cooling water temperature and deducting the background value from the iron ion concentration to obtain a corrected concentration; S3. Constructing a wear quantification model based on the longitudinal distance change, the corrected spacing change, and the corrected concentration, wherein the model includes at least a temperature index term, an iron ion term, and a cross term, and each term is provided with a weight coefficient; S4. Using a recursive algorithm with a forgetting factor to update the weight coefficient, using an iron ion concentration sample to calibrate and verify the error of the weight coefficient, and determining the weight coefficient; S5. Detecting the wear condition of the tail bearing based on the wear quantification model after determining the weight coefficient.

[0021] Through step S1, the coordinated work of the ultrasonic sensor, temperature sensor, and iron ion sensor enables the collection of multi-dimensional data on sealing ring spacing, longitudinal distance, cooling water temperature, and iron ion concentration, providing rich and accurate basic data for subsequent precise analysis.

[0022] The temperature compensation correction and iron ion concentration correction in step S2 effectively eliminate the influence of environmental factors on the measurement data, ensuring the accuracy and reliability of the data.

[0023] Preferably, step S3 further includes: The corrected spacing transformation, longitudinal distance change, cooling water temperature, and corrected concentration are converted into dimensionless parameters and processed based on the design benchmark value, allowable temperature range, and historical statistical mean, respectively. The corrected data are converted into dimensionless parameters and standardized, which not only facilitates the comparison and analysis between different data, but also improves the generalization ability of the model.

[0024] The model specifically includes linear, logarithmic, temperature exponential, iron ion, and cross terms, each with its own weighting coefficient. A small constant is added to the logarithmic term to prevent anomalies. The constructed comprehensive wear quantification model fully considers the various factors influencing tail bearing wear. By introducing linear, logarithmic, temperature exponential, iron ion, and cross terms, and assigning them appropriate weighting coefficients, the model more accurately reflects the actual wear of the tail bearing. Furthermore, the small constant added to the logarithmic term effectively prevents interference from outliers, while the nonlinear adjustment coefficient introduced into the exponential term enhances the model's adaptability to complex changes.

[0025] The recursive algorithm with a forgetting factor in step S4 updates weights online, enabling dynamic adjustment of model parameters. This allows the model to continuously adapt to new operating environments and wear characteristics, maintaining the accuracy and timeliness of the evaluation results. Regular iron ion sample calibration and error verification further ensure the stability and reliability of the model.

[0026] The implementation of the recursive algorithm with forgetting factor to update weights online includes: The forgetting factor ranges from 0.95 to 0.99. The initial weight coefficient vector contains preset parameter components. The input feature vector is composed of the feature items of the original data after logarithmic transformation, exponential transformation and difference operation, where the logarithm is a non-negative value after small constant compensation. The weights are iteratively updated by recursive least squares method, specifically: According to the covariance matrix of the previous moment and the current input eigenvector, the updated covariance matrix is ​​calculated according to the forgetting factor ratio, and the deviation between the actual wear amount and the predicted value is used as the error term. The weight coefficient is adjusted in combination with the covariance matrix and the input eigenvector; The initial value of the covariance matrix is ​​a unit matrix with diagonal elements of 1000, and the actual wear amount is obtained through offline measurement.

[0027] This high-speed stern bearing wear detection method offers the advantage of precise wear monitoring and intelligent early warning through multi-dimensional data collection and a dynamic model update mechanism. Ultrasonic sensors capture changes in seal ring spacing and longitudinal distance, combined with temperature and iron ion concentration sensor data to comprehensively capture key wear factors. Temperature compensation is applied to the spacing change, and background iron ion concentration is deducted, effectively eliminating environmental interference and ensuring data reliability.

[0028] In terms of model construction, the corrected multi-source parameters are converted into dimensionless quantities and standardized based on the design benchmark, temperature range, and statistical mean to enhance data comparability. The comprehensive wear quantification model introduces linear terms, logarithmic terms (including small constants to prevent anomalies), temperature exponential terms (with nonlinear adjustment coefficients), and cross terms. The weights are dynamically updated through a recursive algorithm with a forgetting factor. The forgetting factor is in the range of 0.95-0.99 to balance the weights of historical data and real-time information. The covariance matrix is ​​initialized to an identity matrix with 1000 diagonal elements. It is iteratively updated using recursive least squares, correcting predicted deviations based on actual wear measured offline. Specifically, the amplitude of a single weight adjustment is limited to less than 10% to avoid parameter mutations. Errors are verified offline every 24 hours. If the threshold exceeds 0.15, the parameters are rolled back and manually verified. The iron ion sensor is calibrated and corrected every 72 hours, forming a closed-loop quality control system.

[0029] The step S5 further includes: Set threshold intervals and alarm mechanisms, including: After the wear quantification model outputs wear detection data, the wear detection result is compared with the threshold interval. When the wear detection data exceeds the threshold interval, an alarm is triggered and a log containing time, the wear detection data and environmental parameters is generated.

[0030] Specifically, the threshold range is set between 0.8 and 1.2. Three consecutive exceeding of the threshold triggers an audible and visual alarm and records environmental parameters. A sustained 120% upper limit for two minutes results in a system shutdown, ensuring both sensitivity and robustness. Trend curves are generated and annotated with alarm events at the minute level, and weight change records are structured and stored for 30 days to support long-term data analysis. The manual intervention interface allows for adjustment of the threshold within a 20% range, password verification ensures operational security, and an audit database tracks change history. This design leverages algorithmic adaptability to improve monitoring accuracy, while ensuring system reliability through multiple verification mechanisms. It also empowers operators with emergency adjustment capabilities, forming an intelligent monitoring system that collaborates with humans and machines.

[0031] The working mode switching strategy includes: If the fluctuation of the cooling water temperature and the fluctuation of the iron ion concentration both meet the preset stable conditions, it is determined to be a steady-state operating condition, and the detection period of the wear quantification model is set to a first preset time length; Otherwise, it is determined to be a warning condition, and the detection period of the wear quantification model is set to a second preset time length, which is shorter than the first preset time length.

[0032] Specifically, when the temperature fluctuation does not exceed 2 degrees Celsius per hour and the iron ion concentration change rate does not exceed 0.1ppm per minute, it is determined to be a steady-state operating condition, using a 10-minute basic update cycle; When the temperature suddenly changes by more than 5 degrees Celsius per minute or the iron ion concentration reaches twice the initial value, it switches to the 1-minute high-frequency update mode and delays the reset cycle for 15 minutes after steady state recovery; The system also includes a manual reset command, which restores the weights to factory defaults and resets the covariance matrix. While locked, the system verifies the predicted wear value every hour, forcing the system to unlock and issue an alarm if the deviation exceeds 20%. In high-frequency mode, the data sampling rate is increased to 10Hz, and a Kalman filter is used for noise reduction, discarding abnormal data packets at the end of each cycle. This flexible operating mode switching strategy significantly benefits from the ability to dynamically adjust the monitoring frequency and data processing method based on the actual tail bearing wear under different operating conditions, effectively improving system efficiency and responsiveness while ensuring monitoring accuracy.

[0033] The wear state of the tail bearing is affected by a variety of complex factors, including temperature and iron ion concentration, and the changing characteristics of these factors vary under different operating conditions. Under steady-state conditions, wear changes relatively slowly, and adopting a longer update cycle can reduce system energy consumption while ensuring the continuity and stability of monitoring. When the temperature or iron ion concentration suddenly changes, it often indicates that the wear state may change significantly. At this time, switching to a high-frequency update mode can capture the dynamic changes of wear more promptly and provide early warning information to maintenance personnel. In addition, the manual reset and parameter lock functions enhance the controllability and safety of the system, ensuring that the system can quickly return to a stable state under special circumstances. The implementation of this strategy not only reflects a deep understanding of the wear characteristics of the tail bearing, but also demonstrates the level of intelligence and refinement in ensuring the safe operation of ships.

[0034] In summary, the present invention provides a high-speed ship stern bearing wear detection system and method, which realizes real-time monitoring of the wear degree of high-speed ship stern bearings, improves the accuracy and timeliness of detection, can effectively avoid safety hazards caused by sudden abnormal wear, and ensure the navigation safety and power efficiency of the ship; by fusing multi-source heterogeneous data, including spacing data, temperature data and iron ion concentration data, and using advanced algorithm models for comprehensive analysis and processing, the system can more comprehensively evaluate the wear status of the stern bearing, reduce misjudgment caused by single data, improve the accuracy and reliability of diagnosis, and also has intelligent early warning and alarm functions, which can issue early warning signals in time according to changes in the degree of wear, reminding relevant personnel to take measures to deal with it, and avoid serious failures and downtime losses caused by increased wear.

[0035] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's description and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A high-speed stern bearing wear detection system, characterized in that: include: Sensor module, data acquisition module, data processing module and early warning module; The sensor module is used to collect size data, temperature data and iron ion concentration data; The data acquisition module is electrically connected to the sensor module and is used to receive data collected by the sensor module and transmit it to the data processing module; The data processing module is electrically connected to the data acquisition module and is used to analyze and process the received data to determine the wear degree of the tail bearing; The early warning module is electrically connected to the data processing module and is used to issue a corresponding early warning signal according to the degree of wear determined by the data processing module.

2. A high-speed ship stern bearing wear detection system according to claim 1, characterized in that: The sensor module includes: ultrasonic sensor unit, temperature sensor unit and iron ion monitoring sensor unit; The ultrasonic sensing unit is arranged between the first sealing ring and the second sealing ring of the tail bearing, and is used to measure the distance data between the first sealing ring and the second sealing ring; The ultrasonic sensing unit is further arranged below the tail bearing and is used to measure the distance data between the tail bearing and the first sealing ring; The temperature sensing unit is used to monitor the temperature data of the tail bearing cooling water in real time; The iron ion monitoring sensor unit is used to monitor the concentration data of iron ions in cooling water in real time; The data processing module receives the spacing data, the temperature data, and the iron ion concentration data, and processes the data collected by the ultrasonic sensor unit using ultrasonic imaging technology to form a 2D image of the internal sealing ring of the tail bearing.

3. A high-speed ship stern bearing wear detection system according to claim 2, characterized in that: When determining the degree of wear of the tail bearing, the data processing module evaluates the degree of wear based on changes in the spacing data between the first sealing ring and the second sealing ring, and changes in the spacing data from the first sealing ring to the bottom of the tail bearing, obtained from the 2D image, combined with the temperature data and the iron ion concentration data. During actual use, the weighting of each parameter is continuously corrected based on actual detection data.

4. A high-speed ship stern bearing wear detection system according to claim 3, characterized in that: The specific steps for evaluating the degree of wear are as follows: Based on the distance data, the temperature data, and the iron ion concentration data acquired by the sensor module, normalizing the various data; Assign weight coefficients to various data and calculate the comprehensive quantitative value of the tail bearing wear degree. The calculation formula is: ； Where, is the comprehensive quantitative value of the tail bearing wear degree, is the distance change between the first sealing ring and the second sealing ring, is the distance change from the first sealing ring to the ultrasonic sensor below the tail bearing, is the temperature value detected by the temperature sensor, The iron ion concentration detected by the iron ion monitoring sensor. is the temperature change, is the baseline value of iron ion concentration, 、 、 、 、 is the weight coefficient, is the material thermal expansion correction coefficient, is the nonlinear adjustment coefficient affected by temperature.

5. A high-speed ship stern bearing wear detection system according to claim 3, characterized in that: The method for continuously correcting the proportion of each parameter based on actual detection data is as follows: An initial parameter weight model based on historical data was established. After deployment, sensors were used to continuously collect seal displacement, temperature, and iron ion concentration data, and correlated these with actual wear. After preprocessing the collected data by filtering outliers, correcting temperature compensation, and normalizing dimensions, a sliding window mechanism is used to retain valid data for a preset time period. The preprocessed data is input into the recursive least squares algorithm with a forgetting factor to iteratively update the weight coefficients, while constraining the single parameter adjustment amplitude to not exceed the set threshold; Regularly compare the prediction error with historical data through offline verification. When the error exceeds the limit, the weight parameters are rolled back and manual verification is triggered. An independent injection calibration experiment is performed for the iron ion concentration parameter to eliminate environmental interference. The parameter update frequency is dynamically adjusted according to the stability of the operating conditions, and real-time high-frequency updates are initiated when a sudden change in temperature or iron ion concentration is detected. At the same time, a manual intervention interface is set to force a weight reset or impose empirical constraints.

6. A high-speed ship stern bearing wear detection system according to claim 1, characterized in that: The early warning module is provided with a normal wear degree range. When the wear degree Q value calculated by the data processing module deviates from the normal wear degree range, the early warning module sends a wear alarm signal.

7. A method for detecting wear of a high-speed ship stern bearing, characterized in that: The following steps are involved: S1. Measure the change in the spacing between the sealing rings and the change in the longitudinal distance to the tail bearing using an ultrasonic sensor, and collect the cooling water temperature and iron ion concentration using a temperature sensor and an iron ion sensor, respectively; S2. performing temperature compensation correction on the distance variation based on the cooling water temperature and deducting the background value from the iron ion concentration to obtain a corrected concentration; S3. Constructing a wear quantification model based on the longitudinal distance change, the corrected spacing change, and the corrected concentration, wherein the model includes at least a temperature index term, an iron ion term, and a cross term, and each term is provided with a weight coefficient; S4. Using a recursive algorithm with a forgetting factor to update the weight coefficient, using an iron ion concentration sample to calibrate and verify the error of the weight coefficient, and determining the weight coefficient; S5. Detecting the wear condition of the tail bearing based on the wear quantification model after determining the weight coefficient.

8. A high-speed ship stern bearing wear detection method according to claim 7, characterized in that: The implementation of the recursive algorithm with forgetting factor to update weights includes: The forgetting factor ranges from 0.95 to 0.

99. The initial weight coefficient vector contains preset parameter components. The input feature vector is composed of the feature items of the original data after logarithmic transformation, exponential transformation and difference operation, where the logarithm is a non-negative value after small constant compensation. The weights are iteratively updated by recursive least squares method, specifically: According to the covariance matrix of the previous moment and the current input eigenvector, the updated covariance matrix is ​​calculated according to the forgetting factor ratio, and the deviation between the actual wear amount and the predicted value is used as the error term. The weight coefficient is adjusted in combination with the covariance matrix and the input eigenvector; The initial value of the covariance matrix is ​​a unit matrix with diagonal elements of 1000, and the actual wear amount is obtained through offline measurement.

9. A high-speed ship stern bearing wear detection method according to claim 7, characterized in that: The step S5 further includes: Set threshold intervals and alarm mechanisms, including: After the wear quantification model outputs wear detection data, the wear detection result is compared with the threshold interval. When the wear detection data exceeds the threshold interval, an alarm is triggered and a log containing time, the wear detection data and environmental parameters is generated.

10. A high-speed ship stern bearing wear detection method according to claim 7, characterized in that: The step S5 further includes: setting the working condition switching mode, specifically including: If the fluctuation of the cooling water temperature and the fluctuation of the iron ion concentration both meet the preset stable conditions, it is determined to be a steady-state operating condition, and the detection period of the wear quantification model is set to a first preset time length; Otherwise, it is determined to be a warning condition, and the detection period of the wear quantification model is set to a second preset time length, which is shorter than the first preset time length.

Images