Self-checking method and system of comprehensive ship bridge system

Abstract

The invention discloses a self-checking method and system for a comprehensive bridge system, and relates to the technical field of intelligent ship management, and the method comprises the steps: collecting bridge equipment data in the comprehensive bridge system through a data collector, carrying out the clustering processing of the bridge equipment data with a data function label, and obtaining a bridge equipment data cluster; according to a clustering result, carrying out modular division on the comprehensive ship bridge system, constructing self-inspection models of different modules of the comprehensive ship bridge system based on a data driving model, and training the self-inspection models by using the training data; and acquiring actual ship bridge equipment data of the comprehensive ship bridge system, matching a corresponding self-checking model by using the data function label, constructing an abnormal classifier in the self-checking model based on binary logic, and outputting a self-checking result by estimating a residual error between the ship bridge equipment data and the actual ship bridge equipment data. According to the invention, automatic abnormity discrimination of the comprehensive ship bridge system is realized through multi-source signal processing, early warning of ship abnormity is realized, and long-term stable operation of the comprehensive ship bridge system is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent ship management, and more specifically, to a self-inspection method and system for an integrated ship bridge system. Background Art

[0002] The integrated bridge system (IBS) is an integrated system for implementing ship navigation and control. It interconnects and centralizes sensor information, commands, and controls from workstations to improve the safety and efficiency of ship management by operators. Built on the ship's bridge, the IBS integrates navigation, control, radar collision avoidance, and navigation management. Leveraging multidisciplinary modern technologies such as computers, automatic control, networking, and information fusion, it centrally acquires sensor information, monitors it, sensitively detects and rapidly responds to events, and promptly issues various control commands for effective navigation and control. It serves as the ship's information center and command and control center.

[0003] An integrated bridge system can be composed of one or more multifunctional workstations, each of which provides users with a multi-task operating interface consisting of ECDIS, radar, and Conning, and can switch between them. The main function of ECDIS is to provide route planning and navigation monitoring tasks based on electronic charts, AIS data, and sensor data. The main function of radar is to provide collision avoidance tasks based on radar echo data, AIS data, and sensor data. Conning mainly provides navigation information, navigation control information, central alarm management, and system status monitoring tasks. Since ships include a large number of mechanical and electronic equipment, fault monitoring of mechanical and electronic equipment is required and timely elimination is required to ensure the normal operation of the equipment. Therefore, the system has certain self-detection and automatic repair capabilities. When an anomaly is detected, the system can issue a system alarm in advance to remind relevant personnel to intervene manually, realizing statistical monitoring and early warning functions to ensure the long-term stable operation of the entire system. Summary of the Invention

[0004] In order to solve the above technical problems, the present invention proposes a self-inspection method and system for an integrated ship bridge system, the purpose of which is to realize automatic abnormality discrimination of the integrated ship bridge system through multi-source signal processing and achieve early warning of ship abnormalities.

[0005] A first aspect of the present invention provides a self-checking method for an integrated ship bridge system, comprising the following steps: collecting bridge equipment data in the integrated bridge system through a data collector, matching the bridge equipment data with data functions, and performing data preprocessing on the matched data; The bridge equipment data with data function labels are clustered, and the integrated bridge system is divided into modules based on the clustering results. The characteristic factors corresponding to different modules are obtained by feature extraction. Extracting normal operation data of the integrated bridge system, expanding the dimension of the normal operation data, constructing training data based on characteristic factors, constructing self-checking models of different modules of the integrated bridge system based on a data-driven model, and using the training data to train the self-checking models; The actual bridge equipment data of the integrated bridge system is obtained, and the corresponding self-inspection model is matched with the data function label. An anomaly classifier is constructed in the self-inspection model based on binary logic, and the self-inspection result is output by estimating the residual between the bridge equipment data and the actual bridge equipment data.

[0006] In this solution, the data of the bridge equipment in the integrated bridge system is collected by a data collector, the bridge equipment data is matched with the data function, and the matched data is preprocessed, specifically: using preset sensors in the integrated bridge system as data collectors to collect multi-sensor signals in the integrated bridge system, evaluating the multi-sensor signals based on functional responses of relevant bridge equipment, and selecting multi-sensor signals that meet preset evaluation criteria as bridge equipment data; Match the selected bridge equipment data with the data functions, perform data noise reduction and data cleaning preprocessing on the matched bridge equipment data, and eliminate missing bridge equipment data; Reading historical alarm information from the integrated bridge system, extracting the data type involved in the historical alarm information, using the data type to perform data verification in the remaining bridge equipment data, and when the remaining bridge equipment data does not contain a given data type, performing data expansion based on the data type; When the data verification is passed, the pre-processed bridge equipment data is divided according to the data function to generate bridge equipment data corresponding to different data function labels.

[0007] In this scheme, the bridge equipment data with data function labels are clustered, and the comprehensive bridge system is modularized according to the clustering results, specifically: The K-means algorithm is used to cluster the bridge equipment data with data function labels. The K-means algorithm is optimized using the optimized genetic algorithm to select the initial cluster center. The Mahalanobis distance is used as the metric function to calculate the membership between the bridge equipment data sample points and the initial cluster center. Preset the soft clustering threshold and hard clustering threshold corresponding to the integrated bridge system. When the degree of membership between the bridge equipment data sample point and the initial cluster center is not less than the hard clustering threshold, the bridge equipment data sample point is classified into the corresponding initial cluster center to generate the corresponding cluster cluster; When the degree of membership between a bridge equipment data sample point and the initial cluster center is greater than or equal to the soft clustering threshold and less than the hard clustering threshold, the bridge equipment data sample point is assigned to all initial cluster centers that meet the conditions. After all bridge equipment data sample points are assigned, the cluster center is continuously updated in the iterative clustering process, and the clustering result of the last clustering is output; According to the output clustering results, the corresponding bridge equipment data subset is obtained, and the bridge equipment corresponding to each bridge equipment data subset is regarded as a comprehensive bridge system module.

[0008] In this solution, feature extraction is used to obtain the characteristic factors corresponding to different modules, specifically: Obtain bridge equipment data subsets for different integrated bridge system modules, extract historical fault data of bridge equipment based on big data methods, and use Spearman correlation analysis to obtain bridge equipment data variables in the bridge equipment data subset whose correlation meets the preset standards; Constructing a one-dimensional convolutional neural network as a feature extraction network, importing the bridge equipment data variables into the one-dimensional convolutional neural network, using convolution operations to extract deep features of the data, and introducing an attention mechanism in the fully connected layer to obtain the attention weight of the one-dimensional feature vector; The one-dimensional feature vectors are sorted using the attention weights, and a preset number of one-dimensional feature vectors are intercepted according to the sorting results. When redundant features exist in the intercepted one-dimensional feature vectors, the redundant features are removed and a corresponding number of one-dimensional feature vectors are continuously selected from the sorting results. When no redundant vectors exist in the selected one-dimensional feature vectors, they are used as characteristic factors of the integrated bridge system module.

[0009] In this solution, the normal operation data of the integrated bridge system is extracted, the normal operation data is expanded, and training data is constructed based on characteristic factors, specifically: Filter normal operation data from the bridge equipment data subsets corresponding to different integrated bridge system modules, and use the compressed sensing algorithm to expand the dimension of the filtered normal operation data to improve data quality; Characteristic factors of different integrated bridge system modules are obtained as data indicators, and characteristic parameters are extracted from the normal operating data after dimensional expansion using the data indicators. Noise is randomly added through additive injection during the extraction process to generate noise-enhanced characteristic parameters. The characteristic parameters corresponding to each data indicator and the noise-enhanced characteristic parameters are aggregated to generate a data set, in which training data and test data are divided based on a preset ratio.

[0010] In this solution, a self-check model of different modules of the integrated bridge system is constructed based on a data-driven model, and the self-check model is trained using the training data, specifically: A self-test model for different integrated bridge system modules is constructed using an LSTM autoencoder network. Training data for different integrated bridge system modules is imported into the self-test model for training. The time series features of the training data are obtained using the memory unit and gating mechanism of the LSTM network. Data is encoded based on the time series features to achieve data reconstruction. Decoding the reconstructed training data through the decoder module, mapping the decoded training data to the sample label space for comparison, obtaining a reconstruction error, and guiding the self-checking model to perform iterative training based on the reconstruction error; The generalization ability of the self-checking model of the test sample is used for training, and the residual threshold of the abnormal classifier in the self-checking model is output. When the classification performance of the self-checking model reaches the preset standard, the self-checking model of the current integrated bridge system module is output.

[0011] In this solution, the actual bridge equipment data of the integrated bridge system is obtained, and the corresponding self-test model is matched with the data function label. An anomaly classifier is constructed in the self-test model based on binary logic. The self-test results are output by estimating the residual between the equipment data and the actual equipment data. Specifically, obtaining actual bridge equipment data of the integrated bridge system, determining a corresponding self-test model based on data function labels corresponding to the actual bridge equipment data, passing the actual bridge equipment data through a feature extraction network, and then importing the output of the feature extraction network into the self-test model to obtain estimated bridge equipment data under normal operation of the bridge equipment data; The residual data between the estimated bridge equipment data and the actual bridge equipment data is obtained through the self-check model. The probability distribution of the integrated bridge system at the current moment is obtained in the anomaly classifier based on the residual data. The probability distribution at the current moment is compared with the probability distribution corresponding to the residual threshold to obtain a deviation value. When the deviation value is greater than the preset deviation threshold, it proves that there is an abnormality in the integrated bridge system and an abnormality warning is generated.

[0012] A second aspect of the present invention provides a self-checking system for an integrated ship bridge system, the system comprising a characteristic factor analysis unit, an integrated ship bridge system self-checking unit, a ship bridge equipment data acquisition unit, and a self-checking abnormality warning unit; The characteristic factor analysis unit uses the bridge equipment data to cluster and divide the comprehensive bridge system into modules, and uses feature extraction to obtain characteristic factors corresponding to different modules; The integrated bridge system self-check unit extracts normal operating data of the integrated bridge system, constructs training data based on characteristic factors, constructs self-check models of different modules of the integrated bridge system based on a data-driven model, constructs an anomaly classifier based on binary logic in the self-check model, and trains the model using the training data; The bridge equipment data acquisition unit acquires actual bridge equipment data of the integrated bridge system and performs data preprocessing; The self-inspection abnormality warning unit imports the pre-processed actual bridge equipment data into the self-inspection model, outputs the self-inspection result by estimating the residual between the bridge equipment data and the actual bridge equipment data, and generates a self-inspection abnormality warning.

[0013] Compared with the prior art, the present invention has the following beneficial effects: The present invention features automatic fault diagnosis for the integrated bridge system, automatically locating the location of system faults and alerting on-duty personnel and equipment maintenance personnel through alarms. When an anomaly is detected through self-diagnosis, a system alarm is issued in advance, prompting personnel to intervene manually. This implements statistical monitoring and early warning functions to ensure the long-term stable operation of the entire system.

[0014] The integrated bridge system's self-diagnosis system utilizes a modular design. If one module experiences an anomaly, the other modules remain operational. A functioning module detects an anomaly in another associated module and monitors in real time whether the associated module has returned to normal. Once that occurs, the system resumes normal operation immediately. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments or exemplary embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or exemplary descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained according to these drawings without paying any creative work.

[0016] Figure 1 A flow chart showing a self-checking method for an integrated ship bridge system is shown; Figure 2 A flow chart showing the use of feature screening to obtain feature factors corresponding to different modules; Figure 3 A flow chart showing the use of the self-check model to obtain the self-check results of the integrated bridge system is shown; Figure 4 A block diagram of a self-checking system for an integrated ship bridge system is shown. DETAILED DESCRIPTION

[0017] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0018] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0019] like Figure 1 As shown, a first embodiment of the present invention provides a self-checking method for an integrated ship bridge system, comprising: S102, collecting bridge equipment data in the integrated bridge system through a data collector, matching the bridge equipment data with data functions, and performing data preprocessing on the matched data; S104, clustering the bridge equipment data with data function labels, dividing the integrated bridge system into modules based on the clustering results, and obtaining characteristic factors corresponding to different modules through feature extraction; S106, extracting normal operating data of the integrated bridge system, expanding the dimension of the normal operating data, constructing training data based on characteristic factors, constructing self-checking models for different modules of the integrated bridge system based on a data-driven model, and training the self-checking models using the training data; S108, obtaining actual bridge equipment data of the integrated bridge system, matching the corresponding self-checking model with the data function label, constructing an anomaly classifier in the self-checking model based on binary logic, and outputting the self-checking result by estimating the residual between the bridge equipment data and the actual bridge equipment data.

[0020] It should be noted that the pre-set sensors in the integrated bridge system are used as data collectors to collect multi-sensor signals in the integrated bridge system. The multi-sensor signals are evaluated based on the functional response of the relevant bridge equipment, and the multi-sensor signals that meet the pre-set evaluation criteria are selected as the bridge equipment data. Preferably, the evaluation is performed based on the response of the data function, and the sensor signals that successfully respond are selected as the bridge equipment data. For example, the successful generation of a measurement display during onboard equipment monitoring proves that the response is successful. The selected bridge equipment data is matched with the data function, and the matched bridge equipment data is pre-processed by data noise reduction and data cleaning to eliminate missing bridge equipment data. The historical alarm information in the integrated bridge system is read, and the data type involved in the historical alarm information is extracted. The data type is used to perform data verification in the remaining bridge equipment data. When the remaining bridge equipment data does not contain a given data type, the data is expanded according to the data type. When the data verification passes, the pre-processed bridge equipment data is divided according to the data function to generate bridge equipment data corresponding to different data function labels.

[0021] It should be noted that the K-means algorithm is used to cluster the bridge equipment data with data function labels, the optimized genetic algorithm is used to optimize the K-means algorithm, the genetic algorithm is optimized and improved, and the chaos algorithm is used to make the initial population chaotic so that the initial population distribution is more uniform. According to the clustering silhouette coefficient as the fitness function, in the iteration of the genetic algorithm, the bridge equipment data samples with high fitness values ​​are selected for replication to obtain the search direction of the initial cluster center, and the initial cluster center is selected by searching in the direction of high importance. The Mahalanobis distance is used as the metric function to calculate the membership between the bridge equipment data sample points and the initial cluster center; the soft clustering threshold and hard clustering threshold corresponding to the comprehensive bridge system are preset. When the degree of membership between the bridge equipment data sample point and the initial clustering center is not less than the hard clustering threshold, the bridge equipment data sample point is classified into the corresponding initial clustering center to generate the corresponding cluster cluster; when the degree of membership between the bridge equipment data sample point and the initial clustering center is greater than or equal to the soft clustering threshold and less than the hard clustering threshold, the bridge equipment data sample point is classified into all initial clustering centers that meet the conditions. After the allocation of all bridge equipment data sample points is completed, the clustering center is continuously updated in the iterative clustering process, and the clustering result of the last clustering is output; the corresponding bridge equipment data subset is obtained according to the output clustering result, and the bridge equipment corresponding to each bridge equipment data subset is regarded as a comprehensive bridge system module.

[0022] According to an embodiment of the present invention, feature extraction is used to obtain feature factors corresponding to different modules, specifically: S202, obtaining bridge equipment data subsets of different integrated bridge system modules, extracting historical fault data of the bridge equipment based on big data methods, and using Spearman correlation analysis to obtain bridge equipment data variables in the bridge equipment data subsets whose correlations meet preset standards; S204: constructing a one-dimensional convolutional neural network as a feature extraction network, importing the bridge equipment data variables into the one-dimensional convolutional neural network, extracting deep features of the data using a convolution operation, and introducing an attention mechanism in the fully connected layer to obtain attention weights of the one-dimensional feature vectors; S206, sorting the one-dimensional feature vectors using the attention weights, intercepting a preset number of one-dimensional feature vectors based on the sorting results, removing redundant features if any, and continuing to select a corresponding number of one-dimensional feature vectors from the sorting results, and using the selected one-dimensional feature vectors as feature factors for the integrated bridge system module if no redundant vectors exist in the selected one-dimensional feature vectors.

[0023] It should be noted that the feature extraction network consists of an input layer, a convolutional module, and a fully connected layer. The convolutional module includes a convolutional layer, a pooling layer, and an activation layer. The pooling layer uses max pooling to filter features, and the activation layer uses the ReLU activation function to increase the nonlinearity of the network. After the convolutional module extracts the deep features of the data, the fully connected layer performs an attention-weighted operation on the input features, and the characteristic factors of the integrated bridge system module are selected in the fully connected layer.

[0024] Normal operating data is screened from the subset of bridge equipment data corresponding to different integrated bridge system modules, and the screened normal operating data is expanded using a compressed sensing algorithm. Compressed sensing can recover and reconstruct the original signal from a small amount of normal operating data while retaining all the characteristics of the original signal. This dimension expansion improves data quality, makes it easier to extract signal features, and helps improve self-test accuracy. Characteristic factors of different integrated bridge system modules are obtained as data indicators, and these data indicators are used to extract characteristic parameters from the expanded normal operating data. During the extraction process, noise is randomly added through additive injection to generate noise-enhanced characteristic parameters. The characteristic parameters corresponding to each data indicator and the noise-enhanced characteristic parameters are aggregated to generate a data set. Within this data set, training data and test data are divided based on a preset ratio. The noise injection method is used to enhance the predictive performance of the subsequent self-test network.

[0025] It should be noted that the self-inspection models of different integrated bridge system modules are constructed using LSTM autoencoder networks, preferably using multi-layer LSTM autoencoders to construct the self-inspection models of different integrated bridge system modules, which can better capture, learn and mine the features contained in the data and obtain effective data encoding. The LSTM autoencoder network includes an LSTM encoder module network and an LSTM decoder module. The training data of different integrated bridge system modules are imported into the self-inspection model for training. The memory unit and gating mechanism of the LSTM network are used to obtain the time series features of the training data. The data is encoded according to the time series features of the data to achieve data reconstruction; the training data reconstructed by the decoder module is decoded, and the decoded training data is mapped to the sample label space for comparison to obtain the reconstruction error. The self-inspection model is guided to iterative training according to the reconstruction error; the generalization ability of the test sample self-inspection model is used for training, and the residual threshold of the abnormal classifier in the self-inspection model is output. When the classification performance of the self-inspection model reaches the preset standard, the self-inspection model of the current integrated bridge system module is output.

[0026] Figure 3 A flow chart of obtaining the self-check results of the integrated bridge system using the self-check model is shown.

[0027] According to an embodiment of the present invention, actual bridge equipment data of an integrated bridge system is obtained, and the corresponding self-test model is matched using data function labels. An anomaly classifier is constructed in the self-test model based on binary logic, and the self-test results are output by estimating the residual between the equipment data and the actual equipment data. Specifically, S302, obtaining actual bridge equipment data of the integrated bridge system, determining a corresponding self-checking model based on data function labels corresponding to the actual bridge equipment data, passing the actual bridge equipment data through a feature extraction network, and then importing the output of the feature extraction network into the self-checking model to obtain estimated bridge equipment data under normal operation of the bridge equipment data; S304, obtaining residual data between the estimated bridge equipment data and the actual bridge equipment data through the self-checking model, and obtaining the probability distribution of the integrated bridge system at the current moment based on the residual data in the anomaly classifier; S306, comparing the probability distribution at the current moment with the probability distribution corresponding to the residual threshold to obtain a deviation value. When the deviation value is greater than the preset deviation threshold, it is proved that there is an abnormality in the integrated bridge system, and an abnormality warning is generated.

[0028] It should be noted that the operating status of the integrated bridge system is monitored in real time based on the bridge equipment data of the online integrated bridge system to achieve automatic detection and accurate diagnosis of integrated bridge system faults. Normally, when the integrated bridge system is operating normally, the deviation of the residual probability distribution remains within a relatively stable and small fluctuation range. When the integrated bridge system is operating abnormally, the deviation of the residual probability distribution gradually increases and fluctuates violently, gradually deviating from the stable value under normal operating conditions. By monitoring the changing trend of the residual probability distribution deviation in real time, automatic detection of the degradation state of the integrated bridge system is achieved. Based on the alarm thresholds set in the trained self-test model and the bridge equipment data of the integrated bridge system, real-time monitoring is performed to promptly and effectively detect abnormal changes in the online integrated bridge system status and issue warning information, thus achieving early warning of online integrated bridge system faults.

[0029] A second aspect of the present invention provides a self-checking system for an integrated ship bridge system, the system comprising a characteristic factor analysis unit, an integrated ship bridge system self-checking unit, a ship bridge equipment data acquisition unit, and a self-checking abnormality warning unit; The characteristic factor analysis unit uses the bridge equipment data to cluster and divide the comprehensive bridge system into modules, and uses feature extraction to obtain characteristic factors corresponding to different modules; The integrated bridge system self-check unit extracts normal operating data of the integrated bridge system, constructs training data based on characteristic factors, constructs self-check models of different modules of the integrated bridge system based on a data-driven model, constructs an anomaly classifier based on binary logic in the self-check model, and trains the model using the training data; The bridge equipment data acquisition unit acquires actual bridge equipment data of the integrated bridge system and performs data preprocessing; The self-inspection abnormality warning unit imports the pre-processed actual bridge equipment data into the self-inspection model, outputs the self-inspection result by estimating the residual between the bridge equipment data and the actual bridge equipment data, and generates a self-inspection abnormality warning.

[0030] The device embodiments described above are merely illustrative. For example, the division of units described herein is merely a logical functional division. Actual implementations may employ alternative divisions, such as combining multiple units or components, integrating them into another system, or omitting or disabling certain features. Furthermore, the coupling, direct coupling, or communication connection between components shown or discussed may be through interfaces. Indirect coupling or communication connections between devices or units may be electrical, mechanical, or other in nature.

[0031] The units described above as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units may be selected based on actual needs to achieve the purpose of the present embodiment. In addition, the functional units in the various embodiments of the present invention may all be integrated into one processing unit, or each unit may be independently configured as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0032] Those skilled in the art will understand that all or part of the steps of the above-mentioned method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above-mentioned method embodiment; and the aforementioned storage medium includes: mobile storage devices, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disks or optical disks, and other media that can store program codes.

[0033] The above description is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A self-checking method for an integrated ship bridge system, characterized in that: The following steps are involved: collecting bridge equipment data in the integrated bridge system through a data collector, matching the bridge equipment data with data functions, and performing data preprocessing on the matched data; The bridge equipment data with data function labels are clustered, and the integrated bridge system is divided into modules based on the clustering results. The characteristic factors corresponding to different modules are obtained by feature extraction. Extracting normal operation data of the integrated bridge system, expanding the dimension of the normal operation data, constructing training data based on characteristic factors, constructing self-checking models of different modules of the integrated bridge system based on a data-driven model, and using the training data to train the self-checking models; The actual bridge equipment data of the integrated bridge system is obtained, and the corresponding self-inspection model is matched with the data function label. An anomaly classifier is constructed in the self-inspection model based on binary logic, and the self-inspection result is output by estimating the residual between the bridge equipment data and the actual bridge equipment data.

2. The self-inspection method of an integrated ship bridge system according to claim 1, characterized in that: The data of the bridge equipment in the integrated bridge system is collected by a data collector, the bridge equipment data is matched with the data function, and the matched data is pre-processed, specifically: using preset sensors in the integrated bridge system as data collectors to collect multi-sensor signals in the integrated bridge system, evaluating the multi-sensor signals based on functional responses of relevant bridge equipment, and selecting multi-sensor signals that meet preset evaluation criteria as bridge equipment data; Match the selected bridge equipment data with the data functions, perform data noise reduction and data cleaning preprocessing on the matched bridge equipment data, and eliminate missing bridge equipment data; Reading historical alarm information from the integrated bridge system, extracting the data type involved in the historical alarm information, using the data type to perform data verification in the remaining bridge equipment data, and when the remaining bridge equipment data does not contain a given data type, performing data expansion based on the data type; When the data verification is passed, the pre-processed bridge equipment data is divided according to the data function to generate bridge equipment data corresponding to different data function labels.

3. The self-inspection method of an integrated bridge system according to claim 1, characterized in that: The bridge equipment data with data function labels are clustered, and the comprehensive bridge system is modularized according to the clustering results, specifically: The K-means algorithm is used to cluster the bridge equipment data with data function labels. The K-means algorithm is optimized using the optimized genetic algorithm to select the initial cluster center. The Mahalanobis distance is used as the metric function to calculate the membership between the bridge equipment data sample points and the initial cluster center. Preset the soft clustering threshold and hard clustering threshold corresponding to the integrated bridge system. When the degree of membership between the bridge equipment data sample point and the initial cluster center is not less than the hard clustering threshold, the bridge equipment data sample point is classified into the corresponding initial cluster center to generate the corresponding cluster cluster; When the degree of membership between a bridge equipment data sample point and the initial cluster center is greater than or equal to the soft clustering threshold and less than the hard clustering threshold, the bridge equipment data sample point is assigned to all initial cluster centers that meet the conditions. After all bridge equipment data sample points are assigned, the cluster center is continuously updated in the iterative clustering process, and the clustering result of the last clustering is output; According to the output clustering results, the corresponding bridge equipment data subset is obtained, and the bridge equipment corresponding to each bridge equipment data subset is regarded as a comprehensive bridge system module.

4. The self-inspection method of an integrated ship bridge system according to claim 1, characterized in that: Feature extraction is used to obtain the characteristic factors corresponding to different modules, specifically: Obtain bridge equipment data subsets for different integrated bridge system modules, extract historical fault data of bridge equipment based on big data methods, and use Spearman correlation analysis to obtain bridge equipment data variables in the bridge equipment data subset whose correlation meets the preset standards; Constructing a one-dimensional convolutional neural network as a feature extraction network, importing the bridge equipment data variables into the one-dimensional convolutional neural network, using convolution operations to extract deep features of the data, and introducing an attention mechanism in the fully connected layer to obtain the attention weight of the one-dimensional feature vector; The one-dimensional feature vectors are sorted using the attention weights, and a preset number of one-dimensional feature vectors are intercepted according to the sorting results. When redundant features exist in the intercepted one-dimensional feature vectors, the redundant features are removed and a corresponding number of one-dimensional feature vectors are continuously selected from the sorting results. When no redundant vectors exist in the selected one-dimensional feature vectors, they are used as characteristic factors of the integrated bridge system module.

5. The self-inspection method of an integrated ship bridge system according to claim 1, characterized in that: Extract the normal operation data of the integrated bridge system, expand the dimension of the normal operation data, and construct training data based on characteristic factors, specifically: Filter normal operation data from the bridge equipment data subsets corresponding to different integrated bridge system modules, and use the compressed sensing algorithm to expand the dimension of the filtered normal operation data to improve data quality; Characteristic factors of different integrated bridge system modules are obtained as data indicators, and characteristic parameters are extracted from the normal operating data after dimensional expansion using the data indicators. Noise is randomly added through additive injection during the extraction process to generate noise-enhanced characteristic parameters. The characteristic parameters corresponding to each data indicator and the noise-enhanced characteristic parameters are aggregated to generate a data set, in which training data and test data are divided based on a preset ratio.

6. The self-inspection method of an integrated ship bridge system according to claim 1, characterized in that: Based on the data-driven model, a self-checking model of different modules of the integrated bridge system is constructed, and the self-checking model is trained using the training data, specifically: A self-test model for different integrated bridge system modules is constructed using an LSTM autoencoder network. Training data for different integrated bridge system modules is imported into the self-test model for training. The time series features of the training data are obtained using the memory unit and gating mechanism of the LSTM network. Data is encoded based on the time series features to achieve data reconstruction. Decoding the reconstructed training data through the decoder module, mapping the decoded training data to the sample label space for comparison, obtaining a reconstruction error, and guiding the self-checking model to perform iterative training based on the reconstruction error; The generalization ability of the self-checking model of the test sample is used for training, and the residual threshold of the abnormal classifier in the self-checking model is output. When the classification performance of the self-checking model reaches the preset standard, the self-checking model of the current integrated bridge system module is output.

7. The self-inspection method of an integrated ship bridge system according to claim 1, characterized in that: The actual bridge equipment data of the integrated bridge system is obtained, and the corresponding self-test model is matched with the data function label. An anomaly classifier is constructed in the self-test model based on binary logic. The self-test results are output by estimating the residual between the equipment data and the actual equipment data. Specifically, obtaining actual bridge equipment data of the integrated bridge system, determining a corresponding self-test model based on data function labels corresponding to the actual bridge equipment data, passing the actual bridge equipment data through a feature extraction network, and then importing the output of the feature extraction network into the self-test model to obtain estimated bridge equipment data under normal operation of the bridge equipment data; The residual data between the estimated bridge equipment data and the actual bridge equipment data is obtained through the self-check model. The probability distribution of the integrated bridge system at the current moment is obtained in the anomaly classifier based on the residual data. The probability distribution at the current moment is compared with the probability distribution corresponding to the residual threshold to obtain a deviation value. When the deviation value is greater than the preset deviation threshold, it proves that there is an abnormality in the integrated bridge system and an abnormality warning is generated.

8. A self-checking system for an integrated ship bridge system, characterized in that: Implementing a self-inspection method for an integrated bridge system according to any one of claims 1 to 7, the system comprising a characteristic factor analysis unit, an integrated bridge system self-inspection unit, a bridge equipment data acquisition unit, and a self-inspection abnormality warning unit; The characteristic factor analysis unit uses the bridge equipment data to cluster and divide the comprehensive bridge system into modules, and uses feature extraction to obtain characteristic factors corresponding to different modules; The integrated bridge system self-check unit extracts normal operating data of the integrated bridge system, constructs training data based on characteristic factors, constructs self-check models of different modules of the integrated bridge system based on a data-driven model, constructs an anomaly classifier based on binary logic in the self-check model, and trains the model using the training data; The bridge equipment data acquisition unit acquires actual bridge equipment data of the integrated bridge system and performs data preprocessing; The self-inspection abnormality warning unit imports the pre-processed actual bridge equipment data into the self-inspection model, outputs the self-inspection result by estimating the residual between the bridge equipment data and the actual bridge equipment data, and generates a self-inspection abnormality warning.

Images