Data interpretation method and system based on rocket characteristic space database

Abstract

The application discloses a data interpretation method and system based on a rocket characteristic space database, relates to the technical field of spaceflight measurement and control and data analysis, and utilizes inherent unified coding and a global time reference of the characteristic space database to support cross-system and cross-parameter associated analysis and time sequence logic checking, can find implicit faults or system mismatching problems that cannot be identified by single parameter interpretation, and greatly improves the depth and accuracy of interpretation.

Description

Technical Field

[0001] This invention relates to the field of aerospace telemetry, tracking and control and data analysis technology, and in particular to a data interpretation method and system based on a rocket characteristic spatial database. Background Technology

[0002] The development, testing, and launch of launch vehicles generate a large amount of heterogeneous data from different systems. Existing interpretation methods primarily rely on human experience, which has the following significant drawbacks: Data transmission channels suffer from severe and random packet loss when communication quality is poor; characteristic data of the same rocket system is scattered across multiple data packets, and the lack of a unified time benchmark alignment between these packets makes data fusion difficult. Furthermore, feature classification and naming rules are inconsistent and non-unique, hindering cross-mission and cross-model feature comparisons and longitudinal source tracing analysis. In addition, existing criteria are fragmented, mostly focusing on threshold checks for single parameters, lacking systematic definition and execution capabilities for complex scenarios such as cross-system parameter correlations and temporal logic.

[0003] Therefore, it is necessary to provide a new technical solution to solve the above problems. Summary of the Invention

[0004] To address the aforementioned technical issues, this application provides a data interpretation method and system based on a rocket feature space database. This method utilizes the inherent unified encoding and global time reference of the feature space database to support cross-system and cross-parameter correlation analysis and timing logic checks. It can detect hidden faults or system mismatches that cannot be identified by single parameter interpretation, significantly improving the depth and accuracy of interpretation.

[0005] A data interpretation method based on a rocket feature space database includes: S1. Construct a rocket feature space database; S2. Using the feature space in the rocket feature space database as a unified data base, construct a multi-level criterion classification library that maps to the dimensions of the feature space; S3. Construct a data cross-analysis engine based on unified feature coding and time benchmark; S4. Data interpretation is performed based on a multi-level criterion classification library and a data cross-analysis engine.

[0006] Optionally, step S1 includes: All key state parameters throughout the entire life cycle of a launch vehicle are mapped to a standardized multidimensional description framework to obtain a feature space, enabling all key state parameters to have unified feature encoding. Establish a global timeline based on launch time zero, convert and interpolate the timestamps of all data onto the global timeline, construct a timing processing framework, and enable all key state parameters to have a unified time reference. Based on unified feature coding and time benchmark, a database supporting the storage and querying of data from multiple models and multiple tasks was designed and constructed.

[0007] Optionally, all key state parameters throughout the launch vehicle's entire lifecycle can be mapped to a standardized multidimensional description framework to obtain a feature space, enabling all key state parameters to have unified feature encoding, including: All key state parameters throughout the entire life cycle of a launch vehicle are defined according to the feature acquisition system, the equipment to which the feature belongs, the feature refinement classification, and the interpretation method. Define a structured encoding for each unique parameter in the feature space, which serves as its globally unique identifier and index key in the database; Map all key state parameters of the launch vehicle throughout its entire life cycle to structured codes in the feature space.

[0008] Optionally, step S2 includes: Define the criteria categories in the multi-level criteria classification library and store each category of criteria in a structured form; Each criterion is strongly correlated with the feature codes in the rocket feature space database to form a traceable link between features and criters.

[0009] Optionally, in defining the criteria categories in the multi-level criteria classification library, the criteria categories include: threshold criteria, state criteria, instruction criteria, time-series criteria, and correlation analysis criteria.

[0010] Optionally, the data cross-analysis engine in step S3 can perform the following schemes: Based on the feature codes involved in the criteria, the engine automatically identifies and associates multiple feature parameters from different systems but belonging to the same physical process or device; The engine relies on a globally unified timeline to perform precise time alignment and comparison of multiple characteristic time series data with causal, sequential, or synchronous relationships, verifying the correctness of their temporal logic. For correlation analysis criteria, the engine automatically extracts data of correlation features, performs specified mathematical operations, logical judgments or model calculations, and verifies whether the results meet expectations.

[0011] Optionally, step S4 includes: Based on the interpretation of mission instructions, the system automatically retrieves the corresponding, time-aligned structured feature data from the rocket feature space database, according to the mission identifier and feature code. Based on the current task stage and target system information, load all relevant criterion templates from the multi-level criterion classification library to form the task criterion set for this interpretation; The scheduling data cross-analysis engine executes each criterion in the task criterion set sequentially or in parallel. Summarize all the judgment results and conduct a comprehensive status assessment.

[0012] According to another aspect of this application, a data interpretation system based on a rocket feature space database is also provided for executing the data interpretation method, including a data access layer, a feature space layer, an interpretation engine layer, and an extended application layer. Among them, the data access layer serves as the data source entry point of the system, providing raw data support for subsequent stages, including the CSV data package module, the CSV configuration file module, and the test, development, and control logs; The feature space layer can map all key state parameters throughout the rocket's entire life cycle into a standardized multidimensional description framework, ensuring the consistency and comparability of data representation. The interpretation engine layer can use data from the feature space database, combined with preset criterion templates, to automatically parse and execute interpretation logic according to data type, and output the final interpretation result. The interpretation engine layer includes a criterion template module, an automatic interpretation engine module, and a interpretation result module. The extended application layer is used to generate visual reports of the results for easy viewing and supports manual interaction to review and verify the results. At the same time, it can carry out the construction of a spectral database based on the data and improve the data resource system. The extended application layer includes a visual report generation module, a manual interaction review module, and a spectral database construction module.

[0013] According to another aspect of this application, a computing device is also provided, comprising: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the data interpretation method.

[0014] According to another aspect of this application, a computer-readable storage medium is also provided, which stores computer instructions that, when executed on a computer, cause the computer to perform the data interpretation method.

[0015] Compared with the prior art, this application has at least the following beneficial effects: 1. This invention realizes the systematization and intelligence of interpretation: by establishing a structured multi-level criterion classification library, expert experience and design specifications are transformed into calculable and manageable digital rules, and deeply integrated with the feature space, so that the interpretation process moves from "manually looking at curves" to "system running rules".

[0016] 2. This invention utilizes the inherent unified encoding and global time benchmark of the feature space database, which naturally supports cross-system and cross-parameter correlation analysis and timing logic checks. It can discover hidden faults or system mismatches that cannot be identified by single parameter interpretation, greatly improving the depth and accuracy of interpretation.

[0017] 3. The automated interpretation process of this invention can quickly process massive amounts of data, reduce reliance on manpower, shorten the interpretation cycle, and avoid inconsistent interpretation results caused by human fatigue or experience differences. It is particularly suitable for batch testing and rapid launch evaluation.

[0018] 4. This invention constructs a continuously evolving knowledge ecosystem for interpretation. Through multi-level criterion classification libraries, interpretation cases, and feature data linked storage, a complete closed loop of "data-rules-conclusions" is formed. This not only facilitates knowledge accumulation and reuse but also provides a high-quality data foundation for the subsequent training of intelligent diagnostic models based on machine learning. Attached Figure Description

[0019] The following sections will describe some specific embodiments of the invention in detail by way of example and not limitation, with reference to the accompanying drawings. The same reference numerals in the drawings denote the same or similar parts or portions. Those skilled in the art should understand that these drawings are not necessarily drawn to scale. In the drawings: Figure 1 This is a schematic diagram of the overall process of the present invention; Figure 2 This is a slice of the feature dimension of the feature space of a certain type of rocket in an embodiment of the present invention; Figure 3 This is an architecture diagram of the data interpretation system; Figure 4 This is the user interface of the data interpretation system of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] A data interpretation method based on a rocket feature space database is proposed. The core innovation of this method lies in: constructing a multi-level structured criterion classification system using the feature space as a unified data foundation; designing a data cross-analysis engine based on feature encoding and a unified time benchmark; and implementing a rule- and process-driven fully automated interpretation closed loop. Ultimately, expert experience is solidified into calculable and reusable digital criteria, significantly improving the efficiency, accuracy, consistency, and intelligence level of the interpretation work.

[0022] like Figure 1 As shown, a data interpretation method based on a rocket feature space database includes the following steps: S1. Construct a rocket feature space database.

[0023] Specifically, step S1 includes: S11. Map all key state parameters throughout the entire life cycle of the launch vehicle to a standardized multi-dimensional description framework to obtain a feature space, so that all key state parameters have a unified feature code.

[0024] Specifically, all key state parameters throughout the launch vehicle's entire lifecycle are defined according to the feature acquisition system, the equipment to which the feature belongs, the feature refinement classification, and the interpretation method. A structured code is defined for each unique parameter in the feature space, serving as its globally unique identifier and index key in the database. All key state parameters throughout the launch vehicle's entire lifecycle are then mapped to the structured codes in the feature space.

[0025] The feature space is the core abstraction of this invention's method. It maps all key state parameters throughout the rocket's entire lifecycle to a standardized, multi-dimensional descriptive framework, ensuring the consistency and comparability of data representation. A slice of the feature dimensions of a certain type of rocket's feature space, i.e., the evolutionary analysis process of the same features evolving over time under different missions, is shown below. Figure 2 As shown. The specific design of the feature space is as follows: The feature acquisition system includes the onboard control system, the ground telemetry and control system, and the measurement system.

[0026] The equipment to which the features belong includes the flight control system, main inertial navigation system, core-level servo mechanism, core-level lithium battery, etc.

[0027] The detailed feature classification includes timing data, servo data, servo data interpretation - working status, bus and communication interface status, etc.

[0028] Interpretation methods can be categorized into threshold-based data, status-based data, instruction-based data, time-series data, and correlation analysis data.

[0029] Define a structured encoding for each unique parameter in the feature space, serving as its globally unique identifier and index key in the database. An example of the encoding rule is as follows: "Feature acquisition system_Feature-related device_Feature category_Feature name".

[0030] For example: "Onboard Control System_Flight Control Combination_Timing Data_Second Stage Engine Ignition".

[0031] This encoding integrates multi-dimensional information, achieves standardized naming of features, and naturally supports rapid retrieval and aggregation analysis by model, stage, system, and other dimensions.

[0032] S12. Establish a global timeline based on launch time zero, convert and interpolate the timestamps of all data onto the global timeline, construct a timing processing framework, and ensure that all key state parameters have a unified time reference.

[0033] By establishing a unified time-series processing framework, it is possible to achieve accurate alignment and comparison of time-series data across systems and tasks.

[0034] To address the issues of independent time stamps and inconsistent sampling rates in the raw data of each subsystem, a unified time stamp mapping and time interpolation technique is adopted to convert and interpolate the timestamps of all data onto the global time axis, ensuring that the parameters of each system can be synchronously correlated and analyzed at any given time.

[0035] S13. Based on unified feature coding and time benchmark, design and build a database that supports the storage and querying of data from multiple models and multiple tasks.

[0036] S2. Using the feature space in the rocket feature space database as a unified data base, construct a multi-level criterion classification library that maps to the dimensions of the feature space.

[0037] To address the complexity of rocket data interpretation, a multi-level criterion classification system that maps to the dimensions of the feature space can be designed to ensure that every interpretation intent can be accurately defined and managed.

[0038] Specifically, step S2 includes: S21. Define the criteria categories in the multi-level criteria classification library and store each category of criteria in a structured form.

[0039] In defining the criteria categories in the multi-level criteria classification library, the criteria categories include: threshold criteria, state criteria, instruction criteria, time-series criteria, and correlation analysis criteria.

[0040] Among them, threshold-based criteria: set upper and lower limits, safety ranges or rate of change thresholds for continuous quantity characteristics for inspection, such as "the engine combustion chamber pressure should be between P_min and P_max".

[0041] State-related criteria: Check whether the value of the discrete state variable is equal to the expected state enumeration value, such as "the landing system state should be 'unlocked'".

[0042] Command-type criteria: Check whether the sequence of characteristic values ​​of the command signal completely matches the expected command sequence to be sent.

[0043] Time-series criteria: Define the logical order and time interval requirements of characteristic events on the time axis, such as "after event A occurs, event B must be detected within ΔT time".

[0044] Correlation analysis criteria: Define the mathematical relationship, physical constraint or statistical correlation between multiple feature parameters, such as "the ratio of parameter X to parameter Y should be maintained within the interval R".

[0045] S22. Strongly correlate each type of criterion with the feature code in the rocket feature space database to form a traceable link between features and criterions.

[0046] Specifically, each criterion in step S21 is stored in a structured form (such as JSON, XML, or database records) and strongly correlated with the feature encoding in the feature space to form a traceable link of "feature-criteria".

[0047] S3. Construct a data cross-analysis engine based on unified feature encoding and time benchmark.

[0048] The data cross-analysis engine fully leverages the two core advantages of the feature space database—unified feature encoding and global time benchmark—to achieve in-depth "data cross" analysis.

[0049] Specifically, the data cross-analysis engine can perform the following operations: Dimensional crossover: Based on the feature codes involved in the criteria, the engine automatically identifies and associates multiple feature parameters from different systems (such as control systems and power systems) that belong to the same physical process or equipment.

[0050] Time-series crossover: The engine relies on a globally unified time axis to perform precise time alignment and comparison of multiple characteristic time series data with causal, sequential, or synchronous relationships, verifying the correctness of their time-series logic.

[0051] Association calculation and verification: For association analysis criteria, the engine automatically extracts the data of association features, performs the specified mathematical operations, logical judgments or model calculations, and verifies whether the results meet expectations.

[0052] S4. Data interpretation is performed based on a multi-level criterion classification library and a data cross-analysis engine.

[0053] Specifically, step S4 includes: S41. Based on the interpretation of mission instructions, automatically retrieve the corresponding, time-aligned structured feature data from the rocket feature space database according to the mission identifier and feature code.

[0054] S42. Based on the current task stage and target system information, load all relevant criterion templates from the multi-level criterion classification library to form the task criterion set for this interpretation.

[0055] S43, the scheduling data cross-analysis engine, executes each criterion in the task criterion set sequentially or in parallel.

[0056] Specifically, single-feature criteria are calculated directly, while multi-feature, time-series, and correlation-based criteria are automatically triggered through a data cross-analysis process.

[0057] S44. Summarize all judgment results and conduct a comprehensive status assessment.

[0058] The overall status assessment results can include "normal", "warning", and "abnormal".

[0059] Furthermore, when summarizing all the judgment results for comprehensive status assessment, the process also includes: automatically retrieving corresponding feature data from historical missions and comparing it with the judgment results to assist in locating deviations. Historical missions can be previously successfully flown missions.

[0060] Furthermore, after step S44, the following steps are also included: S45. Automatically generate structured interpretation reports.

[0061] The interpretation report includes the interpretation conclusion, details of the triggered anomaly criteria, and snapshots of related data. All interpretation processes and results are fed back to the database and stored in association with the original feature data, forming traceable and analyzable interpretation knowledge cases.

[0062] Furthermore, such as Figure 3 As shown, the present invention also provides a data interpretation system based on a rocket feature space database for the aforementioned data interpretation method, comprising a data access layer, a feature space layer, an interpretation engine layer, and an extended application layer.

[0063] The data access layer serves as the system's data source entry point, providing raw data support for subsequent stages. It includes a CSV data package module, a CSV configuration file module, and test, launch, and control logs.

[0064] The feature space layer can map all key state parameters throughout the rocket's entire life cycle into a standardized multidimensional description framework, ensuring the consistency and comparability of data representation.

[0065] The interpretation engine layer, based on data from the feature space database and combined with preset criterion templates, automatically parses and executes interpretation logic according to data type, outputting the final interpretation result. The interpretation engine layer includes a criterion template module, an automatic interpretation engine module, and a interpretation result module.

[0066] The extended application layer is used to expand the applications of the interpretation results, namely, generating visual reports for easy viewing, supporting manual interaction for review and verification of the interpretation results, and building a spectral database based on the data to improve the data resource system. The extended application layer includes a visual report generation module, a manual interaction review module, and a spectral database construction module.

[0067] The user interface of the data interpretation system based on the rocket feature space database is as follows: Figure 4 As shown.

[0068] Practical verification shows that the solution of this invention can shorten the data interpretation work of the entire rocket, which originally required several person-hours, to half an hour, and automatically generate standardized interpretation reports, improving efficiency by more than 90%. Simultaneously, it achieves automated coverage of over 80% of interpretable features, eliminating subjective differences in manual interpretation and achieving 100% consistency in results. Through feature cross-analysis, it successfully discovered several deep-seated potential problems such as timing mismatches and signal correlations between system interfaces that could not be identified by single-parameter interpretation, improving fault early warning capabilities. Furthermore, it forms digital assets that can be continuously accumulated and optimized as the model evolves.

[0069] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0070] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in sequences other than those illustrated or described herein.

[0071] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A data interpretation method based on a rocket feature space database, characterized in that, include: S1. Construct a rocket feature space database; S2. Using the feature space in the rocket feature space database as a unified data base, construct a multi-level criterion classification library that maps to the dimensions of the feature space; S3. Construct a data cross-analysis engine based on unified feature coding and time benchmark; S4. Data interpretation is performed based on a multi-level criterion classification library and a data cross-analysis engine.

2. The data interpretation method as described in claim 1, characterized in that, Step S1 includes: All key state parameters throughout the entire life cycle of a launch vehicle are mapped to a standardized multidimensional description framework to obtain a feature space, enabling all key state parameters to have unified feature encoding. Establish a global timeline based on launch time zero, convert and interpolate the timestamps of all data onto the global timeline, construct a timing processing framework, and enable all key state parameters to have a unified time reference. Based on unified feature coding and time benchmark, a database supporting the storage and querying of data from multiple models and multiple tasks was designed and constructed.

3. The data interpretation method as described in claim 2, characterized in that, Map all key state parameters throughout the entire lifecycle of a launch vehicle to a standardized, multi-dimensional description framework. The feature space is obtained, enabling all key state parameters to have unified feature encoding, including: All key state parameters throughout the entire life cycle of a launch vehicle are defined according to the feature acquisition system, the equipment to which the feature belongs, the feature refinement classification, and the interpretation method. Define a structured encoding for each unique parameter in the feature space, which serves as its globally unique identifier and index key in the database; Map all key state parameters of the launch vehicle throughout its entire life cycle to structured codes in the feature space.

4. The data interpretation method as described in claim 3, characterized in that, Step S2 includes: Define the criteria categories in the multi-level criteria classification library and store each category of criteria in a structured form; Each criterion is strongly correlated with the feature codes in the rocket feature space database to form a traceable link between features and criters.

5. The data interpretation method as described in claim 4, characterized in that, In defining the criteria categories in the multi-level criteria classification library, the criteria categories include: threshold criteria, state criteria, instruction criteria, time-series criteria, and correlation analysis criteria.

6. The data interpretation method as described in claim 5, characterized in that, The data cross-analysis engine in step S3 can perform the following schemes: Based on the feature codes involved in the criteria, the engine automatically identifies and associates multiple feature parameters from different systems but belonging to the same physical process or device; The engine relies on a globally unified timeline to perform precise time alignment and comparison of multiple characteristic time series data with causal, sequential, or synchronous relationships, verifying the correctness of their temporal logic. For correlation analysis criteria, the engine automatically extracts data of correlation features, performs specified mathematical operations, logical judgments or model calculations, and verifies whether the results meet expectations.

7. The data interpretation method as described in claim 6, characterized in that, Step S4 includes: Based on the interpretation of mission instructions, the system automatically retrieves the corresponding, time-aligned structured feature data from the rocket feature space database, according to the mission identifier and feature code. Based on the current task stage and target system information, load all relevant criterion templates from the multi-level criterion classification library to form the task criterion set for this interpretation; The scheduling data cross-analysis engine executes each criterion in the task criterion set sequentially or in parallel. Summarize all the judgment results and conduct a comprehensive status assessment.

8. A data interpretation system based on a rocket feature space database, characterized in that, The method for performing the data interpretation method according to any one of claims 1 to 7 includes a data access layer, a feature space layer, an interpretation engine layer, and an extended application layer; Among them, the data access layer serves as the data source entry point of the system, providing raw data support for subsequent stages, including the CSV data package module, the CSV configuration file module, and the test, development, and control logs; The feature space layer can map all key state parameters throughout the rocket's entire life cycle into a standardized multidimensional description framework, ensuring the consistency and comparability of data representation. The interpretation engine layer can use data from the feature space database, combined with preset criterion templates, to automatically parse and execute interpretation logic according to data type, and output the final interpretation result. The interpretation engine layer includes a criterion template module, an automatic interpretation engine module, and a interpretation result module. The extended application layer is used to generate visual reports of the results for easy viewing and supports manual interaction to review and verify the results. At the same time, it can carry out the construction of a spectral database based on the data and improve the data resource system. The extended application layer includes a visual report generation module, a manual interaction review module, and a spectral database construction module.

9. A computing device, characterized in that, include: The processor and the memory storing a computer program, wherein the computer program, when executed by the processor, performs the data interpretation method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed on a computer, cause the computer to perform the data interpretation method as described in any one of claims 1 to 7.

Images