A method and system for monitoring and evaluating marine navigation aids based on data analysis

Through data analysis-based methods, real-time monitoring and evaluation of the status of maritime beacons is solved, and the high cost and delayed response problems caused by traditional manual inspection dependence are achieved, and the good maintenance of beacons under harsh conditions is achieved, and accident risk and maintenance needs are reduced.

CN117493791BActive Publication Date: 2024-12-06WENZHOU NAVIGATION MARK OFFICE EAST CHINA SEA NAVIGATION SUPPORT CENT MINISTRY OF TRANSPORT
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311305199.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-10-10
Publication Date
2024-12-06
Estimated Expiration
2043-10-10

AI Technical Summary

Technical Problem

Traditional maritime navigation beacon monitoring and evaluation methods rely on manual inspection, resulting in high costs, delayed response time, subjective factors and the inability to respond to changes in the maritime environment in real time, affecting the reliability of the assessment results.

Method used

By using data analysis, the actual operating data of the beacon is obtained and processed, dimensionality reduction processing, noise point screening, and evaluation results are compared and analyzed. The big data network and gray correlation analysis method are used to monitor and evaluate the beacon status in real time, and automatically Implementation of regulatory measures.

Benefits of technology

Real-time tracking of maritime beacon status and timely correction of problems, reduce dependence on manual inspection, save human resources and costs, improve the visibility and availability of beacons, ensure the good condition of beacons under harsh conditions, and reduce accidents Risk and maintenance requirements.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117493791B_ABST
    Figure CN117493791B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of marine navigation technology, and in particular to a marine navigation monitoring and evaluation method and system based on data analysis. The actual operation data of the target navigation mark is obtained within a preset time period, and the actual operation data is subjected to dimensionality reduction processing to obtain vector data corresponding to each actual operation data, and each vector data is collected to obtain a vector data set; the characteristics of each vector data in the vector data set are obtained, noise points are retrieved based on the characteristics of each vector data, and the noise points are removed to filter out the outlier actual operation data to obtain the filtered actual operation data; if the evaluation result is the first evaluation result, the actual operation data of the target navigation mark is regulated, and the marine navigation monitoring and evaluation method based on data analysis can improve the safety, efficiency and sustainability of maritime navigation, reduce maintenance costs, and reduce accident risks.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of marine navigation marks, and in particular to a marine navigation mark monitoring and evaluation method and system based on data analysis. Background Art

[0002] Marine aids to navigation are a key navigation tool used to guide ships through waterways safely, warn of dangerous areas, and mark port entrances. They play an indispensable role in global trade, shipping operations, and maritime safety. However, traditional methods for monitoring and evaluating marine aids to navigation have a series of challenges. First, traditional aids to navigation require regular manual monitoring and evaluation, which not only requires a lot of human resources, but also incurs high costs, especially in remote or harsh marine environmental conditions. Second, monitoring methods that rely on manual patrols may not be able to detect problems or anomalies immediately, thereby delaying the response time for regulation and maintenance, which may increase the risk of accidents. Third, manual monitoring is susceptible to subjective factors, such as the skill level and attention of the patrolman, which may lead to inconsistent and inaccurate data. Fourth, marine environmental conditions may change rapidly, including meteorological, tidal, and hydrological conditions, and traditional methods often cannot respond to these changes in real time, thus affecting the reliability of the evaluation results. In order to overcome these challenges, this application proposes a method and system for monitoring and evaluating marine aids to navigation based on data analysis. Summary of the invention

[0003] The present invention overcomes the deficiencies of the prior art and provides a method and system for monitoring and evaluating marine navigation aids based on data analysis.

[0004] To achieve the above-mentioned purpose, the technical solution adopted by the present invention is:

[0005] The first aspect of the present invention discloses a method for monitoring and evaluating marine navigation aids based on data analysis, comprising the following steps:

[0006] Acquiring an actual working environment of the target navigation mark within a preset time period, and determining preset operating data of the target navigation mark within the preset time period according to the actual working environment;

[0007] Acquire actual operation data of the target navigation mark within a preset time period, perform dimensionality reduction processing on the actual operation data to obtain vector data corresponding to each actual operation data, and collect each vector data to obtain a vector data set;

[0008] Acquire features of each vector data in the vector data set, retrieve noise points based on the features of each vector data, and remove the noise points to filter out outlier actual operation data to obtain filtered actual operation data;

[0009] The preset operating data is compared and analyzed with the screened actual operating data to obtain a first evaluation result or a second evaluation result; if the evaluation result is the first evaluation result, the actual operating data of the target beacon is regulated; if the evaluation result is the second evaluation result, the actual operating data of the target beacon is not regulated.

[0010] Furthermore, in a preferred embodiment of the present invention, the actual working environment of the target navigation mark is obtained within a preset time period, and the preset operating data of the target navigation mark within the preset time period is determined according to the actual working environment, specifically:

[0011] Acquire standard operation data of target navigation marks under various preset working environment conditions through a big data network, construct a database, and import the standard operation data of navigation marks under various preset working environment conditions into the database to obtain a standard database;

[0012] Acquire the actual working environment of the target navigation mark within a preset time period, import the actual working environment into the standard database, calculate the similarity between the actual working environment and various preset working environments by grey correlation analysis method, and obtain multiple similarities;

[0013] Constructing a sorting table, importing the plurality of similarities into the sorting table for sorting based on numerical values, and extracting the maximum similarity after the sorting is completed; obtaining a preset working environment corresponding to the maximum similarity, and constructing a search tag according to the preset working environment corresponding to the maximum similarity;

[0014] The standard operating data of the target navigation mark under the actual working environment is retrieved in the standard database according to the retrieval tag, and the standard operating data of the target navigation mark under the actual working environment is marked as the preset operating data within the preset time period.

[0015] Furthermore, in a preferred embodiment of the present invention, the actual operation data of the target navigation mark is obtained within a preset time period, and the actual operation data is subjected to dimensionality reduction processing to obtain vector data corresponding to each actual operation data, specifically:

[0016] Extracting features of each actual operation data, and performing standardization on the actual operation data to ensure that the mean of each feature is 0 and the variance is 1, thereby obtaining standardized actual operation data; wherein the features include numerical features, category features, time features, and text features;

[0017] Obtaining a covariance matrix of the standardized actual operation data, and performing eigendecomposition on the covariance matrix to obtain eigenvalues ​​and eigenvectors; wherein the eigenvalues ​​represent the importance of the principal components, and the eigenvectors represent the directions of the principal components;

[0018] Arrange the eigenvalues ​​in descending order to determine the order of the principal components. The eigenvector corresponding to the largest eigenvalue is the first principal component, and so on to obtain several principal components.

[0019] The number of principal components to be retained is determined according to a preset threshold to obtain the filtered principal components; a projection matrix is ​​constructed according to the eigenvectors corresponding to the filtered principal components, and the standardized actual operation data is multiplied by the projection matrix to project the standardized actual operation data into a new low-dimensional space to obtain vector data corresponding to each standardized actual operation data.

[0020] Furthermore, in a preferred embodiment of the present invention, the features of each vector data in the vector data set are obtained, noise points are retrieved based on the features of each vector data, and the noise points are removed to filter out the outlier actual operation data to obtain the filtered actual operation data, specifically:

[0021] Acquire the features of each vector data in the vector data set, and calculate the mean of the features of each vector data to obtain a mean vector containing a plurality of means; wherein the mean vector is the data center of the vector data set;

[0022] Taking the data center as the coordinate origin, constructing a three-dimensional coordinate system according to the coordinate origin, importing the vector data set into the three-dimensional coordinate system, and obtaining the three-dimensional coordinate value corresponding to each vector data in the vector data set in the three-dimensional coordinate system;

[0023] Calculate the Euclidean distance between each vector data and the data center according to the three-dimensional coordinate value, and compare the Euclidean distance between each vector data and the data center with a preset Euclidean distance;

[0024] Marking vector data corresponding to a Euclidean distance greater than a preset Euclidean distance as noise points, and removing each noise point from the vector data set to obtain vector data after outlier screening;

[0025] The actual operation data corresponding to the vector data after outlier screening is extracted to screen the outlier actual operation data to obtain screened actual operation data.

[0026] Furthermore, in a preferred embodiment of the present invention, the preset operation data is compared and analyzed with the screened actual operation data to obtain a first evaluation result or a second evaluation result, specifically:

[0027] Sorting the preset operation data based on the time series to obtain the preset operation data based on the time series, and constructing a preset operation data curve chart according to the preset operation data based on the time series;

[0028] Sorting the filtered actual operation data based on time series to obtain the filtered actual operation data based on time series, and constructing an actual operation data curve chart according to the filtered actual operation data based on time series;

[0029] Constructing a virtual pairing space, and importing the preset operation data curve graph and the actual operation data curve graph into the virtual pairing space for pairing;

[0030] After the pairing is completed, the matching degree between the preset operation data and the screened actual operation data is calculated based on the length of the curve that overlaps and the length of the curve that does not overlap between the preset operation data curve graph and the actual operation data curve graph;

[0031] If the matching degree is not greater than the preset matching degree, it means that the operation state of the target navigation mark within the preset time period is abnormal, and a first evaluation result is generated;

[0032] If the matching degree is greater than the preset matching degree, it means that the operating status of the target navigation mark within the preset time period is normal, and a second evaluation result is generated.

[0033] Furthermore, in a preferred embodiment of the present invention, if the evaluation result is the first evaluation result, the actual operation data of the target navigation mark is regulated, specifically:

[0034] Obtain the optimal control measures corresponding to the target navigation mark under various preset abnormal operation data combination conditions through the big data network, construct a knowledge graph, and import the optimal control measures corresponding to the target navigation mark under various preset abnormal operation data combination conditions into the knowledge graph;

[0035] If the matching degree is not greater than the preset matching degree, then obtaining the non-overlapping curve area in the preset operation data curve graph and the actual operation data curve graph, marking the non-overlapping curve area as the abnormal operation period, and marking the actual operation data corresponding to the abnormal operation period in the actual operation data curve graph as abnormal operation data;

[0036] Importing the abnormal operation data into the knowledge graph, and calculating the similarity between the abnormal operation data and various preset abnormal operation data combinations by grey correlation analysis to obtain multiple similarities;

[0037] A maximum similarity is extracted from multiple similarities, a preset operation abnormality data combination corresponding to the maximum similarity is obtained, and the optimal control measures of the target beacon when the current operation state is abnormal are determined based on the preset operation abnormality data combination corresponding to the maximum similarity, and the optimal control measures are transmitted to the control terminal of the target beacon.

[0038] The second aspect of the present invention discloses a marine navigation mark monitoring and evaluation system based on data analysis, the marine navigation mark monitoring and evaluation system comprises a memory and a processor, the memory stores a marine navigation mark monitoring and evaluation method program, when the marine navigation mark monitoring and evaluation method program is executed by the processor, the following steps are implemented:

[0039] Acquiring an actual working environment of the target navigation mark within a preset time period, and determining preset operating data of the target navigation mark within the preset time period according to the actual working environment;

[0040] Acquire actual operation data of the target navigation mark within a preset time period, perform dimensionality reduction processing on the actual operation data to obtain vector data corresponding to each actual operation data, and collect each vector data to obtain a vector data set;

[0041] Acquire features of each vector data in the vector data set, retrieve noise points based on the features of each vector data, and remove the noise points to filter out outlier actual operation data to obtain filtered actual operation data;

[0042] The preset operating data is compared and analyzed with the screened actual operating data to obtain a first evaluation result or a second evaluation result; if the evaluation result is the first evaluation result, the actual operating data of the target beacon is regulated; if the evaluation result is the second evaluation result, the actual operating data of the target beacon is not regulated.

[0043] Furthermore, in a preferred embodiment of the present invention, the actual working environment of the target navigation mark is obtained within a preset time period, and the preset operating data of the target navigation mark within the preset time period is determined according to the actual working environment, specifically:

[0044] Acquire standard operation data of target navigation marks under various preset working environment conditions through a big data network, construct a database, and import the standard operation data of navigation marks under various preset working environment conditions into the database to obtain a standard database;

[0045] Acquire the actual working environment of the target navigation mark within a preset time period, import the actual working environment into the standard database, calculate the similarity between the actual working environment and various preset working environments by grey correlation analysis method, and obtain multiple similarities;

[0046] Constructing a sorting table, importing the plurality of similarities into the sorting table for sorting based on numerical values, and extracting the maximum similarity after the sorting is completed; obtaining a preset working environment corresponding to the maximum similarity, and constructing a search tag according to the preset working environment corresponding to the maximum similarity;

[0047] The standard operating data of the target navigation mark under the actual working environment is retrieved in the standard database according to the retrieval tag, and the standard operating data of the target navigation mark under the actual working environment is marked as the preset operating data within the preset time period.

[0048] Furthermore, in a preferred embodiment of the present invention, the preset operation data is compared and analyzed with the screened actual operation data to obtain a first evaluation result or a second evaluation result, specifically:

[0049] Sorting the preset operation data based on the time series to obtain the preset operation data based on the time series, and constructing a preset operation data curve chart according to the preset operation data based on the time series;

[0050] Sorting the filtered actual operation data based on time series to obtain the filtered actual operation data based on time series, and constructing an actual operation data curve chart according to the filtered actual operation data based on time series;

[0051] Constructing a virtual pairing space, and importing the preset operation data curve graph and the actual operation data curve graph into the virtual pairing space for pairing;

[0052] After the pairing is completed, the matching degree between the preset operation data and the screened actual operation data is calculated based on the length of the curve that overlaps and the length of the curve that does not overlap between the preset operation data curve graph and the actual operation data curve graph;

[0053] If the matching degree is not greater than the preset matching degree, it means that the operation state of the target navigation mark within the preset time period is abnormal, and a first evaluation result is generated;

[0054] If the matching degree is greater than the preset matching degree, it means that the operating status of the target navigation mark within the preset time period is normal, and a second evaluation result is generated.

[0055] Furthermore, in a preferred embodiment of the present invention, if the evaluation result is the first evaluation result, the actual operation data of the target navigation mark is regulated, specifically:

[0056] Obtain the optimal control measures corresponding to the target navigation mark under various preset abnormal operation data combination conditions through the big data network, construct a knowledge graph, and import the optimal control measures corresponding to the target navigation mark under various preset abnormal operation data combination conditions into the knowledge graph;

[0057] If the matching degree is not greater than the preset matching degree, then obtaining the non-overlapping curve area in the preset operation data curve graph and the actual operation data curve graph, marking the non-overlapping curve area as the abnormal operation period, and marking the actual operation data corresponding to the abnormal operation period in the actual operation data curve graph as abnormal operation data;

[0058] Importing the abnormal operation data into the knowledge graph, and calculating the similarity between the abnormal operation data and various preset abnormal operation data combinations by grey correlation analysis to obtain multiple similarities;

[0059] A maximum similarity is extracted from multiple similarities, a preset operation abnormality data combination corresponding to the maximum similarity is obtained, and the optimal control measures of the target beacon when the current operation state is abnormal are determined based on the preset operation abnormality data combination corresponding to the maximum similarity, and the optimal control measures are transmitted to the control terminal of the target beacon.

[0060] The present invention solves the technical defects existing in the background technology, and the present invention has the following beneficial effects: By analyzing the operating data of the target navigation mark, the method can track the status and performance of the marine navigation mark in real time, which helps to discover and correct problems in a timely manner, and ensure that the navigation mark maintains good visibility and availability under bad weather and other challenging conditions. The monitoring and evaluation system based on data analysis can reduce the dependence on manual inspections, thereby saving human resources and costs. By analyzing the status of the navigation mark, a more effective maintenance plan can be formulated, which helps to extend the life of the navigation mark equipment, reduce unplanned maintenance needs, and reduce maintenance costs. In general, the marine navigation mark monitoring and evaluation method based on data analysis can improve the safety, efficiency and sustainability of maritime navigation, reduce maintenance costs, reduce accident risks, and better meet the needs of maritime traffic, which plays an important role in ensuring the smoothness of maritime traffic and the safety of navigation. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0062] Figure 1 A first method flow chart of a method for monitoring and evaluating marine aids to navigation based on data analysis;

[0063] Figure 2 A second method flow chart of a method for monitoring and evaluating marine aids to navigation based on data analysis;

[0064] Figure 3 A third method flow chart of a method for monitoring and evaluating marine aids to navigation based on data analysis;

[0065] Figure 4 This is a system block diagram of a marine navigation aid monitoring and evaluation system based on data analysis. DETAILED DESCRIPTION

[0066] In order to more clearly understand the above-mentioned purpose, 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 the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0067] 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 protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0068] like Figure 1 As shown, the first aspect of the present invention discloses a method for monitoring and evaluating marine navigation aids based on data analysis, comprising the following steps:

[0069] S102: Acquire the actual working environment of the target navigation mark within a preset time period, and determine preset operating data of the target navigation mark within the preset time period according to the actual working environment;

[0070] S104: acquiring actual operation data of the target navigation mark within a preset time period, performing dimensionality reduction processing on the actual operation data to obtain vector data corresponding to each actual operation data, and aggregating each vector data to obtain a vector data set;

[0071] S106: Acquire features of each vector data in the vector data set, retrieve noise points based on the features of each vector data, and remove the noise points to filter out outlier actual operation data, thereby obtaining filtered actual operation data;

[0072] S108: Compare and analyze the preset operating data with the screened actual operating data to obtain a first evaluation result or a second evaluation result; if the evaluation result is the first evaluation result, adjust the actual operating data of the target beacon; if the evaluation result is the second evaluation result, do not adjust the actual operating data of the target beacon.

[0073] It should be noted that this method uses modern sensing technology, automation systems and data analysis algorithms to collect, process and analyze relevant data of marine aids in real time, including information such as location, status, brightness and power supply, and can monitor the performance and status of aids in real time to ensure that the target aids are in good working condition; once the analysis detects a problem or anomaly, the system can automatically trigger a maintenance alarm, reducing the response time and helping to improve the reliability and safety of aids; data analysis technology can provide more accurate information, reduce the risk of human error, and improve the consistency of monitoring; by reducing the demand for human resources, timely maintenance and reducing the risk of accidents, methods based on data analysis can reduce the overall cost of aid monitoring and evaluation.

[0074] like Figure 2 As shown, the actual working environment of the target navigation mark is obtained within a preset time period, and the preset operating data of the target navigation mark within the preset time period is determined according to the actual working environment, specifically:

[0075] S202: acquiring standard operation data of the target navigation mark under various preset working environment conditions through a big data network, constructing a database, and importing the standard operation data of the navigation mark under various preset working environment conditions into the database to obtain a standard database;

[0076] S204: acquiring an actual working environment of a target navigation mark within a preset time period, importing the actual working environment into the standard database, and calculating similarities between the actual working environment and various preset working environments by using a grey correlation analysis method to obtain multiple similarities;

[0077] S206: constructing a sorting table, importing the plurality of similarities into the sorting table for sorting based on numerical values, and after sorting, extracting the maximum similarity; obtaining a preset working environment corresponding to the maximum similarity, and constructing a search tag according to the preset working environment corresponding to the maximum similarity;

[0078] S208: Retrieve the standard operating data of the target navigation mark under the actual working environment in the standard database according to the search tag, and mark the standard operating data of the target navigation mark under the actual working environment as preset operating data within a preset time period.

[0079] The working environment includes but is not limited to temperature, humidity, salinity, light intensity and wind pressure.

[0080] It should be noted that Grey Relational Analysis (GRA) is a statistical method used to study the degree of association between multiple variables. The basic idea of ​​Grey Relational Analysis is to convert the data sequence between the various variables of the research object into a grey series, and then analyze it by comparing the degree of association between these grey series.

[0081] It should be noted that the operating parameters (operating data) of the marine beacon under different working environment conditions are different. For example, at low temperatures, the available capacity of the battery in the beacon is usually reduced. This is because low temperatures slow down the ion migration speed in the electrolyte and reduce the reaction efficiency between the positive and negative electrodes of the battery, resulting in a reduction in the energy that the battery can release at the same discharge rate. The temperature difference at sea is large at different times. Therefore, in order to eliminate the impact of the working environment on the monitoring and evaluation results, the actual working environment of the target beacon needs to be taken into account during the monitoring and evaluation of the target beacon. The actual working environment information of the target beacon is obtained through a series of sensors, such as temperature sensors, humidity sensors, etc., and then the collected signals are sent to the management platform wirelessly. Through the above steps, the preset operating data of the target beacon, that is, the standard operating data, such as the standard discharge voltage and discharge rate of the battery under the current working environment conditions, can be inferred in combination with the actual working environment of the target beacon, which can eliminate the impact of the environment on the monitoring and evaluation results, thereby improving the accuracy and reliability of the monitoring and evaluation results and avoiding misjudgment.

[0082] like Figure 3 As shown, the actual operation data of the target navigation mark is obtained within a preset time period, and the actual operation data is subjected to dimensionality reduction processing to obtain vector data corresponding to each actual operation data, specifically:

[0083] S302: extracting features of each actual operation data, and performing standardization processing on the actual operation data to ensure that the mean of each feature is 0 and the variance is 1, thereby obtaining standardized actual operation data; wherein the features include numerical features, category features, time features, and text features;

[0084] S304: Obtaining a covariance matrix of the standardized actual operation data, and performing eigendecomposition on the covariance matrix to obtain eigenvalues ​​and eigenvectors; wherein the eigenvalues ​​represent the importance of the principal component, and the eigenvectors represent the direction of the principal component;

[0085] S306: Arrange the eigenvalues ​​in descending order to determine the order of the principal components. The eigenvector corresponding to the largest eigenvalue is the first principal component, and so on to obtain a plurality of principal components.

[0086] S308: Determine the number of principal components to be retained according to a preset threshold to obtain the filtered principal components; construct a projection matrix according to the eigenvectors corresponding to the filtered principal components, and multiply the standardized actual operation data by the projection matrix to project the standardized actual operation data into a new low-dimensional space to obtain vector data corresponding to each standardized actual operation data.

[0087] The principal component refers to the most important feature direction or feature vector in the data set in principal component analysis.

[0088] Among them, the characteristics of actual operating data refer to the various attributes or variables in the data set, which describe different aspects or properties in the data. Features are usually used to describe the various dimensions of the data for use in data analysis, machine learning, and statistical modeling. Features can be attributes, properties, measurements, or descriptions of the data.

[0089] It should be noted that the actual operation data of the target beacon is obtained through a series of sensors set inside the target beacon within a preset time period, such as obtaining the change in the power of the battery in the target beacon through a power sensor. Then, the actual operation data collected is subjected to dimensionality reduction processing to obtain vector data corresponding to each actual operation data. Dimensionality reduction can map high-dimensional data to low-dimensional space, making the data easier to visualize and understand, which helps to discover patterns and structures in the data, and better explore the characteristics of the operation data of the target beacon within a preset time period. Specifically, in this step, the actual operation data collected is subjected to dimensionality reduction processing by principal component analysis, and the data is standardized to ensure that the mean of each feature is 0 and the variance is 1. This is a prerequisite for PCA because PCA is affected by the feature scale. The covariance matrix is ​​calculated for the standardized data, and the covariance matrix describes the linear relationship between the data features. Then the covariance matrix is ​​subjected to eigenvalue decomposition; the eigenvalue represents the importance of the principal component, and the eigenvector represents the direction of the principal component; this can be done by using a linear algebra library or a PCA algorithm. According to the size of the eigenvalue, select the number of principal components to be retained. Usually, the number of principal components is determined according to the percentage of the total variance retained. For example, you can choose to retain 95% of the total variance. Then build a projection matrix that maps the original data to a new low-dimensional space. Multiply the standardized original data by the projection matrix to project the data to a new low-dimensional space, which will generate vector data after dimensionality reduction. Through the above steps, the collected high-dimensional operating data can be reduced to a low dimension, reducing data complexity and noise, while retaining the main information for creating scatter plots or data visualization to more easily understand the structure of the operating data, and can lay the foundation for subsequent detection of whether each operating data is an outlier.

[0090] Among them, the characteristics of each vector data in the vector data set are obtained, noise points are retrieved based on the characteristics of each vector data, and the noise points are removed to filter out the outlier actual operation data to obtain the filtered actual operation data, specifically:

[0091] Acquire the features of each vector data in the vector data set, and calculate the mean of the features of each vector data to obtain a mean vector containing a plurality of means; wherein the mean vector is the data center of the vector data set;

[0092] Taking the data center as the coordinate origin, constructing a three-dimensional coordinate system according to the coordinate origin, importing the vector data set into the three-dimensional coordinate system, and obtaining the three-dimensional coordinate value corresponding to each vector data in the vector data set in the three-dimensional coordinate system;

[0093] Calculate the Euclidean distance between each vector data and the data center according to the three-dimensional coordinate value, and compare the Euclidean distance between each vector data and the data center with a preset Euclidean distance;

[0094] Marking vector data corresponding to a Euclidean distance greater than a preset Euclidean distance as noise points, and removing each noise point from the vector data set to obtain vector data after outlier screening;

[0095] The actual operation data corresponding to the vector data after outlier screening is extracted to screen the outlier actual operation data to obtain screened actual operation data.

[0096] The characteristics of vector data refer to the numerical representations that describe the attributes or properties of data points or objects.

[0097] It should be noted that after reducing the high-dimensional operating data to a low dimension, by calculating the average value of each feature in the data set, for each feature, the feature values ​​of all samples are added and then divided by the number of samples, which will generate an average value for each feature to form a vector, which is the center of the data set. Then use the three-dimensional software to construct a three-dimensional coordinate system with the center of the data as the coordinate origin, then import the vector data set into the three-dimensional coordinate system, and obtain the three-dimensional coordinate values ​​corresponding to each vector data in the vector data set in the three-dimensional coordinate system, so as to calculate the Euclidean distance between each vector data and the data center based on the three-dimensional coordinate values ​​of each vector data. If the Euclidean distance between a certain vector data and the data center is greater than the preset Euclidean distance, it means that the data is a noise point (i.e., an outlier). At this time, the corresponding actual operating data needs to be eliminated to obtain the filtered actual operating data. It should also be noted that the reason for the outliers is that in the process of collecting the actual operation data of the target beacon, the collected data is affected by the collection environment and the equipment collection accuracy, which leads to data drift and distortion, resulting in outliers. It may also be due to erroneous collection, such as collecting current data as voltage data. Through the above steps, the actual operation data after dimensionality reduction can be mapped to the three-dimensional coordinate system, and then the corresponding data point can be quickly judged as an outlier by distance measurement, so as to filter out the data that does not meet the requirements, improve the reliability of the data, and further improve the reliability of subsequent monitoring and evaluation results.

[0098] The preset operation data is compared and analyzed with the screened actual operation data to obtain the first evaluation result or the second evaluation result, specifically:

[0099] Sorting the preset operation data based on the time series to obtain the preset operation data based on the time series, and constructing a preset operation data curve chart according to the preset operation data based on the time series;

[0100] Sorting the filtered actual operation data based on time series to obtain the filtered actual operation data based on time series, and constructing an actual operation data curve chart according to the filtered actual operation data based on time series;

[0101] Constructing a virtual pairing space, and importing the preset operation data curve graph and the actual operation data curve graph into the virtual pairing space for pairing;

[0102] After the pairing is completed, the matching degree between the preset operation data and the screened actual operation data is calculated based on the length of the curve that overlaps and the length of the curve that does not overlap between the preset operation data curve graph and the actual operation data curve graph;

[0103] If the matching degree is not greater than the preset matching degree, it means that the operation state of the target navigation mark within the preset time period is abnormal, and a first evaluation result is generated;

[0104] If the matching degree is greater than the preset matching degree, it means that the operating status of the target navigation mark within the preset time period is normal, and a second evaluation result is generated.

[0105] It should be noted that a virtual pairing space is constructed by software such as SolidWorks and CAD, and then the preset operation data curve graph and the actual operation data curve graph are imported into the virtual pairing space for pairing. During pairing, two or more identical operation data at the same time point are first retrieved, and then the two or more identical operation data at the same time point are used as pairing reference points. If the preset operation data curve graph and the actual operation data curve graph have the same operation data at the 5th second and the 30th second, these two reference points can be used as pairing references to pair the preset operation data curve graph and the actual operation data curve graph. After pairing is completed, the matching degree between the preset operation data and the screened actual operation data is calculated based on the length of the curve that overlaps with the actual operation data curve graph and the length of the curve that does not overlap. If the matching degree is not greater than the preset matching degree, it means that the actual operating data of the target beacon within the preset time period deviates greatly from the preset operating data, which means that the operating status of the target beacon within the preset time period is abnormal, and a first evaluation result is generated, and an alarm message is generated; if the matching degree is greater than the preset matching degree, it means that the actual operating data of the target beacon within the preset time period is highly consistent with the preset operating data, which means that the operating status of the target beacon within the preset time period is normal, and a second evaluation result is generated.

[0106] If the evaluation result is the first evaluation result, the actual operation data of the target navigation mark is adjusted, specifically:

[0107] Obtain the optimal control measures corresponding to the target navigation mark under various preset abnormal operation data combination conditions through the big data network, construct a knowledge graph, and import the optimal control measures corresponding to the target navigation mark under various preset abnormal operation data combination conditions into the knowledge graph;

[0108] If the matching degree is not greater than the preset matching degree, then obtaining the non-overlapping curve area in the preset operation data curve graph and the actual operation data curve graph, marking the non-overlapping curve area as the abnormal operation period, and marking the actual operation data corresponding to the abnormal operation period in the actual operation data curve graph as abnormal operation data;

[0109] Importing the abnormal operation data into the knowledge graph, and calculating the similarity between the abnormal operation data and various preset abnormal operation data combinations by grey correlation analysis to obtain multiple similarities;

[0110] A maximum similarity is extracted from multiple similarities, a preset operation abnormality data combination corresponding to the maximum similarity is obtained, and the optimal control measures of the target beacon when the current operation state is abnormal are determined based on the preset operation abnormality data combination corresponding to the maximum similarity, and the optimal control measures are transmitted to the control terminal of the target beacon.

[0111] It should be noted that when an abnormal operating state occurs in the target navigation mark, the above steps can be used to timely and automatically adjust the target navigation mark so that it can return to normal operating state, such as adjusting the light brightness, position offset, etc., thereby improving the safety, efficiency and sustainability of maritime navigation, reducing the risk of accidents, and better meeting the needs of maritime traffic, which is of great significance for ensuring smooth maritime navigation and safe navigation.

[0112] In addition, a method for monitoring and evaluating marine aids to navigation based on data analysis also includes the following steps:

[0113] Acquire data information transmitted by each sensor in the target navigation mark through the remote receiving terminal within a preset time period;

[0114] Determine whether the data information transmitted by each sensor can be received within a preset time period. If not, the sensor that cannot receive the data information is recorded as an abnormal sensor;

[0115] If yes, it is further determined whether the delay speed of the data information transmitted by each sensor in the preset time period is greater than the delay speed;

[0116] If the delay speed is greater than the delay speed, the sensor corresponding to the delay speed greater than the delay speed is recorded as an abnormal sensor.

[0117] It should be noted that this method can quickly detect whether any abnormal operation has occurred in any sensor in the target beacon. When an abnormal operation occurs, the maintenance personnel will be informed in time to carry out maintenance, so that each sensor can operate effectively, and each sensor can effectively monitor and collect the operating data of the target beacon, thereby improving the reliability of the collected operating data.

[0118] In addition, the method for monitoring and evaluating marine navigation aids based on data analysis also includes the following steps:

[0119] Acquire the types of meteorological disasters that can evolve under various meteorological factor combinations through a big data network, construct a second database, and obtain a meteorological database by adding the types of meteorological disasters that can evolve under various meteorological factor combinations to the second database;

[0120] Acquire the real-time meteorological factors of the target navigation mark within a preset time period, import the real-time meteorological factors into the meteorological database, calculate the similarity between the real-time meteorological factors and various meteorological factor combinations by grey correlation analysis method, and obtain multiple similarities;

[0121] Extract the maximum similarity from multiple similarities, obtain the meteorological factor combination corresponding to the maximum similarity, and predict the meteorological disaster type of the target navigation mark after a preset time period based on the meteorological factor combination corresponding to the maximum similarity;

[0122] Determine whether the meteorological disaster type of the target navigation mark after the preset time period is the preset meteorological disaster type. If so, generate warning information.

[0123] It should be noted that the preset meteorological disaster types include storms and thunderstorms. When it is predicted that the preset meteorological disaster type will occur in the area where the target navigation mark is located after the preset time period, the target navigation mark may be affected by the meteorological disaster and become ineffective. At this time, an early warning information is generated, such as marking the area as a no-go zone, and ships are prevented from passing through during the time period to avoid the probability of a maritime accident caused by the target navigation mark.

[0124] In addition, the method for monitoring and evaluating marine navigation aids based on data analysis also includes the following steps:

[0125] Obtain the energy consumption rate of the battery in the target flight table under various meteorological factor combinations through the big data network;

[0126] A prediction model is constructed based on a convolutional neural network, and the energy consumption rate of the battery in the target flight table under various meteorological factor combinations is introduced into the prediction model for training to obtain a trained prediction model;

[0127] Acquire actual meteorological factors of the target navigation mark within a preset time period, import the actual meteorological factors into the trained prediction model for prediction, and predict the actual energy consumption rate of the battery in the target navigation mark within the preset time period;

[0128] Acquire the actual energy value of the battery in the target flight schedule, and calculate the remaining effective working time of the battery in the target flight schedule based on the actual energy value and the actual energy consumption rate;

[0129] The remaining effective working time is compared with the preset working time; if the remaining effective working time is less than the preset working time, electric energy warning information is generated and the electric energy warning information is sent to the user end.

[0130] It should be noted that due to the complex and changeable climate environment at sea, the rate of power consumption in the battery is different under different climate and environmental conditions, so the effective working time of the battery in the target navigation table is variable. In actual use, the battery of the target navigation table is charged regularly. Therefore, in this process, it is possible that the battery will be low before the battery is charged regularly, thereby causing the target navigation table to fail. Therefore, through the above steps, it can be determined whether the battery of the target navigation table will be low before the battery is charged regularly. If so, an energy warning message is generated to remind the user to charge in advance to ensure the effective operation of the target navigation table.

[0131] like Figure 4 As shown, the second aspect of the present invention discloses a marine navigation mark monitoring and evaluation system based on data analysis, the marine navigation mark monitoring and evaluation system comprises a memory 12 and a processor 13, the memory 12 stores a marine navigation mark monitoring and evaluation method program, when the marine navigation mark monitoring and evaluation method program is executed by the processor 13, the following steps are implemented:

[0132] Acquiring an actual working environment of the target navigation mark within a preset time period, and determining preset operating data of the target navigation mark within the preset time period according to the actual working environment;

[0133] Acquire actual operation data of the target navigation mark within a preset time period, perform dimensionality reduction processing on the actual operation data to obtain vector data corresponding to each actual operation data, and collect each vector data to obtain a vector data set;

[0134] Acquire features of each vector data in the vector data set, retrieve noise points based on the features of each vector data, and remove the noise points to filter out outlier actual operation data to obtain filtered actual operation data;

[0135] The preset operating data is compared and analyzed with the screened actual operating data to obtain a first evaluation result or a second evaluation result; if the evaluation result is the first evaluation result, the actual operating data of the target beacon is regulated; if the evaluation result is the second evaluation result, the actual operating data of the target beacon is not regulated.

[0136] Furthermore, in a preferred embodiment of the present invention, the actual working environment of the target navigation mark is obtained within a preset time period, and the preset operating data of the target navigation mark within the preset time period is determined according to the actual working environment, specifically:

[0137] Acquire standard operation data of target navigation marks under various preset working environment conditions through a big data network, construct a database, and import the standard operation data of navigation marks under various preset working environment conditions into the database to obtain a standard database;

[0138] Acquire the actual working environment of the target navigation mark within a preset time period, import the actual working environment into the standard database, calculate the similarity between the actual working environment and various preset working environments by grey correlation analysis method, and obtain multiple similarities;

[0139] Constructing a sorting table, importing the plurality of similarities into the sorting table for sorting based on numerical values, and extracting the maximum similarity after the sorting is completed; obtaining a preset working environment corresponding to the maximum similarity, and constructing a search tag according to the preset working environment corresponding to the maximum similarity;

[0140] The standard operating data of the target navigation mark under the actual working environment is retrieved in the standard database according to the retrieval tag, and the standard operating data of the target navigation mark under the actual working environment is marked as the preset operating data within the preset time period.

[0141] Furthermore, in a preferred embodiment of the present invention, the preset operation data is compared and analyzed with the screened actual operation data to obtain a first evaluation result or a second evaluation result, specifically:

[0142] Sorting the preset operation data based on the time series to obtain the preset operation data based on the time series, and constructing a preset operation data curve chart according to the preset operation data based on the time series;

[0143] Sorting the filtered actual operation data based on time series to obtain the filtered actual operation data based on time series, and constructing an actual operation data curve chart according to the filtered actual operation data based on time series;

[0144] Constructing a virtual pairing space, and importing the preset operation data curve graph and the actual operation data curve graph into the virtual pairing space for pairing;

[0145] After the pairing is completed, the matching degree between the preset operation data and the screened actual operation data is calculated based on the length of the curve that overlaps and the length of the curve that does not overlap between the preset operation data curve graph and the actual operation data curve graph;

[0146] If the matching degree is not greater than the preset matching degree, it means that the operation state of the target navigation mark within the preset time period is abnormal, and a first evaluation result is generated;

[0147] If the matching degree is greater than the preset matching degree, it means that the operating status of the target navigation mark within the preset time period is normal, and a second evaluation result is generated.

[0148] Furthermore, in a preferred embodiment of the present invention, if the evaluation result is the first evaluation result, the actual operation data of the target navigation mark is regulated, specifically:

[0149] Obtain the optimal control measures corresponding to the target navigation mark under various preset abnormal operation data combination conditions through the big data network, construct a knowledge graph, and import the optimal control measures corresponding to the target navigation mark under various preset abnormal operation data combination conditions into the knowledge graph;

[0150] If the matching degree is not greater than the preset matching degree, then obtaining the non-overlapping curve area in the preset operation data curve graph and the actual operation data curve graph, marking the non-overlapping curve area as the abnormal operation period, and marking the actual operation data corresponding to the abnormal operation period in the actual operation data curve graph as abnormal operation data;

[0151] Importing the abnormal operation data into the knowledge graph, and calculating the similarity between the abnormal operation data and various preset abnormal operation data combinations by grey correlation analysis to obtain multiple similarities;

[0152] A maximum similarity is extracted from multiple similarities, a preset operation abnormality data combination corresponding to the maximum similarity is obtained, and the optimal control measures of the target beacon when the current operation state is abnormal are determined based on the preset operation abnormality data combination corresponding to the maximum similarity, and the optimal control measures are transmitted to the control terminal of the target beacon.

[0153] Those skilled in the art can understand that: all or part of the steps of implementing the above method embodiments 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 method embodiments; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), disks or optical disks, and other media that can store program codes.

[0154] Alternatively, if the above-mentioned integrated unit of the present invention is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, RAM, magnetic disks or optical disks.

[0155] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for monitoring and evaluating marine navigation aids based on data analysis, characterized in that: The following steps are involved: Acquiring an actual working environment of the target navigation mark within a preset time period, and determining preset operating data of the target navigation mark within the preset time period according to the actual working environment; Acquire actual operation data of the target navigation mark within a preset time period, perform dimensionality reduction processing on the actual operation data to obtain vector data corresponding to each actual operation data, and collect each vector data to obtain a vector data set; Acquire features of each vector data in the vector data set, retrieve noise points based on the features of each vector data, and remove the noise points to filter out outlier actual operation data to obtain filtered actual operation data; Comparing and analyzing the preset operation data with the screened actual operation data to obtain a first evaluation result or a second evaluation result; If the evaluation result is the first evaluation result, the actual operation data of the target navigation mark is adjusted; if the evaluation result is the second evaluation result, the actual operation data of the target navigation mark is not adjusted; The actual working environment of the target navigation mark is obtained within a preset time period, and the preset operating data of the target navigation mark within the preset time period is determined according to the actual working environment, specifically: Acquire standard operation data of target navigation marks under various preset working environment conditions through a big data network, construct a database, and import the standard operation data of navigation marks under various preset working environment conditions into the database to obtain a standard database; Acquire the actual working environment of the target navigation mark within a preset time period, import the actual working environment into the standard database, calculate the similarity between the actual working environment and various preset working environments by grey correlation analysis method, and obtain multiple similarities; Constructing a sorting table, importing the plurality of similarities into the sorting table for sorting based on numerical values, and extracting the maximum similarity after the sorting is completed; obtaining a preset working environment corresponding to the maximum similarity, and constructing a search tag according to the preset working environment corresponding to the maximum similarity; Retrieving the standard operating data of the target navigation mark under the actual working environment in the standard database according to the retrieval tag, and marking the standard operating data of the target navigation mark under the actual working environment as the preset operating data within the preset time period; The actual operation data of the target navigation mark is obtained within a preset time period, and the actual operation data is subjected to dimensionality reduction processing to obtain vector data corresponding to each actual operation data, specifically: Extracting features of each actual operation data, and performing standardization on the actual operation data to ensure that the mean of each feature is 0 and the variance is 1, thereby obtaining standardized actual operation data; wherein the features include numerical features, category features, time features, and text features; Obtaining a covariance matrix of the standardized actual operation data, and performing eigendecomposition on the covariance matrix to obtain eigenvalues ​​and eigenvectors; wherein the eigenvalues ​​represent the importance of the principal components, and the eigenvectors represent the directions of the principal components; Arrange the eigenvalues ​​in descending order to determine the order of the principal components. The eigenvector corresponding to the largest eigenvalue is the first principal component, and so on to obtain several principal components. Determine the number of principal components to be retained according to a preset threshold value to obtain the filtered principal components; construct a projection matrix according to the eigenvectors corresponding to the filtered principal components, and multiply the standardized actual operation data by the projection matrix to project the standardized actual operation data into a new low-dimensional space to obtain vector data corresponding to each standardized actual operation data; Among them, the characteristics of each vector data in the vector data set are obtained, noise points are retrieved based on the characteristics of each vector data, and the noise points are removed to filter out the outlier actual operation data to obtain the filtered actual operation data, specifically: Acquire the features of each vector data in the vector data set, and calculate the mean of the features of each vector data to obtain a mean vector containing a plurality of means; wherein the mean vector is the data center of the vector data set; Taking the data center as the coordinate origin, constructing a three-dimensional coordinate system according to the coordinate origin, importing the vector data set into the three-dimensional coordinate system, and obtaining the three-dimensional coordinate value corresponding to each vector data in the vector data set in the three-dimensional coordinate system; Calculate the Euclidean distance between each vector data and the data center according to the three-dimensional coordinate value, and compare the Euclidean distance between each vector data and the data center with a preset Euclidean distance; Marking vector data corresponding to a Euclidean distance greater than a preset Euclidean distance as noise points, and removing each noise point from the vector data set to obtain vector data after outlier screening; Extracting actual operation data corresponding to the vector data after outlier screening, so as to screen the outlier actual operation data and obtain screened actual operation data; The preset operation data is compared and analyzed with the screened actual operation data to obtain the first evaluation result or the second evaluation result, specifically: Sorting the preset operation data based on the time series to obtain the preset operation data based on the time series, and constructing a preset operation data curve chart according to the preset operation data based on the time series; Sorting the filtered actual operation data based on time series to obtain the filtered actual operation data based on time series, and constructing an actual operation data curve chart according to the filtered actual operation data based on time series; Constructing a virtual pairing space, and importing the preset operation data curve graph and the actual operation data curve graph into the virtual pairing space for pairing; After the pairing is completed, the matching degree between the preset operation data and the screened actual operation data is calculated based on the length of the curve that overlaps and the length of the curve that does not overlap between the preset operation data curve graph and the actual operation data curve graph; If the matching degree is not greater than the preset matching degree, it means that the operation state of the target navigation mark within the preset time period is abnormal, and a first evaluation result is generated; If the matching degree is greater than the preset matching degree, it means that the operation status of the target navigation mark within the preset time period is normal, and a second evaluation result is generated; If the evaluation result is the first evaluation result, the actual operation data of the target navigation mark is adjusted, specifically: Obtain the optimal control measures corresponding to the target navigation mark under various preset abnormal operation data combination conditions through the big data network, construct a knowledge graph, and import the optimal control measures corresponding to the target navigation mark under various preset abnormal operation data combination conditions into the knowledge graph; If the matching degree is not greater than the preset matching degree, then obtaining the non-overlapping curve area in the preset operation data curve graph and the actual operation data curve graph, marking the non-overlapping curve area as the abnormal operation period, and marking the actual operation data corresponding to the abnormal operation period in the actual operation data curve graph as abnormal operation data; Importing the abnormal operation data into the knowledge graph, and calculating the similarity between the abnormal operation data and various preset abnormal operation data combinations by grey correlation analysis to obtain multiple similarities; A maximum similarity is extracted from multiple similarities, a preset operation abnormality data combination corresponding to the maximum similarity is obtained, and the optimal control measures of the target beacon when the current operation state is abnormal are determined based on the preset operation abnormality data combination corresponding to the maximum similarity, and the optimal control measures are transmitted to the control terminal of the target beacon.

2. A marine navigation aid monitoring and evaluation system based on data analysis, characterized in that: The marine navigation mark monitoring and evaluation system includes a memory and a processor, wherein the marine navigation mark monitoring and evaluation method program is stored in the memory. When the marine navigation mark monitoring and evaluation method program is executed by the processor, the steps of the marine navigation mark monitoring and evaluation method as described in claim 1 are implemented.