A mooring line monitoring method and apparatus

Abstract

This invention relates to a method and apparatus for monitoring mooring lines, belonging to the field of ship safety technology. The method includes: acquiring image data, stress data, and environmental data of the mooring line; extracting features from the image data, stress data, and environmental data respectively to obtain image features, stress features, and environmental features; fusing the image features, stress features, and environmental features to obtain fused features; and inputting the fused features into a preset mooring line state classifier to obtain a state classification result for the mooring line. This invention, by fusing image data, stress data, and environmental data to monitor the state of the mooring line, can improve the accuracy of monitoring.

Description

Technical Field

[0001] This invention relates to the field of ship safety technology, and in particular to a method and device for monitoring mooring lines. Background Technology

[0002] As a crucial component for the safe berthing of ships, the operational status of mooring lines directly impacts the stability and safety of vessels during berthing. Current methods of mooring line safety monitoring largely rely on single-type sensors, such as stress sensors. While this method represents a significant improvement over manual inspection and provides continuous monitoring data, it struggles to accurately detect safety hazards caused by non-stress factors such as surface damage and wear of mooring lines in complex working environments. This limits the comprehensiveness and accuracy of the monitoring system. Summary of the Invention

[0003] In view of this, it is necessary to provide a method and apparatus for monitoring mooring lines to solve the problem that existing technologies using a single sensor are unable to accurately monitor the status of mooring lines.

[0004] To address the aforementioned problems, in a first aspect, the present invention provides a method for monitoring mooring cables, comprising:

[0005] Acquire image data, stress data, and environmental data of the mooring cable;

[0006] Feature extraction is performed on the image data, the stress data, and the environmental data respectively to obtain image features, stress features, and environmental features;

[0007] The image features, stress features, and environmental features are fused to obtain fused features;

[0008] The fused features are input into a preset mooring cable state classifier to obtain the state classification result of the mooring cable.

[0009] Optionally, the step of extracting features from the image data to obtain image features includes:

[0010] The image data is used to detect mooring lines using a trained YOLO model to obtain the mooring line image region.

[0011] The image features are obtained by extracting features from the mooring cable image region using a trained VGG model.

[0012] Optionally, the step of extracting features from the stress data to obtain stress features includes:

[0013] Perform a fast Fourier transform on the stress data to obtain the first stress data;

[0014] Wavelet transform is performed on the stress data to obtain second stress data at different time scales;

[0015] Stress features are generated based on the first stress data and the second stress data.

[0016] Optionally, acquiring the environmental data of the mooring cable includes:

[0017] First environmental data of the mooring cable is acquired at a first preset time interval; wherein, the first environmental data includes one of wind speed, wind direction and wave height;

[0018] Second environmental data of the mooring cable is acquired according to a second preset time interval; wherein the second environmental data includes at least one of air pressure, humidity and seawater temperature; and the second preset time interval is greater than the first preset time interval.

[0019] Optionally, the feature fusion of the image features, stress features, and environmental features to obtain fused features includes:

[0020] The image features, stress features, and environmental features are input into a multilayer perceptron to obtain the attention weights of the image features, stress features, and environmental features;

[0021] Based on the attention weights of the image features, stress features, and environmental features, the image features, stress features, and environmental features are weighted and fused to obtain fused features.

[0022] Optionally, the method further includes:

[0023] The wear condition of the mooring cable is determined based on the image features;

[0024] Determine whether the attention weight of the image feature exceeds the image attention weight threshold;

[0025] When the attention weight of the image feature exceeds the image attention weight threshold, a stress alarm threshold is determined based on the wear condition;

[0026] An alarm message is generated when the stress data exceeds the stress alarm threshold.

[0027] Optionally, the method further includes:

[0028] Determine whether the attention weight of the environmental data exceeds the environmental attention weight threshold;

[0029] When the attention weight of the environmental data exceeds the environmental attention weight threshold, a stress alarm threshold is determined based on the environmental data.

[0030] An alarm message is generated when the stress data exceeds the stress alarm threshold.

[0031] Optionally, the state classification result includes a health state and the probability corresponding to the health state; different health states correspond to different probability thresholds; the method further includes:

[0032] Based on the health status output by the mooring cable status classifier, determine the target probability threshold corresponding to the health status;

[0033] An alarm message is generated when the probability output by the mooring cable status classifier is greater than the target probability threshold.

[0034] Optionally, the method further includes:

[0035] Based on the material properties of the mooring cables and the opinions of experts in the field, probability thresholds are set for each health state.

[0036] Secondly, the present invention also provides a mooring cable monitoring device, comprising:

[0037] The data acquisition module is used to acquire image data, stress data, and environmental data of the mooring cable.

[0038] The feature extraction module is used to extract features from the image data, the stress data, and the environmental data respectively to obtain image features, stress features, and environmental features.

[0039] The feature fusion module is used to fuse the image features, stress features, and environmental features to obtain fused features.

[0040] The state determination module is used to input the fused features into a preset mooring cable state classifier to obtain the state classification result of the mooring cable.

[0041] The beneficial effects of this invention are:

[0042] This invention improves the acquisition of image data, stress data, and environmental data of mooring lines. It extracts features from each of these data separately, obtaining image features, stress features, and environmental features. These features are then fused to obtain fused features. The fused features are input into a pre-defined mooring line state classifier to obtain a state classification result for the mooring line. By fusing image data, stress data, and environmental data to monitor the state of the mooring line, it can accurately detect safety hazards caused by non-stress factors such as surface damage and wear, providing comprehensive monitoring and improving monitoring accuracy. Attached Figure Description

[0043] Figure 1 A flowchart illustrating an embodiment of the mooring cable monitoring method provided by the present invention;

[0044] Figure 2 A data acquisition flowchart provided by the present invention;

[0045] Figure 3 A framework diagram for monitoring mooring cables provided by the present invention;

[0046] Figure 4 A data fusion flowchart provided by the present invention;

[0047] Figure 5 This invention provides another framework diagram for monitoring mooring cables;

[0048] Figure 6 This is a schematic diagram of one embodiment of the mooring cable monitoring device provided by the present invention. Detailed Implementation

[0049] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0050] In the description of the embodiments of this invention, unless otherwise stated, "a plurality of" means two or more. The terms "first," "second," etc., used in the embodiments of this invention are used to distinguish similar objects, and are not used to describe a specific order or sequence, nor to indicate or imply their relative importance or implicitly specify the number of indicated technical features. It should be understood that such data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class, and the number of objects is not limited; for example, a first object can be one or more.

[0051] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0052] Reference Figure 1The diagram illustrates a flowchart of an embodiment of the mooring cable monitoring method provided by the present invention, the method comprising:

[0053] S101, acquire image data, stress data, and environmental data of the mooring cable.

[0054] Regarding image data acquisition, cameras can be installed at multiple key locations on the vessel, such as the bow, stern, both sides of the ship, and above the mooring points, ensuring 360-degree coverage and real-time acquisition of image data of the mooring lines from all directions. Specifically, cameras with night vision and infrared capabilities should be selected to ensure clear image capture of the mooring lines even in low-visibility environments. Furthermore, the cameras should be waterproof, dustproof, and corrosion-resistant, with an enclosure protection rating of at least IP67 to withstand the salt spray corrosion and frequent humidity changes of the marine environment. Next, a dedicated bracket should be used to secure the camera, ensuring stable installation and preventing the ship's movement from affecting image quality. The camera angle should be adjusted to ensure the mooring lines are centered in the camera's view for subsequent image analysis. Then, the cameras should be connected to the ship's internal network to ensure real-time transmission of image data via wired or wireless means. For cameras located far from the main network, wireless transmission technologies such as Wi-Fi or LoRaWan can be used to ensure stable signal transmission. Finally, the camera is controlled to capture at least 30 frames per second to capture the dynamic changes of the cable, ensuring clear images are obtained even when the cable is moving rapidly or the ship is rocking.

[0055] After acquiring the image data, a portion of it can be stored locally in case of network interruption. Simultaneously, the image data is transmitted in real-time to the data processing center via the ship's internal network or satellite communication to ensure data real-time performance and continuity.

[0056] In addition, the system can monitor the camera's online status, storage space, and network connectivity in real time, ensuring the stability and continuity of the image data acquisition process. Regularly cleaning salt and dirt from the lens and checking the camera housing's seal ensures long-term stable operation of the equipment.

[0057] Regarding stress data acquisition, stress sensors can be installed at multiple key locations on the mooring line, such as the fixed end of the mooring line, the contact point between the mooring line and the mooring facility, and potential wear areas on the mooring line, to acquire stress data in real time. Specifically, firstly, high-precision strain gauges or pressure sensors with good temperature stability, electromagnetic interference resistance, and long-term stability can be selected as stress sensors. Then, the stress sensors are calibrated to ensure their measurement accuracy meets requirements. During calibration, the sensitivity and zero-point drift of the stress sensors are recorded for compensation during later data analysis. Next, using specialized fixing clamps and high-viscosity, weather-resistant adhesives, the stress sensors are firmly attached to the surface of the mooring line to prevent displacement or damage caused by vibration or impact from ship movement. Finally, preliminary performance tests are conducted on the stress sensors, including static and dynamic response tests, to verify their stability and accuracy under actual working conditions. Once the performance tests are passed, the pressure sensors can be put into formal use.

[0058] Regarding the acquisition of environmental data, real-time environmental data of the port where the ship is located can be obtained through ship positioning. Alternatively, environmental data of the mooring lines can be obtained through equipment such as temperature and humidity sensors and wind direction monitors installed on the ship. Environmental data can include: wind speed and direction (affecting the wind pressure on the mooring lines), air pressure (affecting the atmospheric environment), humidity (affecting the performance of mooring line materials), seawater temperature (affecting seawater density and the physical properties of the mooring lines), wave height and period (directly affecting the dynamic stress on the mooring lines), etc.

[0059] Furthermore, the update frequency of environmental data can be reasonably set according to the required monitoring accuracy of the mooring line and the data processing capability of the mooring line health device. For example, first environmental data of the mooring line can be acquired at a first preset time interval; the first environmental data may include one of wind speed, wind direction, and wave height; then, second environmental data of the mooring line can be acquired at a second preset time interval; the second environmental data may include at least one of air pressure, humidity, and seawater temperature; the second preset time interval is longer than the first preset time interval.

[0060] Reference Figure 2 This diagram illustrates a data acquisition flowchart provided by the present invention. Image data is acquired through an image acquisition device, stress data is acquired through a stress sensor, and a connection is established with a meteorological service provider via an API to automatically request and receive real-time meteorological and environmental data from the port where the ship is located.

[0061] S102, feature extraction is performed on image data, stress data and environmental data respectively to obtain image features, stress features and environmental features.

[0062] S103 performs feature fusion on image features, stress features, and environmental features to obtain fused features.

[0063] Based on the attention mechanism, feature fusion can be performed on image features, stress features, and environmental features to obtain fused features.

[0064] S104. Input the fused features into the preset mooring cable state classifier to obtain the mooring cable state classification result.

[0065] A classifier is a model that can classify data based on its input features. The input to a mooring line condition classifier can be the aforementioned fused features, and the output can be the health status of the mooring line and the probability of that health status. Health status can include healthy, slightly worn, severely worn, and dangerous.

[0066] This embodiment monitors the condition of mooring lines by fusing image data, stress data, and environmental data. It can accurately capture safety hazards caused by non-stress factors such as surface damage and wear of the mooring lines, providing comprehensive monitoring and improving the accuracy of the monitoring.

[0067] Reference Figure 3 The diagram illustrates a mooring cable monitoring framework provided by this invention. After acquiring image data, stress data, and environmental data, features are extracted and fused from these three types of data to obtain fused features. The health status of the mooring cable is then identified based on these fused features.

[0068] In one embodiment, after acquiring image data, stress data, and environmental data, these three types of data can be preprocessed separately, and then feature extraction can be performed on these three types of data after the preprocessing is completed.

[0069] Specifically, image data preprocessing includes: first, denoising the acquired image, for example, using Gaussian filtering. Then, the image is uniformly resized to a specific size, such as 256×256 pixels, and normalized to ensure that the image pixel values ​​are distributed within a suitable range, reducing differences caused by factors such as lighting.

[0070] Stress data preprocessing includes: firstly, denoising the stress data, such as using a low-pass filter to remove high-frequency noise; then, standardizing the stress data to conform to a standard normal distribution for easier subsequent analysis.

[0071] Environmental data preprocessing includes cleaning the environmental data and removing outliers.

[0072] In one embodiment, the step of extracting features from image data to obtain image features may include: detecting mooring lines in the image data using a trained YOLO model to obtain a mooring line image region; and extracting features from the mooring line image region using a trained VGG model to obtain image features.

[0073] The YOLO model can be the YOLO v8 model. As a highly efficient object detection model, YOLO v8 can comprehensively analyze images and quickly and accurately locate the position of mooring cables within an image. The VGG model can be the VGG16 model. The VGG16 model consists of multiple convolutional layers, pooling layers, and fully connected layers. In the convolutional layers, convolution operations are performed by sliding the convolution kernel across the image region, extracting features at different levels layer by layer, from simple low-level features such as edges and textures to more abstract features of the mooring cable structure (such as diameter and weave pattern) and damage features (such as wear and cracks). Then, after downsampling by the pooling layer, the feature size is compressed while the main features are preserved. Finally, the extracted features are synthesized in the fully connected layer to obtain a fixed-length feature vector, which is the image feature vector. (dimension is) ).

[0074] In one embodiment, the step of extracting stress features from stress data may include: performing a fast Fourier transform on the stress data to obtain first stress data; performing a wavelet transform on the stress data to obtain second stress data at different time scales; and generating stress features based on the first stress data and the second stress data.

[0075] The raw stress data is subject to environmental interference and noise during acquisition, so denoising is the first step. A common denoising method is to use a low-pass filter, which allows low-frequency signals to pass through while effectively suppressing high-frequency noise, making the stress data smoother and more accurate. After denoising, spectral analysis can be performed on the stress data, with the Fast Fourier Transform (FFT) being a key operation. The FFT transforms the denoised stress data from the time domain to the frequency domain, obtaining the first stress data. By performing frequency domain analysis on the first stress data, the frequency components can be clearly identified. For example, frequency peaks can be found, which are often related to the vibration frequency of the cable under specific working conditions or the external excitation frequency. Furthermore, wavelet transform can be performed. By selecting appropriate wavelet basis functions, the stress data can be decomposed into components with different frequencies and time resolutions, i.e., second stress data. Information reflecting the variation characteristics of the second stress data can be extracted from it, such as the stress fluctuation at different time scales (maximum, minimum, and mean values ​​of stress data at different time scales) and stress abrupt change points. The identified frequency components and stress fluctuations at different time scales are then processed to obtain the stress feature vector. (dimension is) )

[0076] In one embodiment, an LSTM neural network can also be used to extract features from the denoised stress data to generate stress features.

[0077] In one embodiment, the step of extracting environmental features from environmental data may include: First, due to potential sensor malfunctions or interference, the collected environmental data may contain outliers and missing values, necessitating data cleaning. For example, outliers can be detected and removed using statistical outlier detection algorithms, and missing values ​​can be filled using linear interpolation to ensure the integrity and reliability of the environmental data. Then, based on domain knowledge, key variables are selected from the cleaned environmental data, such as wind speed and wave height, which directly affect the stress on mooring cables, and temperature and humidity, which affect cable material properties (e.g., corrosion rate). For these key variables, feature engineering is performed based on physical principles and practical experience. For example, the combined force of wind and waves on the cable can be represented by calculating the product of wind speed and wave height, or the ratio of temperature to humidity can be calculated to reflect the combined environmental impact on cable corrosion. Finally, the product of wind speed and wave height and the ratio of temperature to humidity obtained after the above processing are used as environmental feature vectors. (dimension is) ).

[0078] In one embodiment, after obtaining image features, stress features, and environmental features, these three features can be stitched together to obtain stitched features. The dimension of the splicing feature is Then, the splicing features can be... In a Multilayer Perceptron (MLP), the input has two layers. The first layer uses ReLU activation, and the second layer has an output dimension of 3 (corresponding to the image, stress, and environment modalities). Assume the second layer of the MLP calculates... Then, the attention weights for image features, stress features, and environmental features can be calculated using the softmax function:

[0079] Attention weights for image features: .

[0080] Attention weights for stress features: .

[0081] Attention weights for environmental features: .

[0082] By applying attention weights to image features, stress features, and environmental features, a weighted fusion of these features can be obtained. .

[0083] In one embodiment, the mooring line status classification result output by the classifier includes the health status of the mooring line and the probability corresponding to the health status. The health status includes healthy, relatively healthy, slightly worn, severely worn, dangerous, etc., and different health statuses correspond to different probability thresholds. The mooring line monitoring method further includes: determining a target probability threshold corresponding to the health status based on the health status output by the mooring line status classifier; and generating an alarm message when the probability output by the mooring line status classifier is greater than the target probability threshold.

[0084] For example, when the classifier outputs the status of the mooring cable as severely worn, an alarm mechanism is triggered and an alarm message is generated if the probability of severe wear exceeds 50%; when the classifier outputs the status of the mooring cable as dangerous, an alarm mechanism is triggered and an alarm message is generated if the probability of danger exceeds 30%.

[0085] The probability thresholds for each health status can be set based on the material properties of the mooring cable and in conjunction with the opinions of experts in the field.

[0086] In one embodiment, in addition to determining whether to trigger the alarm mechanism based on the classifier's output, the determination can also be based on the results of auxiliary judgments. Specifically, it can also: determine the wear condition of the mooring cable based on image features; determine whether the attention weight of the image features exceeds the image attention weight threshold; when the attention weight of the image features exceeds the image attention weight threshold, determine a stress alarm threshold based on the wear condition; when the stress data exceeds the stress alarm threshold, generate alarm information.

[0087] For example, if the attention weights of image features Attention weights for stress features Both are relatively high (e.g., >0.3 and The threshold is >0.3), and the classifier outputs a high probability of cable wear or abnormalities. The focus is on whether the fiber breakage and stress data of the mooring cable, identified based on image data, are close to or exceed the rated load-bearing capacity. For example, if image analysis shows that more than 3% of the cable's fibers are broken and the stress data exceeds 70% of the rated load-bearing capacity, an early warning mechanism is triggered.

[0088] In one embodiment, the auxiliary judgment further includes: determining whether the attention weight of the environmental data exceeds the environmental attention weight threshold; when the attention weight of the environmental data exceeds the environmental attention weight threshold, determining the stress alarm threshold based on the environmental data; and when the stress data exceeds the stress alarm threshold, generating alarm information.

[0089] For example, when the attention weight of environmental features is high (e.g., When the stress level is greater than 0.3, environmental data is used to assist in assessing the health status of the mooring lines. A linear or nonlinear regression model is established between mooring line stress and environmental factors (such as wind speed and wave height) through historical data analysis. For example, if the wind speed exceeds 15 m / s, and the regression model predicts a 20% increase in line stress, then if the stress data exceeds 80% of the rated load capacity at this point, an early warning mechanism is triggered.

[0090] In one embodiment, the auxiliary judgment further includes: setting different stress level thresholds based on the material properties of the mooring cable and historical stress data, combined with the opinions of experts in the field. For example, the stress level threshold can be set to 80% of the rated load capacity for a relatively healthy condition, and to 40% of the rated load capacity for a dangerous condition.

[0091] Reference Figure 4 This diagram illustrates a data fusion flowchart provided by the present invention. First, data fusion rules are defined, which are also auxiliary judgment rules. Then, multimodal data is fused, or correlated, based on the fusion rules.

[0092] In one embodiment, the training process for the classifier, YOLO model, and VGG model may include the following steps:

[0093] ① Dataset partitioning: The collected labeled data (including images, stress, and environmental data under normal and abnormal conditions) are randomly partitioned to ensure consistent data distribution. Typically, the training set accounts for 70%-80%, while the validation and test sets each account for 10%-15%.

[0094] ② Initialization Parameters: Use weight initialization based on the pre-trained model. When using weight initialization based on the pre-trained model, for example, when processing image data using a model pre-trained on a large image dataset (such as ImageNet) (VGG16), the weights of the pre-trained model are loaded into the current model. If the current model structure is not exactly the same as the pre-trained model, for example, the current model may only need some layers from the pre-trained model (such as convolutional layers), then the corresponding layer weights can be selected for initialization. For newly added layers (such as custom fully connected layers), the weights of these layers can be initialized randomly.

[0095] ③ Choose an appropriate loss function (such as cross-entropy loss) and optimization algorithm (such as Adam), and set the initial learning rate and decay strategy. The cross-entropy loss function has a natural advantage in handling classification problems. It effectively measures the difference between the probability distribution predicted by the model and the true class. The higher the model's predicted probability for the correct class, the smaller the cross-entropy loss value, and vice versa. This helps guide the model to continuously adjust parameters during training, improving the accuracy of classifying different health states. The Adam optimization algorithm combines the advantages of momentum and RMSProp, and can adaptively adjust the learning rate. When dealing with complex neural network models, the update frequency of different parameters may differ. Adam can automatically adjust the learning rate of each parameter based on the gradient history information, making the training process more stable and efficient. In the model training for cable health status assessment, since the model may be complex (involving multimodal data fusion, multi-layer neural networks, etc.), the Adam optimization algorithm can, to some extent, avoid getting trapped in local optima, accelerate the model's convergence speed, and thus find better model parameters more quickly.

[0096] ④ Hyperparameter Tuning: Using techniques such as Bayesian optimization, hyperparameters such as learning rate, batch size, and regularization coefficient are adjusted to find the optimal configuration. Bayesian optimization is a model-based hyperparameter tuning method that uses a probabilistic model to guide the hyperparameter search. The advantage of Bayesian optimization is that it can find better hyperparameter combinations with fewer evaluation iterations, making it particularly suitable for situations where hyperparameter evaluation is costly (e.g., long training time).

[0097] ⑤ Model Evaluation: Use the test set to evaluate model performance, focusing on metrics such as accuracy, recall, and F1 score to ensure the model's generalization ability and stability.

[0098] Once trained, the model can be deployed to the production environment to monitor the health of the mooring lines in real time. Model performance can be checked periodically, and retraining or optimization can be performed as needed.

[0099] In one embodiment, alarm information can be disseminated to all crew members via onboard radio, intercom, or a dedicated alarm system. Simultaneously, satellite communication or the internet can be used to send emails or text messages to shore-based operations centers, maintenance teams, and management, ensuring that relevant shore-based parties receive the information synchronously. Maritime safety applications can also be developed or integrated to push real-time alerts to designated personnel via mobile phones or tablets, ensuring rapid information dissemination. Alarm information can clearly indicate the severity of the alarm for rapid identification and response; provide the specific location information of mooring lines for on-site inspection and maintenance; and outline preliminary response strategies, such as reducing speed, changing course, and activating backup mooring lines.

[0100] In one embodiment, an alarm command can be automatically issued upon receiving a hazard warning. In extreme risk situations, contact with the port terminal management office is automatically initiated to request immediate assistance. Upon triggering the alarm, an emergency response team is immediately activated, organizing professionals to conduct a comprehensive inspection and assessment of the cables. Based on a pre-established emergency manual, appropriate cable replacement, reinforcement, or temporary repair work is carried out. A command center is established to collect on-site information, maintain close communication with decision-makers, and ensure that all actions receive high-level approval and support.

[0101] Reference Figure 5 This diagram illustrates another mooring cable monitoring framework provided by the present invention. Image data, stress data, and environmental data are collected. These three types of data are preprocessed separately. After preprocessing, feature extraction and feature fusion are performed on these three types of data. The health status of the mooring cable is identified based on the fused features. Whether to trigger an alarm mechanism is determined based on auxiliary judgment rules and the mooring cable status identification results.

[0102] In summary, the present invention has the following beneficial effects:

[0103] ① Significantly improves monitoring efficiency and accuracy. Traditional manual inspection methods are limited by manpower and time, making continuous monitoring impossible. This invention, by integrating various sensors and video surveillance, can collect mooring cable status information 24 / 7, ensuring data real-time performance and completeness. Especially under complex environmental conditions, such as at night, in severe weather, or when the cable is under high stress, multimodal data fusion technology can integrate information from various aspects, providing more comprehensive analysis results. This avoids misjudgments that may arise from a single data source, greatly improving the accuracy and reliability of monitoring.

[0104] ② This invention enhances the intelligence level of mooring line condition assessment. By applying deep learning and machine learning algorithms, it can automatically identify surface damage, wear levels, and stress change trends of the mooring lines, and even predict potential failure risks. This intelligent assessment method not only saves labor costs but also provides early warnings, giving port management personnel ample time to take preventative measures, effectively avoiding safety accidents caused by mooring line failures, and greatly improving the safety of port operations.

[0105] ③ It promotes more refined and predictable maintenance work. Traditional maintenance strategies are often based on experience or periodic inspections, while this invention, through continuous monitoring and intelligent analysis, can grasp the health status of mooring lines in real time, helping port managers to develop more scientific and reasonable maintenance plans. For example, when the stress on the mooring line exceeds the safety threshold or surface damage reaches a certain level, it can automatically trigger a maintenance reminder, guiding staff to carry out timely repairs or replacements, avoiding over-maintenance or under-maintenance, thereby reducing maintenance costs and extending the service life of the mooring lines.

[0106] ④ It provides strong support for the digital transformation of port operations. By integrating data from various sensors, a digital twin model of the mooring line status was constructed, providing a rich and accurate data foundation for the port's information management. This data can be used not only for real-time monitoring but also for historical data analysis, failure mode research, and future trend prediction, providing valuable decision-making support for port operation and management and helping the port develop towards intelligence and efficiency.

[0107] Reference Figure 6 The diagram shows a structural schematic of an embodiment of the mooring cable monitoring device provided by the present invention. The device 60 includes:

[0108] The data acquisition module 601 is used to acquire image data, stress data, and environmental data of the mooring cable.

[0109] The feature extraction module 602 is used to extract features from image data, stress data and environmental data respectively to obtain image features, stress features and environmental features;

[0110] The feature fusion module 603 is used to fuse image features, stress features, and environmental features to obtain fused features.

[0111] The state determination module 604 is used to input the fused features into a preset mooring cable state classifier to obtain the state classification result of the mooring cable.

[0112] It should be noted that the implementation principles or processes of the above modules can be referred to the aforementioned implementation examples of the mooring cable monitoring method, and will not be elaborated here.

[0113] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0114] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for monitoring mooring cables, characterized in that, include: Acquire image data, stress data, and environmental data of the mooring cable; Feature extraction is performed on the image data, the stress data, and the environmental data respectively to obtain image features, stress features, and environmental features; The image features, stress features, and environmental features are fused to obtain fused features; The fused features are input into a preset mooring cable state classifier to obtain the state classification result of the mooring cable; The step of extracting features from the stress data to obtain stress features includes: Perform a fast Fourier transform on the stress data to obtain the first stress data; Wavelet transform is performed on the stress data to obtain second stress data at different time scales; Based on the first stress data and the second stress data, stress characteristics are generated; The feature fusion of the image features, stress features, and environmental features to obtain fused features includes: The image features, stress features, and environmental features are input into a multilayer perceptron to obtain the attention weights of the image features, stress features, and environmental features; Based on the attention weights of the image features, stress features, and environmental features, the image features, stress features, and environmental features are weighted and fused to obtain fused features; The method further includes: The wear condition of the mooring cable is determined based on the image features; Determine whether the attention weight of the image feature exceeds the image attention weight threshold; When the attention weight of the image feature exceeds the image attention weight threshold, a stress alarm threshold is determined based on the wear condition; When the stress data exceeds the stress alarm threshold, an alarm message is generated; The method further includes: Determine whether the attention weight of the environmental data exceeds the environmental attention weight threshold; When the attention weight of the environmental data exceeds the environmental attention weight threshold, a stress alarm threshold is determined based on the environmental data. An alarm message is generated when the stress data exceeds the stress alarm threshold.

2. The mooring cable monitoring method according to claim 1, characterized in that, The step of extracting features from the image data to obtain image features includes: The image data is used to detect mooring lines using a trained YOLO model to obtain the mooring line image region. The image features are obtained by extracting features from the mooring cable image region using a trained VGG model.

3. The mooring cable monitoring method according to claim 1, characterized in that, The acquisition of environmental data of the mooring cable includes: First environmental data of the mooring cable is acquired at a first preset time interval; wherein, the first environmental data includes one of wind speed, wind direction and wave height; Second environmental data of the mooring cable is acquired according to a second preset time interval; wherein the second environmental data includes at least one of air pressure, humidity and seawater temperature; and the second preset time interval is greater than the first preset time interval.

4. The mooring cable monitoring method according to claim 1, characterized in that, The state classification result includes a health status and the probability corresponding to the health status; Different health states correspond to different probability thresholds; the method further includes: Based on the health status output by the mooring cable status classifier, determine the target probability threshold corresponding to the health status; An alarm message is generated when the probability output by the mooring cable status classifier is greater than the target probability threshold.

5. The mooring cable monitoring method according to claim 1, characterized in that, The method further includes: Based on the material properties of the mooring cables and the opinions of experts in the field, probability thresholds are set for each health state.

6. A mooring cable monitoring device, characterized in that, include: The data acquisition module is used to acquire image data, stress data, and environmental data of the mooring cable. The feature extraction module is used to extract features from the image data, the stress data, and the environmental data respectively to obtain image features, stress features, and environmental features. The feature fusion module is used to fuse the image features, stress features, and environmental features to obtain fused features. The state determination module is used to input the fused features into a preset mooring cable state classifier to obtain the state classification result of the mooring cable; The step of extracting features from the stress data to obtain stress features includes: Perform a fast Fourier transform on the stress data to obtain the first stress data; Wavelet transform is performed on the stress data to obtain second stress data at different time scales; Based on the first stress data and the second stress data, stress characteristics are generated; The feature fusion of the image features, stress features, and environmental features to obtain fused features includes: The image features, stress features, and environmental features are input into a multilayer perceptron to obtain the attention weights of the image features, stress features, and environmental features; Based on the attention weights of the image features, stress features, and environmental features, the image features, stress features, and environmental features are weighted and fused to obtain fused features; The device is also used for: The wear condition of the mooring cable is determined based on the image features; Determine whether the attention weight of the image feature exceeds the image attention weight threshold; When the attention weight of the image feature exceeds the image attention weight threshold, a stress alarm threshold is determined based on the wear condition; When the stress data exceeds the stress alarm threshold, an alarm message is generated; The device is also used for: Determine whether the attention weight of the environmental data exceeds the environmental attention weight threshold; When the attention weight of the environmental data exceeds the environmental attention weight threshold, a stress alarm threshold is determined based on the environmental data. An alarm message is generated when the stress data exceeds the stress alarm threshold.

Images