Scientific data analysis platform for fast radio burst

Abstract

The application discloses a kind of scientific research data analysis platform for fast radio burst, including data management module: for the structured, standardized, standardization management fast radio burst related scientific research data;Fast radio burst searching module: for searching fast radio burst in radio telescope observation data, and intercept fast radio burst sample storage in database;Fast radio burst sample analysis module: based on astronomical data analysis algorithm component, fast radio burst sample is in-depth analyzed, and analysis result is transmitted to visual module;Visual module: based on real galaxy distribution constructs 3D celestial sphere model, and fast radio burst data and analysis result are shown on the corresponding position of celestial sphere model.The application provides fast radio burst data acquisition, management and sharing function, constructs rich standardization data analysis capability, realizes analysis result visualization, greatly improves data acquisition and analysis efficiency, promotes astronomical achievement sharing, accelerates scientific research achievement output.

Description

Technical Field

[0001] This invention relates to the field of data processing and analysis technology, specifically a scientific research data analysis platform for fast radio bursts. Background Technology

[0002] Currently, radio astronomy has entered the era of ultra-large-scale sky surveys, resulting in an explosive growth in the amount of data that needs to be processed. Especially with the deployment of large radio telescopes such as FAST (Five-hundred-meter Aperture Spherical Telescope), the demand for big data management and computation in radio astronomy is increasing daily. Fast radio bursts (FRBs) are one of the hottest frontiers in radio astronomy today, and there is still no universally accepted explanation for their origin. Clarifying their origin could have a revolutionary impact on physics and astronomy. Taking FAST as an example, it can generate hundreds of gigabytes of data daily during observations, and this amount will increase significantly with equipment upgrades. Physically transporting data by copying it to hard drives is inefficient, and developing data processing methods that rely on stand-alone machines for execution is difficult to guarantee in terms of code quality and processing capacity that cannot keep up with such a massive data volume. Therefore, researchers in the field of radio astronomy have a strong need for data processing platforms based on big data technologies.

[0003] However, existing big data analytics platforms primarily target fields such as the internet, industry, and finance, and cannot conduct targeted data analysis for scientific purposes. They do not support data access and management from radio telescopes, nor do they support the operation of scientific algorithms in the astronomical field, thus affecting the efficiency of scientific research in radio astronomy. Searching for fast radio bursts from massive amounts of observational data and extracting effective information and scientific conclusions presents a significant challenge to researchers. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention discloses a scientific data analysis platform for fast radio bursts, which improves the efficiency of researchers in acquiring and analyzing fast radio burst data and accelerates the output of scientific research results.

[0005] To achieve the above objectives, this invention discloses a scientific data analysis platform for fast radio bursts, which includes:

[0006] Data Management Module: Used to manage the raw observation data of fast radio bursts, the intermediate data generated during processing, and the result data obtained after analysis;

[0007] Fast Radio Burst Search Module: Used to search for the existence of fast radio bursts in the observation data, locate the burst time coordinates of the searched bursts, collect fast radio burst samples based on the time coordinates, and store them in the data management module;

[0008] Fast Radio Burst Sample Analysis Module: Based on the astronomical data analysis algorithm component, it analyzes the fast radio burst samples obtained by the fast radio burst search module, transmits the analysis results to the visualization module, and stores the intermediate data and result data obtained by the fast radio burst sample analysis module in the data management module.

[0009] Visualization module: Based on the actual distribution of galaxies, a 3D celestial sphere model is constructed, and fast radio burst data and analysis results are displayed at the corresponding positions in the celestial sphere model.

[0010] Furthermore, the data management module consists of the following sub-modules, including:

[0011] The data statistics submodule is used to collect statistics on the data storage and usage within the platform, including trends in total data volume, data type distribution, platform access frequency, data download frequency, data capacity ratio, and platform registered user number.

[0012] The data retrieval submodule returns target data that meets the user's input of professional fields related to fast radio bursts.

[0013] The data read / write submodule is used for external modules to read data and write result data, and supports manual correction of stored data.

[0014] The database submodule is used to categorize and store relevant data generated at different stages of research on fast radio bursts.

[0015] Furthermore, the fast radio burst search module consists of the following sub-modules, including:

[0016] The interference cancellation submodule is used to eliminate interference signals in the original observation data;

[0017] The achromatic submodule is used to eliminate the dispersion effect caused by the spatial dispersion of free electrons during the propagation of radio pulse signals;

[0018] The feature extraction submodule extracts the corresponding features based on the search algorithm selected by the platform user and transmits them to the candidate recognition submodule. The features that can be extracted include periodic features, pulse contour description features, image features, and convolutional time features.

[0019] The candidate identification submodule trains one or more binary classifiers based on deep learning algorithms, inputs the received features into the binary classifiers for scoring, combines one or more scoring results, and uses those results that are higher than a threshold as fast radio burst candidates, which are then transmitted to the candidate discrimination submodule.

[0020] The candidate discrimination submodule sets discrimination rules based on the values ​​of professional fields in the field of fast radio bursts, filters out candidates that meet the value range of the fields in the rules, and marks them as real fast radio bursts;

[0021] The data sampling submodule extracts the corresponding raw observation data sampling time segments based on the identified burst time coordinates of the real fast radio bursts and stores them in the data management module.

[0022] Furthermore, the fast radio burst sample analysis module consists of the following sub-modules, including:

[0023] The algorithm component submodule is used to build data analysis algorithm components for different scientific research scenarios. The data analysis algorithm components include: calibration analysis, scattering analysis, polarization analysis, microscopic information calculation and macroscopic information calculation for a single fast radio burst pulse; energy distribution analysis, time distribution analysis and period analysis for multiple fast radio burst pulses; and pulse width distribution analysis, energy distribution analysis, polarization distribution analysis, correlation analysis and cluster analysis for fast radio burst groups.

[0024] The pulse analysis submodule performs in-depth analysis on the preprocessed fast radio burst pulse data by calling algorithm components, and classifies and stores the analyzed data according to the fast radio burst source coordinates.

[0025] Furthermore, the visualization module consists of the following sub-modules, including:

[0026] The visual processing submodule is used to perform visual rendering on the astronomical data and analysis results stored in the platform, and to display the rendered visualization results on the user interface.

[0027] The human-computer interaction submodule is used to perform interactive display according to the user interaction processing needs of the platform. The interaction processing includes scene scaling, scene positioning, and layer transformation.

[0028] Furthermore, the discrimination rules for candidates can be configured by adjusting the value range of professional fields in the field of fast radio bursts based on the user's own experience. Whether a candidate is a real fast radio burst sample can be determined based on whether the candidate meets the discrimination rules configured by the user.

[0029] Furthermore, users can select the data to be analyzed and the algorithm components to be invoked through the platform interface, generate corresponding data analysis tasks according to user needs, and display the analysis results to users through the interface.

[0030] Furthermore, the fast radio burst sample analysis module allows users to view the progress of the data analysis task at any time during the data analysis process, pause or restart the analysis task, and modify the parameter values ​​of the algorithm components through the user interface to obtain analysis results under different parameters. The optimal result can be obtained by comparing the results under different parameters.

[0031] Furthermore, the fast radio burst sample analysis module can execute multiple data analysis tasks simultaneously. When multiple tasks need to be executed, the module will allocate computing resources to the analysis tasks based on the availability of computing resources, thereby enabling parallel execution of multiple tasks.

[0032] Furthermore, the algorithm components in the fast radio burst sample analysis module run user-compiled algorithm components and, based on the platform's computing power resources, call the user-compiled algorithm components within the platform to perform data analysis.

[0033] The beneficial effects of this invention are as follows:

[0034] This invention provides a fast radio burst (FRB) scientific data analysis platform with strong processing capabilities and ease of data analysis, management, and mining. It overcomes the limitations of general-purpose big data analysis platforms, which cannot handle FRB-specific scientific data analysis tasks and do not support radio astronomy observation data formats. This invention categorizes and stores FRB-related scientific data, allowing users to easily and conveniently upload or download data through a user interface, enabling the acquisition, management, and sharing of radio astronomy data. Furthermore, the data analysis platform encapsulates commonly used data analysis algorithm components in the FRB field, providing users with rich data analysis capabilities. Additionally, this invention simulates real galaxy distributions and constructs a celestial sphere model commonly used in astronomical research, enabling the visualization of observational data and analysis results. In summary, this invention significantly improves data acquisition and analysis efficiency, promotes the sharing of astronomical achievements, and accelerates the output of scientific research results. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 A structural diagram of a scientific data analysis platform for fast radio bursts;

[0037] Figure 2 This is a structural diagram of the data management module in the data analysis platform.

[0038] Figure 3 This is a structural diagram of the fast radio burst search module in the data analysis platform.

[0039] Figure 4 This is a structural diagram of the fast radio burst sample analysis module in the data analysis platform. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] Example 1

[0042] This embodiment provides a scientific data analysis platform for fast radio bursts. Please refer to [link / reference]. Figure 1 ,include:

[0043] The data management module is used to manage the raw observation data, intermediate data generated during processing, and the results data obtained after analysis of fast radio bursts in a structured, standardized, and normalized manner. This invention uses MongoDB to classify, store, and manage fast radio burst-related data at different stages. Since the data types generated at different stages are different, this invention adopts a data classification and management mode for different stages.

[0044] The Fast Radio Burst (FRB) search module searches for the presence of FRBs in observational data, locates the time coordinates of the found bursts, collects FRB samples based on the time coordinates, and stores them in the data management module. Supported search algorithms include PRESTO, HEIMDALL, and a self-developed deep learning algorithm based on YOLOv5. Observational data refers to the pulse signal data received by the radio telescope, which can be plotted as a pulse signal graph. A FRB is a brief and intense burst of radio waves, which appears as a short, bright, sloping curve in the pulse signal graph. The FRB search involves searching the observational data for signals that match the characteristics of the current observation, determining whether they are genuine FRB samples based on configured manual rules, and finally storing genuine FRB samples in the data management module while discarding false candidate samples.

[0045] The Fast Radio Burst (FRB) sample analysis module, based on astronomical data analysis algorithm components, performs in-depth analysis of FRB samples and transmits the analysis results to the visualization module. FRB samples present different analysis needs, and the platform has built-in corresponding algorithm components for these needs. Users of the data development platform can select different algorithm components for analysis according to specific requirements. Intermediate and final data obtained by the FRB sample analysis module are stored in the data management module.

[0046] The visualization module constructs a 3D celestial model based on the actual distribution of galaxies, displaying fast radio burst data and analysis results on the corresponding positions of the celestial model. The celestial model is a sphere defined in astronomy as having the same center as Earth, the same axis of rotation, and an infinitely large radius. All objects in the sky can be considered as projections onto the celestial sphere. Earth's equator and geographic poles projected onto the celestial sphere become the celestial equator and celestial pole; to accurately describe the positions of celestial bodies, astronomers have developed a coordinate system to mark their positions on the celestial sphere. By projecting astronomical observation data and analysis results onto the celestial sphere, researchers can more intuitively understand the relative distribution and correlations between data.

[0047] As one embodiment of the present invention, as shown in the appendix Figure 2 As shown, the data management module consists of the following sub-modules, including:

[0048] The data statistics submodule is used to statistically analyze the data storage and usage within the platform, including trends in total data volume, data type distribution, platform access frequency, data download frequency, data volume ratio, and platform registered user number; it is mainly used to assess the platform's development status.

[0049] The data retrieval submodule returns target data that meets the user's input of professional fields in the field of fast radio bursts. The query conditions are all professional fields in the field of radio astronomy, including: telescope observation time, telescope name, data type, data name, right ascension, declination, and host galaxy.

[0050] The data read / write submodule is used to read data from external modules and write result data, and supports manual correction of stored data. Due to the differences in format and fields between the raw observation data, intermediate data generated during processing, and result data obtained after analysis of fast radio bursts, the read / write submodule needs to be called and data type parameters need to be passed in during read / write operations to correctly read and write the target data. The data types include: fast radio burst observation data (FRB fits), fast radio burst sample data (FRB sample), and fast radio burst pulse data (FRB pulse).

[0051] The database submodule is used to classify and store relevant data generated at different stages of research on fast radio bursts; these research stages include: fast radio burst observation data acquisition, fast radio burst search, and fast radio burst sample analysis.

[0052] As one embodiment of the present invention, as shown in the appendix Figure 3 As shown, the fast radio burst search module consists of the following sub-modules, including:

[0053] The interference reduction submodule is used to eliminate interference signals in the original observation data. First, it marks strong narrowband interference and short-term broadband interference in the observation data. Without processing the data of the marked channels, it reduces interference by removing interference frequency channel data and converting some channel data through methods such as waveform clipping.

[0054] The dedispersion submodule is used to eliminate the dispersion effect caused by spatially dispersed free electrons during the propagation of radio pulse signals. Since the dispersion value of the searched fast radio bursts is unknown, a blind search experiment for dispersion values ​​is required for each observation data point. First, an optimal DM step size selection scheme is planned based on information such as the center frequency and sampling time of the observation data to obtain a dispersion grid. Then, based on the generated dispersion grid, dedispersion processing is performed on each observation data point to finally obtain the observation data after eliminating the dispersion effect.

[0055] The feature extraction submodule extracts corresponding features based on the search algorithm selected by the platform user and transmits them to the candidate recognition submodule. Supported features include periodic features, pulse contour description features, image features, and cross-correlation time series features. Different extraction methods correspond to different search features, and the platform automatically extracts the corresponding features based on the search algorithm selected by the user. The purpose of feature extraction is to find some significant feature points in the observation data to describe the characteristics of the observation data, which facilitates subsequent feature matching and finds fast radio burst candidates that match the observation features.

[0056] The candidate identification submodule trains one or more binary classifiers based on deep learning algorithms. The received features are input into the binary classifiers for scoring. The results of one or more scoring methods are combined, and those with results higher than a threshold are identified as fast radio burst candidates and transmitted to the candidate discrimination submodule. The binary classifiers are mainly used to filter out candidates with higher true probabilities, thereby accelerating the search speed and reducing the limitations of relying entirely on manual rule-based discrimination. The purpose of using multiple binary classifiers is to reduce the probability of misclassification by a single classifier.

[0057] The candidate discrimination submodule sets discrimination rules based on the values ​​of professional fields in the field of fast radio bursts (FRBs), and filters out candidates that meet the value range of the fields in the rules, marking them as real FRBs. Since the binary classifier trained by the deep learning algorithm cannot guarantee 100% accuracy, to reduce the waste of data analysis resources caused by misclassification of FRBs, an additional layer of filtering rules based on human experience is added to remove samples with high scores but significantly false results. At the same time, due to different user experiences and analysis scenarios, the platform supports user-defined filtering rules. Commonly used professional fields in the rules include dispersion value and signal-to-noise ratio.

[0058] The data sampling submodule extracts the corresponding raw observation data sampling time segments based on the identified burst time coordinates of the real fast radio bursts and stores them in the database submodule of the fast radio burst sample analysis module.

[0059] As one embodiment of the present invention, as shown in the appendix Figure 4 As shown, the fast radio burst sample analysis module consists of the following sub-modules, including:

[0060] The algorithm component submodule is used to build data processing and analysis algorithm components for different scientific research scenarios. Among them, the data analysis algorithm components include: calibration analysis, scattering analysis, polarization analysis, microscopic information calculation, and macroscopic information calculation for a single fast radio burst pulse; energy distribution analysis, time distribution analysis, and periodic analysis for multiple fast radio burst pulses; and pulse width distribution analysis, energy distribution analysis, polarization distribution analysis, correlation analysis, and cluster analysis for fast radio burst groups. The scientific research scenarios of fast radio bursts are mainly divided into single-pulse analysis, multi-pulse analysis, and group analysis. The platform has built-in algorithm components required for each scenario, and allows users to view the purpose, principle, and core code of each algorithm component through the interface, so that they can select and call them according to their needs during analysis.

[0061] The pulse analysis submodule performs in-depth analysis of preprocessed fast radio burst pulse data by calling algorithm components, and classifies and stores the analyzed data according to the source coordinates of the fast radio bursts. Users select the data to be analyzed and the algorithm components to be called through the platform interface. The pulse analysis submodule generates the corresponding data analysis task according to the user's needs and displays the analysis results to the user through the interface.

[0062] In this embodiment, if it is necessary to process and analyze the observation data of the FAST radio telescope on January 1, 2022, the user can select the telescope = "FAST" and the observation time = "January 1, 2022" to perform a fast radio burst search. After the search is completed, if a fast radio burst is found, the fast radio burst sample data will be added to the database. After finding the corresponding sample data, the user can select the corresponding algorithm component and create an analysis task according to the subsequent scientific research analysis needs. Taking flux calibration in single-pulse analysis as an example, the user selects the fast radio burst sample data of the telescope = "FAST" and the observation time = "January 1, 2022", and selects the algorithm component = "single-pulse analysis / flux calibration" to create the task. After successful creation, the task will start to execute automatically. After execution, the analysis results will be displayed on the interface and mapped to the corresponding position of the fast radio burst source in the celestial sphere.

[0063] Example 2

[0064] At the implementation level, based on Example 1, this example refers to... Figure 1 The following provides a further detailed description of the scientific data analysis platform for fast radio bursts in Example 1. The data management module also supports users uploading their own data via cloud drive address or file import, and using the platform's algorithm components for data analysis. Before uploading data, platform users can obtain templates and specifications for different types of data from the platform to prevent misalignment or failure during data parsing, which could affect subsequent analysis results. Simultaneously, users can choose whether to make the uploaded data public; non-public data is only visible to the uploaded user to ensure data ownership.

[0065] The algorithm component submodule in the fast radio burst sample analysis module can also run user-compiled algorithm components and perform analysis within the platform based on the platform's computing resources. Custom algorithm components support platform users writing functions for specific analysis scenarios using Python. After a user submits the function code, the platform automatically extracts parameters and encapsulates the code into an algorithm component, providing an interface for calling external modules. The function code can call general Python components or directly call some or all of the functions in the platform's built-in algorithm components as a step in implementing the custom function. Simultaneously, users can choose whether to make the algorithm components public when submitting code. For non-public algorithm components, only the user who wrote the code can see them, ensuring the user's code ownership.

[0066] The pulse analysis submodule in the fast radio burst sample analysis module allows users to view the task execution progress and intermediate results at any time during the analysis task execution, pause or restart the task, and modify the parameter values ​​of the algorithm components through the user interface to obtain analysis results under different parameters. Users can obtain the optimal result by comparing the results under different parameters.

[0067] In this embodiment, as in the single-pulse analysis-flow calibration task described in monitoring example 1, if it is found that the model parameters need to be adjusted, the user can stop the task, bring up the parameter adjustment interface, fill in the new parameters, and restart the task; finally, the analysis result of the task is the analysis result after parameter adjustment.

[0068] Example 3

[0069] At the implementation level, based on Example 2, this example refers to... Figure 1 The following is a further detailed description of the research data development platform for fast radio bursts in Example 2. The visualization module consists of the following sub-modules, including:

[0070] The visual processing submodule is used to perform visual rendering on the astronomical data and analysis results stored in the platform, and display the rendered visualization results on the user interface. The analysis results exist in two forms: visual charts and numerical values. The visual charts will display different forms of charts depending on the analysis model, and support customization of colors, chart elements, and chart titles to adapt to different scientific research needs and visualization effect requirements.

[0071] The human-computer interaction submodule is used to provide interactive displays based on the specific needs of platform users. The interactive processing includes scene scaling, scene positioning, and layer transformation. Scene scaling is used to zoom in and out of the celestial sphere model, allowing platform users to adjust the scene range, enabling both overall observation and focused detail. Scene positioning automatically locates the corresponding position on the celestial sphere based on the data coordinates, galaxy names, and data names input by the platform user, and displays the corresponding attributes, including the results of the analysis task, astronomical data observed in the galaxy and corresponding events, and data fields. Layer transformation addresses the need to display different radio wavelengths in different scientific research scenarios, thus meeting the specific research goals of platform users.

[0072] In this embodiment, if the fast radio burst search step described in Example 1 is successful and a fast radio burst is found, the visualization module will receive the time-frequency map corresponding to the found fast radio burst. The user can customize the color and chart elements of the time-frequency map. At the same time, the user can see the source location of the fast radio burst and the distribution of surrounding galaxies in the celestial sphere.

[0073] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it; obviously, those skilled in the art will understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A research data analysis platform for fast radio bursts, characterized by, This scientific research data analysis platform includes: Data Management Module: Used to manage raw observation data of fast radio bursts, intermediate data generated during processing, and result data obtained after analysis; the data management module consists of the following sub-modules, including: The data statistics submodule is used to collect statistics on the data storage and usage within the platform, including trends in total data volume, data type distribution, platform access frequency, data download frequency, data capacity ratio, and platform registered user number. The data retrieval submodule returns target data that meets the user's input of professional fields related to fast radio bursts. The data read / write submodule is used for external modules to read data and write result data, and supports manual correction of stored data. The database submodule is used to classify and store relevant data generated at different stages of research on fast radio bursts; Fast Radio Burst (FRB) Search Module: This module searches for the presence of FRBs in the observation data, locates the burst time coordinates, collects FRB samples based on the time coordinates, and stores them in the data management module. The FRB search module consists of the following sub-modules: The interference cancellation submodule is used to eliminate interference signals in the original observation data; The achromatic submodule is used to eliminate the dispersion effect caused by the spatial dispersion of free electrons during the propagation of radio pulse signals; The feature extraction submodule extracts the corresponding features based on the search algorithm selected by the platform user and transmits them to the candidate recognition submodule. The features that can be extracted include periodic features, pulse contour description features, image features, and convolutional time features. The candidate identification submodule trains one or more binary classifiers based on deep learning algorithms, inputs the received features into the binary classifiers for scoring, combines one or more scoring results, and uses those results that are higher than a threshold as fast radio burst candidates, which are then transmitted to the candidate discrimination submodule. The candidate discrimination submodule sets discrimination rules based on the values ​​of professional fields in the field of fast radio bursts, filters out candidates that meet the value range of the fields in the rules, and marks them as real fast radio bursts; The data sampling submodule extracts the corresponding raw observation data sampling time segments based on the identified burst time coordinates of the real fast radio bursts and stores them in the data management module; Fast Radio Burst (FRB) Sample Analysis Module: Based on astronomical data analysis algorithm components, this module analyzes FRB samples obtained by the FRB search module and transmits the analysis results to the visualization module. Intermediate and final data obtained by the FRB sample analysis module are stored in the data management module. The FRB sample analysis module consists of the following sub-modules: The algorithm component submodule is used to build data analysis algorithm components for different scientific research scenarios. Among them, the data analysis algorithm components include: calibration analysis, scattering analysis, polarization analysis, microscopic information calculation and macroscopic information calculation for a single fast radio burst pulse; energy distribution analysis, time distribution analysis and period analysis for multiple fast radio burst pulses; and pulse width distribution analysis, energy distribution analysis, polarization distribution analysis, correlation analysis and cluster analysis for fast radio burst groups. The pulse analysis submodule calls the algorithm components selected by the user to perform in-depth analysis of fast radio burst samples and classifies and stores the analyzed data according to the source coordinates of the fast radio bursts. Visualization Module: Based on the actual distribution of galaxies, a 3D celestial model is constructed, displaying fast radio burst data and analysis results at the corresponding positions on the celestial model; the visualization module consists of the following sub-modules, including: The visual processing submodule is used to perform visual rendering on the astronomical data and analysis results stored in the platform, and to display the rendered visualization results on the user interface. The human-computer interaction submodule is used to perform interactive display according to the user interaction processing needs of the platform. The interaction processing includes scene scaling, scene positioning, and layer transformation.

2. The research data analysis platform for fast radio burst of claim 1, wherein, The discrimination rules for candidates are configured by adjusting the value range of professional fields in the field of fast radio bursts based on the user's own experience. The system determines whether a candidate is a real fast radio burst sample based on whether it meets the discrimination rules configured by the user.

3. The research data analysis platform for fast radio burst of claim 1, wherein, Users select the data to be analyzed and the algorithm components to be invoked through the platform interface. The platform generates corresponding data analysis tasks based on user needs and displays the analysis results to the user through the interface.

4. The research data analysis platform for fast radio burst of claim 1, wherein, The fast radio burst sample analysis module allows users to view the progress of the data analysis task at any time during the analysis process, pause or restart the analysis task, and modify the parameter values ​​of the algorithm components through the user interface based on the intermediate results generated by the data analysis to obtain analysis results under different parameters. The optimal result can be obtained by comparing the results under different parameters.

5. The research data analysis platform for fast radio burst of claim 1, wherein, The fast radio burst sample analysis module can execute multiple data analysis tasks simultaneously. When multiple tasks need to be executed, the module will allocate computing resources to the analysis tasks based on the availability of computing resources, thereby enabling the parallel execution of multiple tasks.

6. The scientific data analysis platform for fast radio bursts as described in claim 1, characterized in that, The fast radio burst sample analysis module runs user-compiled algorithm components and, based on the platform's computing power resources, calls these user-compiled algorithm components to perform data analysis within the platform.

Images