Method for monitoring abnormal running state of software component of ship control system

Abstract

The invention discloses a ship control system software component operation state abnormity monitoring method comprising the following steps: obtaining a software component of a ship control system, determining a plurality of to-be-monitored parameter characteristics, and forming a characteristic description set; multiple business scenes are simulated to obtain learning data, the business scenes comprise normal operation scenes, abnormal scenes and failure scenes with different load proportions, and characteristic values of parameters in the characteristic description set in the scenes are recorded; intercepting characteristic values in a time interval to form a characteristic description sequence of the software component; performing training and fine tuning on the artificial intelligence large model by using the feature description sequence; and performing monitoring and abnormality judgment on the real-time operation state of the ship control system software component by using the trained and fine-tuned artificial intelligence large model. According to the method provided by the embodiment of the invention, accurate distinguishing and intelligent judgment of the operation state of the software component of the ship control system can be realized, and the automation level and judgment reliability of state monitoring are improved.

Description

Technical Field

[0001] This application relates to the field of computer software technology, and more specifically, to a method for monitoring abnormal operating status of software components in a ship control system. Background Technology

[0002] With the continuous improvement of ship automation and intelligence, ship control systems are becoming increasingly complex, undertaking real-time control of many critical tasks such as ship navigation, management, and safety. To enhance the independence, reusability, and scalability of software development, software componentization technology has become an important implementation method for advanced systems such as ship distributed control systems and service-oriented architectures. As reusable units encapsulating specific functions, the health of software components directly affects the stability and reliability of the entire control system. Therefore, effective and accurate monitoring of the operational status of various software components in ship control systems is a crucial technical foundation for ensuring the safe and efficient operation of ships and is an urgent practical need.

[0003] Currently, the methods for monitoring the operational status of software components in ship control systems are relatively limited and outdated. Existing technologies mainly rely on basic system tools such as the process and task managers built into the operating system. These tools typically only provide very limited static indicators such as process existence and CPU and memory usage. However, in actual operation, ship control systems may experience brief, intermittent bursts of high workload due to handling unexpected tasks. This is a normal high-load phenomenon, but its high resource consumption characteristics are very similar to genuine software anomalies. Existing technologies lack the ability to comprehensively analyze multi-dimensional, time-series characteristic parameters, making it impossible to effectively distinguish between this normal high load and genuine anomalies. This results in high false alarm and false negative rates, potentially overlooking real safety hazards or causing confusion for operators due to frequent false alarms, severely compromising the accuracy and practicality of monitoring. Summary of the Invention

[0004] In view of at least one defect or improvement need in the prior art, the present invention provides a method for monitoring the abnormal operation status of ship control system software components, which can solve at least one of the problems existing in the background art.

[0005] To achieve the above objectives, according to a first aspect of the present invention, a method for monitoring abnormal operating status of ship control system software components is provided, comprising the following steps: The software components of the ship control system are acquired, and multiple monitoring parameter features corresponding to the operating status of the software components are determined to form a feature description set. Run the software component to simulate various business scenarios to obtain learning data. The business scenarios include normal operation scenarios, abnormal scenarios, and failure scenarios with different load ratios. Record the feature values ​​of the parameters in the feature description set under each scenario. The feature values ​​within a certain time interval are extracted to form the feature description sequence of the software component; The aforementioned feature description sequences are used to train and fine-tune large-scale artificial intelligence models; The trained and finely tuned large-scale artificial intelligence model is used to monitor the real-time operating status and identify anomalies of the ship control system software components.

[0006] Furthermore, in the above-mentioned method for monitoring abnormal operating status of ship control system software components, the characteristics of the parameters to be monitored include at least one of system resource parameters, network performance parameters, and service quality parameters.

[0007] Furthermore, in the above-mentioned method for monitoring abnormal operating status of ship control system software components, the system resource parameters include CPU utilization, memory usage, process switching frequency, CPU interrupt frequency, process physical memory, process virtual memory, disk throughput, and process disk I / O usage. The network performance parameters include data latency, network packet loss rate, and process bandwidth utilization. The service quality parameters include service call success rate, service response time, number of logs, and log addition frequency.

[0008] Furthermore, the above-mentioned method for monitoring abnormal operating status of ship control system software components, specifically includes the following: Training and fine-tuning the large-scale artificial intelligence model using the feature description sequence. The feature description sequence is divided into a training set and a validation set according to a certain ratio; The large artificial intelligence model is initially trained using the training set. During training, a feedback reinforcement learning mechanism is introduced to correct and guide the model's output, thereby fine-tuning the model. The fine-tuned model is validated using the validation set. Training is complete when the validation accuracy reaches a preset threshold.

[0009] Furthermore, in the above-mentioned method for monitoring abnormal operating status of ship control system software components, the software components include signal acquisition software components, including signal filtering, preprocessing, and signal conversion software components, and operation monitoring software components, including curve display, parameter display, and data refresh software components.

[0010] Furthermore, in the above-mentioned method for monitoring abnormal operating status of ship control system software components, the abnormal scenarios include communication abnormalities, input / output abnormalities, and display abnormalities.

[0011] According to a second aspect of the present invention, a device for monitoring abnormal operating status of ship control system software components is also provided, comprising: The component acquisition module is used to acquire the software components of the ship control system, determine multiple monitoring parameter features corresponding to the operating status of the software components, and form a feature description set. The scenario simulation module is used to run the software component and simulate various business scenarios to obtain learning data. The business scenarios include normal operation scenarios, abnormal scenarios and failure scenarios with different load ratios, and record the feature values ​​of the parameters in the feature description set under each scenario. The feature sequence generation module is used to extract the feature values ​​within a time interval to form the feature description sequence of the software component. The large model training module is used to train and fine-tune the large artificial intelligence model using the feature description sequence; The monitoring module is used to monitor the real-time operating status and identify anomalies of the ship control system software components using a trained and fine-tuned large artificial intelligence model.

[0012] According to a third aspect of the present invention, a device for monitoring abnormal operating status of a ship control system software component is also provided, comprising at least one processing unit and at least one storage unit, wherein the storage unit stores a computer program, and when the computer program is executed by the processing unit, the processing unit performs the steps of any of the methods described above.

[0013] According to a fourth aspect of the present invention, a storage medium is also provided, which stores a computer program executable by a ship control system software component operation status anomaly monitoring device, wherein when the computer program is run on the ship control system software component operation status anomaly monitoring device, the ship control system software component operation status anomaly monitoring device performs the steps of any of the above-described methods.

[0014] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects: This application provides a method for monitoring the abnormal operating status of ship control system software components. By selecting multi-dimensional feature parameters that can comprehensively reflect the operating status of the software components, a temporal feature description sequence is constructed. An artificial intelligence large model is used to perform deep training and learning on massive sequence data covering normal, abnormal, and failure scenarios with different load ratios. Finally, the mature large model is applied to real-time data stream analysis, which can accurately distinguish and intelligently judge the operating status of ship control system software components, especially the normal state of high load such as intermittent sudden busy, from the real abnormal state. This effectively overcomes the technical bottleneck of low accuracy and high false alarm rate when relying on the operating system task manager for monitoring. It significantly improves the automation level and reliability of status monitoring and can ensure the stable and safe operation of the ship control system. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart illustrating a method for monitoring abnormal operating status of a ship control system software component, as provided in an embodiment of this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0018] The terms "first," "second," "third," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0019] Figure 1 This application provides a flowchart illustrating a method for monitoring abnormal operating status of a ship control system software component, as shown in the embodiments below. Figure 1As shown in the embodiment of this application, a method for monitoring abnormal operating status of a ship control system software component includes the following steps: The software components of the ship control system are acquired, and multiple monitoring parameter features corresponding to the operating status of the software components are determined to form a feature description set. Run the software component to simulate various business scenarios to obtain learning data. The business scenarios include normal operation scenarios, abnormal scenarios, and failure scenarios with different load ratios. Record the feature values ​​of the parameters in the feature description set under each scenario. The feature values ​​within a certain time interval are extracted to form the feature description sequence of the software component; The aforementioned feature description sequences are used to train and fine-tune large-scale artificial intelligence models; The trained and finely tuned large-scale artificial intelligence model is used to monitor the real-time operating status and identify anomalies of the ship control system software components.

[0020] Specifically, the first step is to clearly define the scope of software components to be monitored, which should cover the key functional modules of the ship's control system. Then, a feature set that comprehensively reflects the operational status of these components needs to be constructed. This feature set should include parameters across multiple dimensions, such as system resource usage, network communication status, and service response performance. The selection of these parameters must consider their correlation with the component's operational status, monitorability, and real-time requirements to ensure accurate capture of changes in the component's health status.

[0021] To train an AI model with strong generalization capabilities, training data covering various operating conditions is required. This involves systematically collecting operational data of software components under different load conditions by simulating various working scenarios that a ship's control system might encounter. Specifically, this includes normal operating states from low to high load, various typical abnormal operating conditions, and severe fault states. For each scenario, it is necessary to completely record the numerical changes of all parameters in the feature description set and accurately label the data with state information, forming a high-quality supervised learning sample set.

[0022] Considering the significant temporal correlation of the operating states of software components, discrete feature values ​​are organized into a feature sequence according to time order. This temporal processing can effectively capture the dynamic changes in the operating states of components, providing a data foundation for subsequent temporal pattern recognition. The length of the temporal window needs to be reasonably selected based on the operating characteristics of the component and the real-time monitoring requirements, ensuring coverage of the complete state change cycle while avoiding introducing excessive monitoring delays.

[0023] Employing a large-scale artificial intelligence model as the core of the analysis, the training process first utilizes massive feature sequence data for model pre-training to establish preliminary state recognition capabilities. Subsequently, an expert feedback mechanism is introduced for reinforcement learning fine-tuning, using the knowledge guidance of domain experts to improve the model's judgment accuracy and reliability. This data-driven and knowledge-guided training strategy ensures that the model can both learn complex patterns from data and incorporate the experiential knowledge of domain experts.

[0024] The trained intelligent model is deployed into the actual ship control system to build a real-time status monitoring and early warning platform. This platform continuously collects operational characteristic parameters of software components and organizes them into time-series sequences, which are then input into the trained model. By analyzing the pattern characteristics contained in the sequences, the model outputs a judgment on the current state of the components and generates different levels of status warnings based on confidence levels, providing decision support for system maintenance personnel.

[0025] This application provides a method for monitoring the abnormal operating status of ship control system software components. By selecting multi-dimensional feature parameters that can comprehensively reflect the operating status of the software components, a temporal feature description sequence is constructed. An artificial intelligence large model is used to perform deep training and learning on massive sequence data covering normal, abnormal, and failure scenarios with different load ratios. Finally, the mature large model is applied to real-time data stream analysis, which can accurately distinguish and intelligently judge the operating status of ship control system software components, especially the normal state of high load such as intermittent sudden busy, from the real abnormal state. This effectively overcomes the technical bottleneck of low accuracy and high false alarm rate when relying on the operating system task manager for monitoring. It significantly improves the automation level and reliability of status monitoring and can ensure the stable and safe operation of the ship control system.

[0026] Optionally, the method for monitoring abnormal operating status of ship control system software components provided in this application embodiment includes at least one of system resource parameters, network performance parameters, and service quality parameters as the monitored parameters.

[0027] Optionally, the method for monitoring abnormal operating status of ship control system software components provided in this application embodiment includes system resource parameters such as CPU utilization, memory usage, process switching frequency, CPU interrupt frequency, process physical memory, process virtual memory, disk throughput, and process disk I / O usage. The network performance parameters include data latency, network packet loss rate, and process bandwidth utilization. The service quality parameters include service call success rate, service response time, number of logs, and log addition frequency.

[0028] Specifically, for a selected typical software component, such as a curve display software component, a series of monitoring parameters corresponding to the operating status of the software component are selected to form a set of feature descriptions for the curve display software component. A typical set of feature descriptions at a certain moment includes CPU utilization, process switching frequency, CPU interrupt frequency, memory utilization, process physical memory, process virtual memory, data latency, network packet loss rate, process bandwidth utilization, disk throughput, process disk I / O utilization, service call success rate, service response time, number of logs, and log increase frequency.

[0029] Optionally, the method for monitoring abnormal operating status of ship control system software components provided in this application embodiment, wherein the step of training and fine-tuning the large artificial intelligence model using the feature description sequence specifically includes: The feature description sequence is divided into a training set and a validation set according to a certain ratio; The large artificial intelligence model is initially trained using the training set. During training, a feedback reinforcement learning mechanism is introduced to correct and guide the model's output, thereby fine-tuning the model. The fine-tuned model is validated using the validation set. Training is complete when the validation accuracy reaches a preset threshold.

[0030] Specifically, the data partitioning process first divides the labeled feature description sequences into training and validation sets according to a preset ratio, thus constructing a fair model performance evaluation mechanism. The training set is specifically used for learning and adjusting model parameters, while the validation set is used to evaluate the model's generalization ability on unseen data, effectively preventing the model from overfitting to noise and specific patterns in the training data and losing its universality.

[0031] The initial training phase uses training set data to perform basic training on a selected large-scale artificial intelligence model. During this phase, the model learns complex patterns inherent in feature description sequences through deep neural networks, initially establishing a mapping relationship from input sequences to output states (normal, abnormal, and specific load ratios, etc.). This process enables the model to acquire basic state recognition capabilities.

[0032] The core innovation of this claim is fine-tuning through feedback reinforcement learning, which is also a key step in improving the model's practicality. This mechanism is not simple supervised learning, but rather incorporates domain knowledge from an expert database to adjust the model. After initial training, the model makes predictions on the validation set data, and its output (such as state judgments and load estimates) is submitted to a domain expert database for review. Based on its deep expertise, the expert database assesses the accuracy of the model's judgments and provides corrective feedback. This feedback is transformed into reward signals that the reinforcement learning algorithm can understand. For example, when the model underestimates the load ratio of a sequence, the expert database provides a negative reward, guiding the model to adjust its internal parameters in subsequent iterations, bringing the output closer to the true value. Through multiple rounds of prediction-feedback-fine-tuning cycles, the model can gradually absorb experience that is difficult to obtain from pure data, thereby continuously improving the accuracy and reliability of its judgments.

[0033] The model, after multiple rounds of fine-tuning, is finally evaluated using a validation set. Training is considered complete when the model's overall accuracy on the validation set reaches a preset threshold (e.g., 95%). This objective standard ensures that the final deployed model achieves the performance level required for industrial applications.

[0034] In summary, the training fine-tuning method defined in this claim effectively combines data-driven and knowledge-guided approaches through a systematic process design and innovative feedback mechanism, ensuring that the final artificial intelligence model possesses both powerful learning capabilities and high practicality.

[0035] Optionally, the ship control system software component abnormal operation monitoring method provided in this application embodiment includes signal acquisition software components, including signal filtering, preprocessing, and signal conversion software components, and operation monitoring software components, including curve display, parameter display, and data refresh software components.

[0036] Optionally, the abnormal monitoring method for the operating status of ship control system software components provided in this application embodiment includes abnormal scenarios such as communication settings, input / output, and display.

[0037] Optionally, for a set of feature descriptions of a typical software component, the acquisition frequency of different parameter features may be inconsistent, which can be ignored. All features within a certain time interval are selected to form a feature description sequence of a typical software component within a certain time interval. A typical time interval can be 30 seconds or 60 seconds.

[0038] Optionally, for a typical software component, the relevant equipment is started and the software component is run. Different business scenarios are simulated through the test equipment and software to obtain the characteristic parameters of the normal operation scenario, minor abnormal scenario, and error failure scenario of the functional software component. For the normal operation scenario, different load ratios can be set, such as 10%, 30%, 50%, 70%, 90%, etc., to obtain the characteristic parameters of the normal operation scenario under different loads. For abnormal scenarios, different abnormalities such as communication abnormality, input / output abnormality, and display abnormality can also be set to obtain the parameter characteristics of the abnormal scenario. The failure scenario is similar.

[0039] Optionally, for a set of feature descriptions of a typical software component, the acquisition frequency of different parameter features may be inconsistent, which can be ignored. All features within a certain time interval are selected to form a feature description sequence of a typical software component within a certain time interval. A typical time interval can be 30 seconds or 60 seconds.

[0040] Data from different task scenarios and with different load ratios are divided into training data and validation data according to a certain ratio. One option is to use 70% of the data for training and 30% for validation.

[0041] The trained and fine-tuned large model is validated using 30% of the data. When the overall accuracy is greater than a certain threshold, the training and fine-tuning are complete. One threshold setting is to achieve an accuracy of 95% or higher.

[0042] This application embodiment also provides a device for monitoring abnormal operating status of ship control system software components, including: The component acquisition module is used to acquire the software components of the ship control system, determine multiple monitoring parameter features corresponding to the operating status of the software components, and form a feature description set. The scenario simulation module is used to run the software component and simulate various business scenarios to obtain learning data. The business scenarios include normal operation scenarios, abnormal scenarios and failure scenarios with different load ratios, and record the feature values ​​of the parameters in the feature description set under each scenario. The feature sequence generation module is used to extract the feature values ​​within a time interval to form the feature description sequence of the software component. The large model training module is used to train and fine-tune the large artificial intelligence model using the feature description sequence; The monitoring module is used to monitor the real-time operating status and identify anomalies of the ship control system software components using a trained and fine-tuned large artificial intelligence model.

[0043] This application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method. The computer-readable storage medium may include, but is not limited to, any type of disk, including floppy disks, optical disks, DVDs, CD-ROMs, microdrives, as well as magneto-optical disks, ROMs, RAMs, EPROMs, EEPROMs, DRAMs, VRAMs, flash memory devices, magnetic cards or optical cards, nanosystems (including molecular memory ICs), or any type of medium or device suitable for storing instructions and / or data.

[0044] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0045] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0046] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some service interface; the indirect coupling or communication connection between devices or units may be electrical or other forms.

[0047] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0048] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0049] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0050] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0051] The foregoing description is merely an exemplary embodiment of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Those skilled in the art will readily conceive of embodiments of this disclosure upon considering the specification and practicing the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described herein. The specification and embodiments are to be considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

[0052] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0053] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for monitoring abnormal operating status of ship control system software components, characterized in that, Includes the following steps: The software components of the ship control system are acquired, and multiple monitoring parameter features corresponding to the operating status of the software components are determined to form a feature description set. Run the software component to simulate various business scenarios to obtain learning data. The business scenarios include normal operation scenarios, abnormal scenarios, and failure scenarios with different load ratios. Record the feature values ​​of the parameters in the feature description set under each scenario. The feature values ​​within a certain time interval are extracted to form the feature description sequence of the software component; The aforementioned feature description sequences are used to train and fine-tune large-scale artificial intelligence models; The trained and finely tuned large-scale artificial intelligence model is used to monitor the real-time operating status and identify anomalies of the ship control system software components.

2. The method for monitoring abnormal operating status of ship control system software components as described in claim 1, characterized in that, The characteristics of the parameters to be monitored include at least one of system resource parameters, network performance parameters, and service quality parameters.

3. The method for monitoring abnormal operating status of ship control system software components as described in claim 1, characterized in that, The system resource parameters include CPU utilization, memory usage, process switching frequency, CPU interrupt frequency, process physical memory, process virtual memory, disk throughput, and process disk I / O usage. The network performance parameters include data latency, network packet loss rate, and process bandwidth utilization. The service quality parameters include service call success rate, service response time, number of logs, and log addition frequency.

4. The method for monitoring abnormal operating status of ship control system software components as described in claim 1, characterized in that, The process of training and fine-tuning a large-scale artificial intelligence model using the feature description sequence specifically includes: The feature description sequence is divided into a training set and a validation set according to a certain ratio; The large artificial intelligence model is initially trained using the training set. During training, a feedback reinforcement learning mechanism is introduced to correct and guide the model's output, thereby fine-tuning the model. The fine-tuned model is validated using the validation set. Training is complete when the validation accuracy reaches a preset threshold.

5. The method for monitoring abnormal operating status of ship control system software components as described in claim 1, characterized in that, The software components include signal acquisition software components, including signal filtering, preprocessing, and signal conversion software components, as well as operation monitoring software components, including curve display, parameter display, and data refresh software components.

6. The method for monitoring abnormal operating status of ship control system software components as described in claim 1, characterized in that, The abnormal scenarios include communication errors, input / output errors, and display errors.

7. A device for monitoring abnormal operating status of ship control system software components, characterized in that, include: The component acquisition module is used to acquire the software components of the ship control system, determine multiple monitoring parameter features corresponding to the operating status of the software components, and form a feature description set. The scenario simulation module is used to run the software component and simulate various business scenarios to obtain learning data. The business scenarios include normal operation scenarios, abnormal scenarios and failure scenarios with different load ratios, and record the feature values ​​of the parameters in the feature description set under each scenario. The feature sequence generation module is used to extract the feature values ​​within a time interval to form the feature description sequence of the software component. The large model training module is used to train and fine-tune the large artificial intelligence model using the feature description sequence; The monitoring module is used to monitor the real-time operating status and identify anomalies of the ship control system software components using a trained and fine-tuned large artificial intelligence model.

8. A device for monitoring abnormal operating status of ship control system software components, characterized in that, It includes at least one processing unit and at least one storage unit, wherein the storage unit stores a computer program that, when executed by the processing unit, causes the processing unit to perform the steps of the method according to any one of claims 1 to 6.

9. A storage medium, characterized in that, It stores a computer program that can be executed by a ship control system software component operation status anomaly monitoring device. When the computer program is run on the ship control system software component operation status anomaly monitoring device, the ship control system software component operation status anomaly monitoring device performs the steps of the method according to any one of claims 1 to 6.

Images