Ship alarm compression and return method and system based on time series clustering

Abstract

This invention discloses a method and system for ship alarm compression and backhaul based on time-series clustering. The method includes: acquiring multiple alarm messages sent by the ship within a preset historical time window; clustering all the alarm messages based on the time characteristics corresponding to the alarm messages to obtain multiple information sets; determining a structured compression algorithm corresponding to each information set based on the information characteristics corresponding to each information set; compressing and integrating each information set according to the structured compression algorithm to obtain alarm aggregate data corresponding to the ship; and sending the alarm aggregate data to the shore-based equipment corresponding to the ship. It is evident that this invention can achieve time-series clustering and structured compression of ship alarm data, improving data transmission efficiency and storage optimization, and reducing transmission pressure and processing delay risks caused by alarm data redundancy.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for compressed transmission of ship alarms based on time series clustering. Background Technology

[0002] With the expansion of intelligent monitoring systems for ships and the rapid growth of alarm data, shipping companies and ship operators are increasingly focusing on improving transmission efficiency and storage optimization through intelligent processing of alarm data. A key technical issue is how to achieve data simplification to reduce transmission pressure and processing delay risks. Existing technologies typically acquire multiple alarm messages sent from the ship, process the data using simple time filtering or fixed threshold deduplication, and perform overall compression based on a general compression algorithm to support data transmission and storage. However, existing solutions lack clustering analysis of the temporal characteristics of alarm information within a preset historical time window, dynamic selection of structured compression algorithms for different information sets, and joint processing of compression and integration. This makes it difficult to effectively eliminate redundancy in alarm data, resulting in insufficient data transmission efficiency and storage optimization. Redundancy in alarm data can easily lead to increased transmission pressure and processing delays, limiting the real-time performance and overall operational efficiency of the ship's alarm system. Therefore, existing technologies have shortcomings that urgently need to be addressed. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a ship alarm compression and backhaul method and system based on time series clustering, which can realize time series clustering and structured compression of ship alarm data, improve data transmission efficiency and storage optimization effect, and reduce transmission pressure and processing delay risk caused by alarm data redundancy.

[0004] To address the aforementioned technical problems, the first aspect of this invention discloses a method for compressed backhaul of ship alarms based on time series clustering, the method comprising: Acquire multiple alarm messages sent by the ship within a preset historical time window; Based on the time characteristics corresponding to the alarm information, all the alarm information is clustered to obtain multiple information sets; Based on the information features corresponding to each information set, a structured compression algorithm corresponding to each information set is determined. The structured compression algorithm is used to compress and integrate each set of information to obtain the alarm aggregation data corresponding to the ship end; the alarm aggregation data is used to send to the shore-side equipment corresponding to the ship end.

[0005] As an optional implementation, in the first aspect of the invention, the step of clustering all the alarm information based on the time characteristics corresponding to the alarm information to obtain multiple information sets includes: Identify multiple dangerous alarm messages among all the aforementioned alarm messages; For each alarm message, calculate the weighted summation average of the time differences between the trigger time of the alarm message and the trigger time of each danger alarm message to obtain the time characteristic parameter corresponding to the alarm message; All the alarm information is clustered and divided based on the time feature parameters to obtain multiple information sets; each information set includes multiple alarm information where the parameter difference between the time feature parameters is less than a preset difference threshold.

[0006] As an optional implementation, in the first aspect of the invention, determining multiple danger alarm messages among all the alarm messages includes: For each alarm message, the alarm message is input into the trained classifier to obtain the corresponding alarm type; The alarm information is input into the trained hazard prediction model to obtain the corresponding predicted hazard level; The number of times each alarm type appears in the classification results corresponding to all alarm information is counted to obtain the hazard weight; The product of the predicted risk level and the risk weight is calculated to obtain the risk level parameter of the alarm information; Based on the aforementioned risk parameter, all alarm messages are sorted from largest to smallest to obtain an information sequence; The first preset proportion of the alarm information in the information sequence is determined as multiple danger alarm information.

[0007] As an optional implementation, in the first aspect of the present invention, when calculating the time feature parameters, the calculation weight corresponding to each time difference is the product of a first weight and a second weight; the first weight is proportional to the content similarity between the alarm information and the corresponding danger alarm information; the second weight is proportional to the similarity between the alarm information and the alarm device type of the corresponding danger alarm information.

[0008] As an optional implementation, in the first aspect of the present invention, determining the structured compression algorithm corresponding to each information set based on the information features corresponding to each information set includes: For each of the aforementioned information sets, each of the aforementioned alarm information in the information set is input into a trained data structure prediction model to obtain the data structure corresponding to each of the aforementioned alarm information; the data structure prediction model is trained using a training dataset that includes multiple training alarm information and corresponding information data structure annotations; Based on the data structure corresponding to all the alarm information, determine the set structure characteristics corresponding to the information set; Among multiple candidate compression algorithms, the compression algorithm corresponding to the set structure features is determined, and thus the structured compression algorithm corresponding to the information set is determined.

[0009] As an optional implementation, in the first aspect of the present invention, determining the set structure features corresponding to the information set based on the data structure corresponding to all the alarm information includes: For each preset structure type, count the number of times that structure type appears in all the data structures corresponding to the alarm information; Determine the order of the occurrence count in the calculated frequency sequence; the frequency sequence is obtained by arranging all the structure types in descending order of their occurrence counts. When the ordered position precedes the preset reference position, the structure type is determined as the preferred structure type; The matrix consisting of all the preferred structure types and the corresponding data content in each alarm message is determined as the set structure feature corresponding to the information set.

[0010] As an optional implementation, in the first aspect of the present invention, determining the compression algorithm corresponding to the set structure features from a plurality of candidate compression algorithms, so as to determine the structured compression algorithm corresponding to the information set, includes: The set structure features are input into the trained algorithm parameter prediction model to obtain the corresponding prediction algorithm parameters; the prediction algorithm parameters include at least one of data splitting parameters, encryption parameters, data combination parameters, and algorithm model architecture parameters; the algorithm parameter prediction model is trained on a training dataset that includes multiple training set structure features and corresponding algorithm parameter annotations; Calculate the parameter similarity between the historical algorithm application parameters and the predicted algorithm parameters for each candidate compression algorithm; The candidate compression algorithm with the highest parameter similarity is determined as the structured compression algorithm corresponding to the information set.

[0011] As an optional implementation, in the first aspect of the present invention, the step of compressing and integrating each information set according to the structured compression algorithm to obtain the alarm aggregation data corresponding to the ship end includes: Based on a preset feature processing algorithm, the feature value corresponding to the set structure feature of each information set is determined; An encryption key corresponding to the feature value is generated according to a preset key generation algorithm; the negotiated public key information of the key generation algorithm is stored in the shore device for use in generating the decryption key; Each set of information is compressed using the corresponding structured compression algorithm to obtain the corresponding compressed data. Based on the encryption key of the information set corresponding to each set of compressed data, the set of compressed data is encrypted to obtain set of encrypted data; All the encrypted data in the set and the corresponding feature values ​​are combined to obtain the alarm aggregation data corresponding to the ship end.

[0012] A second aspect of this invention discloses a ship alarm compression and feedback system based on time series clustering, the system comprising: The acquisition module is used to acquire multiple alarm messages sent by the ship within a preset historical time window; The clustering module is used to cluster all the alarm information based on the time characteristics corresponding to the alarm information to obtain multiple information sets; The determination module is used to determine the structured compression algorithm corresponding to each information set based on the information features corresponding to each information set. The processing module is used to compress and integrate each of the information sets according to the structured compression algorithm to obtain the alarm aggregation data corresponding to the ship end; the alarm aggregation data is used to be sent to the shore-side equipment corresponding to the ship end.

[0013] As an optional implementation, in a second aspect of the invention, the specific method by which the clustering module clusters all the alarm information based on the time characteristics corresponding to the alarm information to obtain multiple information sets includes: Identify multiple dangerous alarm messages among all the aforementioned alarm messages; For each alarm message, calculate the weighted summation average of the time differences between the trigger time of the alarm message and the trigger time of each danger alarm message to obtain the time characteristic parameter corresponding to the alarm message; All the alarm information is clustered and divided based on the time feature parameters to obtain multiple information sets; each information set includes multiple alarm information where the parameter difference between the time feature parameters is less than a preset difference threshold.

[0014] As an optional implementation, in a second aspect of the invention, the clustering module determines the specific method by which it identifies multiple danger alarm messages among all the alarm messages, including: For each alarm message, the alarm message is input into the trained classifier to obtain the corresponding alarm type; The alarm information is input into the trained hazard prediction model to obtain the corresponding predicted hazard level; The number of times each alarm type appears in the classification results corresponding to all alarm information is counted to obtain the hazard weight; The product of the predicted risk level and the risk weight is calculated to obtain the risk level parameter of the alarm information; Based on the aforementioned risk parameter, all alarm messages are sorted from largest to smallest to obtain an information sequence; The first preset proportion of the alarm information in the information sequence is determined as multiple danger alarm information.

[0015] As an optional implementation, in the second aspect of the present invention, when calculating the time feature parameters, the calculation weight corresponding to each time difference is the product of a first weight and a second weight; the first weight is proportional to the content similarity between the alarm information and the corresponding danger alarm information; the second weight is proportional to the similarity between the alarm information and the alarm device type of the corresponding danger alarm information.

[0016] As an optional implementation, in a second aspect of the invention, the determining module determines the specific method of the structured compression algorithm corresponding to each information set based on the information features corresponding to each information set, including: For each of the aforementioned information sets, each of the aforementioned alarm information in the information set is input into a trained data structure prediction model to obtain the data structure corresponding to each of the aforementioned alarm information; the data structure prediction model is trained using a training dataset that includes multiple training alarm information and corresponding information data structure annotations; Based on the data structure corresponding to all the alarm information, determine the set structure characteristics corresponding to the information set; Among multiple candidate compression algorithms, the compression algorithm corresponding to the set structure features is determined, and thus the structured compression algorithm corresponding to the information set is determined.

[0017] As an optional implementation, in a second aspect of the invention, the determining module determines the specific method by which it determines the set structure features corresponding to the information set based on the data structure corresponding to all the alarm information, including: For each preset structure type, count the number of times that structure type appears in all the data structures corresponding to the alarm information; Determine the order of the occurrence count in the calculated frequency sequence; the frequency sequence is obtained by arranging all the structure types in descending order of their occurrence counts. When the ordered position precedes the preset reference position, the structure type is determined as the preferred structure type; The matrix consisting of all the preferred structure types and the corresponding data content in each alarm message is determined as the set structure feature corresponding to the information set.

[0018] As an optional implementation, in a second aspect of the invention, the determining module determines the compression algorithm corresponding to the set structure features from a plurality of candidate compression algorithms, and the specific method of determining the structured compression algorithm corresponding to the information set includes: The set structure features are input into the trained algorithm parameter prediction model to obtain the corresponding prediction algorithm parameters; the prediction algorithm parameters include at least one of data splitting parameters, encryption parameters, data combination parameters, and algorithm model architecture parameters; the algorithm parameter prediction model is trained on a training dataset that includes multiple training set structure features and corresponding algorithm parameter annotations; Calculate the parameter similarity between the historical algorithm application parameters and the predicted algorithm parameters for each candidate compression algorithm; The candidate compression algorithm with the highest parameter similarity is determined as the structured compression algorithm corresponding to the information set.

[0019] As an optional implementation, in a second aspect of the invention, the specific method by which the processing module compresses and integrates each information set according to the structured compression algorithm to obtain the alarm aggregation data corresponding to the ship includes: Based on a preset feature processing algorithm, the feature value corresponding to the set structure feature of each information set is determined; An encryption key corresponding to the feature value is generated according to a preset key generation algorithm; the negotiated public key information of the key generation algorithm is stored in the shore device for use in generating the decryption key; Each set of information is compressed using the corresponding structured compression algorithm to obtain the corresponding compressed data. Based on the encryption key of the information set corresponding to each set of compressed data, the set of compressed data is encrypted to obtain set of encrypted data; All the encrypted data in the set and the corresponding feature values ​​are combined to obtain the alarm aggregation data corresponding to the ship end.

[0020] A third aspect of this invention discloses another ship alarm compression and feedback system based on time series clustering, the system comprising: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute some or all of the steps in the ship alarm compression and backhaul method based on time series clustering disclosed in the first aspect of the present invention.

[0021] The fourth aspect of the present invention discloses a computer storage medium storing computer instructions, which, when invoked, are used to execute some or all of the steps in the ship alarm compression and backhaul method based on time series clustering disclosed in the first aspect of the present invention.

[0022] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: This invention acquires multiple alarm messages sent by the ship within a preset historical time window, clusters the alarm messages based on time features to obtain multiple information sets, determines a structured compression algorithm based on the information features of each information set, and compresses and integrates the information sets to obtain aggregated alarm data. This enables time-series clustering and structured compression of ship alarm data, improves data transmission efficiency and storage optimization, and reduces transmission pressure and processing delay risks caused by alarm data redundancy. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart illustrating a ship alarm compression and feedback method based on time series clustering disclosed in an embodiment of the present invention.

[0025] Figure 2 This is a schematic diagram of a ship alarm compression and feedback system based on time series clustering disclosed in an embodiment of the present invention.

[0026] Figure 3 This is a schematic diagram of another ship alarm compression and feedback system based on time series clustering disclosed in an embodiment of the present invention. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0029] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0030] This invention discloses a method and system for compressing and transmitting ship alarm data based on time-series clustering. By acquiring multiple alarm messages sent by the ship within a preset historical time window, the alarm messages are clustered based on time characteristics to obtain multiple information sets. A structured compression algorithm is determined according to the information characteristics of each information set, and the information sets are compressed and integrated to obtain aggregated alarm data. This enables time-series clustering and structured compression of ship alarm data, improving data transmission efficiency and storage optimization, and reducing transmission pressure and processing delay risks caused by alarm data redundancy. Detailed explanations follow.

[0031] Example 1 Please see Figure 1 , Figure 1 This is a flowchart illustrating a ship alarm compression and feedback method based on time series clustering disclosed in an embodiment of the present invention. Figure 1 The described time-series clustering-based ship alarm compression and feedback method can be applied to data processing systems / data processing equipment / data processing servers (including local processing servers or cloud processing servers). Figure 1As shown, the ship alarm compression and feedback method based on time series clustering may include the following operations: 101. Obtain multiple alarm messages sent by the ship within a preset historical time window.

[0032] Optionally, the ship end can be an integrated engine room alarm monitoring system (AMS), an integrated bridge system (IBS), an onboard automatic identification system (AIS) gateway installed in the ship's backbone network, or an edge computing node of an intelligent ship data center deployed in the main control room. This invention does not limit the scope of the invention.

[0033] Optionally, the preset historical time window can be the most recent 10 minutes, 1 hour, 24 hours of the rolling slip, or an adaptive dynamic time period automatically triggered when an emergency or abnormal event occurs on the ship. This invention does not limit the time window.

[0034] Optionally, the alarm information may be an alarm for excessive temperature of diesel engine main bearing, an alarm for low oil pressure of fuel separator, an alarm for electronic chart corridor crossing safety warning, or a discrete fault code for reverse power trip of marine generator. This invention does not limit the alarm information.

[0035] 102. Based on the time characteristics corresponding to the alarm information, all alarm information is clustered to obtain multiple information sets.

[0036] Optionally, the time feature can be an absolute high-precision timestamp of the alarm trigger, the millisecond difference between the clocks of two adjacent alarm messages, or the alarm pulse trigger frequency within a set observation period; this invention does not impose any limitations.

[0037] Optionally, the clustering can be performed using a one-dimensional K-Means clustering algorithm, a density-based DBSCAN time-series clustering algorithm, or a dynamic time warping (DTW) adjacency matrix hierarchical clustering algorithm; this invention does not limit the specific clustering method.

[0038] Optionally, the information set can be a cluster of high-incidence fault data divided according to time sequence, or a sequence of concurrent alarm information triggered by the same chain reaction; the present invention does not limit this.

[0039] 103. Based on the information features corresponding to each information set, determine the structured compression algorithm corresponding to each information set.

[0040] Optionally, the information feature can be the average character length of the message string in the information set, the tree-like nesting depth of the message data structure, or the statistical probability distribution of repeated key-value pairs. This invention does not limit the features.

[0041] Optionally, the structured compression algorithm can be the Zstandard algorithm with a dynamically pre-compiled dictionary, the Protocol Buffers streaming compression mechanism for key-value pair serialization, adaptive Huffman coding, or a run-length compression scheme based on columnar storage; this invention does not limit the specific algorithm.

[0042] 104. Compress and integrate each information set according to the structured compression algorithm to obtain the corresponding alarm aggregation data at the ship end.

[0043] Optionally, the alarm aggregation data is used to send to the corresponding shore-based equipment on the ship.

[0044] Optionally, the alarm aggregation data can be a compressed file that has been encapsulated in binary streaming, a highly simplified structured JSON data stream, or a specific protocol encrypted message with a self-verified checksum; this invention does not impose any limitations.

[0045] Optionally, the shore-based equipment can be a remote technical support center server deployed at the shipping company's headquarters, a Vessel Traffic Management System (VTS) base station of the Maritime Safety Administration, or a cloud-based distributed data center that provides global navigation safety data hosting services; this invention does not limit the scope of the equipment.

[0046] As can be seen, the above-described embodiments of the invention acquire multiple alarm messages sent by the ship within a preset historical time window, cluster the alarm messages based on time features to obtain multiple information sets, determine a structured compression algorithm based on the information features of each information set, and compress and integrate the information sets to obtain alarm aggregate data. This enables time-series clustering and structured compression of ship alarm data, improves data transmission efficiency and storage optimization, and reduces transmission pressure and processing delay risks caused by alarm data redundancy.

[0047] As an optional embodiment, in the above steps, based on the time characteristics corresponding to the alarm information, all alarm information is clustered to obtain multiple information sets, including: Identify multiple hazardous alarm messages among all alarm messages; For each alarm message, calculate the weighted sum of the time differences between the trigger time of the alarm message and the trigger time of each danger alarm message to obtain the time characteristic parameters corresponding to the alarm message; All alarm information is clustered based on time feature parameters to obtain multiple information sets.

[0048] Optionally, each information set includes multiple alarm messages where the parameter difference between the time feature parameters of each other is less than a preset difference threshold.

[0049] Optionally, the time characteristic parameter can be a scalar evaluation value reflecting the temporal proximity between the alarm information and all hazard alarm sources, or a distance mapping score in a nonlinear time decay space; this invention does not limit this.

[0050] Optionally, the difference threshold can be a fixed numerical threshold such as 0.05, or a dynamic bandwidth variance value retrieved from the system configuration file based on the current danger level encountered by the ship. This invention does not impose any limitations on this.

[0051] As can be seen, through the above optional embodiments, by identifying multiple dangerous alarm messages among all alarm messages, calculating the weighted sum of the time differences between each alarm message and the dangerous alarm messages to obtain the time feature parameter, and clustering the alarm messages based on the time feature parameter to obtain the information set, accurate alarm clustering based on the time sequence correlation of dangerous alarms is achieved, improving the rationality and relevance of the information set division, and reducing the risk of clustering bias caused by ignoring time features.

[0052] As an optional embodiment, in the above steps, identifying multiple hazardous alarm messages among all alarm messages includes: For each alarm message, the alarm message is input into the trained classifier to obtain the corresponding alarm type; The alarm information is input into the trained hazard prediction model to obtain the corresponding predicted hazard level; The number of times each alarm type appears in the classification results corresponding to all alarm information is counted to obtain the hazard weight; The product of the predicted risk level and the risk weight is calculated to obtain the risk parameter of the alarm information; All alarm messages are sorted from largest to smallest based on the hazard level parameter to obtain an information sequence; The first preset proportion of alarm messages in the information sequence are identified as multiple dangerous alarm messages.

[0053] Optionally, the classifier can be a support vector machine (SVM) multi-classification model, a Naive Bayes text classifier, a semantic-level text classification network based on a lightweight BERT architecture, or a cascaded decision tree model; this invention does not impose any limitations.

[0054] Optionally, the alarm type can be mechanical component fatigue, electrical circuit short circuit, navigation safety deviation, or hydraulic system failure; this invention does not limit the types.

[0055] Optionally, the hazard prediction model can be a deep multilayer perceptron (MLP) regression network, an extreme gradient boosting tree (XGBoost) risk quantification model, or a long short-term memory (LSTM) neural network; this invention does not limit the specific model.

[0056] Optionally, the predicted risk level can be a continuous risk probability factor mapped in the interval [0, 1], or a quantified entropy value representing the probability of the system running out of control. This invention does not limit the specific risk level.

[0057] Optionally, the preset ratio can be a fixed segmentation ratio, such as 5% or 10%, or a boundary segmentation point obtained by dynamic convergence calculation based on the total number of alarms. This invention does not limit the specific ratio.

[0058] As can be seen, through the above optional embodiments, by inputting each alarm information into a classifier to obtain the alarm type, inputting it into a hazard prediction model to obtain the predicted hazard level, counting the frequency of alarm types to obtain the hazard weight, and calculating the product of the predicted hazard level and the hazard weight to obtain the hazard parameter, and sorting and filtering hazard alarm information based on the hazard parameter, accurate hazard alarm identification based on the fusion of type frequency and hazard level is achieved, improving the accuracy and objectivity of hazard alarm screening, and reducing the risk of hazard omission due to single-dimensional judgment.

[0059] As an optional embodiment, in the above steps, when calculating the time feature parameters, the calculation weight corresponding to each time difference is the product of the first weight and the second weight; the first weight is proportional to the content similarity between the alarm information and the corresponding danger alarm information; the second weight is proportional to the similarity between the alarm device type of the alarm information and the corresponding danger alarm information.

[0060] Optionally, the content similarity can be the text literal coverage calculated using the edit distance algorithm, or the cosine similarity calculated based on the word vector space; this invention does not limit the specific similarity.

[0061] Optionally, the similarity between the alarm device types can be the inverse of the shortest path between two device nodes calculated based on a preset ship equipment topology tree, or the classification association score determined based on the degree of overlap of functional attributes. This invention does not limit the similarity.

[0062] As can be seen, through the above optional embodiments, by using a product weight in which the first weight is proportional to the content similarity and the second weight is proportional to the alarm device type similarity when calculating the time feature parameters, adaptive time difference weighting based on content and device type is achieved, which improves the rationality and pertinence of time feature parameter calculation and reduces the risk of time correlation deviation caused by improper weight allocation.

[0063] As an optional embodiment, the step above, determining the structured compression algorithm corresponding to each information set based on the information features corresponding to each information set, includes: For each information set, each alarm information in the information set is input into the trained data structure prediction model to obtain the data structure corresponding to each alarm information; optionally, the data structure prediction model is trained using a training dataset that includes multiple training alarm information and corresponding information data structure annotations. Based on the data structure corresponding to all alarm information, determine the set structure characteristics corresponding to this information set; Among multiple candidate compression algorithms, the compression algorithm corresponding to the set's structural features is determined, thus identifying the structured compression algorithm corresponding to the information set.

[0064] Optionally, the data structure prediction model can be a text pattern recognition network based on autoregressive sequence generation, a template matching engine with a rule tree parser, or a probabilistic graphical neural network model for inferring JSON Schema; this invention does not limit the model.

[0065] Optionally, the data structure can be a binary fixed-length frame structure with a fixed offset, an XML nested structure with parent-child node relationships, or a sparse JSON multilevel dictionary structure containing nested arrays; this invention does not limit the types of structures.

[0066] Optionally, the set structure feature can be a high-dimensional tensor representing the similarity of various data structures within the information set, or an abstractly extracted general metadata pattern feature matrix; this invention does not impose any limitations.

[0067] As can be seen, through the above optional embodiments, by inputting the alarm information in each information set into the data structure prediction model to obtain the data structure, determining the set structure characteristics based on all data structures, and determining the corresponding structured compression algorithm from the candidate compression algorithms, intelligent compression algorithm selection based on structure characteristics is realized, improving the adaptability and efficiency of compression processing, and reducing the risk of poor compression effect due to algorithm mismatch.

[0068] As an optional embodiment, the step above, determining the set structure characteristics corresponding to the information set based on the data structure corresponding to all alarm information, includes: For each preset structure type, count the number of times that structure type appears in the data structure corresponding to all alarm information; Determine the order of occurrence counts within the calculated frequency sequence; optionally, the frequency sequence can be obtained by arranging all structure types in descending order of frequency. When the sequential position precedes the preset reference position, the structure type is determined as the preferred structure type; The matrix consisting of all preferred structure types and their corresponding data content in each alarm message is determined as the set structure feature corresponding to the information set.

[0069] Optionally, the structure type can be a floating-point sensor value array, an enumerated status character item, a long text fault description segment, or a nested timestamp sequence; this invention does not impose any limitations.

[0070] Optionally, the reference position can be a fixed ranking threshold index, such as the top 3, top 5, or a Pareto truncation position whose cumulative frequency reaches 75% of the total. This invention does not limit this.

[0071] Optionally, the matrix composed of the data content can be a standardized two-dimensional character feature matrix after row or column alignment, or a high-dimensional dense numerical tensor after missing items have been filled in; this invention does not impose any limitations.

[0072] As can be seen, through the above optional embodiments, by counting the number of times each structure type appears in the alarm information data structure, determining its order position in the frequency sequence, and selecting the preferred structure type, a matrix is ​​formed as the set structure feature, realizing the preferred extraction of structure features based on the frequency of occurrence, improving the representativeness and refinement of the set structure features, and reducing the risk of feature redundancy caused by low-frequency structure interference.

[0073] As an optional embodiment, the step described above, determining the compression algorithm corresponding to the set's structural features from multiple candidate compression algorithms to identify the structured compression algorithm corresponding to the information set, includes: The set structure features are input into the trained algorithm parameter prediction model to obtain the corresponding prediction algorithm parameters; Calculate the parameter similarity between the historical algorithm application parameters and the predicted algorithm parameters for each candidate compression algorithm; The candidate compression algorithm with the highest parameter similarity is determined as the structured compression algorithm corresponding to this information set.

[0074] Optionally, the prediction algorithm parameters include at least one of the following: data splitting parameters, encryption parameters, data combination parameters, and algorithm model architecture parameters.

[0075] Optionally, the algorithm parameter prediction model is trained using a training dataset that includes multiple training set structural features and corresponding algorithm parameter annotations.

[0076] Optionally, the parameter prediction model of the algorithm can be a feedforward neural network regressor based on a multi-task learning architecture, or a gradient boosting decision tree network that incorporates a nonlinear transformation layer; this invention does not limit the model.

[0077] Optionally, the data splitting parameter in the prediction algorithm can be the block size, the encryption parameter can be the initialization vector (IV) length, the data combination parameter can be the multi-field concatenation order, and the algorithm model architecture parameter can be the maximum search sliding window depth of the compressed dictionary. This invention does not impose any limitations on these parameters.

[0078] Optionally, the parameters applied by the historical algorithm can be the baseline vector of the optimal compression core operating parameters stored in the system's historical log database and invoked in multiple successful transmission cases in the past; this invention does not impose any limitations on this.

[0079] Optionally, the similarity parameter can be the negative Manhattan distance, Pearson correlation coefficient, or the inverse of the normalized Euclidean distance calculated in the algorithm parameter space; this invention does not limit the specific parameters.

[0080] As can be seen, through the above optional embodiments, by inputting the set structure features into the algorithm parameter prediction model to obtain the predicted algorithm parameters, calculating the similarity with the historical application parameters of the candidate compression algorithms and selecting the highest one as the structured compression algorithm, accurate algorithm matching based on structural features and historical applications is achieved, improving the reliability and optimality of compression algorithm selection, and reducing the risk of insufficient compression performance due to algorithm parameter deviation.

[0081] As an optional embodiment, the above steps, including compressing and integrating each information set according to a structured compression algorithm to obtain the corresponding alarm aggregation data at the ship end, include: Based on the preset feature processing algorithm, the feature value corresponding to the set structure feature of each information set is determined; Generate an encryption key corresponding to the feature value based on the preset key generation algorithm; Each information set is compressed using the corresponding structured compression algorithm to obtain the corresponding compressed set data; Based on the encryption key of the information set corresponding to each set of compressed data, the compressed data of the set is encrypted to obtain the encrypted data of the set; All encrypted data sets and their corresponding feature values ​​are combined to obtain the corresponding alarm aggregation data from the ship's end.

[0082] Optionally, the negotiated public key information of the key generation algorithm is stored in the shore device for use in generating the decryption key.

[0083] Optionally, the feature processing algorithm can be the salted secure hash algorithm SHA-256, the message digest algorithm MD5, or the SM3 national cryptographic hash function with collision resistance. This invention does not limit the specific algorithm.

[0084] Optionally, the feature value can be a fixed-length hash string in hexadecimal form or a fixed-length binary digest byte array; this invention does not limit the type of hash value.

[0085] Optionally, the key generation algorithm can be a stream key derivation algorithm based on elliptic curve Diffie-Hellman key exchange (ECDH), a commercial cryptographic key derivation function HKDF with timestamp seed, or a session key negotiation mechanism based on the national cryptographic SM2 asymmetric algorithm. This invention does not limit the specific algorithm.

[0086] Optionally, the decryption key can be a symmetric encryption key obtained by extracting a pre-stored private key from the shore-side device and performing calculations with the negotiated public key information and feature value information received from the satellite communication link. This invention does not impose any limitations on this.

[0087] Optionally, the compressed data in this set can be a compact, high-density byte stream that has been compressed by pruning a structured dictionary and stripping away redundant key names; this invention does not limit this.

[0088] As can be seen, through the above optional embodiments, by determining the feature values ​​corresponding to the structural features of the set based on the feature processing algorithm, generating encryption keys, performing structured compression and encryption on the information set to obtain encrypted set data, and combining all the encrypted set data and feature values ​​to obtain alarm aggregate data, the structured compression, encryption and aggregate transmission of alarm data is realized, improving data security and transmission efficiency, and reducing the risk of data leakage or transmission overhead caused by unencrypted or uncompressed data.

[0089] Example 2 Please see Figure 2 , Figure 2 This is a schematic diagram of a ship alarm compression and feedback system based on time series clustering, as disclosed in an embodiment of the present invention. Figure 2 The described time-series clustering-based ship alarm compression and feedback system can be applied to data processing systems / data processing equipment / data processing servers (including local processing servers or cloud processing servers). For example... Figure 2 As shown, the ship alarm compression and feedback system based on time series clustering may include: The acquisition module 201 is used to acquire multiple alarm messages sent by the ship within a preset historical time window.

[0090] Clustering module 202 is used to cluster all alarm information based on the time characteristics corresponding to the alarm information to obtain multiple information sets.

[0091] The determination module 203 is used to determine the structured compression algorithm corresponding to each information set based on the information features corresponding to each information set.

[0092] The processing module 204 is used to compress and integrate each information set according to the structured compression algorithm to obtain the corresponding alarm aggregation data at the ship end.

[0093] Optionally, the alarm aggregation data is used to send to the corresponding shore-based equipment on the ship.

[0094] As can be seen, the above-described embodiments of the invention acquire multiple alarm messages sent by the ship within a preset historical time window, cluster the alarm messages based on time features to obtain multiple information sets, determine a structured compression algorithm based on the information features of each information set, and compress and integrate the information sets to obtain alarm aggregate data. This enables time-series clustering and structured compression of ship alarm data, improves data transmission efficiency and storage optimization, and reduces transmission pressure and processing delay risks caused by alarm data redundancy.

[0095] As an optional implementation, the clustering module clusters all alarm information based on the time characteristics corresponding to the alarm information to obtain multiple information sets in the following specific ways: Identify multiple hazardous alarm messages among all alarm messages; For each alarm message, calculate the weighted sum of the time differences between the trigger time of the alarm message and the trigger time of each danger alarm message to obtain the time characteristic parameters corresponding to the alarm message; All alarm information is clustered based on time feature parameters to obtain multiple information sets; each information set includes multiple alarm information where the parameter difference between the time feature parameters is less than a preset difference threshold.

[0096] As can be seen, through the above optional embodiments, by identifying multiple dangerous alarm messages among all alarm messages, calculating the weighted sum of the time differences between each alarm message and the dangerous alarm messages to obtain the time feature parameter, and clustering the alarm messages based on the time feature parameter to obtain the information set, accurate alarm clustering based on the time sequence correlation of dangerous alarms is achieved, improving the rationality and relevance of the information set division, and reducing the risk of clustering bias caused by ignoring time features.

[0097] As an optional implementation, the clustering module determines the specific method by which multiple hazardous alarm messages are identified among all alarm messages, including: For each alarm message, the alarm message is input into the trained classifier to obtain the corresponding alarm type; The alarm information is input into the trained hazard prediction model to obtain the corresponding predicted hazard level; The number of times each alarm type appears in the classification results corresponding to all alarm information is counted to obtain the hazard weight; The product of the predicted risk level and the risk weight is calculated to obtain the risk parameter of the alarm information; All alarm messages are sorted from largest to smallest based on the hazard level parameter to obtain an information sequence; The first preset proportion of alarm messages in the information sequence are identified as multiple dangerous alarm messages.

[0098] As can be seen, through the above optional embodiments, by inputting each alarm information into a classifier to obtain the alarm type, inputting it into a hazard prediction model to obtain the predicted hazard level, counting the frequency of alarm types to obtain the hazard weight, and calculating the product of the predicted hazard level and the hazard weight to obtain the hazard parameter, and sorting and filtering hazard alarm information based on the hazard parameter, accurate hazard alarm identification based on the fusion of type frequency and hazard level is achieved, improving the accuracy and objectivity of hazard alarm screening, and reducing the risk of hazard omission due to single-dimensional judgment.

[0099] As an optional embodiment, when calculating the time feature parameters, the calculation weight corresponding to each time difference is the product of the first weight and the second weight; the first weight is proportional to the content similarity between the alarm information and the corresponding danger alarm information; the second weight is proportional to the similarity between the alarm device type of the alarm information and the corresponding danger alarm information.

[0100] As can be seen, through the above optional embodiments, by using a product weight in which the first weight is proportional to the content similarity and the second weight is proportional to the alarm device type similarity when calculating the time feature parameters, adaptive time difference weighting based on content and device type is achieved, which improves the rationality and pertinence of time feature parameter calculation and reduces the risk of time correlation deviation caused by improper weight allocation.

[0101] As an optional embodiment, the determining module determines the specific method of the structured compression algorithm corresponding to each information set based on the information features corresponding to each information set, including: For each information set, each alarm information in the information set is input into the trained data structure prediction model to obtain the data structure corresponding to each alarm information; optionally, the data structure prediction model is trained using a training dataset that includes multiple training alarm information and corresponding information data structure annotations. Based on the data structure corresponding to all alarm information, determine the set structure characteristics corresponding to this information set; Among multiple candidate compression algorithms, the compression algorithm corresponding to the set's structural features is determined, thus identifying the structured compression algorithm corresponding to the information set.

[0102] As can be seen, through the above optional embodiments, by inputting the alarm information in each information set into the data structure prediction model to obtain the data structure, determining the set structure characteristics based on all data structures, and determining the corresponding structured compression algorithm from the candidate compression algorithms, intelligent compression algorithm selection based on structure characteristics is realized, improving the adaptability and efficiency of compression processing, and reducing the risk of poor compression effect due to algorithm mismatch.

[0103] As an optional embodiment, the determining module determines the specific method of the set structure characteristics corresponding to the information set based on the data structure corresponding to all alarm information, including: For each preset structure type, count the number of times that structure type appears in the data structure corresponding to all alarm information; Determine the order of occurrence counts within the calculated frequency sequence; optionally, the frequency sequence can be obtained by arranging all structure types in descending order of frequency. When the sequential position precedes the preset reference position, the structure type is determined as the preferred structure type; The matrix consisting of all preferred structure types and their corresponding data content in each alarm message is determined as the set structure feature corresponding to the information set.

[0104] As can be seen, through the above optional embodiments, by counting the number of times each structure type appears in the alarm information data structure, determining its order position in the frequency sequence, and selecting the preferred structure type, a matrix is ​​formed as the set structure feature, realizing the preferred extraction of structure features based on the frequency of occurrence, improving the representativeness and refinement of the set structure features, and reducing the risk of feature redundancy caused by low-frequency structure interference.

[0105] As an optional embodiment, the determining module determines the compression algorithm corresponding to the set structure features from multiple candidate compression algorithms, and the specific method for determining the structured compression algorithm corresponding to the information set includes: The set structure features are input into the trained algorithm parameter prediction model to obtain the corresponding prediction algorithm parameters; optionally, the prediction algorithm parameters include at least one of data splitting parameters, encryption parameters, data combination parameters and algorithm model architecture parameters; the algorithm parameter prediction model is trained on a training dataset that includes multiple training set structure features and corresponding algorithm parameter annotations; Calculate the parameter similarity between the historical algorithm application parameters and the predicted algorithm parameters for each candidate compression algorithm; The candidate compression algorithm with the highest parameter similarity is determined as the structured compression algorithm corresponding to this information set.

[0106] As can be seen, through the above optional embodiments, by inputting the set structure features into the algorithm parameter prediction model to obtain the predicted algorithm parameters, calculating the similarity with the historical application parameters of the candidate compression algorithms and selecting the highest one as the structured compression algorithm, accurate algorithm matching based on structural features and historical applications is achieved, improving the reliability and optimality of compression algorithm selection, and reducing the risk of insufficient compression performance due to algorithm parameter deviation.

[0107] As an optional embodiment, the processing module compresses and integrates each information set according to a structured compression algorithm to obtain the specific method of the corresponding alarm aggregation data at the ship end, including: Based on the preset feature processing algorithm, the feature value corresponding to the set structure feature of each information set is determined; Based on a preset key generation algorithm, an encryption key corresponding to the feature value is generated; optionally, the negotiated public key information of the key generation algorithm is stored in the shore device for use in generating the decryption key. Each information set is compressed using the corresponding structured compression algorithm to obtain the corresponding compressed set data; Based on the encryption key of the information set corresponding to each set of compressed data, the compressed data of the set is encrypted to obtain the encrypted data of the set; All encrypted data sets and their corresponding feature values ​​are combined to obtain the corresponding alarm aggregation data from the ship's end.

[0108] As can be seen, through the above optional embodiments, by determining the feature values ​​corresponding to the structural features of the set based on the feature processing algorithm, generating encryption keys, performing structured compression and encryption on the information set to obtain encrypted set data, and combining all the encrypted set data and feature values ​​to obtain alarm aggregate data, the structured compression, encryption and aggregate transmission of alarm data is realized, improving data security and transmission efficiency, and reducing the risk of data leakage or transmission overhead caused by unencrypted or uncompressed data.

[0109] Example 3 Please see Figure 3 , Figure 3 This is another ship alarm compression and feedback system based on time series clustering disclosed in the embodiments of the present invention. Figure 3 The described time-series clustering-based ship alarm compression and feedback system is applied in a data processing system / data processing equipment / data processing server (wherein, the server includes a local processing server or a cloud processing server). For example... Figure 3 As shown, the ship alarm compression and feedback system based on time series clustering may include: Memory 301 storing executable program code; Processor 302 coupled to memory 301; The processor 302 calls the executable program code stored in the memory 301 to execute the steps of the ship alarm compression and feedback method based on time series clustering described in Embodiment 1.

[0110] Example 4 This invention discloses a computer read storage medium that stores a computer program for electronic data interchange, wherein the computer program causes a computer to execute the steps of the ship alarm compression and backhaul method based on time series clustering described in Embodiment 1.

[0111] Example 5 This invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to perform the steps of the ship alarm compression and backhaul method based on time series clustering described in Embodiment 1.

[0112] The foregoing has described specific embodiments of this specification; other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims may be performed in a different order than those shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily have to follow the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0113] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0114] For ease of description, the above devices are described in terms of function, divided into various units. Of course, in implementing this specification, the functions of each unit can be implemented in one or more software and / or hardware components.

[0115] Those skilled in the art will understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, the embodiments of this specification can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the embodiments of this specification can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0116] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this specification. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0117] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0118] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0119] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0120] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0121] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0122] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0123] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0124] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0125] Finally, it should be noted that the ship alarm compression and feedback method and system based on time series clustering disclosed in the embodiments of the present invention are merely preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for compressed transmission of ship alarms based on time series clustering, characterized in that, The method includes: Acquire multiple alarm messages sent by the ship within a preset historical time window; Based on the time characteristics corresponding to the alarm information, all the alarm information is clustered to obtain multiple information sets; Based on the information features corresponding to each information set, a structured compression algorithm corresponding to each information set is determined. The structured compression algorithm is used to compress and integrate each set of information to obtain the alarm aggregation data corresponding to the ship end; the alarm aggregation data is used to send to the shore-side equipment corresponding to the ship end.

2. The ship alarm compression and feedback method based on time series clustering according to claim 1, characterized in that, Based on the time characteristics corresponding to the alarm information, all the alarm information is clustered to obtain multiple information sets, including: Identify multiple dangerous alarm messages among all the aforementioned alarm messages; For each alarm message, calculate the weighted summation average of the time differences between the trigger time of the alarm message and the trigger time of each danger alarm message to obtain the time characteristic parameter corresponding to the alarm message; All the alarm information is clustered and divided based on the time feature parameters to obtain multiple information sets; each information set includes multiple alarm information where the parameter difference between the time feature parameters is less than a preset difference threshold.

3. The ship alarm compression and feedback method based on time series clustering according to claim 2, characterized in that, The determination of multiple dangerous alarm messages among all the alarm messages includes: For each alarm message, the alarm message is input into the trained classifier to obtain the corresponding alarm type; The alarm information is input into the trained hazard prediction model to obtain the corresponding predicted hazard level; The number of times each alarm type appears in the classification results corresponding to all alarm information is counted to obtain the hazard weight; The product of the predicted risk level and the risk weight is calculated to obtain the risk level parameter of the alarm information; Based on the aforementioned risk parameter, all alarm messages are sorted from largest to smallest to obtain an information sequence; The first preset proportion of the alarm information in the information sequence is determined as multiple danger alarm information.

4. The ship alarm compression and feedback method based on time series clustering according to claim 2, characterized in that, When calculating the time feature parameters, the calculation weight corresponding to each time difference is the product of a first weight and a second weight; the first weight is proportional to the content similarity between the alarm information and the corresponding danger alarm information; the second weight is proportional to the similarity between the alarm information and the alarm device type of the corresponding danger alarm information.

5. The ship alarm compression and feedback method based on time series clustering according to claim 1, characterized in that, The step of determining the structured compression algorithm corresponding to each information set based on the information features corresponding to each information set includes: For each of the aforementioned information sets, each of the aforementioned alarm information in the information set is input into a trained data structure prediction model to obtain the data structure corresponding to each of the aforementioned alarm information; the data structure prediction model is trained using a training dataset that includes multiple training alarm information and corresponding information data structure annotations; Based on the data structure corresponding to all the alarm information, determine the set structure characteristics corresponding to the information set; Among multiple candidate compression algorithms, the compression algorithm corresponding to the set structure features is determined, and thus the structured compression algorithm corresponding to the information set is determined.

6. The ship alarm compression and backhaul method based on time series clustering according to claim 5, characterized in that, The step of determining the set structure characteristics corresponding to the information set based on the data structure corresponding to all the alarm information includes: For each preset structure type, count the number of times that structure type appears in all the data structures corresponding to the alarm information; Determine the order of the occurrence count in the calculated frequency sequence; the frequency sequence is obtained by arranging all the structure types in descending order of their occurrence counts. When the ordered position precedes the preset reference position, the structure type is determined as the preferred structure type; The matrix consisting of all the preferred structure types and the corresponding data content in each alarm message is determined as the set structure feature corresponding to the information set.

7. The ship alarm compression and backhaul method based on time series clustering according to claim 5, characterized in that, The step of determining the compression algorithm corresponding to the set's structural features from multiple candidate compression algorithms, and thus identifying it as the structured compression algorithm corresponding to the information set, includes: The set structure features are input into the trained algorithm parameter prediction model to obtain the corresponding prediction algorithm parameters; the prediction algorithm parameters include at least one of data splitting parameters, encryption parameters, data combination parameters, and algorithm model architecture parameters; the algorithm parameter prediction model is trained on a training dataset that includes multiple training set structure features and corresponding algorithm parameter annotations; Calculate the parameter similarity between the historical algorithm application parameters and the predicted algorithm parameters for each candidate compression algorithm; The candidate compression algorithm with the highest parameter similarity is determined as the structured compression algorithm corresponding to the information set.

8. The ship alarm compression and feedback method based on time series clustering according to claim 5, characterized in that, The step of compressing and integrating each information set according to the structured compression algorithm to obtain the alarm aggregation data corresponding to the ship includes: Based on a preset feature processing algorithm, the feature value corresponding to the set structure feature of each information set is determined; An encryption key corresponding to the feature value is generated according to a preset key generation algorithm; the negotiated public key information of the key generation algorithm is stored in the shore device for use in generating the decryption key; Each set of information is compressed using the corresponding structured compression algorithm to obtain the corresponding compressed data. Based on the encryption key of the information set corresponding to each set of compressed data, the set of compressed data is encrypted to obtain set of encrypted data; All the encrypted data in the set and the corresponding feature values ​​are combined to obtain the alarm aggregation data corresponding to the ship end.

9. A ship alarm compression and feedback system based on time series clustering, characterized in that, The system includes: The acquisition module is used to acquire multiple alarm messages sent by the ship within a preset historical time window; The clustering module is used to cluster all the alarm information based on the time characteristics corresponding to the alarm information to obtain multiple information sets; The determination module is used to determine the structured compression algorithm corresponding to each information set based on the information features corresponding to each information set. The processing module is used to compress and integrate each of the information sets according to the structured compression algorithm to obtain the alarm aggregation data corresponding to the ship end; the alarm aggregation data is used to be sent to the shore-side equipment corresponding to the ship end.

10. A ship alarm compression and feedback system based on time series clustering, characterized in that, The system includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the ship alarm compression and backhaul method based on time series clustering as described in any one of claims 1-8.

Images