Marine water lubrication stern bearing abrasion loss analysis method and system and ship monitoring equipment

Abstract

The invention provides a method and a system for analyzing the abrasion loss of a marine water-lubricated stern bearing and ship monitoring equipment. The method comprises the following steps: acquiring multi-modal data of the marine water-lubricated stern bearing; performing multi-dimensional feature extraction on the multi-modal data to obtain multi-dimensional feature data; inputting the multi-dimensional feature data into a wear quantitative evaluation model to obtain a wear quantitative evaluation result of the marine water-lubricated stern bearing, and determining an operation state of the marine water-lubricated stern bearing based on the wear quantitative evaluation result; and inputting the wear quantitative evaluation result and the historical wear quantitative evaluation result into a wear trend prediction model to obtain a predicted wear trend of the marine water-lubricated stern bearing in a future preset time period. According to the method, the accuracy and the reliability of abrasion loss analysis of the marine water-lubricated stern bearing are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of ship engineering and intelligent monitoring systems, and in particular to a method and system for analyzing the wear of a ship's water-lubricated stern bearing and ship monitoring equipment. Background Art

[0002] Marine water-lubricated stern bearings are key components in the ship's propulsion system, and their performance directly affects the ship's operating efficiency and safety.

[0003] Traditional stern bearing monitoring methods rely primarily on regular manual inspections or data collection from a single sensor. These methods suffer from long monitoring cycles, incomplete data, and limited fault prediction capabilities. For example, while vibration monitoring can detect dynamic bearing changes, it is prone to false alarms in high-noise environments. Temperature monitoring can reflect the bearing's thermal state, but it lacks a direct understanding of wear. Especially in the complex operating environment of a ship, a single sensor often struggles to fully reflect the true wear state of a bearing.

[0004] With the development of the Internet of Things, sensor technology, and artificial intelligence, multimodal data fusion technology is increasingly being applied to the health monitoring of mechanical equipment. By integrating data from various sensors, the operating status of equipment can be more comprehensively and accurately assessed. However, current multimodal monitoring systems for marine water-lubricated stern bearings are still incomplete, particularly lacking dynamic wear monitoring, analysis, and prediction systems based on multimodal data. Therefore, how to comprehensively utilize data from multiple sensors to improve the accuracy and reliability of stern bearing wear monitoring has become an urgent issue. Summary of the Invention

[0005] In view of this, it is necessary to provide a method and system for analyzing the wear of a marine water-lubricated stern bearing and a ship monitoring device to improve the accuracy and reliability of stern bearing wear monitoring.

[0006] In order to achieve the above objectives, in a first aspect, the present invention provides a method for analyzing wear of a water-lubricated stern bearing of a ship, comprising: Acquire multimodal data of a ship's water-lubricated stern bearing; Performing multi-dimensional feature extraction on the multimodal data to obtain multi-dimensional feature data; Inputting the multidimensional feature data into a wear quantitative evaluation model to obtain a wear quantitative evaluation result of the marine water-lubricated stern bearing, and determining an operating state of the marine water-lubricated stern bearing based on the wear quantitative evaluation result; The wear quantitative evaluation result and the historical wear quantitative evaluation result are input into a wear trend prediction model to obtain a predicted wear trend of the marine water-lubricated stern bearing within a future preset time period.

[0007] In a possible implementation, the multimodal data includes real-time shaft speed, real-time torque, real-time axial vibration acceleration, real-time bearing liner strip thickness, real-time bearing specific pressure, and real-time lubrication water temperature of a marine water-lubricated stern bearing.

[0008] In a possible implementation, performing multi-dimensional feature extraction on the multimodal data to obtain multi-dimensional feature data includes: A multi-level feature extraction strategy is used to extract time domain feature data, frequency domain feature data and time-frequency domain feature data corresponding to the multimodal data; The time domain feature data includes the mean, variance and peak factor of the multimodal data, the frequency domain feature data includes the main frequency component, frequency band energy distribution and power spectral density of the multimodal data, and the time-frequency domain feature data includes the energy distribution characteristics of the multimodal data at different time-frequency resolutions.

[0009] In a possible implementation, before performing multi-dimensional feature extraction on the multimodal data, the method further includes: Performing data synchronization, denoising and standardization on the multimodal data; Performing a data synchronization operation on the multimodal data includes: Using timestamp alignment technology to perform time consistency matching on the multimodal data; Performing denoising processing on the multimodal data, comprising: determining a target filtering algorithm based on characteristics of the multimodal data, and performing filtering processing on the multimodal data based on the target filtering algorithm; The multimodal data is subjected to standardization processing, including: A preset normalization method is used to perform scale consistency matching on the multimodal data.

[0010] In a possible implementation, the wear quantification assessment model is an integrated model of a support vector machine, a deep neural network model, and a random forest model.

[0011] In a possible implementation, the wear quantitative assessment result includes a wear degree score; and determining the operating state of the marine water-lubricated stern bearing based on the wear quantitative assessment result includes: Obtaining a correlation between the operating state and the multi-level warning threshold and a matching relationship between the wear degree score and the multi-level warning threshold; The operating status is determined based on the association relationship, the matching relationship, and the wear degree score.

[0012] In one possible implementation, the wear trend prediction model includes a convolutional neural network for extracting local temporal features and a long short-term memory network for capturing asymptotic changes in wear amount.

[0013] In a possible implementation, the method further includes: The wear quantitative evaluation results and the predicted wear trend are presented in a graphical form.

[0014] In a second aspect, the present invention provides a system for analyzing wear of a marine water-lubricated stern bearing, comprising: A data acquisition module, used to acquire multimodal data of a marine water-lubricated stern bearing; A multi-dimensional feature extraction module, configured to extract multi-dimensional features from the multimodal data to obtain multi-dimensional feature data; a wear quantitative evaluation module, configured to input the multidimensional feature data into a wear quantitative evaluation model, obtain a wear quantitative evaluation result of the marine water-lubricated stern bearing, and determine an operating state of the marine water-lubricated stern bearing based on the wear quantitative evaluation result; The wear trend prediction module is used to input the wear quantitative evaluation result and the historical wear quantitative evaluation result into the wear trend prediction model to obtain the predicted wear trend of the marine water-lubricated stern bearing within a future preset time period.

[0015] In a third aspect, the present invention provides a ship monitoring device, comprising a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps of any one of the above-mentioned methods for analyzing wear of a marine water-lubricated stern bearing.

[0016] The present invention has the following beneficial effects: by acquiring multimodal data from a marine water-lubricated stern bearing and evaluating wear based on data from multiple different modalities, the present invention enables a more comprehensive and accurate assessment of wear, thereby improving the accuracy of the wear quantification assessment results. Furthermore, the present invention extracts multidimensional features from the multimodal data after acquisition, further improving the comprehensiveness and diversity of the features, thereby further improving the accuracy of the wear quantification assessment results.

[0017] Furthermore, after obtaining the wear quantitative assessment results, the present invention can also obtain the predicted wear trend of the marine water-lubricated stern bearing within a future preset time period based on the wear quantitative assessment results and historical wear quantitative assessment results, providing a basis for the preventive maintenance of the marine water-lubricated stern bearing and further ensuring the safety of the use of the marine water-lubricated stern bearing. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 A schematic flow chart of an embodiment of a method for analyzing wear of a marine water-lubricated stern bearing provided by the present invention; Figure 2 A schematic diagram of the process of training the wear trend prediction model provided by the present invention; Figure 3 This is a schematic structural diagram of the marine water-lubricated stern bearing wear analysis system provided by the present invention; Figure 4 This is a schematic structural diagram of an embodiment of the ship monitoring equipment provided by the present invention. DETAILED DESCRIPTION

[0020] 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 some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0021] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present invention. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute a separate or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0022] The present invention provides a method and system for analyzing the wear of a water-lubricated stern bearing of a ship, and a ship monitoring device, which are described below respectively.

[0023] Figure 1 A schematic flow chart of an embodiment of a method for analyzing wear of a water-lubricated stern bearing of a ship provided by the present invention is shown in FIG. Figure 1 As shown in the figure, the wear analysis method of marine water-lubricated stern bearing includes: S101: Acquire multimodal data of a water-lubricated stern bearing of a ship.

[0024] Multimodal data generally refers to data from different sources or types. In the present invention, it refers to multiple different types of sensor data. Among them, the multimodal data of the marine water-lubricated stern bearing includes the real-time shaft speed, real-time torque value, real-time axial vibration acceleration, real-time bearing liner strip thickness, real-time bearing specific pressure and real-time lubricating water temperature of the marine water-lubricated stern bearing.

[0025] The specific method of acquiring multimodal data is as follows: based on the shaft speed sensor, the real-time shaft speed of the ship's water-lubricated stern bearing is collected; based on the shaft torque sensor, the real-time torque value of the ship's water-lubricated stern bearing is collected; based on the acceleration sensor, the real-time axial vibration acceleration of the ship's water-lubricated stern bearing is collected; based on the slat residual thickness sensor, the real-time bearing liner slat thickness of the ship's water-lubricated stern bearing is collected; based on the pressure sensor, the real-time bearing specific pressure of the ship's water-lubricated stern bearing is collected; based on the lubricating water temperature sensor, the real-time lubricating water temperature of the ship's water-lubricated stern bearing is collected.

[0026] Specifically, the real-time shaft speed measurement range is 0-500 rpm, with a sampling frequency of 100 Hz, and the real-time torque measurement range is 0-200 kN·m, with a sampling frequency of 50 Hz. Accelerometers monitor the real-time vibration state of the shaft and bearings, with an X / Y / Z axial acceleration measurement range of ±50 g, a sampling frequency of 1 kHz, and a resolution of 0.001 g. A slat thickness sensor monitors the remaining thickness of the bearing lining slats in real time, with a measurement range of 0-50 mm and a resolution of 0.01 mm. A pressure sensor measures the bearing pressure response, with a measurement range of 0-30 MPa and an accuracy of ±0.2%. A lubricating water temperature sensor monitors the bearing lubricating water temperature in real time, with a measurement range of 0-100°C and an accuracy of ±0.1°C. The collected multimodal data is transmitted via a wired or wireless network.

[0027] S102: performing multi-dimensional feature extraction on the multi-modal data to obtain multi-dimensional feature data corresponding to the multi-modal data; Multi-dimensional feature extraction refers to feature extraction in three different dimensions: time domain, frequency domain, and time-frequency domain. Correspondingly, multi-dimensional feature data includes time domain feature data, frequency domain feature data, and time-frequency domain feature data.

[0028] In a specific embodiment of the present invention, step S102 specifically comprises: extracting time domain feature data, frequency domain feature data, and time-frequency domain feature data corresponding to the multimodal data using a multi-level feature extraction strategy. Specifically, the time domain feature data includes the mean, variance, and crest factor of the multimodal data; the frequency domain feature data includes the main frequency component, frequency band energy distribution, and power spectral density of the multimodal data; and the time-frequency domain feature data includes the energy distribution characteristics of the multimodal data at different time-frequency resolutions.

[0029] S103: inputting the multi-dimensional feature data into a wear quantitative evaluation model to obtain a wear quantitative evaluation result of the marine water-lubricated stern bearing, and determining an operating state of the marine water-lubricated stern bearing based on the wear quantitative evaluation result; S104: Inputting the wear quantitative assessment result and the historical wear quantitative assessment result into a wear trend prediction model to obtain a predicted wear trend of the marine water-lubricated stern bearing within a future preset time period.

[0030] Compared to existing technologies, the present invention acquires multimodal data from marine water-lubricated stern bearings and assesses wear based on data from multiple different modalities. This allows for a more comprehensive and accurate assessment of wear, improving the accuracy of wear quantification assessment results. Furthermore, the present invention extracts multidimensional features from the acquired multimodal data, further enhancing the comprehensiveness and diversity of features and, consequently, the accuracy of wear quantification assessment results.

[0031] Furthermore, after obtaining the wear quantitative assessment results, the embodiment of the present invention can also obtain the predicted wear trend of the marine water-lubricated stern bearing within a future preset time period based on the wear quantitative assessment results and historical wear quantitative assessment results, providing a basis for preventive maintenance of the marine water-lubricated stern bearing and further ensuring the safety of the use of the marine water-lubricated stern bearing.

[0032] To ensure the accuracy of multi-dimensional feature extraction, in some embodiments of the present invention, before step S102, the following steps are further included: Perform data synchronization, denoising, and standardization on multimodal data; Perform data synchronization operations on multimodal data, including: Use timestamp alignment technology to perform time consistency matching on multimodal data; Denoising of multimodal data, including: Determine a target filtering algorithm based on the characteristics of the multimodal data, and perform filtering processing on the multimodal data based on the target filtering algorithm; Standardize multimodal data, including: A preset normalization method is used to perform scale consistency matching on multimodal data.

[0033] Timestamp alignment is a key technology for synchronizing multimodal data time series, widely used in fields such as the Internet of Things (IoT), multimedia processing, sensor fusion, financial transactions, and autonomous driving. Its core goal is to ensure temporal consistency of multimodal data by adjusting the timestamps of different data sources to eliminate time deviations caused by device clock differences, network latency, or varying acquisition frequencies. Specifically, an interpolation algorithm is used to achieve millisecond-level temporal consistency matching of multimodal sensor data, ensuring precise alignment of the data's temporal dimension.

[0034] Among them, the target filtering algorithm is wavelet threshold filtering or Kalman filtering method, which can effectively remove high-frequency noise and random interference signals while retaining effective signal characteristics.

[0035] Among them, the preset standardization processing methods include Z-score standardization or minimum-maximum normalization method. By standardizing multimodal data, the influence of different dimensions and magnitudes can be eliminated, and a unified scale conversion of multimodal data can be achieved.

[0036] In some embodiments of the present invention, the wear quantification assessment model is an integrated model of a support vector machine, a deep neural network model, and a random forest model.

[0037] By integrating multiple models into the wear quantification assessment model, its adaptability and accuracy can be improved. Specifically, support vector machines are used for accurate classification in small sample sizes, deep neural networks are used to model complex nonlinear relationships, and random forest ensemble learning methods are used to enhance the model's generalization capabilities.

[0038] In some embodiments of the present invention, the wear quantitative assessment result includes a wear degree score; then determining the operating state of the marine water-lubricated stern bearing based on the wear quantitative assessment result in step S103 includes: Obtain the correlation between the operating status and the multi-level warning thresholds, as well as the matching relationship between the wear degree score and the multi-level warning thresholds; The operating status is determined based on the association relationship, matching relationship and wear degree score.

[0039] Specifically, it is first determined at which warning threshold the wear degree score is located, and the operating state is determined based on the determined warning threshold and the correlation relationship.

[0040] The embodiment of the present invention dynamically compares the wear degree score with the preset multi-level warning threshold value to analyze and judge the operating status of the bearing.

[0041] In some embodiments of the present invention, the wear quantitative assessment result may further include a remaining predicted service life.

[0042] The full life cycle of the marine water-lubricated stern bearing and the predicted remaining service life can be compared to obtain the operating dynamics.

[0043] Since the wear trend prediction is a time series prediction and the wear trend prediction model needs to be trained before use, in some embodiments of the present invention, before step S104, a training process of the wear trend prediction model is further included, specifically: S201: Obtain historical wear data of a water-lubricated stern bearing of a ship, and construct a training sample set based on the historical wear data; Among them, historical wear data is time series data.

[0044] S202: Using the training sample set, adopting the sliding window technology, and combining the cross-validation method to train the initial wear trend prediction model to obtain the wear trend prediction model, wherein the initial wear trend prediction model includes a time series analysis unit, a local feature extraction unit based on a convolutional neural network, and a wear asymptotic change extraction unit based on a long short-term memory network.

[0045] The initial wear trend prediction model is constructed using deep learning models such as time series analysis, convolutional neural networks, and long short-term memory networks. Time series analysis is suitable for capturing trends, seasonality, and periodicity. CNNs are used to extract local time series features, such as short-term fluctuations. LSTMs handle long-term dependencies and are suitable for capturing asymptotic changes in wear. LSTM layers can be stacked or combined with a bidirectional structure to enhance feature extraction. During the training process of the initial wear trend prediction model, historical wear data is first modeled. This historical wear data is then input to construct a training sample set. Finally, a sliding window technique is used in conjunction with cross-validation to train the initial wear trend prediction model. The sliding window technique can capture local features from different time series. Cross-validation allows for repeated partitioning of the dataset to more comprehensively evaluate model performance, thereby improving the performance of the wear trend prediction model.

[0046] It is understandable that the wear trend prediction model also introduces an online learning mechanism, which continuously updates the model parameters of the wear trend prediction model through the real-time wear amount of the ship's water-lubricated stern bearing, thereby realizing dynamic optimization of the wear trend prediction model.

[0047] Furthermore, the prediction results of the wear trend prediction model not only include the predicted wear trend in a specific time period in the future, but also include the failure risk level assessment based on probability statistics and the corresponding confidence interval, providing an optimization basis for preventive maintenance decisions.

[0048] In one embodiment of the present invention, the above method further includes: The wear quantitative assessment results and predicted wear trends are presented in graphical form.

[0049] As you can understand, the user interface visually displays monitoring, analysis, and prediction results in charts and other formats, providing real-time status feedback and maintenance recommendations. The user-friendly interface includes a line graph of real-time monitoring data, a bar chart of wear, a trend chart for forecasting, and status indicators for fault warnings. This interface allows users to view the operating status of bearings in real time, receive warnings, and make maintenance decisions as needed.

[0050] The present invention fuses multimodal data and integrates data from multiple sensors such as temperature, vibration, and pressure to comprehensively reflect the operating status of the stern bearing, thereby improving the accuracy and reliability of monitoring. It adopts data fusion technology and machine learning algorithms to achieve efficient processing and in-depth analysis of complex data, and can accurately identify the degree of wear and operating anomalies. Based on historical data and real-time analysis results, the system can predict future wear trends and provide early warning information to help users take maintenance measures in advance and avoid sudden failures. Finally, the system provides an intuitive visual display, allowing users to easily view monitoring results, analysis reports, and forecast information, thereby improving operational convenience and decision-making efficiency.

[0051] In order to better implement the wear analysis method of the marine water-lubricated stern bearing in the embodiment of the present invention, based on the wear analysis method of the marine water-lubricated stern bearing, correspondingly, Figure 3 As shown, an embodiment of the present invention further provides a marine water-lubricated stern bearing wear analysis system, and the marine water-lubricated stern bearing wear analysis system 300 includes: The data acquisition module 301 is used to acquire multimodal data of a marine water-lubricated stern bearing; A multi-dimensional feature extraction module 302 is used to extract multi-dimensional features from multimodal data to obtain multi-dimensional feature data; A wear quantitative evaluation module 303 is configured to input the multi-dimensional feature data into a wear quantitative evaluation model to obtain a wear quantitative evaluation result of the marine water-lubricated stern bearing, and determine an operating state of the marine water-lubricated stern bearing based on the wear quantitative evaluation result; The wear trend prediction module 304 is used to input the wear quantitative evaluation results and the historical wear quantitative evaluation results into the wear trend prediction model to obtain the predicted wear trend of the marine water-lubricated stern bearing within a future preset time period.

[0052] The marine water-lubricated stern bearing wear analysis system 300 provided in the above embodiment can implement the technical solution described in the above embodiment of the marine water-lubricated stern bearing wear analysis method. The specific implementation principles of the above modules or units can refer to the corresponding contents in the above embodiment of the marine water-lubricated stern bearing wear analysis method, and will not be repeated here.

[0053] like Figure 4As shown, the present invention also provides a ship monitoring device 400. The ship monitoring device 400 includes a processor 401, a memory 402 and a display 403. Figure 4 Only some of the components of the ship monitoring device 400 are shown, but it should be understood that it is not required to implement all of the shown components, and more or fewer components may be implemented instead.

[0054] In some embodiments, the processor 401 may be a central processing unit (CPU), a microprocessor, or other data processing chip, configured to execute program codes or process data stored in the memory 402, such as the wear analysis method for a marine water-lubricated stern bearing according to the present invention.

[0055] In some embodiments, processor 401 may be a single server or a server group. The server group may be centralized or distributed. In some embodiments, processor 401 may be local or remote. In some embodiments, processor 401 may be implemented on a cloud platform. In one embodiment, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, multiple clouds, or any combination thereof.

[0056] In some embodiments, the memory 402 may be an internal storage unit of the ship monitoring device 400, such as a hard disk or memory of the ship monitoring device 400. In other embodiments, the memory 402 may be an external storage device of the ship monitoring device 400, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., equipped on the ship monitoring device 400.

[0057] Furthermore, the memory 402 may include both an internal storage unit of the ship monitoring device 400 and an external storage device. The memory 402 is used to store application software installed in the ship monitoring device 400 and various data.

[0058] In some embodiments, display 403 can be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. Display 403 is used to display information from vessel monitoring device 400 and to present a visual user interface. Components 401-403 of vessel monitoring device 400 communicate with each other via a system bus.

[0059] In one embodiment, when the processor 401 executes the marine water-lubricated stern bearing wear analysis program in the memory 402, the following steps may be implemented: Acquire multimodal data of a ship's water-lubricated stern bearing; Perform multi-dimensional feature extraction on multimodal data to obtain multi-dimensional feature data; Inputting the multi-dimensional feature data into a wear quantitative evaluation model to obtain a wear quantitative evaluation result of the marine water-lubricated stern bearing, and determining the operating state of the marine water-lubricated stern bearing based on the wear quantitative evaluation result; The wear quantitative assessment results and historical wear quantitative assessment results are input into the wear trend prediction model to obtain the predicted wear trend of the marine water-lubricated stern bearing in the future preset time period.

[0060] It should be understood that, when the processor 401 executes the marine water-lubricated stern bearing wear analysis method program in the memory 402 , in addition to the above functions, it can also implement other functions. For details, please refer to the description of the corresponding method embodiment above.

[0061] Furthermore, the embodiment of the present invention does not specifically limit the type of the ship monitoring device 400 mentioned. The ship monitoring device 400 can be a portable ship monitoring device such as a mobile phone, a tablet computer, a personal digital assistant (PDA), a wearable device, a laptop computer, or the like. Exemplary embodiments of portable ship monitoring devices include, but are not limited to, portable ship monitoring devices running iOS, Android, Microsoft, or other operating systems. The aforementioned portable ship monitoring device can also be other portable ship monitoring devices, such as a laptop computer with a touch-sensitive surface (e.g., a touch panel). It should also be understood that in some other embodiments of the present invention, the ship monitoring device 400 may not be a portable ship monitoring device, but rather a desktop computer with a touch-sensitive surface (e.g., a touch panel).

[0062] Accordingly, an embodiment of the present application also provides a computer-readable storage medium, which is used to store computer-readable programs or instructions. When the program or instructions are executed by a processor, the steps or functions of the marine water-lubricated stern bearing wear analysis method provided in the above-mentioned method embodiments can be implemented.

[0063] Those skilled in the art will appreciate that all or part of the process flow of the above-described method embodiment can be implemented by instructing related hardware (such as a processor, controller, etc.) through a computer program, and the computer program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a magnetic disk, an optical disk, a read-only memory, or a random access memory.

[0064] The above is a detailed introduction to the marine water-lubricated stern bearing wear analysis method, system and ship monitoring equipment provided by the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only intended to help understand the method and core ideas of the present invention. At the same time, for those skilled in the art, according to the ideas of the present invention, there may be changes in the specific implementation methods and application scopes. In summary, the contents of this specification should not be understood as limiting the present invention.

Claims

1. A method for analyzing wear of a water-lubricated stern bearing of a ship, characterized in that: include: Acquire multimodal data of a ship's water-lubricated stern bearing; Performing multi-dimensional feature extraction on the multimodal data to obtain multi-dimensional feature data; Inputting the multidimensional feature data into a wear quantitative evaluation model to obtain a wear quantitative evaluation result of the marine water-lubricated stern bearing, and determining an operating state of the marine water-lubricated stern bearing based on the wear quantitative evaluation result; The wear quantitative evaluation result and the historical wear quantitative evaluation result are input into a wear trend prediction model to obtain a predicted wear trend of the marine water-lubricated stern bearing within a future preset time period.

2. The wear analysis method for a marine water-lubricated stern bearing according to claim 1, characterized in that: The multimodal data includes real-time shaft speed, real-time torque, real-time axial vibration acceleration, real-time bearing liner strip thickness, real-time bearing specific pressure and real-time lubrication water temperature of a marine water-lubricated stern bearing.

3. The method for analyzing wear of a marine water-lubricated stern bearing according to claim 1 or 2, characterized in that: The performing multi-dimensional feature extraction on the multimodal data to obtain multi-dimensional feature data includes: A multi-level feature extraction strategy is used to extract time domain feature data, frequency domain feature data and time-frequency domain feature data corresponding to the multimodal data; The time domain feature data includes the mean, variance and peak factor of the multimodal data, the frequency domain feature data includes the main frequency component, frequency band energy distribution and power spectral density of the multimodal data, and the time-frequency domain feature data includes the energy distribution characteristics of the multimodal data at different time-frequency resolutions.

4. The wear analysis method for a marine water-lubricated stern bearing according to claim 1, characterized in that: Before performing multi-dimensional feature extraction on the multimodal data, the method further includes: Performing data synchronization, denoising and standardization on the multimodal data; Performing a data synchronization operation on the multimodal data includes: Using timestamp alignment technology to perform time consistency matching on the multimodal data; Performing denoising processing on the multimodal data, comprising: determining a target filtering algorithm based on characteristics of the multimodal data, and performing filtering processing on the multimodal data based on the target filtering algorithm; The multimodal data is subjected to standardization processing, including: A preset normalization method is used to perform scale consistency matching on the multimodal data.

5. The wear analysis method for a marine water-lubricated stern bearing according to claim 1, characterized in that: The wear quantitative evaluation model is an integrated model of support vector machine, deep neural network model and random forest model.

6. The wear analysis method for a marine water-lubricated stern bearing according to claim 1, characterized in that: The wear quantitative assessment result includes a wear degree score; and determining the operating state of the marine water-lubricated stern bearing based on the wear quantitative assessment result includes: Obtaining a correlation between the operating state and the multi-level warning threshold and a matching relationship between the wear degree score and the multi-level warning threshold; The operating status is determined based on the association relationship, the matching relationship, and the wear degree score.

7. The wear analysis method for a marine water-lubricated stern bearing according to claim 1, characterized in that: The wear trend prediction model includes a convolutional neural network for extracting local temporal features and a long short-term memory network for capturing the asymptotic changes in wear amount.

8. The wear analysis method for a marine water-lubricated stern bearing according to claim 1, characterized in that: The method further comprises: The wear quantitative evaluation results and the predicted wear trend are presented in a graphical form.

9. A marine water-lubricated stern bearing wear analysis system, characterized in that: include: A data acquisition module, used to acquire multimodal data of a marine water-lubricated stern bearing; A multi-dimensional feature extraction module, configured to extract multi-dimensional features from the multimodal data to obtain multi-dimensional feature data; a wear quantitative evaluation module, configured to input the multidimensional feature data into a wear quantitative evaluation model, obtain a wear quantitative evaluation result of the marine water-lubricated stern bearing, and determine an operating state of the marine water-lubricated stern bearing based on the wear quantitative evaluation result; The wear trend prediction module is used to input the wear quantitative evaluation result and the historical wear quantitative evaluation result into the wear trend prediction model to obtain the predicted wear trend of the marine water-lubricated stern bearing within a future preset time period.

10. A ship monitoring device, characterized in that: comprising a memory and a processor, wherein, The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps of the marine water-lubricated stern bearing wear analysis method according to any one of claims 1 to 8.

Images