CN117787945A - Ship facility operation and maintenance method, device, equipment and storage medium - Google Patents

Abstract

According to the ship facility operation and maintenance method, device, equipment and storage medium, the actual operation data of all ship facilities in the preset time period are obtained, the actual operation data of all ship facilities in the preset time period are classified, the actual operation data of all ship facilities in the preset time period are obtained, the actual operation data of the ship facilities in the preset time period are imported into a fault prediction model to be predicted, a prediction result is obtained, personalized recommendation is carried out on the ship facilities with all the operation states being fault states according to the prediction result, maintenance schemes and corresponding maintenance personnel information of all the fault ship facilities are obtained, decision support can be provided for operation and maintenance personnel, reasonable operation and maintenance strategies are formulated, the downtime of the facilities is reduced, and the operation and maintenance rationality and the operation and maintenance efficiency are improved.

Description

Ship facility operation and maintenance method, device, equipment and storage medium

Technical Field

The present disclosure relates to the field of ship maintenance technologies, and in particular, to a method, an apparatus, a device, and a storage medium for operating and maintaining a ship facility.

Background

Marine facility operations refer to the process of maintaining, managing, and servicing various equipment, systems, and facilities on a marine vessel. The ship facility operation and maintenance aims to ensure the normal operation of the ship facility and the facility, prolong the service life of the ship facility and improve the reliability and safety of the ship, and the ship facility operation and maintenance can reduce the faults and the downtime to the greatest extent through the measures of regular overhaul, planning and execution of maintenance plans, fault removal, performance optimization and the like. In the field of marine facility operations and maintenance, data involved in marine facility operations and maintenance analysis often come from a number of different types and sources of data sources, such as sensor data, monitoring data, historical fault data, etc., which often have heterogeneity, making fusion and analysis of the data complex and difficult. In addition, the existing ship facility operation and maintenance method cannot fully utilize information of different data sources, accurately identify and predict facility faults, is low in reliability, cannot formulate effective operation and maintenance measures according to the facility faults, and is low in coordination capability.

Disclosure of Invention

Aiming at the problems, the application provides a ship facility operation and maintenance method, device, equipment and storage medium, which can provide decision support for operation and maintenance personnel, make reasonable operation and maintenance strategies, reduce the downtime of the facility and improve the operation and maintenance rationality and operation and maintenance efficiency.

The embodiment of the application provides a ship facility operation and maintenance method, which comprises the following steps:

acquiring actual operation data of all ship facilities in a preset time period;

classifying actual operation data of all ship facilities in a preset time period to obtain actual operation data of each ship facility in the preset time period;

the actual operation data of the ship facilities in a preset time period is imported into a fault prediction model for prediction, and a prediction result is obtained;

and according to the prediction result, individually recommending all the ship facilities with the running states of fault states, and obtaining the maintenance scheme and corresponding maintenance personnel information of each fault ship facility.

In some embodiments, the method comprises:

acquiring fault characteristic data of ship facilities when various faults occur, clustering the fault characteristic data, constructing a plurality of databases, and storing the clustered fault characteristic data into corresponding databases to obtain characteristic databases;

Acquiring common characteristic rules of fault characteristic data in each characteristic database, and acquiring probability distribution matrixes of nodes in the Bayesian network according to the common characteristic rules;

and constructing a fault prediction model according to the probability distribution matrix of each node in the Bayesian network.

In some embodiments, the obtaining fault feature data when various faults occur in the ship facility, clustering the fault feature data, constructing a plurality of databases, and storing the clustered fault feature data into corresponding databases to obtain feature databases, including:

acquiring historical fault information of the ship facility based on a big data network, and extracting characteristics of the historical fault information of the ship facility to obtain fault characteristic data when various faults of the ship facility occur;

introducing a fuzzy clustering algorithm, initializing fuzzy clustering centers, and calculating the mahalanobis distance between each fault characteristic data and each fuzzy clustering center;

sequencing the Marshall distance between each fault characteristic data and each fuzzy clustering center to obtain a sequencing result, and clustering the fault characteristic data with the Marshall distance smaller than the preset Marshall distance into the corresponding fuzzy clustering center according to the sequencing result;

Constructing a plurality of databases, mapping fault characteristic data in different fuzzy clustering centers into corresponding databases for storage respectively, obtaining a plurality of characteristic databases, and updating each characteristic database periodically.

In some embodiments, the obtaining the common feature rule of the fault feature data in each feature database, and obtaining the probability distribution matrix of each node in the bayesian network according to the common feature rule includes:

an isolated forest algorithm is introduced, an isolated score of each fault characteristic data in each characteristic database is calculated through the isolated forest algorithm, and fault characteristic data with the isolated score being larger than a preset isolated score are screened out in the corresponding characteristic database;

carrying out standardized processing on fault characteristic data in each characteristic database so that the fault characteristic data in each characteristic database have the same measurement, and analyzing common characteristic rules among the fault characteristic data with the same measurement through a gray correlation analysis method;

constructing a Bayesian network, acquiring various fault types of each ship facility through a big data network, determining partition attributes according to the fault types, and dividing the ship facilities into a plurality of nodes in the Bayesian network based on the partition attributes;

And introducing a machine learning algorithm, learning probability distribution of corresponding nodes in the Bayesian network by combining common characteristic rules among fault characteristic data with the same measurement, and constructing a probability distribution matrix of each node according to the probability distribution.

In some embodiments, the constructing the fault prediction model according to the probability distribution matrix of each node in the bayesian network includes:

performing dimension reduction on the probability distribution matrix by an independent component analysis method to obtain a probability distribution matrix consisting of one-dimensional vectors;

calculating the log likelihood function of each matrix data in the probability distribution matrix formed by the one-dimensional vectors through expectation maximization, and determining the expected score of each matrix data in the probability distribution matrix formed by the one-dimensional vectors according to the log likelihood function;

removing matrix data with expected scores not greater than a preset expected score from the probability distribution matrix, reserving matrix data with expected scores greater than the preset expected score in the probability distribution matrix, and generating a processed probability distribution matrix;

and importing the processed probability distribution matrix into the fault prediction model, carrying out iterative learning on the fault prediction model according to the processed probability distribution matrix and a genetic algorithm, and stopping iteration when the fault prediction model meets the preset requirement, and outputting the fault prediction model.

In some embodiments, the classifying the actual operation data of all the ship facilities in the preset time period to obtain the actual operation data of each ship facility in the preset time period includes:

constructing a decision tree trunk, acquiring the facility number of ship facilities, determining segmentation nodes of the decision tree trunk according to the facility number, segmenting the decision tree trunk according to the segmentation nodes, and updating the state of the segmented decision tree trunk to obtain a decision tree model;

acquiring relative position information of each branch in the decision tree model, defining a classified hierarchical structure according to the relative position information of each branch, and acquiring a hierarchical center of each hierarchical structure; wherein the hierarchy is a hierarchical tree structure, each hierarchy representing a particular class;

importing actual operation data in an operation database into the decision tree model, calculating Euclidean distances between each actual operation data and each level center, and sorting the Euclidean distances between each actual operation data and each level center based on numerical values;

after the sorting is completed, mapping each actual operation data to a hierarchical center corresponding to the shortest Euclidean distance so as to classify each actual operation data into the hierarchical structure to which the actual operation data belongs; after all the actual operation data are classified, cutting each branch in the decision tree model to obtain the actual operation data of each ship facility in a preset time period.

In some embodiments, the individually recommending, according to the prediction result, all the ship facilities with the running states being fault states, to obtain the maintenance scheme and the corresponding maintenance personnel information of each fault ship facility, including:

obtaining the running state of each ship facility according to the prediction result, and if the running state of the ship facility is a fault state, obtaining the fault type of the ship facility with the running state being the fault state;

generating a search tag according to the fault type, searching a big data network based on the search tag, searching to obtain a plurality of maintenance schemes, and obtaining the maintenance success rate of each maintenance scheme; sequencing the maintenance success rates of all the maintenance schemes, and recommending the maintenance scheme corresponding to the maximum maintenance success rate to a preset platform for display;

acquiring position information of ship facilities with a fault running state, and searching maintenance personnel on the ship according to the position information to obtain candidate maintenance personnel information;

calculating a hash value between the candidate maintenance personnel information and the fault type through a hash algorithm, and eliminating the candidate maintenance personnel information with the hash value not larger than a preset hash value to obtain the remaining candidate maintenance personnel information;

Acquiring distance values between the residual candidate maintenance personnel information and the position information, sequencing each distance value, and recommending the residual candidate maintenance personnel information corresponding to the minimum distance value to a preset platform for display;

repeating the steps until personalized recommendation is completed on all the ship facilities with the running states of fault states, and obtaining the maintenance scheme and corresponding maintenance personnel information of each fault ship facility.

The embodiment of the application provides a ship facility operation and maintenance device, which comprises:

the acquisition module is used for acquiring actual operation data of all ship facilities in a preset time period;

the classification module is used for classifying the actual operation data of all the ship facilities in the preset time period to obtain the actual operation data of each ship facility in the preset time period;

the prediction module is used for guiding actual operation data of the ship facility in a preset time period into the fault prediction model for prediction to obtain a prediction result;

and the recommendation module is used for individually recommending all the ship facilities with the running states of fault states according to the prediction result to obtain the maintenance scheme of each ship facility with the fault and the corresponding maintenance personnel information.

An embodiment of the present application provides an electronic device, including:

a memory and a processor, the memory having stored thereon a computer program which, when executed by the processor, performs the method of any of the above.

Embodiments of the present application provide a storage medium storing a computer program executable by one or more processors for implementing any one of the above-described marine facility operation and maintenance methods.

According to the ship facility operation and maintenance method, device, equipment and storage medium, the actual operation data of all ship facilities in the preset time period are obtained, the actual operation data of all ship facilities in the preset time period are classified, the actual operation data of all ship facilities in the preset time period are obtained, the actual operation data of the ship facilities in the preset time period are imported into a fault prediction model to be predicted, a prediction result is obtained, personalized recommendation is carried out on the ship facilities with all the operation states being fault states according to the prediction result, maintenance schemes and corresponding maintenance personnel information of all the fault ship facilities are obtained, decision support can be provided for operation and maintenance personnel, reasonable operation and maintenance strategies are formulated, the downtime of the facilities is reduced, and the operation and maintenance rationality and the operation and maintenance efficiency are improved.

Drawings

The present application will be described in more detail hereinafter based on embodiments and with reference to the accompanying drawings.

Fig. 1 is a schematic implementation flow chart of a ship facility operation and maintenance method according to an embodiment of the present application;

fig. 2 is a schematic structural diagram of a ship facility operation and maintenance device according to an embodiment of the present application;

fig. 3 is a schematic diagram of a composition structure of an electronic device according to an embodiment of the present application.

In the drawings, like parts are given like reference numerals, and the drawings are not drawn to scale.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the present application more apparent, the present application will be described in further detail with reference to the accompanying drawings, and the described embodiments should not be construed as limiting the present application, and all other embodiments obtained by those skilled in the art without making any inventive effort are within the scope of the present application.

In the following description, reference is made to "some embodiments" which describe a subset of all possible embodiments, but it is to be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments and can be combined with one another without conflict.

If a similar description of "first\second\third" appears in the application document, the following description is added, in which the terms "first\second\third" are merely distinguishing between similar objects and do not represent a particular ordering of the objects, it being understood that the "first\second\third" may be interchanged in a particular order or precedence, where allowed, so that the embodiments of the application described herein can be implemented in an order other than that illustrated or described herein.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of the present application only and is not intended to be limiting of the present application.

Based on the problems existing in the related art, the embodiment of the present application provides a ship facility operation and maintenance method, and an execution subject of the ship facility operation and maintenance method may be an electronic device. The electronic device may be a notebook computer, a tablet computer, a desktop computer, a set-top box, a mobile device (e.g., a mobile phone, a portable music player, a personal digital assistant, a dedicated messaging device, a portable game device), or other various types of terminals, and may also be implemented as a server. The server may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), basic cloud computing services such as big data and artificial intelligent platforms, and the like.

The functions realized by the ship facility operation and maintenance method provided by the embodiment of the application can be realized by calling program codes by a processor of electronic equipment, wherein the program codes can be stored in a computer storage medium.

An embodiment of the present application provides a method for operating and maintaining a ship facility, and fig. 1 is a schematic implementation flow diagram of the method for operating and maintaining a ship facility provided in the embodiment of the present application, as shown in fig. 1, including:

step S1: acquiring actual operation data of all ship facilities in a preset time period;

in the embodiment of the application, the actual operation data of all the ship facilities can be acquired through a series of sensors, monitoring equipment and the like. The method comprises the steps of constructing an operation database, and transmitting actual operation data of all ship facilities in a preset time period to the operation database.

Step S2: classifying actual operation data of all ship facilities in a preset time period to obtain actual operation data of each ship facility in the preset time period;

in the embodiment of the application, the actual operation data can be classified by combining a hierarchical classification method and a decision tree algorithm. And if the actual operation data are stored in the operation database, classifying the actual operation data in the operation database.

Step S3: the actual operation data of the ship facilities in a preset time period is imported into a fault prediction model for prediction, and a prediction result is obtained;

step S4: and according to the prediction result, individually recommending all the ship facilities with the running states of fault states, and obtaining the maintenance scheme and corresponding maintenance personnel information of each fault ship facility.

In the embodiment of the application, when the prediction result of the ship facility is a fault state, a maintenance scheme can be recommended according to a fault type corresponding to the fault state, wherein a mapping relationship exists between the maintenance scheme and the fault type, the fault type corresponds to at least one maintenance scheme, and when a plurality of maintenance schemes exist, the maintenance scheme can be recommended in a random mode or the maintenance scheme with the highest maintenance success rate can be adopted. And when the prediction result of the ship facility is in a fault state, acquiring the position information of the ship facility, searching maintenance personnel on the ship according to the position information to obtain candidate maintenance personnel information, screening the candidate maintenance personnel information according to the fault type of the ship facility to obtain residual candidate maintenance personnel information, acquiring a distance value between the residual candidate maintenance personnel information and the position information, and taking the residual candidate maintenance personnel information corresponding to the minimum distance value as recommended maintenance personnel information. Or the candidate serviceman information can be determined according to the fault type, then the distance value between the candidate serviceman information and the position information is acquired by the root, and the remaining candidate serviceman information corresponding to the minimum distance value is used as recommended serviceman information.

In summary, by acquiring actual operation data of all the ship facilities in the preset time period, classifying the actual operation data of all the ship facilities in the preset time period to obtain the actual operation data of each ship facility in the preset time period, guiding the actual operation data of the ship facilities in the preset time period into a fault prediction model to predict, obtaining a prediction result, and performing personalized recommendation on the ship facilities with all the operation states being fault states according to the prediction result to obtain maintenance schemes and corresponding maintenance personnel information of each fault ship facility, decision support can be provided for operation personnel, reasonable operation and maintenance strategies are formulated, the downtime of the facilities is reduced, and the operation and maintenance rationality and the operation and maintenance efficiency are improved.

In some embodiments, the method comprises:

step S10: acquiring fault characteristic data of ship facilities when various faults occur, clustering the fault characteristic data, constructing a plurality of databases, and storing the clustered fault characteristic data into corresponding databases to obtain characteristic databases;

step S20: acquiring common characteristic rules of fault characteristic data in each characteristic database, and acquiring probability distribution matrixes of nodes in the Bayesian network according to the common characteristic rules;

Step S30: and constructing a fault prediction model according to the probability distribution matrix of each node in the Bayesian network.

In the embodiment of the application, a large amount of fault characteristic data can be obtained through a large data network, and the fault characteristic data have heterogeneity, so that clustering processing is needed for the fault characteristic data, and the fault characteristic data are data corresponding to ship facilities. Constructing a plurality of databases, and storing the clustered fault characteristic data into the corresponding databases to obtain the characteristic databases. By acquiring common feature rules among the fault feature data to learn probability distribution of corresponding nodes in the Bayesian network, a probability distribution matrix of each node can be constructed according to the probability distribution, wherein the probability distribution matrix is a form of probability distribution describing random variables, and each element represents probability when a specific value is measured corresponding to the random variables. And then, constructing a fault prediction model according to the probability distribution matrix of each node in the Bayesian network, so that the accuracy of the fault prediction model can be improved.

In some embodiments, the step S10 includes:

step S101: acquiring historical fault information of the ship facility based on a big data network, and extracting characteristics of the historical fault information of the ship facility to obtain fault characteristic data when various faults of the ship facility occur;

Step S102: introducing a fuzzy clustering algorithm, initializing fuzzy clustering centers, and calculating the mahalanobis distance between each fault characteristic data and each fuzzy clustering center;

step S103: sequencing the Marshall distance between each fault characteristic data and each fuzzy clustering center to obtain a sequencing result, and clustering the fault characteristic data with the Marshall distance smaller than the preset Marshall distance into the corresponding fuzzy clustering center according to the sequencing result;

step S104: constructing a plurality of databases, mapping fault characteristic data in different fuzzy clustering centers into corresponding databases for storage respectively, obtaining a plurality of characteristic databases, and updating each characteristic database periodically.

In the embodiment of the application, the fault characteristic data, such as voltage data, current data, transmitting power data and the like, corresponding to each ship facility when the communication equipment breaks down are obtained. And because of the heterogeneity of a large amount of fault characteristic data obtained through a large data network, the fault characteristic data are clustered through a fuzzy clustering algorithm, so that the data corresponding to the fault characteristic data are further distinguished, the fault characteristic data of different ship facilities are respectively mapped into corresponding databases for storage, the subsequent reading and comparison analysis of the corresponding data are facilitated, and the operation and maintenance speed is improved.

In some embodiments, the step S20 includes:

step S201: an isolated forest algorithm is introduced, an isolated score of each fault characteristic data in each characteristic database is calculated through the isolated forest algorithm, and fault characteristic data with the isolated score being larger than a preset isolated score are screened out in the corresponding characteristic database;

step S202: carrying out standardized processing on fault characteristic data in each characteristic database so that the fault characteristic data in each characteristic database have the same measurement, and analyzing common characteristic rules among the fault characteristic data with the same measurement through a gray correlation analysis method;

step S203: constructing a Bayesian network, acquiring various fault types of each ship facility through a big data network, determining partition attributes according to the fault types, and dividing the ship facilities into a plurality of nodes in the Bayesian network based on the partition attributes;

step S204: and introducing a machine learning algorithm, learning probability distribution of corresponding nodes in the Bayesian network by combining common characteristic rules among fault characteristic data with the same measurement, and constructing a probability distribution matrix of each node according to the probability distribution.

In the embodiment of the application, in the process of data storage or reading, some data may drift and distortion, so that a singular data phenomenon occurs, and therefore, before a fault prediction model is constructed through fault characteristic data in a database, the data in the database needs to be detected through an isolated forest algorithm so as to screen out the singular data, thereby improving the reliability of the constructed fault prediction model. Various fault types of each ship facility are obtained through a big data network, partition attributes are determined according to the fault types, a plurality of nodes are obtained by partitioning in the Bayesian network based on the partition attributes, specifically, variables to be considered are determined according to collected fault type information, the variables possibly comprise equipment parameters, running time, maintenance records and the like, and each variable represents one node in the network so as to partition the Bayesian network. And then learning probability distribution of corresponding nodes in the Bayesian network by means of machine learning algorithms such as logistic regression, support vector machines and K nearest neighbors and combining common characteristic rules among fault characteristic data with the same metric, and constructing a probability distribution matrix of each node according to the probability distribution, wherein the probability distribution matrix is a form of probability distribution describing random variables, and each element represents probability when the corresponding random variable takes a specific value.

In some embodiments, the step S30 includes:

step S301: performing dimension reduction on the probability distribution matrix by an independent component analysis method to obtain a probability distribution matrix consisting of one-dimensional vectors;

step S302: calculating the log likelihood function of each matrix data in the probability distribution matrix formed by the one-dimensional vectors through expectation maximization, and determining the expected score of each matrix data in the probability distribution matrix formed by the one-dimensional vectors according to the log likelihood function;

step S303: removing matrix data with expected scores not greater than a preset expected score from the probability distribution matrix, reserving matrix data with expected scores greater than the preset expected score in the probability distribution matrix, and generating a processed probability distribution matrix;

step S304: and importing the processed probability distribution matrix into the fault prediction model, carrying out iterative learning on the fault prediction model according to the processed probability distribution matrix and a genetic algorithm, and stopping iteration when the fault prediction model meets the preset requirement, and outputting the fault prediction model.

In the embodiment of the application, the probability distribution matrix is subjected to dimension reduction processing by an independent component analysis method (ICA algorithm) to obtain a probability distribution matrix consisting of one-dimensional vectors, wherein the probability distribution matrix possibly comprises a plurality of related variables, and the ICA aims at finding a group of uncorrelated independent variables (independent components), and potential independent features in the data can be found by carrying out ICA dimension reduction on the probability distribution matrix, extracted from the original data, mapped into a feature space which is easier to understand and interpret, thereby being beneficial to analyzing potential factors of the data, reducing storage and calculation cost and simplifying the subsequent data analysis and model establishment process. And then, correcting the probability distribution matrix through expected maximization to remove redundant data, so that the probability distribution matrix can describe the distribution characteristics of the data more accurately, and the accuracy of the model is improved.

In some embodiments, the step 2 includes:

step S21: constructing a decision tree trunk, acquiring the facility number of ship facilities, determining segmentation nodes of the decision tree trunk according to the facility number, segmenting the decision tree trunk according to the segmentation nodes, and updating the state of the segmented decision tree trunk to obtain a decision tree model;

step S22: acquiring relative position information of each branch in the decision tree model, defining a classified hierarchical structure according to the relative position information of each branch, and acquiring a hierarchical center of each hierarchical structure; wherein the hierarchy is a hierarchical tree structure, each hierarchy representing a particular class;

step S23: importing actual operation data in an operation database into the decision tree model, calculating Euclidean distances between each actual operation data and each level center, and sorting the Euclidean distances between each actual operation data and each level center based on numerical values;

step S24: after the sorting is completed, mapping each actual operation data to a hierarchical center corresponding to the shortest Euclidean distance so as to classify each actual operation data into the hierarchical structure to which the actual operation data belongs; after all the actual operation data are classified, cutting each branch in the decision tree model to obtain the actual operation data of each ship facility in a preset time period.

In the embodiment of the application, the actual operation data of each ship facility are collected in a preset time through a series of sensors, monitoring equipment and the like, an operation database is constructed, and the collected actual operation data are transmitted to the operation database. Hierarchical classification is based on dividing data into different hierarchical structures, so that classification is finer and more accurate, and different categories and sub-categories can be distinguished more accurately by layering the data according to different characteristics and attributes. The decision tree algorithm classifies the data based on the importance of the features and the splitting criterion, so that the data can be effectively segmented according to the importance of the features, and the classification accuracy is improved. The decision tree algorithm represents the classification rules in a tree structure, each node representing a feature, and each branch representing a classification result. The visual representation form enables the classification process to be more visual and easy to understand, can explain the basis of classification decisions, can provide more accurate classification results by combining a hierarchical classification method and a decision tree algorithm for classifying data, has higher interpretability, and simultaneously has better expandability and flexibility, improves the speed of data processing and improves the robustness of a system. The method can classify and process the acquired massive actual operation data so as to rapidly distinguish the actual operation data of each ship facility, and is convenient for predicting the fault state of each ship facility.

In some embodiments, the step S4 includes:

step S41: obtaining the running state of each ship facility according to the prediction result, and if the running state of the ship facility is a fault state, obtaining the fault type of the ship facility with the running state being the fault state;

step S42: generating a search tag according to the fault type, searching a big data network based on the search tag, searching to obtain a plurality of maintenance schemes, and obtaining the maintenance success rate of each maintenance scheme; sequencing the maintenance success rates of all the maintenance schemes, and recommending the maintenance scheme corresponding to the maximum maintenance success rate to a preset platform for display;

step S43: acquiring position information of ship facilities with a fault running state, and searching maintenance personnel on the ship according to the position information to obtain candidate maintenance personnel information;

step S44: calculating a hash value between the candidate maintenance personnel information and the fault type through a hash algorithm, and eliminating the candidate maintenance personnel information with the hash value not larger than a preset hash value to obtain the remaining candidate maintenance personnel information;

step S45: acquiring distance values between the residual candidate maintenance personnel information and the position information, sequencing each distance value, and recommending the residual candidate maintenance personnel information corresponding to the minimum distance value to a preset platform for display;

Step S46: repeating the steps until personalized recommendation is completed on all the ship facilities with the running states of fault states, and obtaining the maintenance scheme and corresponding maintenance personnel information of each fault ship facility.

In the embodiment of the application, the prediction result is obtained by guiding the actual operation data of each ship facility in the preset time period into the fault prediction model for prediction, and the prediction result comprises the fault state and the fault type of the ship facility. And then, according to the prediction result, individually recommending all the ship facilities with the running states of fault states, and obtaining the maintenance scheme and corresponding maintenance personnel information of each fault ship facility. By the method, the maintenance scheme and maintenance personnel which are suitable for the fault condition and the fault type of each facility can be recommended, the rationality in the operation and maintenance process is improved, and the operation and maintenance efficiency can be effectively improved.

In some embodiments, the method further comprises:

step S501: obtaining standard three-dimensional model diagrams corresponding to parts of different types in each ship facility through a big data network, constructing a standard part library, and importing the standard three-dimensional model diagrams corresponding to the parts of different types into the standard part library;

Step S502: acquiring image information of a fault part of a ship facility, identifying a signal of the fault part according to the image information, and searching the standard part library according to the signal of the fault part to obtain a standard three-dimensional model diagram corresponding to the fault part;

step S503: constructing a three-dimensional model diagram of the fault part according to the image information; establishing a three-dimensional coordinate system, and importing the three-dimensional model diagram of the fault part and the corresponding standard three-dimensional model diagram thereof into the three-dimensional coordinate system;

step S504: the positioning reference surfaces of the three-dimensional model diagrams of the fault part and the corresponding standard three-dimensional model diagrams are mutually overlapped, so that the three-dimensional model diagrams of the fault part and the corresponding standard three-dimensional model diagrams are aligned;

step S5: after alignment is finished, rejecting model areas where the three-dimensional model diagrams of the fault part and the corresponding standard three-dimensional model diagrams are overlapped in the three-dimensional coordinate system, and reserving model areas where the three-dimensional model diagrams of the fault part and the corresponding standard three-dimensional model diagrams are not overlapped to obtain a deviation model;

step S506: calculating a model volume value of the deviation model based on a grid method in the three-dimensional coordinate system, and if the model volume value is larger than a preset volume value, scrapping the fault part; and if the model volume value is not greater than the preset volume value, repairing the fault part.

In the embodiment of the application, if maintenance personnel have the condition of changing parts after carrying out operation and maintenance on each ship facility, the parts that can change at this moment are collected, then acquire the image information of these parts to compare these parts with standard part, obtain the deviation model, if the model volume value is not greater than predetermineeing the volume value, indicate that this part is not big with the difference of standard part, then repair this trouble part, if some parts probably are changed out because of the plane degree is inequality, but after polishing, this part is normally used, and this trouble part is repair this moment and is handled, can reduce the rejection cost for the resource can make full use of.

In some embodiments, the method further comprises:

step S11: acquiring a three-dimensional layout model diagram of a ship, retrieving the three-dimensional layout model diagram to obtain position information of strong electric facilities in the ship, and acquiring the facility types of the strong electric facilities;

step S12: acquiring the electromagnetic radiation intensity and the electromagnetic radiation range of the strong electric facility from big data according to the facility type of the strong electric facility, and constructing an electromagnetic radiation distribution diagram according to the electromagnetic radiation intensity and the electromagnetic radiation range;

Step S13: integrating the electromagnetic radiation distribution diagram of each strong electric facility with the three-dimensional layout model diagram to obtain an electromagnetic-terrain dynamic three-dimensional model diagram;

step S14: acquiring position node information of each ship facility to be acquired, acquiring position node information of a cloud platform, and marking the position node information of each ship facility to be acquired and the position node information of the cloud platform in the electromagnetic-terrain dynamic three-dimensional model diagram; on the basis of a ray tracing method, a plurality of data acquisition paths are simulated and planned;

step S15: acquiring electromagnetic radiation intensity values of all the data acquisition paths, and removing the data acquisition paths with the electromagnetic radiation intensity values larger than a preset intensity value to obtain the rest data acquisition paths;

step S16: acquiring path length values of all the residual data acquisition paths, sequencing the path length values of all the residual data acquisition paths to obtain the shortest acquisition path, and taking the residual data acquisition path corresponding to the shortest acquisition path as a final acquisition path;

step S17: repeating the steps until the final collection path of each ship facility to be collected is planned, and collecting the actual operation data of each ship facility within the preset time based on the final collection path of each ship facility to be collected.

In embodiments of the present application, ray tracing methods emanate from a signal source by simulating a large number of rays and track the propagation paths of these rays in the environment. During the propagation process, the phenomena such as reflection, refraction and diffraction occur when the ray intersects with objects such as barriers, buildings and the like, and the influence of the phenomena on signals is considered by a ray tracing method. By simulating and tracking the propagation paths of a large number of rays, the ray tracing method can generate simulation results of signal coverage and intensity distribution, thereby helping to plan a wireless signal acquisition path. The designed path can cover the target area to the greatest extent, and the influence of environmental factors on signal propagation is considered, so that the acquisition efficiency and the acquisition precision are improved. By the method, a reasonable wireless signal acquisition path can be planned, so that the phenomena of data drift and distortion in the data acquisition process are reduced, and the data reliability is improved.

Based on the foregoing embodiments, the embodiments of the present application provide a ship facility operation and maintenance device, where each module included in the device and each unit included in each module may be implemented by a processor in a computer device; of course, the method can also be realized by a specific logic circuit; in practice, the processor may be a central processing unit (CPU, central Processing Unit), a microprocessor (MPU, microprocessor Unit), a digital signal processor (DSP, digital Signal Processing), or a field programmable gate array (FPGA, field Programmable Gate Array), or the like.

An embodiment of the present application provides a ship facility operation and maintenance device, and fig. 2 is a schematic structural diagram of the ship facility operation and maintenance device provided in the embodiment of the present application, as shown in fig. 2, including:

the acquisition module is used for acquiring actual operation data of all ship facilities in a preset time period;

the classification module is used for classifying the actual operation data of all the ship facilities in the preset time period to obtain the actual operation data of each ship facility in the preset time period;

the prediction module is used for guiding actual operation data of the ship facility in a preset time period into the fault prediction model for prediction to obtain a prediction result;

and the recommendation module is used for individually recommending all the ship facilities with the running states of fault states according to the prediction result to obtain the maintenance scheme of each ship facility with the fault and the corresponding maintenance personnel information.

In some embodiments, the marine facility operation and maintenance device includes:

the clustering module is used for acquiring fault characteristic data when various faults occur to the ship facilities, carrying out clustering processing on the fault characteristic data, constructing a plurality of databases, and storing the clustered fault characteristic data into the corresponding databases to obtain characteristic databases;

The second acquisition module is used for acquiring the common characteristic rule of the fault characteristic data in each characteristic database and acquiring the probability distribution matrix of each node in the Bayesian network according to the common characteristic rule;

the construction module is used for constructing a fault prediction model according to the probability distribution matrix of each node in the Bayesian network.

In some embodiments, the clustering module comprises:

the characteristic extraction unit is used for acquiring historical fault information of the ship facility based on the big data network, and carrying out characteristic extraction on the historical fault information of the ship facility to obtain fault characteristic data when various faults occur to the ship facility;

the computing unit is used for introducing a fuzzy clustering algorithm, initializing fuzzy clustering centers and computing the mahalanobis distance between each fault characteristic data and each fuzzy clustering center;

the clustering unit is used for sequencing the Marshall distance between each fault characteristic data and each fuzzy clustering center to obtain a sequencing result, and clustering the fault characteristic data with the Marshall distance smaller than the preset Marshall distance into the corresponding fuzzy clustering center according to the sequencing result;

the construction unit is used for constructing a plurality of databases, mapping fault characteristic data in different fuzzy clustering centers into corresponding databases respectively for storage to obtain a plurality of characteristic databases, and updating each characteristic database periodically.

In some embodiments, the second acquisition module includes:

the screening unit is used for introducing an isolated forest algorithm, calculating the isolated score of each fault characteristic data in each characteristic database through the isolated forest algorithm, and screening out the fault characteristic data with the isolated score being larger than the preset isolated score in the corresponding characteristic database;

the analysis unit is used for carrying out standardization processing on the fault characteristic data in each characteristic database so that the fault characteristic data in each characteristic database have the same measurement, and analyzing the common characteristic rule among the fault characteristic data with the same measurement through a gray correlation analysis method;

the attribute dividing unit is used for constructing a Bayesian network, acquiring various fault types of each ship facility through a big data network, determining dividing attributes according to the fault types, and dividing the plurality of nodes in the Bayesian network based on the dividing attributes;

the second construction unit is used for introducing a machine learning algorithm, learning probability distribution of corresponding nodes in the Bayesian network by combining common characteristic rules among fault characteristic data with the same measurement, and constructing probability distribution matrixes of all the nodes according to the probability distribution.

In some embodiments, the building block comprises:

the dimension reduction processing unit is used for carrying out dimension reduction processing on the probability distribution matrix through an independent component analysis method to obtain a probability distribution matrix composed of one-dimensional vectors;

a determining unit, configured to determine, by desiring to maximize computation of a log-likelihood function of each matrix data in the probability distribution matrix composed of one-dimensional vectors, a desirability score of each matrix data in the probability distribution matrix composed of one-dimensional vectors according to the log-likelihood function;

the generation unit is used for eliminating matrix data with expected scores not greater than a preset expected score in the probability distribution matrix, reserving the matrix data with expected scores greater than the preset expected score in the probability distribution matrix, and generating a processed probability distribution matrix;

and the iteration unit is used for importing the processed probability distribution matrix into the fault prediction model, carrying out iterative learning on the fault prediction model according to the processed probability distribution matrix and a genetic algorithm, stopping iteration after the fault prediction model meets the preset requirement, and outputting the fault prediction model.

In some embodiments, the classification module comprises:

The segmentation unit is used for constructing a decision tree trunk, acquiring the facility number of ship facilities, determining segmentation nodes of the decision tree trunk according to the facility number, segmenting the decision tree trunk according to the segmentation nodes, and updating the state of the segmented decision tree trunk to obtain a decision tree model;

the acquisition unit is used for acquiring the relative position information of each branch in the decision tree model, defining a classified hierarchical structure according to the relative position information of each branch and acquiring the hierarchical center of each hierarchical structure; wherein the hierarchy is a hierarchical tree structure, each hierarchy representing a particular class;

the sorting unit is used for importing actual operation data in the operation database into the decision tree model, calculating Euclidean distances between each actual operation data and each level center, and sorting the Euclidean distances between each actual operation data and each level center based on the values;

the cutting unit is used for mapping each actual operation data to a hierarchy center corresponding to the shortest Euclidean distance after the sequencing is completed so as to classify each actual operation data into the hierarchy structure to which the actual operation data belongs; after all the actual operation data are classified, cutting each branch in the decision tree model to obtain the actual operation data of each ship facility in a preset time period.

In some embodiments, the recommendation module includes:

the second obtaining unit is used for obtaining the running state of each ship facility according to the prediction result, and obtaining the fault type of the ship facility with the running state being the fault state if the running state of the ship facility is the fault state;

the recommending unit is used for generating a search tag according to the fault type, searching the big data network based on the search tag, searching to obtain a plurality of maintenance schemes, and obtaining the maintenance success rate of each maintenance scheme; sequencing the maintenance success rates of all the maintenance schemes, and recommending the maintenance scheme corresponding to the maximum maintenance success rate to a preset platform for display;

the searching unit is used for acquiring the position information of the ship facility with the running state being the fault state, searching maintenance personnel on the ship according to the position information, and obtaining candidate maintenance personnel information;

the rejecting unit is used for calculating the hash value between the candidate maintenance personnel information and the fault type through a hash algorithm, and rejecting the candidate maintenance personnel information with the hash value not larger than a preset hash value to obtain the remaining candidate maintenance personnel information;

the second recommending unit is used for acquiring distance values between the residual candidate maintenance personnel information and the position information, sequencing the distance values, recommending the residual candidate maintenance personnel information corresponding to the minimum distance value to a preset platform and displaying the residual candidate maintenance personnel information;

And the repeating unit is used for repeating the steps until personalized recommendation is completed on all the ship facilities with the running states of fault states, and obtaining the maintenance scheme and corresponding maintenance personnel information of each fault ship facility.

In some embodiments, the marine facility operation and maintenance device further comprises:

the second construction module is used for acquiring standard three-dimensional model diagrams corresponding to parts of different types in each ship facility through a big data network, constructing a standard part library, and importing the standard three-dimensional model diagrams corresponding to the parts of different types into the standard part library;

the retrieval module is used for acquiring image information of the fault parts of the ship facility, identifying signals of the fault parts according to the image information, and retrieving the standard part library according to the signals of the fault parts to obtain a standard three-dimensional model diagram corresponding to the fault parts;

the third construction module is used for constructing a three-dimensional model diagram of the fault part according to the image information; establishing a three-dimensional coordinate system, and importing the three-dimensional model diagram of the fault part and the corresponding standard three-dimensional model diagram thereof into the three-dimensional coordinate system;

the alignment module is used for enabling the positioning reference surfaces of the three-dimensional model diagrams of the fault part and the corresponding standard three-dimensional model diagrams to coincide with each other so as to perform alignment processing on the three-dimensional model diagrams of the fault part and the corresponding standard three-dimensional model diagrams;

The deviation module is used for removing the model area where the three-dimensional model image of the fault part and the corresponding standard three-dimensional model image are overlapped in the three-dimensional coordinate system after alignment is finished, and reserving the model area where the three-dimensional model image of the fault part and the corresponding standard three-dimensional model image are not overlapped to obtain a deviation model;

the calculation module is used for calculating a model volume value of the deviation model based on a grid method in the three-dimensional coordinate system, and if the model volume value is larger than a preset volume value, the fault part is scrapped; and if the model volume value is not greater than the preset volume value, repairing the fault part.

In some embodiments, the acquisition module further comprises:

the second retrieval unit is used for acquiring a three-dimensional layout model diagram of the ship, retrieving the three-dimensional layout model diagram to obtain the position information of the strong electric facilities in the ship, and acquiring the facility types of the strong electric facilities;

the third construction unit is used for acquiring the electromagnetic radiation intensity and the electromagnetic radiation range of the strong electric facility from big data according to the facility type of the strong electric facility and constructing an electromagnetic radiation distribution diagram according to the electromagnetic radiation intensity and the electromagnetic radiation range;

The integration unit is used for integrating the electromagnetic radiation distribution diagram of each strong electric facility with the three-dimensional layout model diagram to obtain an electromagnetic-terrain dynamic three-dimensional model diagram;

the simulation unit is used for acquiring the position node information of each ship facility to be acquired, acquiring the position node information of the cloud platform and marking the position node information of each ship facility to be acquired and the position node information of the cloud platform in the electromagnetic-terrain dynamic three-dimensional model diagram; on the basis of a ray tracing method, a plurality of data acquisition paths are simulated and planned;

the second eliminating unit is used for obtaining electromagnetic radiation intensity values of all the data acquisition paths, eliminating the data acquisition paths with the electromagnetic radiation intensity values larger than the preset intensity values, and obtaining residual data acquisition paths;

the second sorting unit is used for obtaining the path length value of each residual data acquisition path, sorting the path length value of each residual data acquisition path to obtain the shortest acquisition path, and taking the residual data acquisition path corresponding to the shortest acquisition path as the final acquisition path;

and the second repeating unit is used for repeating the steps until the final acquisition path of each ship facility to be acquired is planned, and acquiring the actual operation data of each ship facility within the preset time based on the final acquisition path of each ship facility to be acquired.

In the embodiment of the present application, if the above-mentioned method for operating and maintaining the ship facility is implemented in the form of a software functional module, the method may also be stored in a computer readable storage medium when sold or used as a separate product. Based on such understanding, the technical solutions of the embodiments of the present application may be essentially or partly contributing to the prior art, and the computer software product may be stored in a storage medium, and include several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read Only Memory (ROM), a magnetic disk, an optical disk, or other various media capable of storing program codes. Thus, embodiments of the present application are not limited to any specific combination of hardware and software.

Accordingly, an embodiment of the present application provides a storage medium having a computer program stored thereon, characterized in that the computer program, when executed by a processor, implements the steps in the marine facility operation and maintenance method provided in the above embodiment.

The embodiment of the application provides electronic equipment; fig. 3 is a schematic diagram of a composition structure of an electronic device according to an embodiment of the present application, as shown in fig. 3, the electronic device 400 includes: a processor 401, at least one communication bus 402, a user interface 403, at least one external communication interface 404, a memory 405. Wherein communication bus 402 is configured to enable connected communications between these components. The user interface 403 may include a display screen, and the external communication interface 404 may include a standard wired interface and a wireless interface, among others. The processor 401 is configured to execute a program of the ship facility operation and maintenance method stored in the memory to implement the steps in the ship facility operation and maintenance method provided in the above-described embodiment.

It should be noted here that: the description of the storage medium, the electronic device embodiments, and the description of the method embodiments above are similar, with similar benefits as the method embodiments. For technical details not disclosed in the embodiments of the storage medium and the apparatus of the present application, please refer to the description of the method embodiments of the present application for understanding.

It should be appreciated that reference throughout this specification to "one embodiment" or "an embodiment" means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the present application. Thus, the appearances of the phrases "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. It should be understood that, in various embodiments of the present application, the sequence numbers of the foregoing processes do not mean the order of execution, and the order of execution of the processes should be determined by the functions and internal logic thereof, and should not constitute any limitation on the implementation process of the embodiments of the present application. The foregoing embodiment numbers of the present application are merely for describing, and do not represent advantages or disadvantages of the embodiments.

It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, object or apparatus comprising such element.

In the several embodiments provided in this application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above described device embodiments are only illustrative, e.g. the division of the units is only one logical function division, and there may be other divisions in practice, such as: multiple units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. In addition, the various components shown or discussed may be coupled or directly coupled or communicatively coupled to each other via some interface, whether indirectly coupled or communicatively coupled to devices or units, whether electrically, mechanically, or otherwise.

The units described above as separate components may or may not be physically separate, and components shown as units may or may not be physical units; can be located in one place or distributed to a plurality of network units; some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present application may be integrated in one processing unit, or each unit may be separately used as one unit, or two or more units may be integrated in one unit; the integrated units may be implemented in hardware or in hardware plus software functional units.

Those of ordinary skill in the art will appreciate that: all or part of the steps for implementing the above method embodiments may be implemented by hardware related to program instructions, and the foregoing program may be stored in a computer readable storage medium, where the program, when executed, performs steps including the above method embodiments; and the aforementioned storage medium includes: a mobile storage device, a Read Only Memory (ROM), a magnetic disk or an optical disk, or the like, which can store program codes.

Alternatively, the integrated units described above may be stored in a computer readable storage medium if implemented in the form of software functional modules and sold or used as a stand-alone product. Based on such understanding, the technical solutions of the embodiments of the present application may be essentially or partly contributing to the prior art, embodied in the form of a software product stored in a storage medium, comprising several instructions for causing a controller to execute all or part of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a removable storage device, a ROM, a magnetic disk, or an optical disk.

The foregoing is merely an embodiment of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think about changes or substitutions within the technical scope of the present application, and the changes and substitutions are intended to be covered in the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims (10)

1. A method of operating and maintaining a marine facility, comprising:

Acquiring actual operation data of all ship facilities in a preset time period;

classifying actual operation data of all ship facilities in a preset time period to obtain actual operation data of each ship facility in the preset time period;

the actual operation data of the ship facilities in a preset time period is imported into a fault prediction model for prediction, and a prediction result is obtained;

and according to the prediction result, individually recommending all the ship facilities with the running states of fault states, and obtaining the maintenance scheme and corresponding maintenance personnel information of each fault ship facility.

2. The method according to claim 1, characterized in that the method comprises:

acquiring fault characteristic data of ship facilities when various faults occur, clustering the fault characteristic data, constructing a plurality of databases, and storing the clustered fault characteristic data into corresponding databases to obtain characteristic databases;

acquiring common characteristic rules of fault characteristic data in each characteristic database, and acquiring probability distribution matrixes of nodes in the Bayesian network according to the common characteristic rules;

and constructing a fault prediction model according to the probability distribution matrix of each node in the Bayesian network.

3. The method according to claim 2, wherein the obtaining fault feature data of the ship facility when various faults occur, clustering the fault feature data, constructing a plurality of databases, and storing the clustered fault feature data in the corresponding databases to obtain feature databases, includes:

acquiring historical fault information of the ship facility based on a big data network, and extracting characteristics of the historical fault information of the ship facility to obtain fault characteristic data when various faults of the ship facility occur;

introducing a fuzzy clustering algorithm, initializing fuzzy clustering centers, and calculating the mahalanobis distance between each fault characteristic data and each fuzzy clustering center;

sequencing the Marshall distance between each fault characteristic data and each fuzzy clustering center to obtain a sequencing result, and clustering the fault characteristic data with the Marshall distance smaller than the preset Marshall distance into the corresponding fuzzy clustering center according to the sequencing result;

constructing a plurality of databases, mapping fault characteristic data in different fuzzy clustering centers into corresponding databases for storage respectively, obtaining a plurality of characteristic databases, and updating each characteristic database periodically.

4. The method according to claim 2, wherein the obtaining the common feature rule of the fault feature data in each feature database, and obtaining the probability distribution matrix of each node in the bayesian network according to the common feature rule, includes:

an isolated forest algorithm is introduced, an isolated score of each fault characteristic data in each characteristic database is calculated through the isolated forest algorithm, and fault characteristic data with the isolated score being larger than a preset isolated score are screened out in the corresponding characteristic database;

carrying out standardized processing on fault characteristic data in each characteristic database so that the fault characteristic data in each characteristic database have the same measurement, and analyzing common characteristic rules among the fault characteristic data with the same measurement through a gray correlation analysis method;

constructing a Bayesian network, acquiring various fault types of each ship facility through a big data network, determining partition attributes according to the fault types, and dividing the ship facilities into a plurality of nodes in the Bayesian network based on the partition attributes;

and introducing a machine learning algorithm, learning probability distribution of corresponding nodes in the Bayesian network by combining common characteristic rules among fault characteristic data with the same measurement, and constructing a probability distribution matrix of each node according to the probability distribution.

5. The method according to claim 2, wherein the constructing a fault prediction model according to the probability distribution matrix of each node in the bayesian network comprises:

performing dimension reduction on the probability distribution matrix by an independent component analysis method to obtain a probability distribution matrix consisting of one-dimensional vectors;

calculating the log likelihood function of each matrix data in the probability distribution matrix formed by the one-dimensional vectors through expectation maximization, and determining the expected score of each matrix data in the probability distribution matrix formed by the one-dimensional vectors according to the log likelihood function;

removing matrix data with expected scores not greater than a preset expected score from the probability distribution matrix, reserving matrix data with expected scores greater than the preset expected score in the probability distribution matrix, and generating a processed probability distribution matrix;

and importing the processed probability distribution matrix into the fault prediction model, carrying out iterative learning on the fault prediction model according to the processed probability distribution matrix and a genetic algorithm, and stopping iteration when the fault prediction model meets the preset requirement, and outputting the fault prediction model.

6. The method according to claim 1, wherein classifying the actual operation data of all the ship facilities in the preset time period to obtain the actual operation data of each ship facility in the preset time period includes:

Constructing a decision tree trunk, acquiring the facility number of ship facilities, determining segmentation nodes of the decision tree trunk according to the facility number, segmenting the decision tree trunk according to the segmentation nodes, and updating the state of the segmented decision tree trunk to obtain a decision tree model;

acquiring relative position information of each branch in the decision tree model, defining a classified hierarchical structure according to the relative position information of each branch, and acquiring a hierarchical center of each hierarchical structure; wherein the hierarchy is a hierarchical tree structure, each hierarchy representing a particular class;

importing actual operation data in an operation database into the decision tree model, calculating Euclidean distances between each actual operation data and each level center, and sorting the Euclidean distances between each actual operation data and each level center based on numerical values;

after the sorting is completed, mapping each actual operation data to a hierarchical center corresponding to the shortest Euclidean distance so as to classify each actual operation data into the hierarchical structure to which the actual operation data belongs; after all the actual operation data are classified, cutting each branch in the decision tree model to obtain the actual operation data of each ship facility in a preset time period.

7. The method according to claim 1, wherein the individually recommending all the ship facilities with the running states being the fault states according to the prediction result to obtain the maintenance scheme and the corresponding maintenance personnel information of each fault ship facility, comprises:

obtaining the running state of each ship facility according to the prediction result, and if the running state of the ship facility is a fault state, obtaining the fault type of the ship facility with the running state being the fault state;

generating a search tag according to the fault type, searching a big data network based on the search tag, searching to obtain a plurality of maintenance schemes, and obtaining the maintenance success rate of each maintenance scheme; sequencing the maintenance success rates of all the maintenance schemes, and recommending the maintenance scheme corresponding to the maximum maintenance success rate to a preset platform for display;

acquiring position information of ship facilities with a fault running state, and searching maintenance personnel on the ship according to the position information to obtain candidate maintenance personnel information;

calculating a hash value between the candidate maintenance personnel information and the fault type through a hash algorithm, and eliminating the candidate maintenance personnel information with the hash value not larger than a preset hash value to obtain the remaining candidate maintenance personnel information;

Acquiring distance values between the residual candidate maintenance personnel information and the position information, sequencing each distance value, and recommending the residual candidate maintenance personnel information corresponding to the minimum distance value to a preset platform for display;

repeating the steps until personalized recommendation is completed on all the ship facilities with the running states of fault states, and obtaining the maintenance scheme and corresponding maintenance personnel information of each fault ship facility.

8. The utility model provides a marine facility fortune dimension device which characterized in that is applied to heat preservation device includes:

the acquisition module is used for acquiring actual operation data of all ship facilities in a preset time period;

the classification module is used for classifying the actual operation data of all the ship facilities in the preset time period to obtain the actual operation data of each ship facility in the preset time period;

the prediction module is used for guiding actual operation data of the ship facility in a preset time period into the fault prediction model for prediction to obtain a prediction result;

and the recommendation module is used for individually recommending all the ship facilities with the running states of fault states according to the prediction result to obtain the maintenance scheme of each ship facility with the fault and the corresponding maintenance personnel information.

9. An electronic device, comprising:

a memory and a processor, the memory having stored thereon a computer program which, when executed by the processor, performs the method of any of claims 1 to 7.

10. A storage medium storing a computer program executable by one or more processors for implementing a method as claimed in any one of claims 1 to 7.