Multi-database fusion access method, device and system applied to scientific investigation ship

Abstract

The invention relates to the technical field of databases, and provides a multi-database fusion access method, device and system applied to a scientific investigation ship. According to the method, characteristics of collected data of a scientific investigation ship are mapped into data tags, the data tags comprise frequency-variant FB-Type, slow-variant MB-Type and stable ST-Type, and different data path strategies are matched for the data tags; the collected data of different data labels can move to respective database engines along a preset strategy; the method comprises the following steps: importing collected data with an FB-Type data tag into a time sequence database; the collected data with the MB-Type data tag is imported into a document type database; and the collected data with the ST-Type data tag is imported into the relational database, so that the problem of poor practicability of heterogeneous data stored in a single database architecture is solved.

Description

Technical Field

[0001] The present invention relates to the field of database technology, and in particular to a multi-database fusion access method, device and system applied to a scientific research vessel. Background Art

[0002] In smart ships, industrial control, and multi-source perception systems, data types are often characterized by high-frequency acquisition, heterogeneous structures, and diverse transmission. For example, in the case of an oceanographic research vessel, the system must simultaneously process continuous time-series data from onboard sensors, structured instruction data generated by task scheduling, log data generated by equipment health monitoring, and cached status data for temporary responses. This data is highly heterogeneous, requiring varying access strategies and storage methods.

[0003] Current systems typically use a single database to manage all types of data, which presents the following problems: (1) mismatched structural constraints; (2) limited access efficiency; (3) difficulty coupling multi-dimensional indicators; and (4) redundant storage resource configuration. Therefore, existing data access methods are unable to support intelligent data flow scheduling in complex scenarios.

[0004] In view of this, overcoming the defects of the prior art is an urgent problem to be solved in this technical field. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a multi-database fusion access method, device and system applied to scientific research vessels.

[0006] The present invention adopts the following technical solutions: In a first aspect, the present invention provides a multi-database fusion access method applied to a scientific research vessel; the characteristics of the collected data of the scientific research vessel are mapped into data tags, wherein the data tags include: frequency-varying type FB-Type, slowly varying type MB-Type and stable type ST-Type, and different data path strategies are matched to each of the data tags, including: The collected data with different data tags will be sent to their respective database engines according to the preset strategy; Among them, the collected data with frequency-varying FB-Type data tags will be imported into the time series database to support high-frequency writing and window aggregation queries; the collected data with slowly varying MB-Type data tags will be imported into the document database to support flexible structures and version updates; the collected data with stable ST-Type data tags will be imported into the relational database to store table-building task information and indicator analysis results.

[0007] Further, the frequency-variant data includes data collected from one or more of an acoustic Doppler current profiler, a conductivity temperature depth device, and a weather station; Slowly changing data include one or more of the following: the attitude of the accompanying ship, the trajectory of the autonomous underwater vehicle, and the change of the tide level; Stable data includes one or more of the engine operating status, the start and end time of the station operation task, the station operation task coordinates, and the station operation task type.

[0008] Furthermore, by default, the frequency-varying data is not remotely synchronized between the research vessel and the roadbed server, and is only cached in the time series database of the local node. The method further includes: When the deviation between the currently collected frequency-varying data and the first-level callable data exceeds the deviation threshold, a remote synchronization request is triggered; Among them, the first-level callable data specifically refers to the equipment detection data performed before the scientific research vessel leaves the site, as well as the dynamic interval values ​​of the frequency-variable data pre-generated according to the scientific research tasks to be completed.

[0009] Furthermore, the method further comprises: Based on one or more of the task priority, sensor association strength and load limit factor, the data label of the collected data is dynamically adjusted to switch from the frequency-varying FB-Type to the slowly varying MB-Type.

[0010] Furthermore, the method specifically includes: For task priorities lower than a preset level and involving collected data from strongly associated sensors, when the difference between the network bandwidth load and the load limit factor is less than a first preset value, the data label of the collected data is switched from a frequency-varying FB-Type to a slowly varying MB-Type; The collected data involving the strongly associated sensor is collected data whose dependence on other sensors is greater than a second preset value.

[0011] Furthermore, the slowly varying MB-Type data is subjected to stable structure projection, and the method further includes: A trajectory envelope is formed for the accompanying trajectory, and the envelope is used as an input parameter of the stable ST-Type data tag.

[0012] Furthermore, the method further comprises: Perform preview template comparison on stable data and encode the deviation into the data packet.

[0013] In a second aspect, the present invention further provides a multi-database fusion access system for a scientific research vessel, comprising: The ship collects continuous data on conductivity, temperature, and depth, identifies it as frequency-varying data, and only temporarily stores it in a local time series database; When the ship returns to the station, the system writes the stable data into the relational database for the dispatch center to call; wherein, the stable data includes the station number and operation plan; If energy consumption anomalies occur along the ship route, the deviation between the structural operation and environmental data will determine whether to dispatch the corresponding frequency-variable data and feed it back to the roadbed server; If the results of multiple consecutive tasks converge, some frequency-varying data will be written into the document-based database as the analysis trajectory envelope to participate in subsequent analysis and recalibration.

[0014] Furthermore, the frequency-varying data is not synchronized remotely between the research vessel and the roadbed server by default, and is only cached in the time series database of the local node, and further includes: When the deviation between the currently collected frequency-varying data and the first-level callable data exceeds the deviation threshold, a remote synchronization request is triggered; Among them, the first-level callable data specifically refers to the equipment detection data performed before the scientific research vessel leaves the site, as well as the dynamic interval values ​​of the frequency-variable data pre-generated according to the scientific research tasks to be completed.

[0015] In a third aspect, the present invention further provides a multi-database fusion access device for a scientific research vessel, for implementing the multi-database fusion access method for a scientific research vessel described in the first aspect, the device comprising: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the processor to execute the multi-database fusion access method applied to a scientific research vessel as described in the first aspect.

[0016] In a fourth aspect, the present invention further provides a non-volatile computer storage medium, which stores computer-executable instructions, and the computer-executable instructions are executed by one or more processors to complete the multi-database fusion access method applied to scientific research vessels described in the first aspect.

[0017] The present invention provides a multi-database fusion access mechanism that maps the characteristics of data collected on a research vessel into multiple data tags. Different data path strategies are then matched to each data tag according to the data tag. The collected data with different data tags will follow the preset strategy and be sent to their respective database engines. That is, they will be stored in corresponding databases according to the data type. This avoids using a single database to store all types of data, which can lead to problems such as mismatched structural constraints, limited access efficiency, difficulty in coupling multi-dimensional indicators, and redundant storage resource configuration. By identifying data behavior types through data tags, building multi-path access strategies and a dynamic tag mapping mechanism, differentiated scheduling, optimized storage, and intelligent routing of heterogeneous data during the task execution cycle are achieved, enabling the system to perform multi-strategy heterogeneous storage of collected data with different characteristics according to specific logic during operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0019] Figure 1 This is a flow chart of a multi-database fusion access method applied to a scientific research vessel provided by an embodiment of the present invention; Figure 2 This is a flowchart of a specific example of writing collected data according to data tags provided by an embodiment of the present invention; Figure 3 This is a schematic diagram of a specific example of path mapping provided by an embodiment of the present invention; Figure 4 This is a schematic diagram of a specific example of a generation path provided by an embodiment of the present invention; Figure 5 is a schematic diagram of a specific example of another generation path provided by an embodiment of the present invention; Figure 6 This is a schematic diagram of a specific example of comparing first-level callable data with real-time collected data provided by an embodiment of the present invention; Figure 7 This is a flow chart of another multi-database fusion access method applied to a scientific research vessel provided by an embodiment of the present invention; Figure 8 This is a flow chart of another multi-database fusion access method applied to a research vessel provided by an embodiment of the present invention; Figure 9 This is a schematic diagram of the architecture of a multi-database fusion access device applied to a scientific research vessel, provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0021] In the description of the present invention, the terms "inside", "outside", "longitudinal", "lateral", "upper", "lower", "top", "bottom", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings. They are only for the convenience of describing the present invention and do not require that the present invention must be constructed and operated in a specific orientation. Therefore, they should not be understood as limitations on the present invention.

[0022] The terms "first," "second," etc., used in this application are for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature identified with "first," "second," etc. may explicitly or implicitly include one or more of such features. Throughout this application, unless otherwise specified, "plurality" means two or more.

[0023] In this application, unless otherwise specified or limited, the term "connection" should be understood broadly. For example, "connection" can mean fixed connection, detachable connection, or integration; it can mean direct connection or indirect connection through an intermediate medium. In addition, the term "coupling" can refer to the manner in which electrical connection is achieved for signal transmission.

[0024] In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0025] Embodiment 1: In today's intelligent systems, data comes from a wide range of sources, has diverse structures, and changes rapidly. This is especially true in areas like maritime, industrial control, and urban IoT. Data includes not only high-frequency sensor information but also periodically recorded device status, task-oriented analysis results, and instantaneous data caches. The coexistence of these heterogeneous data within the same system poses significant challenges to traditional single-database architectures.

[0026] Conventional data systems usually store all data in a unified database engine. However, this will cause a series of problems such as data type mismatch, low query efficiency, resource waste, and high development costs.

[0027] Therefore, there is an urgent need for a general architecture solution that can dynamically integrate multiple database technologies during system operation, so that the system can autonomously optimize data paths and storage selection based on operating status, data characteristics and environmental requirements, thereby achieving more intelligent and efficient management capabilities.

[0028] Embodiment 1 of the present invention provides a multi-database fusion access method applied to a research vessel, such as Figure 1 As shown, the characteristics of the collected data of the research vessel are mapped into data labels, wherein the data labels include: frequency-varying type FB-Type, slowly varying type MB-Type, and stable type ST-Type, and different data path strategies are matched to each of the data labels. The method includes: The collected data with different data tags will go to their respective database engines along the preset strategy, that is, each data tag corresponds to its own path.

[0029] This embodiment of the present invention maps the data collected by a research vessel into different data labels based on their characteristics. Specifically, these labels include three types: frequency-varying (FB-Type), slowly varying (MB-Type), and stable (ST-Type). The data collected by a research vessel refers to data collected by the equipment on the vessel.

[0030] In one embodiment, the data type needs to be identified first to determine the data label. In this embodiment, the collected data is classified into semantic behaviors in the following manner: Data with the frequency-variant FB-Type data tag is called frequency-variant data. This includes data collected by one or more of the following: acoustic Doppler current profilers, conductivity-temperature-depth instruments, and weather stations. Frequency-variant data typically exhibits strong short-term fluctuations and is only used for evaluation during specific time periods. For example, high-frequency data collected by acoustic Doppler current profilers (ADCPs), conductivity-temperature-depth instruments (CTDs), and weather stations can be used.

[0031] In one embodiment, the frequency-varying data includes: scientific research equipment data; target data obtained by deploying specific equipment in deployment operations, wherein the equipment in deployment operations includes: active source ocean bottom seismometer (OBS), passive source OBS, sediment trap, underwater video detection, lander, in-situ experimental station, long baseline, etc.; ocean physical parameters measured by special equipment in physical oceanography operations, the data directly reflects the state of the ocean environment, wherein the physical oceanography equipment includes: Sea-Bird Electronics 21 (SBE21), ADCP, sea-air flux observation equipment and weather station.

[0032] Collected data with slowly varying MB-Type data tags is called slowly varying data. This data includes one or more of the following: escort vessel attitude, autonomous underwater vehicle (AUV) trajectory, and tidal level changes. Slowly varying data has a more pronounced driving change trajectory and is used for accompanying work vessels. Slowly varying data changes more slowly but still requires process tracking. This includes escort vessel attitude data, autonomous underwater vehicle (AUV) trajectory data, and tidal level change data.

[0033] In one embodiment, the slowly varying data includes: escort operation data, wherein the escort operation includes data collected by AUV, hovering (HOV), underwater glider, etc.

[0034] The collected data with stable ST-Type data tags is called stable data. Stable data includes one or more of the engine operating status, the start and end time of the station operation task, the station operation task coordinates, and the station operation task type.

[0035] Stable data has a fixed structure and stable collection, and is the core benchmark data of the mission, such as engine operating status data and station operation mission parameters (such as start and end time, coordinates and operation type, etc.).

[0036] In one embodiment, stable data include: engine operating status data, station operation data, scientific research vessel deployment and recovery / non-recovery operations and mission status data, etc., wherein the engine operating status data includes the scientific research vessel's; first engine speed, second engine speed, third engine speed, etc.; station operation data includes data collected by water extraction, sediment extraction, rock extraction, fine sampling, and detection operations, etc.; scientific research vessel deployment and recovery and / or non-recovery operations include: active source OBS, passive source OBS, sediment collector, underwater video observation, lander, in-situ experimental station and long baseline operation collection data, etc.; mission status data includes: route segment, station number, planned start and end time, station longitude and latitude, operation steps, etc., wherein the mission status data is used to: select a specific deployment operation according to the scientific research task arrangement, add route segment, station number, planned start and end time, station longitude and latitude, operation steps and other information, confirm this deployment operation task and enter the pending release state.

[0037] In an embodiment of the present invention, each data tag is matched with a different data path strategy. The collected data of different data tags will be sent to their respective database engines along the preset strategy. That is, each data tag corresponds to its own path. The preset strategy is determined by those skilled in the art based on the specific usage scenario. The embodiment of the present invention provides a specific example of the preset strategy as follows: Among them, Figure 2 As shown in the figure, the paths corresponding to the three data tags are: importing the collected data with frequency-varying FB-Type data tags into the time series database to support high-frequency writing and window aggregation query; importing the collected data with slowly varying MB-Type data tags into the document database to support flexible structure and version update; importing the collected data with stable ST-Type data tags into the relational database to store table-building task information and indicator analysis results.

[0038] Among them, the specific implementation methods of the time series database, document database and relational database are determined by those skilled in the art according to the specific usage scenarios; in one embodiment, the time series database can use QuestDB, the document database can use MongoDB, and the relational database can use MySQL database.

[0039] like Figure 3A schematic diagram of a data behavior projection mapping framework for an embodiment of the present invention is shown. In one embodiment, collected data is obtained from various data flow nodes of a research vessel, where data is collected in real time. A state identifier maps the collected data to different data labels based on their characteristics: frequency-varying, slowly varying, and stable. These different types of data labels form a multi-level indexed behavior description. By mapping the running data state into a set of data labels, data decision intent is driven. A mapping routing controller then matches different data path strategies to each data label. Collected data with different data labels will follow the preset strategies and be directed to their respective database engines (i.e., the specific hosting entities of the abstracted target storage nodes). Because the database engines storing different types of collected data are determined by data labels, the system can provide data representation and reflection capabilities for generic tasks without directly controlling how database instructions are executed.

[0040] In one embodiment, the system can maintain a dynamic path map structure, where different data tags are directed to different database engines based on pre-defined conditions (e.g., frequency, structure, and business urgency). In an optional embodiment, Redis caching can also be used to maintain temporary state across stages; for example, using Redis cache to store device online flags and / or task fast retrieval fields.

[0041] The present invention provides a multi-database fusion access mechanism that maps the characteristics of data collected on a research vessel into multiple data tags. Different data path strategies are then matched to each data tag according to the data tag. The collected data with different data tags will follow the preset strategy and be sent to their respective database engines. That is, they will be stored in corresponding databases according to the data type. This avoids using a single database to store all types of data, which can lead to problems such as mismatched structural constraints, limited access efficiency, difficulty in coupling multi-dimensional indicators, and redundant storage resource configuration. By identifying data behavior types through data tags, building multi-path access strategies and a dynamic tag mapping mechanism, differentiated scheduling, optimized storage, and intelligent routing of heterogeneous data during the task execution cycle are achieved, enabling the system to perform multi-strategy heterogeneous storage of collected data with different characteristics according to specific logic during operation.

[0042] In one embodiment, in an actual application scenario, before a research vessel goes offshore to perform a mission, it uses the equipment on each research vessel to pre-rehearse the mission execution, that is, to collect various data, and then simulate the state of the research vessel when performing the mission after going offshore. Figure 4As shown, after collecting the data, each evaluation node classifies and evaluates the collected data according to the data access triggering situation of the collected data of a certain device at each moment. The system will perform two key evaluations: one is to determine whether it is a high-frequency access, and the other is to evaluate whether the structure of the collected data is stable. According to the high-frequency access evaluation result, if it is "yes", candidate path A is selected; if it is "no", candidate path B is selected. At the same time, according to the structural stability evaluation result, if the structure is stable, candidate path C is selected. Finally, the information of these candidate paths will be transmitted to the fusion policy device for comprehensive judgment, so as to complete the path selection. The selected path will be used for heterogeneous storage calls, thereby achieving efficient data access and processing. Among them, the specific method for the fusion policy device to select the path is selected by technical personnel in this field according to the specific usage scenario. Here, selecting the data collected by a device at each moment as candidate path A indicates that the corresponding collected data is frequency-varying data, and the corresponding collected data is stored in a time-series database according to the storage method for frequency-varying data. Selecting the data collected by a device at each moment as candidate path B indicates that the corresponding collected data is slowly varying data, and the corresponding collected data is stored in a document-based database according to the storage method for slowly varying data. Selecting the data collected by a device at each moment as candidate path C indicates that the corresponding collected data is stable data, and the corresponding collected data is stored in a relational database according to the storage method for stable data. After different collected data are stored in heterogeneous databases, the corresponding call method can also vary depending on the characteristics of the corresponding database, which is not limited here.

[0043] In one embodiment, if Figure 5The following illustrates the system architecture and processing flow of an embodiment of the present invention. First, collected data flows into the data access port. The data behavior identifier then analyzes and identifies the collected data, generating corresponding analysis data. The analyzed data then enters the data label mapping module, where it is labeled and classified using different data labels (i.e., frequency-varying type FB-Type, slowly varying type MB-Type, and stable type ST-Type). Subsequently, the corresponding collected data is fed into the path generator, which determines the transmission path for each collected data item based on the corresponding data label. Finally, the fusion scheduling gateway integrates and schedules collected data with the same data label within a specific time period. It then distributes the collected data to different database adapters (i.e., DB-A, DB-B, and DB-C) according to the transmission paths associated with each data label. The data is then stored in the corresponding databases (i.e., time series database, document database, and relational database), enabling final storage or further processing of the collected data. The present invention uses a data semantics-driven archiving and calling method to perform fuzzy parameter evaluation on dimensions such as timeliness, access frequency, and structural complexity of a certain type of data to obtain data labels; at the bottom layer of the system, different database capability units are connected through a certain dynamic mapping logic; therefore, it can achieve no restrictions on database types and is highly compatible with most modern databases.

[0044] In one embodiment, data path selection can be adjusted based on real-time context. The specific strategy is as follows: By default, the frequency-varying data is not remotely synchronized between the research vessel and the roadbed server, and is only cached in the time series database of the local node. The method further includes: When the deviation between the currently collected frequency-variable data and the first-level callable data is greater than the deviation threshold, a remote synchronization request is triggered; wherein, the first-level callable data is specifically the equipment detection data performed before the scientific research vessel leaves the site, and the dynamic interval value of the frequency-variable data pre-generated according to the scientific research task to be completed.

[0045] The term "before the research vessel leaves the station" refers to the period before the research vessel sets out to sea. Equipment detection data refers to data collected during the research vessel's pre-sea rehearsal. The specific values ​​of the deviation threshold and dynamic range are determined by those skilled in the art based on specific usage scenarios and are not limited here. Currently collected frequency-variable data refers to data with the frequency-variable FB-Type data tag.

[0046] Level 1 callable data includes collected data and corresponding preset dynamic range values ​​obtained from pre-sea rehearsals. In one embodiment, during a pre-sea rehearsal, the research vessel records the values ​​of the data collected by each device at each moment. The pre-set thresholds for level 1 callable data (i.e., pre-sea) are the corresponding values ​​of each device at each moment, rather than specific fixed values.

[0047] Frequency-varying data in the embodiment of the present invention is not remotely synchronized by default and is only cached in the local node.

[0048] After the research vessel leaves the coast and goes out to sea, it begins to carry out its mission. During the actual mission execution, the corresponding collected data may deviate from the data collected during the rehearsal. Figure 6 As shown, when there is a serious deviation between the real-time collected data and the first-level callable data, that is, when the deviation between the real-time collected data and the first-level callable data is greater than the deviation threshold, a remote synchronization request is triggered, and the data label of the corresponding real-time collected data is switched to store these real-time collected data in a document-type database (that is, when the data label after switching is a slowly changing type) or a relational database (that is, when the data label after switching is a stable type), and remote synchronization is performed on the serious deviation, thereby significantly reducing bandwidth and cloud storage costs.

[0049] For example, the first-level callable data is a dynamic interval value, which includes a maximum range value and a minimum range value, such as Figure 6 The two dotted lines in the figure show respectively; the frequency-varying data currently collected is as follows Figure 6 The solid line shows this. Before time t1, the deviation between the currently collected frequency-variable data and the dynamic interval value is less than or equal to the deviation threshold. Starting from time t1, the currently collected frequency-variable data deviates significantly from the dynamic interval value, with the deviation between the collected data and the dynamic interval value exceeding the deviation threshold. Until time t2, the corresponding deviation falls back to less than or equal to the deviation threshold and remains within the range of less than or equal to the deviation threshold.

[0050] In one embodiment, during a pre-sea rehearsal, all collected data is labeled as frequency-variable (FB-Type) data to record all simulated values. This data is then stored in a time-series database using the same storage method as frequency-variable data. This allows for subsequent comparison with pre-sea simulation values, storing and transmitting only those values ​​that differ significantly from the simulated values. This reduces the amount of data required for transmission and storage, conserves bandwidth, and improves system practicality. By building and learning from models of historical scientific research data, the collected data from the research vessel after leaving the sea is differentiated and transmitted according to three types: frequency-variable (FB-Type), slowly varying (MB-Type), and stable (ST-Type).

[0051] The land-based server is a non-offshore system platform used to receive some of the data collected by the research vessel before and after it leaves the shore (for example, data that requires remote synchronization). The land-based server is located onshore and acts as a shore-based platform. The land-based server can also store first-level callable data; data that requires remote synchronization can be stable data.

[0052] The embodiment of the present invention switches data labels to flexibly and autonomously optimize data paths and adjust storage selection based on system operating status and environmental requirements (e.g., bandwidth limitations). The method further includes: Based on one or more of the task priority, sensor association strength and load limit factor, the data label of the collected data is dynamically adjusted to switch from the frequency-varying FB-Type to the slowly varying MB-Type.

[0053] In one embodiment, the method specifically includes: Figure 7 As shown, for the collected data whose task priority is lower than the preset level and involves strongly correlated sensors, when the difference between the load of the network bandwidth and the load limit factor is less than a first preset value, the data label of the collected data is switched from the frequency-varying type FB-Type to the slowly varying type MB-Type; wherein the collected data involving the strongly correlated sensors is collected data whose degree of dependence on other sensors is greater than a second preset value.

[0054] Among them, the preset level, the first preset value, the load limiting factor and the second preset value are determined by those skilled in the art according to the specific usage scenario and are not limited here. In the embodiment of the present invention, the importance of each task requirement can be represented by the task priority, and the preset level is a pre-specified threshold value of the task priority. In actual application scenarios, there are often many task requirements that need to be performed on the basis of data collected by other sensors to complete other tasks, wherein other sensors refer to: sensors other than the target device that performs the current task; the degree of dependence on other sensors can be represented by the dependence level, and the second preset value is a pre-specified threshold value of the dependence level. The load limiting factor is a threshold value of the network bandwidth.

[0055] A task can correspond to multiple collected data. For collected data with low task priority, since it does not require frequent writes or immediate participation in window aggregation queries, it does not need to be immediately stored in the time series database. When network bandwidth is nearly full, the system prioritizes the removal of collected data that is both unimportant (i.e., low task priority) and dependent on other sensors (i.e., involving strongly correlated sensors). Because the loss of this collected data does not result in a break in the overall data chain and can be indirectly inferred or compensated for using other sensors, pausing its upload frees up bandwidth, ensuring that mission-critical data (such as alarms and control instructions) is delivered in real time, avoiding system crashes or delays. Therefore, embodiments of the present invention reduce this type of collected data with low task priority and involving strongly correlated sensors from frequency-varying data to slowly-varying data by switching the data label of the collected data from frequency-varying data to slowly-varying data. This data is then processed using the data transmission path of the slowly-varying data to minimize the system's real-time load.

[0056] In one embodiment, if Figure 7 As shown, the slowly varying MB-Type data is subjected to stable structure projection, and the method further includes: forming a trajectory envelope for the accompanying trajectory, and using the envelope as an input parameter of the stable ST-Type data label.

[0057] The accompanying trajectory is typically an ideal or measured relative motion trajectory. Stable structure projection can be achieved by forming a trajectory envelope for the accompanying trajectory (i.e., slowly varying data) and using this trajectory envelope as one of the input parameters for the stable data. Stable structure projection refers to mathematically projecting the accompanying trajectory into a stable coordinate system to obtain a stable morphological curve that varies with time but eliminates high-frequency jitter. A three-dimensional (or two-dimensional) safety envelope is defined around this stable morphological curve, which can be understood as adding a shell to the trajectory. The thickness of this shell is the maximum possible deviation of the accompanying trajectory within a safe range.

[0058] Among them, the envelope is the boundary line of the escort trajectory, indicating the maximum possible deviation range of the escort trajectory. For example, the envelope describes the safe range of the escort trajectory of the research vessel; when the actual escort trajectory exceeds the envelope, it can trigger the system control to make adjustments by prompting potential risks. The envelope is regarded as a stable data, and the corresponding data label is the stable ST-Type data label. In this process, the envelope (i.e., stable data) is input as an input parameter to the controller used for real-time control decision-making. At this time, the escort trajectory serves as a safety constraint, providing boundary conditions for the controller to make real-time control decisions, thereby ensuring that the escort trajectory adjusted by the controller remains within the safe range.

[0059] The above mechanism forms an evolutionary dimensionality reduction path between data types, that is, from frequent change to slow change, and from slow change to stability, thereby supporting the automatic improvement of its structural stability and utilization efficiency as the value of data gradually becomes clearer.

[0060] In one embodiment, in order to improve remote reporting efficiency, the method further includes: performing a preview template comparison on the stable data and encoding the deviation into the data packet.

[0061] Since the characteristic of frequency-varying data is that the data will change frequently according to the environment and business content, the embodiment of the present invention preferentially stores this type of data only in the ship's own local time series database, and exists as secondary callable combined data to serve the primary callable data.

[0062] Level 1 callable data is generated from the aforementioned stable data. This data is characterized by typically collecting relevant data from a broader perspective, such as ocean depth, total weight, and longitude and latitude. It may also include information such as the length of the released cable and engine operating status during the acquisition process. When a research vessel is offshore, this data can be used to calculate a rehearsal template using large-scale model data. The corresponding rehearsal template data (i.e., the level 1 callable data) is then stored on the research vessel's server. The rehearsal template data is then compared and analyzed in real time with the actual, current, real-time data being collected. The analysis results (i.e., the significant deviations resulting from a match or mismatch) serve as the optimal parameters and are appended to the data packet to be sent via satellite to the land platform. The specific methods for calculating the rehearsal template from the large-scale model data and for real-time matching and analysis of the rehearsal template data with the level 1 callable data are selected by those skilled in the art based on the specific application scenario. Because of the rehearsal template, the data packet only needs to carry the level 2 callable data generated in real time that deviates from the template; any consistent or minimally deviating data is not transmitted, significantly reducing data transmission volume.

[0063] During normal navigation, the frequency-varying data collected at this time does not need to occupy the satellite data bandwidth to be sent to the land platform; instead, it is stored in the server as secondary callable data. Its greater value and significance lies in that after the secondary callable data is analyzed with the primary callable data, when it is determined that the deviation exceeds the deviation threshold, the research vessel's own server will retrieve one or more secondary callable data related to the currently analyzed primary callable data, form a data packet and send it to the land platform.

[0064] The present invention uses labels as the central abstraction, taking into account both data structure characteristics and task process logic, to construct a cross-engine and cross-type data operation map, providing infrastructure support for intelligent task management under heterogeneous databases. Data labels are determined by dynamically sensing data attributes to achieve label-level routing and differentiated processing. Through multi-database collaborative scheduling, the task scheduling flow no longer relies on a single structural index, improving retrieval efficiency. Frequency-varying data is stored locally by default, and only key differences are reported remotely, significantly optimizing bandwidth usage. For complex task processes (such as escort scientific expeditions and multi-stage equipment deployment), data-driven and link backtracking capabilities are improved.

[0065] On this basis, an embodiment of the present invention further provides a multi-database fusion access system applied to a scientific research vessel, comprising: The ship collects continuous data on conductivity, temperature and depth, identifies it as frequency-varying data, and only temporarily stores it in a local time-series database. When the ship returns to the station, the system writes the stable data into a relational database for the dispatching center to call. The stable data includes the station number and operation plan. If energy consumption anomalies occur along the ship, the deviation between the structural operation and the environmental data will determine whether to dispatch the corresponding frequency-varying data and feed it back to the roadbed server. If the results of multiple consecutive tasks converge, some of the frequency-varying data will be written into the document database as the analysis trajectory envelope to participate in subsequent analysis and recalibration.

[0066] An embodiment of the present invention provides a data flow rule for data collected at sea and stored on shore in an application scenario of an ocean research vessel: continuous data of conductivity, temperature and depth are collected on board, and the system classifies it as frequency-varying data; this type of data is only temporarily cached in the local time series database on the research vessel and is not immediately transmitted back to avoid occupying satellite bandwidth. In one embodiment, the scheduling middle station can be a management system that needs to use research vessels to collect data. The specific system functions and properties of the scheduling middle station are determined by those skilled in the art based on the specific usage scenario and are not limited here. The scheduling middle station integrates a relational database to realize the writing and calling of stable data, supporting real-time decision-making and resource scheduling.

[0067] When the ship arrives at the pre-planned station again (i.e., the ship returns to the station location), the system writes the stable data, i.e., the station number that does not change over time and the operation plan for this time, into the MySQL relational database; the scheduling center can then directly query these static metadata through the MySQL relational database without having to parse the massive amount of raw conductivity, temperature and depth data. If an energy consumption anomaly occurs during the voyage (i.e., an energy consumption anomaly occurs along the ship's route), such as a sudden increase in propulsion power, the system will compare the structural operation data (e.g., planned speed, load curve) with the environmental data (e.g., measured flow rate, wind and waves) in real time and calculate the deviation between the two; if the deviation exceeds the threshold, the corresponding frequency-variable data field (i.e., the key abnormal fragment) will be immediately transmitted back to the shore platform via a satellite link for remote diagnosis; the threshold is determined by technicians in this field based on the specific usage scenario. If the deviation analysis results of multiple consecutive missions tend to be consistent, it means that certain high-frequency variables (such as turbulence dissipation rate) have become statistically significant; the system then samples and compresses these frequency-varying data into trajectory envelopes (i.e., statistical boundary curves) and writes them to the document database, which can be used for subsequent analysis and recalibration, such as model retraining, sensor drift recalibration, and / or route energy consumption baseline updates.

[0068] In the embodiment of the present invention, only the original conductivity, temperature and depth data are temporarily stored on board. Static metadata is written each time the ship returns to the station. Key segments are only transmitted back in the event of an anomaly. After multiple missions have been verified to be effective, the statistical envelope is deposited into a document database for long-term optimization.

[0069] In addition, in one embodiment, the frequency-varying data is not remotely synchronized between the research vessel and the roadbed server by default, and is only cached in the time series database of the local node, and further includes: When the deviation between the currently collected frequency-variable data and the first-level callable data exceeds the deviation threshold, a remote synchronization request is triggered. Specifically, the first-level callable data includes equipment test data collected before the research vessel left the site, as well as dynamic interval values ​​of pre-generated frequency-variable data according to the research mission to be completed.

[0070] The multi-database fusion access mechanism of the embodiment of the present invention is an abstract data fusion operation logic framework, so it does not rely on a specific scheduling control structure, nor does it limit the precise flow details of the data. It can adapt to multiple types of task processing scenarios and has evolutionary data path management capabilities.

[0071] like Figure 8 As shown, the embodiment of the present invention pre-configures the operation strategy for different types of collected data, and performs a preview after the configuration. According to the behavioral feedback of various collected data, the data characteristics of each type of collected data are determined, and the corresponding data labels are mapped thereby. The paths of various types of collected data to the database engine are distinguished by data labels. And subsequently, the data path optimization and storage selection fine-tuning are performed flexibly and autonomously according to the system operation status and environmental requirements, so as to switch the collected data from one data label to another data label and switch the database type for storing real-time collected data; in one embodiment, when switching, the interface of the old database before the switch can be transformed, and the new database after the switch can be adapted according to the characteristics of the collected data.

[0072] The embodiments of the present invention improve the adaptability of the system in a heterogeneous data environment, avoiding structural bottlenecks caused by database mismatch when a single database is used to store various data; by corresponding data labels to different databases, fuzzy control and intelligent generation of data processing paths are achieved without relying on manual configuration or static templates; it can be adapted to a variety of software and hardware platforms and has versatility in maritime scheduling, industrial equipment management, energy monitoring and other systems; it is designed with a certain degree of module drift tolerance and can support subsequent engine replacement and strategy updates.

[0073] like Figure 9 FIG. 1 is a schematic diagram of the architecture of a multi-database fusion access device applied to a research vessel according to an embodiment of the present invention. The multi-database fusion access device applied to a research vessel according to this embodiment includes one or more processors 21 and a memory 22. Figure 9 A processor 21 is taken as an example.

[0074] The processor 21 and the memory 22 may be connected via a bus or other means. Figure 9 The bus connection is taken as an example.

[0075] The memory 22, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs and non-volatile computer executable programs, such as the multi-database fusion access method applied to a scientific research vessel in Example 1. The processor 21 executes the multi-database fusion access method applied to a scientific research vessel by running the non-volatile software programs and instructions stored in the memory 22.

[0076] The memory 22 may include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state memory device. In some embodiments, the memory 22 may optionally include a memory remotely located relative to the processor 21, and such remote memory may be connected to the processor 21 via a network. Examples of such networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0077] The program instructions / modules are stored in the memory 22, and when executed by the one or more processors 21, the multi-database fusion access method applied to the scientific research vessel in the above-mentioned embodiment 1 is executed, for example, each step of the multi-database fusion access method applied to the scientific research vessel in the above-described embodiment 1 is executed.

[0078] It is worth noting that the information interaction, execution process, etc. between the modules and units within the above-mentioned devices and systems are based on the same concept as the processing method embodiment of the present invention. The specific content can be found in the description of the method embodiment of the present invention and will not be repeated here.

[0079] Those skilled in the art will understand that all or part of the steps in the various methods of the embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, which may include: read-only memory (ROM), random access memory (RAM), a disk or an optical disk, etc.

[0080] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A multi-database fusion access method applied to a scientific research vessel, characterized in that: The characteristics of the collected data of the research vessel are mapped to data labels, wherein the data labels include: frequency-varying type, slowly varying type and stable type, and different data path strategies are matched to each data label, including: The collected data with different data tags will be sent to their respective database engines according to the preset strategy; Among them, the collected data with frequency-varying data tags is imported into the time series database as frequency-varying data to support high-frequency writing and window aggregation queries; the collected data with slow-varying data tags is imported into the document database as slow-varying data to support flexible structures and version updates; the collected data with stable data tags is imported into the relational database as stable data to store table-building task information and indicator analysis results; Among them, frequency-varying data include collected data generated by one or more of the acoustic Doppler current profiler, conductivity temperature depth equipment and meteorological station; slowly varying data include one or more of the escort ship attitude, autonomous underwater vehicle operation trajectory and tide level changes; stable data include one or more of the engine operating status, station operation task start and end time, station operation task coordinates, and station operation task operation type.

2. The multi-database fusion access method applied to a scientific research vessel according to claim 1, characterized in that: By default, the frequency-varying data is not remotely synchronized between the research vessel and the roadbed server, and is only cached in the time series database of the local node. The method further includes: When the deviation between the currently collected frequency-varying data and the first-level callable data exceeds the deviation threshold, a remote synchronization request is triggered; Among them, the first-level callable data specifically refers to the equipment detection data performed before the scientific research vessel leaves the site, as well as the dynamic interval values ​​of the frequency-variable data pre-generated according to the scientific research tasks to be completed.

3. The multi-database fusion access method applied to a scientific research vessel according to claim 1, characterized in that: The method further comprises: In combination with one or more of the task priority, sensor association strength and load limiting factor, the data label of the collected data is dynamically adjusted from a frequent changing type to a slow changing type.

4. The multi-database fusion access method applied to a scientific research vessel according to claim 3, characterized in that: The method specifically includes: For task priorities lower than a preset level and involving collected data from strongly associated sensors, when the difference between the network bandwidth load and the load limit factor is less than a first preset value, the data label of the collected data is switched from a frequency-varying type to a slowly varying type; The collected data involving the strongly associated sensor is collected data whose dependence on other sensors is greater than a second preset value.

5. The multi-database fusion access method applied to a scientific research vessel according to claim 1, characterized in that: Performing stable structural projection on the slowly varying type data, the method further includes: A trajectory envelope is formed for the accompanying trajectory, and the envelope is used as an input parameter of the stable data tag.

6. The multi-database fusion access method applied to a scientific research vessel according to claim 1, characterized in that: The method further comprises: Perform preview template comparison on stable data and encode the deviation into the data packet.

7. A multi-database fusion access system applied to a scientific research vessel, characterized in that: include: The characteristics of the data collected by the research vessel are mapped to data tags, including frequency-varying, slowly varying, and stable data tags. Different data path strategies are matched to each data tag. The collected data with different data tags will be sent to their respective database engines according to the preset strategies. The ship collects continuous data on conductivity, temperature, and depth, identifies it as frequency-varying data, and only temporarily stores it in a local time series database; When the ship returns to the station, the system writes the stable data into the relational database for the dispatch center to call; wherein, the stable data includes the station number and operation plan; If energy consumption anomalies occur along the ship route, the deviation between the structural operation and environmental data will determine whether to dispatch the corresponding frequency-variable data and feed it back to the roadbed server; If the results of multiple consecutive tasks converge, some frequency-varying data will be written into the document-based database as the analysis trajectory envelope to participate in subsequent analysis and recalibration.

8. The multi-database fusion access system applied to a scientific research vessel according to claim 7, characterized in that: By default, the frequency-varying data is not remotely synchronized between the research vessel and the roadbed server, but is cached only in the time series database of the local node. It also includes: When the deviation between the currently collected frequency-varying data and the first-level callable data exceeds the deviation threshold, a remote synchronization request is triggered; Among them, the first-level callable data specifically refers to the equipment detection data performed before the scientific research vessel leaves the site, as well as the dynamic interval values ​​of the frequency-variable data pre-generated according to the scientific research tasks to be completed.

9. A multi-database fusion access device applied to a scientific research vessel, characterized in that: include: at least one processor; And, a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the processor to execute the multi-database fusion access method applied to a scientific research vessel as described in any one of claims 1-6.

Images