Database establishment method and device based on ship vdr data

Abstract

The application relates to a database establishment method and equipment based on ship VDR data, and the method comprises the following steps: acquiring original VDR data in a VDR equipment and an associated storage medium; analyzing the original VDR data to obtain field characteristics of the original VDR data; classifying the original VDR data based on the field characteristics to respectively obtain time series data, structured data and vector data; constructing a collaborative three-level associated database, wherein the three-level associated database comprises a time series database, a structured database and a vector database; and respectively storing the time series data, the structured data and the vector data into corresponding databases. Efficient application of VDR data is realized.

Description

Technical Field

[0001] This invention relates to the field of ship electronic information engineering, and specifically to a method and equipment for establishing a database based on ship VDR data. Background Technology

[0002] Ship navigation data recorders (VDRs) are core devices for recording a ship's navigation status, often referred to as the ship's "black box." The data they generate encompasses various types, including text (including time-series data), audio, and video. Currently, there is no unified application or analysis solution for VDR data. Instead, data is simply downloaded directly from the VDR device, stored on a local hard drive or in a single-type database, and then played back and performed basic statistical analysis using basic playback software or query tools. This approach fails to fully exploit the value of VDR data.

[0003] Therefore, how to fully explore the value of VDR data has become an urgent technical problem to be solved. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a method and apparatus for establishing a database based on ship VDR data, in order to solve the technical problem of how to fully exploit the value of VDR data in the prior art.

[0005] In a first aspect, embodiments of the present invention provide a database establishment method based on ship VDR data, the method comprising: Retrieve raw VDR data from the VDR device and associated storage media; The original VDR data is parsed to obtain the field features of the original VDR data; The original VDR data is classified based on the field features to obtain time-series data, structured data, and vector data, respectively. Construct a collaborative three-level relational database, which includes a time-series database, a structured database, and a vector database; The time-series data, the structured data, and the vector data are stored in their respective databases.

[0006] Furthermore, the parsing of the original VDR data includes: Based on the basic type of the original VDR data, the original VDR data is divided into video data, audio data, and text data. The video data and the audio data are stored on a hard disk, and the audio data is converted into audio vector data. The text data and the speech vector data are temporarily stored as a binary stream; Extract the frame features of the binary stream data, perform bitwise operations to extract the type identifier, and obtain the field features.

[0007] Furthermore, storing the time-series data, the structured data, and the vector data into corresponding types of databases includes: Based on the aforementioned field characteristics, cross-database association identifiers and cross-database transformation rules are constructed for the inbound data; The time-series data, the structured data, and the vector data are stored in their respective target databases, and the cross-database association identifier is stored in a non-target database. A cross-database data conversion middleware is constructed, the cross-database conversion rules are stored in the middleware, and the middleware is connected to the third-level relational database for communication. It is used to receive call requests from non-target databases, pull corresponding data from the target database, and convert the data into target type data in real time according to the cross-database conversion rules.

[0008] Furthermore, the step of storing the time-series data, the structured data, and the vector data into the corresponding target databases, and storing the cross-database association identifier into a non-target database, includes: Based on the field features, multiple types of adapted data are identified in the original VDR data. The multiple types of adapted data are cross-database data that can be called across databases. Quantify the degree of compatibility between each type of multi-type adaptive data and each of the databases; The database with the highest degree of compatibility is selected as the target database; The multi-type adaptation data is stored in the target database, and the cross-database association identifier is stored in the non-target database.

[0009] Furthermore, the identification of multiple types of adaptation data in the original VDR data based on the field features includes: Each original VDR data is determined to possess at least two of the following field features: time-series fields, structured fields, and vector fields. When at least two of the time-series field, the structured field, and the vector field are present, the current data is determined to be the multi-type adapted data.

[0010] Furthermore, quantifying the degree of adaptation between each of the multi-type adaptation data and each of the databases includes: For each type of multi-type adaptive data, multi-dimensional feature analysis is performed to determine the priority weight of data usage. The priority of the query type for each type of multi-type adaptive data is determined based on a pre-defined data query pattern. The degree of adaptation is determined by weighted summation of the multi-type adaptation data based on the usage priority weight and the query type priority.

[0011] Furthermore, multi-dimensional feature analysis is performed on each type of multi-adaptation data to determine the priority weights for data usage, including: Calculate the number and distribution density of timestamps in the multi-type adaptation data to determine the expected frequency of time series usage; Identify entity identifiers and relational attributes in the multi-type adaptive data to determine the probability of statistical structured application. The value density of semantic analysis is determined by detecting the proportion and complexity of natural language content in various types of adapted data.

[0012] Secondly, embodiments of the present invention provide a database establishment apparatus based on ship VDR data, the apparatus comprising: The acquisition module is used to acquire raw VDR data from the VDR device and associated storage media; The parsing module is used to parse the original VDR data to obtain the field features of the original VDR data; The classification module is used to classify the original VDR data based on the field features to obtain time-series data, structured data, and vector data; The database construction module is used to build a collaborative three-level relational database, which includes a time-series database, a structured database, and a vector database. The data storage module is used to store the time-series data, the structured data, and the vector data into the corresponding databases, respectively.

[0013] Thirdly, embodiments of the present invention provide a computer device, including: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the method described in the first aspect or any corresponding embodiment thereof.

[0014] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing computer instructions that cause a computer to perform the method described in the first aspect or any of its corresponding embodiments.

[0015] The method provided in this application has the following beneficial effects: The database establishment method based on ship VDR data in this application involves acquiring raw VDR data from the VDR device and associated storage medium; parsing the raw VDR data to obtain its field features; classifying the raw VDR data based on these field features to obtain time-series data, structured data, and vector data; constructing a collaborative three-level associated database, comprising a time-series database, a structured database, and a vector database; and storing the time-series data, structured data, and vector data in their respective databases. The three-level database architecture is designed based on differentiated data characteristics. The time-series database ensures high-concurrency time-series query efficiency, the structured database simplifies entity association management, and the vector database supports semantic retrieval, meeting the efficient data access needs of different scenarios. It supports diverse functions such as data playback, voyage comparison, and semantic retrieval. Compared to existing solutions that only achieve basic playback and simple comparison, this application can meet the needs of multiple scenarios such as ship operation and maintenance, safety analysis, and training review, enhancing the practical application value of VDR data. By adopting a three-tiered hierarchical governance storage architecture for multi-type heterogeneous data in VDRs, the data characteristics are precisely matched with the database functions, enabling the efficient application of VDR data. Attached Figure Description

[0016] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating a database establishment method based on ship VDR data according to an embodiment of the present invention. Figure 2 This is a structural block diagram of a database establishment device based on ship VDR data according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, 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.

[0019] According to embodiments of the present invention, a method and apparatus for establishing a database based on ship VDR data are provided. It should be noted that the steps shown in the flowcharts in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowcharts, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0020] This embodiment provides a method for establishing a database based on ship VDR data. Figure 1 This is a flowchart of a database establishment method based on ship VDR data according to an embodiment of the present invention, such as... Figure 1 As shown, the process includes the following steps: S101. Obtain raw VDR data from the VDR device and associated storage medium. In this embodiment, based on the hardware characteristics of the VDR device and industry-standard transmission protocols, multiple access interfaces are configured, including CAN port, serial port, and network port, to support data reception of different transmission protocols. A priority queue and flow control mechanism are also designed to allocate transmission bandwidth according to data importance (e.g., alarm data has higher priority than ordinary logs) to avoid data congestion or loss. A download task module is constructed to split large-scale data into multiple single-chip tasks for parallel execution. After downloading, the data is temporarily stored on the hard disk according to type. Video data is stored in directories according to date and time, audio data retains its original file structure, and text data is temporarily stored in a temporary folder. Simultaneously, basic data attributes are recorded, such as ship ID, voyage ID, device number, basic data type, file name, file size, download timestamp, interface type, and data source, providing a basis for subsequent classification.

[0021] S102. The original VDR data is parsed to obtain the field characteristics of the original VDR data. In this embodiment, the original data is first transformed into structured, readable, and accurate data. For example, the original VDR data is cleaned by using a dual cleaning mechanism of rule verification and algorithm detection to filter out invalid data. For example, the original VDR data is compared with the fields required by the preset protocol. If the number of fields is missing, the corresponding data is confirmed to be invalid; if the original VDR data is missing a timestamp, the corresponding data is confirmed to be invalid; if a certain piece of original VDR data is all empty values, the corresponding data is confirmed to be invalid; if the number of data entries at the same time in the original VDR data is greater than a preset amount, the data at the same time needs to be deduplicated, and the duplicate data exceeding the preset amount is confirmed to be invalid data.

[0022] Based on the basic type of the original VDR data, the original VDR data is divided into video data, audio data, and text data; the video data and audio data are stored on the hard disk, and the audio data is converted into audio vector data; the text data and audio vector data are temporarily stored as binary streams; the frame features of the binary stream data are extracted, and the type identifier is extracted by bitwise operations to obtain the field features.

[0023] S103. Classify the original VDR data based on the field features to obtain time-series data, structured data, and vector data. Different data can be classified using field features. For example, time-series data is determined as follows: if the data contains continuous, uniform timestamps (e.g., time interval ≤ 1 second) and contains at least one numerical measurement field (e.g., at least one of speed, heading, voltage, current, and rotational speed), it is classified as time-series data. Structured data is determined as follows: if the data contains at least two entity association fields (e.g., ship ID and equipment ID, voyage ID and alarm ID, equipment ID and parameter ID), and the data is an entity attribute or event record (e.g., ship information, voyage information, equipment parameters, and alarm events), it is classified as structured data. Vector data determination: If the data contains unstructured text fields (e.g., text length ≥ 10 characters), or the text fields are descriptive content (e.g., operation logs, speech-to-text results, alarm descriptions, or abnormal phenomenon descriptions), then it is determined to be vector data.

[0024] S104. Construct a collaborative three-level relational database, which includes a time-series database, a structured database, and a vector database; construct a collaborative three-level database system to efficiently load the classified data into the database in a standardized format.

[0025] S105. The time-series data, structured data, and vector data are stored in their respective databases. Specifically, the time-series database is partitioned by voyage ID and time range, for example, one partition per hour, to avoid excessive data volume in a single partition; a combined index of ship ID, voyage ID, and unified timestamp is established to improve query efficiency by voyage and time range. The structured database is used for storing and managing entity-related data, adapting to the query requirements of ship information, alarm events, and other related data, and adopts a structure of general tables and specialized databases. The general tables may include ship information tables, voyage information tables, equipment information tables, alarm summary tables, and data routing log tables; the specialized databases include generator tables, load tables, and power system alarm tables. The vector database is used to store the semantic retrieval requirements of structured text, converting text data into vectors for storage, adapting to similar content query scenarios. The descriptive fields of texts such as operation logs, speech-to-text results, and alarm descriptions are semantically segmented. The segmented texts are preprocessed, and a pre-trained semantic recognition model is used to convert the processed texts into multi-dimensional vectors. The vector dimensions are consistent with the model output to ensure the accuracy of semantic representation. The vector data is stored in partitions by ship ID and voyage ID to reduce the traversal range during retrieval.

[0026] Furthermore, based on a three-tier database architecture, diverse application functions such as data playback, voyage analysis, and semantic retrieval are implemented to meet the needs of scenarios such as ship operation and maintenance, safety analysis, and training debriefing. For data playback, queries are first performed from the Redis cache; if no match is found, data is read from the time-series database or structured database, parsed, and returned to the front end, supporting segmented display by time axis and adjustable playback speed. Voyage analysis can also be implemented, for example, querying measurement data for a specific voyage through the time-series database to calculate indicators such as average speed and maximum load, and combining this with ship information and alarm data from the structured database to generate a voyage analysis report. Voyage comparison can also be implemented; for example, selecting two voyages and querying corresponding data in the three-tier database using voyage_id, comparing dimensions such as total frame count, time-series trends of key parameters, and alarm type distribution to generate an exportable report. Semantic retrieval receives user-input keywords, such as "generator voltage anomaly," through a vector database, converts them into vectors, performs similarity matching, and returns associated text data, corresponding voyage information, and time-series data for accurate retrieval. The database establishment method based on ship VDR data in this application involves acquiring raw VDR data from the VDR device and associated storage medium; parsing the raw VDR data to obtain its field features; classifying the raw VDR data based on these field features to obtain time-series data, structured data, and vector data; constructing a collaborative three-level associated database, comprising a time-series database, a structured database, and a vector database; and storing the time-series data, structured data, and vector data in their respective databases. The three-level database architecture is designed based on differentiated data characteristics. The time-series database ensures high-concurrency time-series query efficiency, the structured database simplifies entity association management, and the vector database supports semantic retrieval, meeting the efficient data access needs of different scenarios. It supports diverse functions such as data playback, voyage comparison, and semantic retrieval. Compared to existing solutions that only achieve basic playback and simple comparison, this application can meet the needs of multiple scenarios such as ship operation and maintenance, safety analysis, and training review, enhancing the practical application value of VDR data. By adopting a three-tiered hierarchical governance storage architecture for multi-type heterogeneous data in VDRs, the data characteristics are precisely matched with the database functions, enabling the efficient application of VDR data.

[0027] Ship VDR data includes multiple data types such as text, audio, and video. Some of this data simultaneously possesses time-series characteristics, structured relational characteristics, and semantic text characteristics. This type of multi-type adapted data is the core data for ship fault analysis and voyage review. How to differentiate the storage and retrieval of multi-type adapted data is the key to ensuring the improved utilization rate of core data. In addition, storing the same data in different databases for multi-type adapted data, such as storing equipment operation logs in time-series databases, relational databases, and vector databases at the same time, results in serious waste of storage space and difficulty in maintaining data consistency. Furthermore, selecting a single database for storage based on only one main characteristic, such as storing it only in a time-series database based on time-series characteristics, fails to fully utilize the value of other characteristics and leads to low query efficiency.

[0028] Therefore, in this embodiment, cross-database association identifiers and cross-database conversion rules are constructed for the inbound data based on the field features; multiple types of adaptable data are identified in the original VDR data based on the field features, and the multiple types of adaptable data are cross-database data that can be called across databases; the degree of adaptability of each of the multiple types of adaptable data with each of the databases is quantified; the database with the highest degree of adaptability is taken as the target database; the multiple types of adaptable data are stored in the target database, and the cross-database association identifier is stored in the non-target database; a cross-database data conversion middleware is constructed, the cross-database conversion rules are stored in the conversion middleware, and the conversion middleware is connected to the three-level association database for communication, used to receive call requests from non-target databases, pull corresponding data from the target database, and convert it into target type data in real time according to the cross-database conversion rules. A unique main storage database for the multiple types of adaptable data is determined through a multi-dimensional weighted decision algorithm. Based on the constructed cross-database association identifier system, the cross-database data conversion middleware is deployed to realize real-time data format conversion, realize independent storage and real-time association calling of the three-level databases, and solve the problem of storage redundancy and flexible calling of multiple types of adaptable data. For example, the step of identifying multi-type adapted data in the original VDR data based on the field features involves determining whether the field features corresponding to each piece of original VDR data simultaneously possess at least two of the following: time-series fields, structured fields, and vector fields. If at least two of these three fields are simultaneously present, the current data is determined to be multi-type adapted data. In an optional embodiment, the parsed VDR data is first analyzed to determine whether it simultaneously possesses at least two of the three key fields: time-series fields, structured fields, and vector fields, thus classifying it as multi-type adapted data.

[0029] Quantifying the compatibility between each type of multi-type adaptive data and each of the respective databases includes: performing multi-dimensional feature analysis on each type of multi-type adaptive data to determine the data usage priority weight; determining the query type priority of each type of multi-type adaptive data based on a pre-set data query mode; and determining the compatibility degree by weighted summation of the multi-type adaptive data based on the usage priority weight and the query type priority. In this embodiment, the compatibility degree may require consideration of the importance of the multi-type data, field characteristics (e.g., char short text type, text long text, numeric type, time type, binary data stream, enumeration type, etc.), data update cycle, data aging duration (e.g., permanent storage, monthly aging, annual aging, etc.), data volume, and business usage methods of the data (e.g., main data usage, which may include semantic retrieval, fuzzy matching, precise query, association analysis, aggregation analysis, etc.), usage frequency, storage space, and other multi-dimensional analysis for a full evaluation to determine the compatible database type, which better suits the business usage scenario and meets functional and performance requirements.

[0030] For example, raw VDR data is acquired and temporal feature analysis is performed. For instance, by statistically analyzing the number and distribution density of timestamp fields, the continuity and interval patterns of the time series are analyzed to determine the expected frequency of time-series usage. Entity identifiers and relational attributes in the multi-type adaptation data are identified to determine the statistical probability of structured usage. Specifically, entity identifier fields in the data are identified, the complexity of relationships between fields is analyzed, and the probability of structured usage is statistically calculated. The proportion and complexity of natural language content in the multi-type adaptation data are detected to determine the value density of semantic analysis. A weighted sum is then performed based on the expected frequency of time-series usage, the probability of structured usage, and the value density of semantic analysis to determine the usage priority weight.

[0031] The query type priority of each of the multi-type adaptive data is determined based on a pre-defined data query pattern. The degree of adaptation is determined by weighted summation of the multi-type adaptive data based on the usage priority weight and the query type priority.

[0032] The database with the highest compatibility is selected as the target database. The multi-type compatible data is stored in the target database, and the cross-database association identifier is stored in the non-target database. A cross-database data conversion middleware is constructed, the cross-database conversion rules are stored in the middleware, and the middleware is communicatively associated with the third-level association database. This middleware receives call requests from non-target databases, retrieves corresponding data from the target database, and converts it to the target type data in real time according to the cross-database conversion rules.

[0033] The transformation middleware includes a call routing module, a transformation rule management module, a data transformation engine, a caching module, and a consistency verification module. The call routing module parses request parameters and extracts the target database type and query conditions when the application layer initiates a call request. It then queries the metadata of the non-main database to confirm the main database type and supported cross-database transformation rule IDs. After verifying cross-database transformation rule support, it generates a query statement for the main database, sends a request to the main database, and retrieves the raw data. The transformation rule management module stores all transformation rules in Redis, using the transformation rule ID as the key and the rule configuration as the value. The data transformation engine receives the raw data and transformation rule ID from the routing module, retrieves the corresponding configuration from the rule management module, and executes the transformation according to the rules. Multi-threaded parallel processing is used, with synchronous output after each batch processing to avoid memory overflow. The caching module generates cache keys and stores the transformation results in Redis. The consistency verification module performs three-level verification on the results after transformation: field integrity, format validity, and data distortion. If the verification passes, the results are returned to the application layer. The routing module accurately locates the main database and data through cross-database association identifiers; the conversion rules are designed for the characteristics of different databases, ensuring that the converted data can directly adapt to the storage and query requirements of the target database; the caching module reduces redundant conversions and improves response speed; and the verification module ensures that the conversion results are accurate and usable. The entire process allows the application layer to be unaware of which database the data actually resides in; it only needs to specify the target format as required, and the middleware automatically completes the automated calls for finding data, converting formats, and verifying accuracy.

[0034] This embodiment also provides a database establishment device based on ship VDR data. This device is used to implement the above embodiments and preferred embodiments, and details already described will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0035] This embodiment provides a database establishment device based on ship VDR data, such as... Figure 2 As shown, it includes: The acquisition module 201 is used to acquire raw VDR data from the VDR device and associated storage medium; The parsing module 202 is used to parse the original VDR data to obtain the field features of the original VDR data; Classification module 203 is used to classify the original VDR data based on the field features to obtain time series data, structured data and vector data; Database construction module 204 is used to construct a collaborative three-level relational database, which includes a time-series database, a structured database, and a vector database. The data storage module 205 is used to store the time-series data, the structured data, and the vector data into the corresponding databases, respectively.

[0036] Furthermore, the parsing module 202 is also used for: Based on the basic type of the original VDR data, the original VDR data is divided into video data, audio data, and text data. The video data and the audio data are stored on a hard disk, and the audio data is converted into audio vector data. The text data and the speech vector data are temporarily stored as a binary stream; Extract the frame features of the binary stream data, perform bitwise operations to extract the type identifier, and obtain the field features.

[0037] The inbound module 205 is also used for: Based on the aforementioned field characteristics, cross-database association identifiers and cross-database transformation rules are constructed for the inbound data; The time-series data, the structured data, and the vector data are stored in their respective target databases, and the cross-database association identifier is stored in a non-target database. A cross-database data conversion middleware is constructed, the cross-database conversion rules are stored in the middleware, and the middleware is communicatively associated with the third-level relational database. This middleware is used to receive call requests from non-target databases, retrieve corresponding data from the target database, and convert it into target type data in real time according to the cross-database conversion rules. Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 3 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system).

[0038] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0039] The memory 20 stores instructions executable by at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiments.

[0040] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device as shown by a landing page for an app. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, which can be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0041] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0042] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0043] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0044] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and all such modifications and variations fall within the scope defined by the appended claims.

Claims

1. A method for establishing a database based on ship VDR data, characterized in that, The method includes: Retrieve raw VDR data from the VDR device and associated storage media; The original VDR data is parsed to obtain the field features of the original VDR data; The original VDR data is classified based on the field features to obtain time-series data, structured data, and vector data, respectively. Construct a collaborative three-level relational database, which includes a time-series database, a structured database, and a vector database; The process involves storing the time-series data, structured data, and vector data into corresponding databases, including: constructing cross-database association identifiers and cross-database transformation rules for the input data based on the field features; identifying multi-type adapted data in the original VDR data based on the field features, wherein the multi-type adapted data is cross-database data that simultaneously possesses at least two of the field features of time-series fields, structured fields, and vector fields, and is cross-database data that can be accessed across databases; quantifying the compatibility degree between each of the multi-type adapted data and each of the databases, including: performing multi-dimensional feature analysis on each of the multi-type adapted data to determine the priority weight of data usage, including: calculating the number and distribution density of timestamps contained in the multi-type adapted data to determine the expected frequency of time-series usage; and identifying entity identifiers and relationship attributes in the multi-type adapted data to determine the statistical structured usage probability. The process involves: determining the proportion and complexity of natural language content in multi-type adaptation data to determine the value density of semantic analysis; determining the query type priority of each type of adaptation data based on a pre-defined data query pattern; weighting and summing the multi-type adaptation data based on the usage priority weight and the query type priority to determine the degree of adaptation; selecting the database with the highest degree of adaptation as the target database; storing the multi-type adaptation data in the target database and storing the cross-database association identifier in a non-target database; constructing a cross-database data conversion middleware, storing the cross-database conversion rules in the middleware, and communicating with the third-level association database to receive call requests from non-target databases, pull corresponding data from the target database, and convert it into target type data in real time according to the cross-database conversion rules.

2. The database establishment method based on ship VDR data as described in claim 1, characterized in that, The parsing of the original VDR data includes: Based on the basic type of the original VDR data, the original VDR data is divided into video data, audio data, and text data. The video data and the audio data are stored on a hard disk, and the audio data is converted into audio vector data. The text data and the speech vector data are temporarily stored as a binary stream; Extract the frame features of the binary stream data, perform bitwise operations to extract the type identifier, and obtain the field features.

3. The database establishment method based on ship VDR data as described in claim 1, characterized in that, The process of identifying multiple types of adaptation data in the original VDR data based on the field features includes: Each original VDR data is determined to possess at least two of the following field features: time-series fields, structured fields, and vector fields. When at least two of the time-series field, the structured field, and the vector field are present, the current data is determined to be the multi-type adapted data.

4. A database establishment device based on ship VDR data, characterized in that, The device includes: The acquisition module is used to acquire raw VDR data from the VDR device and associated storage media; The parsing module is used to parse the original VDR data to obtain the field features of the original VDR data; The classification module is used to classify the original VDR data based on the field features to obtain time-series data, structured data, and vector data; The database construction module is used to build a collaborative three-level relational database, which includes a time-series database, a structured database, and a vector database. The database entry module is used to store the time-series data, structured data, and vector data into corresponding databases, respectively. The module is also used to construct cross-database association identifiers and cross-database transformation rules for the entered data based on the field features; to identify multi-type adapted data in the original VDR data based on the field features, wherein the multi-type adapted data is cross-database data that simultaneously possesses at least two of the field features of time-series fields, structured fields, and vector fields, and is cross-database data that can be accessed across databases; quantifying the adaptability of each multi-type adapted data to each database includes: performing multi-dimensional feature analysis on each multi-type adapted data to determine the data usage priority weight, including: calculating the number and distribution density of timestamps contained in the multi-type adapted data to determine the expected frequency of time-series usage; identifying entity identifiers and relationship attributes in the multi-type adapted data to determine statistical... The process involves: structuring usage probability; detecting the proportion and complexity of natural language content in multi-type adapted data to determine the value density of semantic analysis; determining the query type priority of each of the multi-type adapted data based on a pre-defined data query pattern; weighting and summing the multi-type adapted data based on the usage priority weight and the query type priority to determine the degree of adaptation; selecting the database with the highest degree of adaptation as the target database; storing the multi-type adapted data in the target database and storing the cross-database association identifier in a non-target database; constructing a cross-database data conversion middleware, storing the cross-database conversion rules in the middleware, and communicating with the third-level association database to receive call requests from non-target databases, pull corresponding data from the target database, and convert it into target type data in real time according to the cross-database conversion rules.

5. A computer device, characterized in that, include: A memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, the processor executing the computer instructions to perform the method of any one of claims 1 to 3.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a computer to perform the method of any one of claims 1 to 3.

Images