Physical condition multi-source data analysis and regulation method for college sports team training

Abstract

The application discloses a physical fitness state multi-source data analysis and regulation method for college sports team training, belongs to the technical field of sports measurement, and comprises the following steps: performing integrity check and time synchronization on multi-source physical fitness data of athletes, ensuring the accuracy and comparability of sensor data, identifying first correlation candidate indexes based on statistical characteristics of each physical fitness index, determining the display dimension of a radar chart, performing axis scale adaptive processing on statistical characteristic parameters of each physical fitness index in different training stages, generating a structured rendering parameter set, simultaneously, acquiring action parameters and physical fitness state data of athletes in real time, dynamically triggering three-dimensional model updating, and performing consistency check on three-dimensional visualization results and actual training data, so that accurate monitoring and visualized presentation of the physical fitness state of athletes are realized, training monitoring data is reliable, display is intuitive, and abnormalities are controllable, training feedback and decision basis are provided for college sports teams.

Description

Technical Field

[0001] This invention relates to the field of motion measurement technology, and in particular to a method for multi-source data analysis and control of physical fitness for university sports teams. Background Technology

[0002] In the training management of university sports teams, the level of refinement is constantly improving. The training process has gradually shifted from the traditional qualitative judgment relying on coach experience to a data-supported quantitative analysis and comprehensive control model. In actual training scenarios, athletes' physical condition is usually characterized by multi-source data such as heart rate, exercise load, movement performance, physiological indicators, and environmental perception. These data sources are characterized by high collection frequency, long time span, and diverse data types. Wearable devices, field sensors, and video acquisition devices are commonly used to continuously monitor athletes' training processes. Based on this, time series data and periodic statistical results of physical condition are constructed to assist coaches and researchers in understanding changes in athletes' training status.

[0003] Typically, data preprocessing, statistical analysis, and visualization techniques are combined to integrate and analyze physical fitness data from different training stages and with different training content. The distribution and trends of athletes' physical fitness characteristics are then visually presented using radar charts, multidimensional statistical graphs, or 3D human body models. Simultaneously, some systems have begun to incorporate rule matching, threshold judgment, or model analysis mechanisms to automatically identify changes in training intensity, load, and physical fitness fluctuations, and to some extent support dynamic adjustments and feedback during the training process.

[0004] For example, Chinese invention patent with publication number CN121264988A discloses an intelligent monitoring system for physical training. This system aims to achieve comprehensive and accurate monitoring of physical training. The intelligent monitoring system for physical training includes the following modules: a multi-dimensional data acquisition module, a data transmission module, a data analysis and processing module, a training plan generation module, and a user interaction module.

[0005] For example, Chinese invention patent CN116158740A discloses a physical fitness assessment method and device based on motion-sensing interaction technology, which relates to the field of data recognition and analysis technology. This physical fitness assessment method based on motion-sensing interaction technology includes the following testing steps: S1: Collecting user information, obtaining various user information data, and constructing a user activity information database based on these data. The user activity data database is then classified and identified to construct a database of individual activities; S2: User information analysis, combining the acquired user information with the user's current and past information data.

[0006] The above-mentioned technology has at least the following technical problems:

[0007] Multi-source physical fitness data is prone to errors during collection, preprocessing, and format conversion, which may lead to inconsistencies between the information presented in charts and the actual physical fitness status, thus affecting the objectivity of training assessment. On the other hand, the excessive number of dimensions in the visualization of physical fitness status, and the often unbalanced distribution of data among different dimensions, may mislead the judgments of coaches and athletes, reduce the efficiency of information transmission, and increase the difficulty of understanding and making decisions regarding the athlete's physical fitness status. Summary of the Invention

[0008] To address the technical problems existing in the prior art, embodiments of the present invention provide a method for multi-source data analysis and regulation of physical fitness status for university sports teams, the technical solution of which is as follows:

[0009] The collected multi-source physical fitness data were subjected to integrity verification and time synchronization processing to screen out the first relevant candidate indicators that characterize the athlete's physical fitness status.

[0010] Based on the candidate indicators of the athlete's primary correlation as the display dimensions of physical fitness status visualization, adaptive processing is performed on the axis scale of the display dimensions of physical fitness status visualization.

[0011] The statistical feature parameters of each physical fitness indicator in the final set of physical fitness indicators are subjected to axis-scale adaptive processing to generate an axis-scale adaptive configuration set of statistical feature parameters for the current stage.

[0012] The system acquires and analyzes the athlete's current motion parameters and physical condition data, triggers real-time rendering updates of the athlete's 3D model, and performs consistency checks on the display results of the athlete's 3D model.

[0013] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0014] 1. The present invention provides a method for multi-source data analysis and control of physical fitness status for university sports teams. By performing integrity verification and time synchronization on multi-source physical fitness data of athletes, the accuracy and comparability of data from various sensors are ensured. Subsequently, based on the statistical characteristics of each physical fitness indicator, the first relevant candidate indicator is identified, and the display dimension of the radar chart is determined accordingly. The statistical characteristic parameters of each physical fitness indicator at different training stages are subjected to axis-scale adaptive processing to generate a structured set of rendering parameters. At the same time, by acquiring the athlete's motion parameters and physical fitness status data in real time, the three-dimensional model can be dynamically updated, and the consistency between the three-dimensional visualization results and the actual training data is verified. This enables precise monitoring and visualization of the athlete's physical fitness status, ensuring that the training monitoring data is reliable, intuitively displayed, and anomalies are controllable, providing university sports teams with scientific physical fitness assessment, training feedback, and decision-making basis.

[0015] 2. This invention constructs time series data and performs instantaneous fluctuation analysis on multi-source physical fitness data of athletes. Combined with timestamp deviation detection of sensor data, it identifies potential unreliable abnormal fluctuations and performs time alignment correction or resampling interpolation when anomalies are detected. This ensures the consistency and reliability of multi-source data in the time dimension and effectively eliminates erroneous fluctuations caused by sensor delays, data loss, or sudden interference. This allows subsequent statistical analysis, physical fitness index evaluation, and visualization to be based on high-quality and reliable data, providing reliable data support and guarantee for the accurate monitoring, training regulation, and scientific decision-making of physical fitness status during the training process of university sports teams.

[0016] 3. This invention identifies first-relevance candidate indicators that are highly correlated with the training phase by statistically analyzing athletes' multi-source physical fitness data during the training phase and matching these statistical results with preset phase association rules. Subsequently, based on indicator ranking and radar chart readability constraints, the most representative subset of physical fitness indicators is selected as the visualization dimension, and the corresponding statistical parameters are recorded. This method can extract core physical fitness characteristics from complex multi-source data, ensuring radar chart readability and information integrity, and providing an intuitive, accurate, and reliable data foundation for training monitoring, physical fitness assessment, and scientific training regulation for university sports teams.

[0017] 4. This invention performs statistical parameter analysis on multi-source physical fitness data of athletes under the current training stage and content, and then adaptively adjusts the radar chart axis range. All axis configuration results and statistical constraints are encapsulated into a structured rendering parameter set, achieving automatic verification of data integrity, distribution rationality, and visualization presentation. Simultaneously, by comparing the statistical feature parameter sequence vectors across time points, changes in the athlete's physical fitness are identified and mapped to the 3D model rendering nodes, triggering real-time updates. This ensures that the 3D visualization accurately reflects the dynamics of physical fitness during training, guaranteeing the accuracy and consistency of multi-dimensional data display. It can intuitively present changes in the athlete's physical fitness, providing reliable data support and visual decision-making basis for physical fitness monitoring, training evaluation, and scientific regulation of university sports teams. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A schematic diagram of a multi-source data analysis and regulation method for physical fitness status in training for university sports teams, provided in an embodiment of this application;

[0020] Figure 2 A flowchart for multi-source physical fitness data processing and generation of first correlation candidate indicators provided in this application embodiment;

[0021] Figure 3 A flowchart of physical condition-driven 3D model update and consistency verification provided in this application embodiment;

[0022] Figure 4 A comparative diagram illustrating typical states of radar data provided in the embodiments of this application;

[0023] Figure 5 A radar chart for comprehensive evaluation of physical fitness for university sports teams, provided in an embodiment of this application;

[0024] Figure 6 A flowchart illustrating the physical condition-driven 3D model update and consistency verification process provided in this application embodiment. Detailed Implementation

[0025] The technical solution provided in this application will now be described with reference to the accompanying drawings.

[0026] To facilitate understanding of the embodiments of this application, the following points will be explained first:

[0027] In this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates an "or" relationship between the preceding and following related objects, but it does not exclude the possibility of indicating an "and" relationship; the specific meaning can be understood in context. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can mean: a, b, c; a and b; a and c; b and c; or a and b and c. Here, a, b, and c can be single or multiple.

[0028] In this application, the use of prefixes such as "first" and "second" is solely for the purpose of distinguishing different things belonging to the same category, and does not constrain the order, size, or quantity of things. For example, "first message" and "second message" are simply different messages, and there is no chronological, size, or priority relationship between them.

[0029] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0030] like Figure 1 The diagram shown is a schematic of a multi-source data analysis and regulation method for physical fitness status in college sports team training provided in an embodiment of this application. The method includes the following steps:

[0031] The collected multi-source physical fitness data were subjected to integrity verification and time synchronization processing to screen out the first relevant candidate indicators that characterize the athlete's physical fitness status.

[0032] Based on the candidate indicators of the athlete's primary correlation as the display dimensions of physical fitness status visualization, adaptive processing is performed on the axis scale of the display dimensions of physical fitness status visualization.

[0033] The statistical feature parameters of each physical fitness indicator in the final set of physical fitness indicators are subjected to axis-scale adaptive processing to generate an axis-scale adaptive configuration set of statistical feature parameters for the current stage.

[0034] The system acquires and analyzes the athlete's current motion parameters and physical condition data, triggers real-time rendering updates of the athlete's 3D model, and performs consistency checks on the display results of the athlete's 3D model.

[0035] like Figure 2 As shown, Figure 2 The flowchart for multi-source physical fitness data processing and generation of first correlation candidate indicators provided in this application embodiment describes a process where athletes' physical fitness data is acquired in real time via sensors, and preliminary data cleaning and time synchronization are performed to ensure data integrity. Next, the system performs statistical analysis on physical fitness indicators at different training stages, extracting key information such as mean and fluctuation range. Based on preset rules and training requirements, highly correlated indicators are selected and constructed as the display dimensions for the radar chart. Finally, after adaptive adjustment of the axis range, the rendering result is used for updating the 3D model and performing consistency verification to ensure the accuracy and rationality of the chart. The preliminary data cleaning and time synchronization include performing integrity verification and time synchronization processing on the collected multi-source physical fitness data. The specific process is as follows:

[0036] Acquire multi-source physical fitness data of athletes, construct a time series graph of multi-source physical fitness data of athletes, and count the instantaneous fluctuation amplitude of each physical fitness data in the time series graph of multi-source physical fitness data of athletes. At the same time, preset a time window, and count the difference between the maximum and minimum timestamps of each sensor data point at the same time in the time window, and record it as the timestamp deviation value of each sensor data point.

[0037] If the timestamp deviation of a certain sensor data point and the instantaneous fluctuation amplitude of each physical fitness data are both higher than or equal to the corresponding deviation threshold and instantaneous fluctuation amplitude, then the fluctuation of the athlete's multi-source physical fitness data within the time window is determined to be an unreliable abnormal fluctuation, and time alignment correction processing or resampling interpolation processing is performed on the data within the time window.

[0038] It should be noted that each type of physical fitness data will be generated into a separate time series plot based on timestamps. An athlete's multi-source physical fitness data may come from different sensors and devices; therefore, each data point (e.g., heart rate, acceleration, cadence, etc.) has its corresponding time series plot. Multi-source data time series plots typically use time as the x-axis and the corresponding physical fitness data values ​​as the y-axis, presenting the changing trends of each data source during training.

[0039] It should be noted that in the context of college sports training, if the timestamp deviation value is lower than the preset dynamic motion data deviation threshold or the timestamp deviation value is lower than the 100ms low-frequency physiological data deviation threshold, then time synchronization is considered acceptable. When the timestamp deviation value is higher than or equal to the time deviation threshold stored in the database, it will usually be reflected in the time series graph as different curves showing obvious misalignment at the same event point (such as the start of a sprint).

[0040] It should be noted that the specific steps for performing time alignment correction or resampling interpolation on the data within this time window are as follows: First, extract the data subsequences corresponding to each sensor within the time window, read the original timestamp information of each data point, and use the sensor data with the smallest timestamp deviation as the reference time series to construct a unified reference time axis; then, calculate the corresponding time correction amount based on the time offset between each non-reference sensor data point and the reference time axis, and perform translation correction on the original timestamps according to the time correction amount, so that the data of each sensor is initially aligned under the same reference time axis; after completing the time alignment, for data with inconsistent sampling intervals, ... For time points where data is missing, the data from each sensor is resampled based on a reference time axis. The physical fitness data values ​​at the missing time points are estimated using linear interpolation, spline interpolation, or interpolation based on the changing trends of adjacent time points, thus forming a multi-source physical fitness data sequence with consistent time intervals and aligned timestamps. After the resampling and interpolation are completed, the instantaneous fluctuation amplitude of the data before and after correction within the time window is compared and analyzed. When the corrected fluctuation amplitude falls back to within the preset instantaneous fluctuation amplitude threshold, the time alignment correction and resampling and interpolation are confirmed to be effective, and the corrected multi-source physical fitness data is used as input data for subsequent physical fitness status analysis and chart visualization generation.

[0041] like Figure 3 As shown, Figure 3This application provides a flowchart for a physical fitness state-driven 3D model update and consistency verification process. Based on training phase identifiers and motion data, the real-time rendering update of the 3D model is triggered. This process first acquires the athlete's current motion parameters and physical fitness state data, and compares them with the data from the previous moment to analyze changes in the athlete's training state. When the change reaches a preset threshold, the relevant rendering area is automatically updated, and consistency verification ensures the consistency between the 3D model and the actual training data, avoiding the impact of rendering errors or deviations on the training results. Ensuring the consistency between the 3D model and the actual training data through consistency verification includes...

[0042] The first relevant candidate indicators for characterizing an athlete's physical fitness status were selected through a screening process as follows:

[0043] After completing the time alignment and numerical correction of multi-source physical fitness data, the athlete's current training content and training phase identifier information are read and mapped to a set of training phase parameters.

[0044] Statistical analysis of the time series of various physical fitness indicators in multi-source physical fitness data of athletes within the corresponding time window yields the statistical results of each physical fitness indicator of the athletes, including the mean, standard deviation rate of change, percentage of effective data, and continuous effective duration of each physical fitness indicator during the training phase.

[0045] The statistical results of each physical fitness indicator of the athlete are matched with the preset stage association rules. When a certain physical fitness indicator matches at least one of the corresponding statistical results with the stage association rule during the training stage, it is determined as the first correlation candidate indicator, thereby obtaining the athlete's first correlation candidate indicators.

[0046] It should be noted that the specific process of mapping to the training phase parameter set is as follows: by reading the athlete's current training content and training phase identifier information recorded in the training plan library, the training type, intensity level, load mode and main physical fitness goals corresponding to the training content are structured and parsed. Based on the pre-established training phase-parameter mapping rule table, the training phase identifier is mapped to a training phase parameter set containing the target load range, the weight of key physical fitness indicators and the allowable fluctuation threshold. This provides a unified parameter constraint and semantic basis for the subsequent phased analysis of multi-source physical fitness data, anomaly detection and training effect evaluation.

[0047] It should be noted that when a certain physical fitness indicator matches at least one of the statistical results of various physical fitness indicators during the training phase, specifically, for a certain physical fitness indicator, within the time window of the corresponding training phase, its statistical result vector is extracted. This vector includes feature vectors such as the indicator's phase mean, standard deviation, peak value or kurtosis, instantaneous fluctuation amplitude, rate of change, and continuous effective duration. The feature vector is composed of the mean component, dispersion component, instantaneous fluctuation component, rate of change component, and continuous effective duration component of the physical fitness indicator within that phase, used to characterize the overall change characteristics of the physical fitness indicator within the current training phase. Subsequently, a vector-level matching operation is performed on the phase statistical feature vector and the phase association rule vector, converting the statistical results of each physical fitness indicator into vector form. These statistical feature vectors are then compared with the predefined association rule vector, that is, each dimension of the feature is compared one by one. For example, the mean component of the statistical feature is compared with the mean range in the association rule, the standard deviation component is compared with its corresponding standard deviation interval, and other features such as fluctuation amplitude and rate of change are also compared in a similar way. For each feature component, if its value falls within the set valid range, its matching status is marked as "valid," indicating that the change in the physical fitness index during this stage meets expectations. If a statistical feature exceeds the set valid domain boundary, its matching identifier is marked as "invalid." By determining whether the result of the mapped statistical feature component falls within the valid domain defined by the corresponding judgment parameter component, the matching identifier value of the statistical feature component is set to either a valid or invalid matching state. Following the original arrangement order of the statistical feature components in the stage statistical feature vector, the corresponding generated matching identifier values ​​are sequentially combined to form the matching result vector of the physical fitness index in the current training stage. When at least one statistical feature component in the matching result vector is determined to satisfy the constraint conditions of the corresponding judgment parameter component in the stage association rule vector, the physical fitness index is considered to have a valid association with the current training stage, and the physical fitness index is identified as the first correlation candidate index.Based on the current training content and stage, a subset of physical fitness indicators highly relevant to this stage is selected from the calibrated dataset. The number of dimensions in the radar chart is determined based on the aforementioned statistical analysis results to avoid visual congestion or information dilution due to excessive dimensions. Subsequently, representative values ​​(such as mean or weighted instantaneous fluctuation values) of each selected physical fitness indicator within the time window are mapped onto radar axes under a unified standardized scale. The maximum and minimum values ​​of the radar axes are automatically set according to the statistical distribution of each indicator, ensuring consistent visual weight across different dimensions. After completing the initial radar chart rendering, the system further correlates the radar chart results with the corresponding time series. The system performs cross-validation on the charts. When a certain dimension in the radar chart shows an abnormal surge or drop, it automatically backtracks to the corresponding time series data to check for interpolation errors, abnormal scale mapping, or unreasonable time window selection. After confirming data consistency, the system locks the radar chart display result. Finally, the system caches the validated charts and radar charts in a versioned manner and marks the corresponding data time windows and processing parameters. This ensures that the data visualization results of the same batch are reproducible and consistent in subsequent training analysis or comparative display, thereby supporting the continuous observation and comparative analysis of changes in physical fitness during the training of university sports teams.

[0048] It should be noted that when ranking the candidate indicators of primary relevance, the ranking operation is not performed directly on their stage statistical feature vectors. Instead, a vector-to-scalar mapping calculation is first performed on the stage statistical feature vectors to generate scalar evaluation values ​​for ranking. Specifically, for each candidate indicator of primary relevance, its stage statistical feature vector formed in the current training stage is read. According to the preset vector mapping rules in the database, the dimensions of the stage statistical feature vector formed in the current training stage are first standardized. Principal component analysis is then used to transform the stage statistical feature vector formed in the previous training stage, extracting the main change characteristics and stable characteristics of the physical fitness indicator in the current training stage. In this transformation process, principal component analysis extracts the principal components with the largest variance from the multi-dimensional statistical features and combines them into a new scalar value, which serves as the comprehensive ranking evaluation scalar for the physical fitness indicator in the current training stage. The ranking evaluation scalar serves as the ranking key value corresponding to the physical fitness indicator, reflecting the overall change intensity and distinguishability of the physical fitness indicator in the current training stage. After obtaining the ranking evaluation scalars corresponding to each primary relevance candidate indicator, the primary relevance candidate indicators are ranked based on these ranking evaluation scalars, forming an ordered sequence of candidate indicators arranged from high to low. During this ranking process, the stage statistical feature vectors themselves remain unchanged; the ranking evaluation scalars are only used as index keys associated with the corresponding physical fitness indicator identifiers to determine the relative position of the candidate indicators in the ordered sequence. After ranking, the stage statistical feature vectors of each candidate indicator are reorganized according to the ranking results, arranging them in the storage structure according to the ranking order, so that they can be directly called in subsequent radar chart dimension filtering and readability constraint judgment processes. Subsequently, when performing radar chart dimension filtering, the system does not re-rank the vectors; instead, based on the already ranked candidate indicator sequence, it sequentially reads the corresponding stage statistical feature vectors according to the ranking evaluation scalars from high to low and inputs them into the readability constraint judgment process. Throughout the entire process, the separation of roles—vectors for representing multidimensional statistical features and scalars for ranking decisions—is maintained. For example, in a certain training phase, the system has identified three primary relevance candidate physical fitness indicators, denoted as indicator A, indicator B, and indicator C. For each physical fitness indicator, statistical analysis is performed within the corresponding time window of the training phase to obtain its phase statistical feature vector. A vector-to-scalar mapping calculation is then performed on each phase statistical feature vector, mapping components of different dimensions to the 0-1 interval to obtain a ranking evaluation scalar. Using the ranking evaluation scalar S as the ranking key, the physical fitness indicators are ranked, resulting in a ranking of A > C > B. In this ranking result, the scalar S, not the phase statistical feature vector itself, is actually involved in the ranking, but each ranking position is bound to the corresponding physical fitness indicator and its phase statistical feature vector. After ranking, the results are used in the radar chart dimension filtering process.For example, if the maximum number of dimensions allowed to be displayed in a radar chart is 2, then indicators A and C are selected in order of sorting, and their corresponding stage statistical feature vectors are used as the input of the display dimensions of the radar chart; indicator B, which is ranked lower, is truncated and does not enter the radar chart display range because it exceeds the readability constraint.

[0049] The process of using the candidate indicators of primary relevance of athletes as the display dimensions for physical fitness status visualization is as follows: sort the candidate indicators of primary relevance from high to low, and truncate and filter the candidate indicators according to the readability constraints of the radar chart stored in the database to obtain the final set of displayed physical fitness indicators.

[0050] The final set of physical fitness indicators will be used as the display dimension of the radar chart, and the corresponding statistical parameters and filtering results will be recorded.

[0051] Adaptive processing is performed on the axis scale of the visualization dimension of physical fitness status. The specific process is as follows:

[0052] Read the statistical parameter sequence formed by the final set of physical fitness indicators in the current training phase, and calculate the maximum and minimum values ​​of the physical fitness indicators of each first correlation candidate indicator in this phase.

[0053] It should be noted that the final set of displayed physical fitness indicators, as formed within the current training phase, refers to a series of numerical characteristics obtained for each final displayed physical fitness indicator within the current training phase, based on multi-source physical fitness data and time windows. Specifically, it is a numerical vector sequence generated by statistical analysis of the raw data of each physical fitness indicator changing over time during the training phase. This sequence contains the key statistical characteristics of the physical fitness indicator at different time points or time periods, such as the mean, standard deviation, rate of change, continuous effective duration, and percentage of effective data within each time window. These characteristics are arranged in chronological order. The process of obtaining the statistical parameter sequence involves first extracting the time series data of each physical fitness indicator within the training phase, then statistically analyzing the data within a preset time window or sliding window, calculating parameters such as the mean, fluctuation amplitude, standard deviation, rate of change, and continuous effective duration within the window. Finally, the statistical parameters corresponding to each time window or time point are combined sequentially to obtain the statistical parameter sequence of that physical fitness indicator during the training phase.

[0054] like Figure 4 As shown, Figure 4The diagrams show typical radar data states. Based on statistical analysis and adaptive axis scale processing of multi-source physical fitness data, the final axis range and effective variation interval for each physical fitness indicator in the current training phase are determined. The normal radar diagram is a direct representation of this situation, where the statistical parameter sequence of each physical fitness indicator has sufficient coverage and a reasonable distribution within its adaptive axis range, corresponding to the case where the structured rendering parameter consistency check passes. The radar diagram with insufficient coverage corresponds to the case where, during the consistency check, the coverage ratio of the statistical parameter values ​​of the physical fitness indicator within the final axis range is lower than the preset threshold. Although the axis range has been adaptively configured, the actual data only occupies a small portion of it. The segment triggered an abnormal display dimension flag; the radar diagram with an excessively narrow range of change directly corresponds to the judgment results of the interquartile range and the effective range of change percentage in the steps, that is, the fluctuation range of the physical fitness index in the current training stage is significantly smaller than the axis span, resulting in insufficient effective range of change percentage, which needs to be corrected by compressing, shifting or reconstructing the axis range to correct the display effect. The percentage in the figure is used to quantify the coverage or numerical range of the data in each dimension of the radar diagram. 0-100% in the normal radar diagram is the normal coverage percentage range of each dimension index. The percentage in the diagram with insufficient coverage is the actual coverage percentage range of the index in this state. The percentage in the diagram with an excessively narrow range of change is the percentage scale corresponding to the index value.

[0055] Based on the maximum and minimum values ​​of the physical fitness indicators of each first-relevance candidate indicator in this stage, the initial axis range of the corresponding physical fitness indicator is constructed, and the interquartile range and concentration of the final displayed physical fitness indicator set in the current training stage are statistically analyzed. The axis range is adaptively adjusted according to the interquartile range and concentration of the final displayed physical fitness indicator set in the current training stage, and all axis configuration results and statistical constraint results formed in this stage are uniformly encapsulated into a structured rendering parameter set of the radar chart.

[0056] Figure 5 As shown, Figure 5 This radar chart is designed to comprehensively evaluate the physical fitness of athletes training for university sports teams. It is derived by adaptively adjusting the axis range based on the interquartile range and concentration of the final set of physical fitness indicators in the current training phase. The chart displays the distribution of athletes' comprehensive physical fitness status based on multi-source physical fitness data in the current training phase. By comparing the polygons of the current phase with those of the previous phase, the changing trends and differences of each physical fitness indicator can be seen intuitively, which helps to analyze the relative improvement, fluctuations, and phased advantages or disadvantages of athletes' physical fitness indicators during training.

[0057] It should be noted that the specific process of adaptively adjusting the axis range based on the interquartile range and concentration of the final set of physical fitness indicators in the current training phase is as follows: When the proportion of the interquartile range of the physical fitness indicator subset in the current training phase relative to the initial axis range is lower than a preset concentration threshold, it indicates that the value fluctuation range of the indicator in this phase is small. The proportion of the interquartile range of the physical fitness indicator subset in the current training phase relative to the initial axis range refers to the percentage of the initial axis range occupied by the interquartile range of the physical fitness indicator (i.e., the difference between the upper and lower quartiles) in this training phase. It can be determined that the indicator has a distribution offset and insufficient utilization of axis space within the current axis range. When the proportion of the interquartile range of the fitness indicator subset within the current training phase relative to the initial axis range is higher than or equal to a pre-set concentration threshold, the system first uses the statistical parameter sequence formed by the indicator within the current training phase as an index condition to retrieve matching reference records from the axis range adjustment database composed of historical training data and visualization parameters. The reference records at least include the axis range adjustment offset and axis span correction amounts used under similar training phases, similar indicator fluctuation characteristics, and similar distribution offset directions. The reference records in the database are stored hierarchically according to indicator type, training phase category, interquartile range proportion interval, and concentration interval. When multiple records that meet the conditions are retrieved, the system performs weighted filtering based on the similarity between the current indicator's distribution concentration position and the corresponding distribution position in the historical records, thereby obtaining the target adjustment value for this axis range adjustment. After obtaining the target adjustment value, the system performs directional adjustment on the original axis range based on the relative position relationship of the main distribution interval of the indicator within the current axis range. When the concentrated distribution interval is located within a preset edge interval near the lower limit of the current axis range, the system expands or shifts the lower limit of the original axis range downwards based on the corresponding lower limit offset adjustment value in the database, while keeping the upper limit unchanged or making a small synchronous adjustment proportionally, so that the central area of ​​the adjusted axis range covers the main distribution interval of the indicator. When the concentrated distribution interval is located within a preset edge interval near the upper limit of the current axis range, the system expands or shifts the upper limit of the original axis range upwards based on the corresponding upper limit offset adjustment value in the database, while keeping the lower limit unchanged or performing a small symmetrical linkage adjustment, thereby avoiding the long-term accumulation of effective data at the edge of the axis range. The target adjustment value is superimposed as a relative correction amount to the boundary of the original axis range, so that the adjusted axis range can actively move towards the direction of data concentration while retaining the extreme value information of the original data. If the adjustment value given in the database also includes an axis span correction parameter, after completing the directional translation of the axis range boundary, the system further compresses or widens the overall axis span based on the correction parameter, so that the final axis range neither truncates the original data distribution nor reduces the proportion of the middle effective interval in visualization or analysis.By obtaining adjustment values ​​from database tables and performing directional translation and scale correction on the original axis ranges, an adaptive redistribution of the physical fitness indicator axis space is achieved, ensuring that the axis ranges remain consistent with the indicator distribution characteristics within the current training phase. After completing the adaptive calculation and correction of the axis ranges corresponding to each display dimension, the axis range parameters and the statistical parameters of the corresponding physical fitness indicators are used together as input conditions for radar chart rendering. The axis range parameters are fixed during the rendering process, ensuring that the radar chart maintains axis scale consistency across multiple refreshes within the same training phase.

[0058] It should be noted that all axis configuration results formed at this stage refer to the final set of coordinate axis display parameters determined for each fitness indicator selected for display in the radar chart after completing statistical analysis and adaptive adjustment of the final set of fitness indicators. Specifically, this axis configuration result includes: for each fitness indicator, the final lower and upper axis values ​​calculated based on its statistical parameter sequence during the current training phase; and the corresponding axis direction sequence number and normalized mapping interval in the radar chart for that fitness indicator, used to ensure consistent numerical mapping rules for different fitness indicators in a unified radar chart space. The statistical constraint result refers to a set of judgment parameters and status identifiers derived from statistical analysis during the adaptive adjustment of the axis range, which have a constraining effect on subsequent chart rendering. These statistical constraints are not directly used as coordinate axis values, but rather to limit the data mapping and rendering behavior during chart generation. They include: a concentration state indicator corresponding to the statistical parameter distribution characteristics of each fitness indicator during the current training phase, indicating whether the data for that indicator is significantly concentrated on one side of the axis range; an interval determination result corresponding to the proportion of the interquartile range to the axis range, indicating whether the indicator belongs to a low-fluctuation, narrow-distribution, or highly discrete state; and an axis adjustment method indicator generated based on the above determination results, such as whether to enable axis range shrinkage, whether to enable unilateral offset, and whether to enable proportional compensation. These indicators will be continuously referenced as constraints in subsequent rendering processes. The axis configuration results and statistical constraint results are not stored separately or calculated temporarily, but are uniformly encapsulated using fitness indicators as basic units to form a structured rendering parameter set for the radar chart. In this structured rendering parameter set, each physical fitness indicator corresponds to an independent rendering parameter record. This record contains the final axis range parameters of the indicator and the associated statistical constraint status information. This allows the subsequent chart generation module to accurately restore the axis configuration status and rendering constraints of each physical fitness indicator in the current training stage simply by reading this structured rendering parameter set without re-performing statistical analysis.

[0059] Adaptive processing is performed on the axis scale of the display dimension for visualizing physical condition, including:

[0060] Read the structured rendering parameter set of the athlete's statistical chart, and extract the sequence length of the statistical parameter of each physical fitness indicator, the coverage ratio of the statistical parameter value of each physical fitness indicator in its corresponding final axis range, and the percentage of the effective change range of each physical fitness indicator in its final axis range.

[0061] Athlete statistical charts refer to data structures generated through statistical analysis and processing for visualization. These data structures contain information such as the statistical parameter sequences, display dimensions, and axis ranges for various athletic performance indicators during the current training phase, supporting visualizations like radar charts.

[0062] It's important to note that the statistical parameter sequence length for each fitness indicator refers to the length of the time-series data associated with each fitness indicator; that is, the number of time points during which that fitness indicator is recorded or collected within the current training phase. Specifically, each fitness indicator (such as heart rate, running speed, acceleration, etc.) changes over a period of time as training progresses, and these changes are recorded using time-series data. Each time series represents the change process of a fitness indicator over a period of time. The statistical parameter sequence length represents the number of timestamps or time windows used to record the data for that fitness indicator.

[0063] If the length of any physical fitness indicator statistical parameter sequence is less than the number of dimension nodes configured in the statistical chart stored in the database, the coverage ratio of a physical fitness indicator statistical parameter value in its corresponding final axis range is less than the preset coverage ratio threshold in the final axis range, and the proportion of effective change intervals of each physical fitness indicator in its final axis range is less than the effective change interval proportion threshold in the final axis range stored in the database, then the display dimension is marked as an abnormal state in the structured rendering parameter set, and the abnormal handling process is triggered.

[0064] It's important to note that the number of dimension nodes refers to the number of data points that should be displayed for each dimension (i.e., each fitness metric) when creating a chart (such as a radar chart). This number is determined by the preset chart configuration, which specifies how many data points are needed on each dimension to support the chart's visualization. Simply put, this number of dimension nodes is a standard in the chart structure, representing the minimum number of data points a fitness metric should have to ensure the chart's completeness and accuracy. Coverage ratio refers to the proportion of actual data covered by the statistical parameter values ​​of the fitness metric. Specifically, the value range of the statistical parameters of a fitness metric (such as heart rate, speed, etc.) is mapped onto the chart's axis, and the coverage ratio is the distribution of actual data within that axis range. For example, if a fitness metric ranges from 0 to 100, and the actual data points are only distributed within the range of 0 to 50, then its coverage ratio is low, indicating that the entire axis space is not being fully utilized. The effective variation range percentage refers to the proportion of the fitness metric's variation range within its axis range over a given time period. The effective range of change refers to the actual range of changes in physical fitness indicators within the statistical period, such as the range of heart rate changes from 60 to 100.

[0065] It should be noted that the structured rendering parameter set for athlete statistical charts includes at least: the data vector length corresponding to each display dimension, the upper and lower limits of the axis range for each dimension, the axis scale resolution, the normalized value range for each dimension, the dimension order index, and the effective identification ratio of statistical parameters used for rendering.

[0066] It should be noted that each statistical chart's display dimension corresponds to a sequence of statistical parameters for a physical fitness indicator in the current training phase. The length of the statistical parameter sequence, the coverage ratio of statistical parameters within the axis range, and the percentage of effective variation intervals for the display dimension are all calculated based on the statistical parameter sequence of that physical fitness indicator. Therefore, when the length of the statistical parameter sequence corresponding to that physical fitness indicator cannot cover the number of dimension nodes configured in the radar chart, it indicates that there is insufficient data integrity for that display dimension.

[0067] It should be noted that the axis-scale adaptive configuration set uses physical fitness indicators as index units, and records the final upper and lower limits of each physical fitness indicator, the axis-scale adjustment status, and the corresponding numerical mapping constraint information determined based on its statistical feature parameters during the current training phase.

[0068] It should be noted that the exception handling process includes issuing warnings, recording the exception type, and generating suggestions based on preset repair strategies in the database to ensure that subsequent chart rendering correctly reflects the statistical characteristics of the physical fitness indicator. Based on this, the system determines whether there are potential rendering anomalies by performing consistency checks on these parameters. For example, if the length of the statistical parameter sequence for a certain display dimension is inconsistent with the number of dimension nodes configured in the radar chart, or if a dimension is selected as a display dimension in the current stage but the percentage of valid data for its corresponding statistical parameters is lower than a preset threshold, the "data integrity parameter" can be used to determine that this dimension has a risk of missing data in the chart structure, thus identifying it as a structural inconsistency error.

[0069] like Figure 6 As shown, Figure 6 The flowchart for updating and verifying the consistency of the 3D model driven by physical fitness status provided in this application embodiment determines whether there are data inconsistencies by checking key indicators such as the length of the statistical parameter sequence and the coverage ratio of the axis range for each dimension. When data anomalies are detected, such as statistical results not meeting preset coverage ratio or variation range requirements, the system marks that dimension as an abnormal state and triggers the anomaly handling process. Simultaneously, by comparing the data with the athlete's current physical fitness status data, the system continuously optimizes the data rendering during training to ensure that the displayed results match the actual training situation, thereby providing coaches with accurate training feedback information. The specific process of acquiring and analyzing the athlete's current movement parameters and physical fitness status data is as follows:

[0070] Extract the athlete's current training content and training stage identifier information, preset several monitoring times, and obtain the statistical feature parameter sequence vector generated by the athlete under the current training stage and training content at each monitoring time. Compare the statistical feature parameter sequence vector generated by the athlete under the current training stage and training content with the statistical feature parameter sequence vector generated by the athlete under the current training stage and training content at the previous time point element by element to obtain the average rate of change of the components of the statistical feature parameter sequence vector generated by the athlete under the current training stage and training content.

[0071] The process of triggering real-time rendering updates of the athlete's 3D model is as follows: The average rate of change of the statistical feature parameter sequence vector components generated by the athlete under the current training stage and training content is compared with the fluctuation allowable threshold stored in the database. If the average rate of change of the statistical feature parameter sequence vector components generated by the athlete under the current training stage and training content is higher than or equal to the fluctuation allowable threshold stored in the database, the statistical feature parameter sequence vector generated by the athlete under the current training stage and training content is mapped to the rendering node set of the 3D model to obtain the updated rendering area of ​​the athlete's 3D model under the current training stage, and the real-time rendering update of the 3D model is automatically triggered in the updated rendering area of ​​the athlete's 3D model under the current training stage.

[0072] The consistency verification of the displayed 3D models of athletes is performed as follows:

[0073] The system reads the rendered values ​​of the three-dimensional models of each athlete's physical fitness indicators within the same time window and calculates the difference between them and the actual values ​​to obtain the error value of the rendered values ​​of the three-dimensional models of each athlete's physical fitness indicators. It then determines whether the rendered values ​​of each athlete's physical fitness indicators are within the allowable error range.

[0074] It's important to note that the rendered values ​​of the 3D model are visualizations generated from the athlete's physical fitness data and motion parameters. They reflect a dynamic simulation of the athlete's condition. The simulation results are calculated and graphically generated based on the input physical fitness data. However, these simulations may be affected by certain computational assumptions and estimation errors, especially when dealing with complex motion data. For example, the athlete's trajectory, the movement of different body parts, and the influence of the external environment can all introduce biases. On the other hand, actual values ​​usually come from direct measurement data, such as physiological data recorded by sensors or motion capture devices. This actual data directly reflects the athlete's real-time performance and is generally more accurate, but it can also be affected by equipment errors, sensor accuracy, and instantaneous changes in the athlete's movements. Therefore, the error between the rendered values ​​of the 3D model and the actual measured values ​​reflects the deviations that occur when dynamically representing the athlete's physical fitness and movements. This error is obtained by comparing the two.

[0075] The specific process for determining whether the rendered values ​​of an athlete's various physical fitness indicators are within the allowable error range is as follows:

[0076] The error values ​​of the 3D model renderings of various physical fitness indicators of the athlete are compared with the allowable deviation range stored in the configuration file. The specific process is as follows:

[0077] If the error value of the 3D model rendering value of a certain physical fitness indicator of an athlete exceeds the allowable deviation range stored in the configuration file, the physical fitness indicator will be marked as abnormal, and rendering correction and data abnormality warnings will be triggered. Otherwise, the physical fitness indicator will not be marked as abnormal.

[0078] It's important to note that if the error value exceeds the deviation range stored in the configuration file, the system will automatically identify and mark the physical fitness indicator as abnormal. Simultaneously, the system will initiate a rendering correction process. This process involves recalculating the rendering data, adjusting rendering algorithm parameters, refitting the model, or applying error compensation techniques to correct rendering deviations in the 3D model. The corrected rendering data will be fed back into the system, updating relevant charts and the 3D model to ensure accuracy. Furthermore, a data anomaly warning will be triggered, notifying relevant personnel (such as coaches or data analysts) that the physical fitness indicator has a deviation. The warning will include the physical fitness indicator exceeding the deviation range, the specific error value, and the potential impact range, helping coaches or analysts take timely action, such as adjusting the training plan or re-collecting data. The anomaly event will also be recorded in a log for future analysis or model optimization.

[0079] The various features and processes described above can be used independently of each other or can be combined in various ways. All possible combinations and sub-combinations are intended to fall within the scope of this disclosure. Furthermore, certain method or process blocks may be omitted in some embodiments. The methods and processes described herein are not limited to any particular order, and the blocks or states associated with them may be performed in other suitable orders. For example, the described blocks or states may be performed in an order different from the order specifically disclosed, or multiple blocks or states may be combined in a single block or state. Example blocks or states may be performed serially, in parallel, or in some other manner. Blocks or states may be added to or removed from the disclosed example embodiments. The exemplary cases and components described herein may be configured differently from those described. For example, elements may be added to, removed from, or rearranged compared to the disclosed example embodiments.

[0080] The various operations of the example methods described herein can be performed at least in part by an algorithm. This algorithm can be contained in program code or instructions stored in memory (e.g., the aforementioned non-transitory computer-readable storage medium). Such an algorithm may include a machine learning algorithm. In some embodiments, the machine learning algorithm may not be explicitly programmed into the computer to perform the function, but can learn from training data to create a predictive model that performs the function.

[0081] The various operations of the example methods described herein can be performed, at least in part, by one or more processors that are temporarily configured (e.g., by software) or permanently configured to perform the relevant operations. Whether temporarily or permanently configured, such processors can constitute the engine of a processor implementation that operates to perform one or more of the operations or functions described herein.

[0082] Similarly, the methods described herein can be implemented at least in part by a processor, where one or more specific processors are examples of hardware. For example, at least some operations of a method can be performed by one or more processors or an engine implemented by a processor. Furthermore, one or more processors can also be operated to support the performance of related operations in a “cloud computing” environment or as “Software as a Service” (SaaS). For example, at least some operations can be performed by a set of computers (as an example of a machine including processors), where these operations are accessible via a network (e.g., the Internet) and via one or more suitable interfaces (e.g., application programming interfaces (APIs)).

[0083] The performance of certain operations can be distributed across processors, residing not only within a single machine but also deployed across multiple machines. In some example embodiments, the processor or processor-implemented engine may reside in a single geographic location (e.g., within a home environment, office environment, or server cluster). In other example embodiments, the processor or processor-implemented engine may be distributed across multiple geographic locations.

[0084] In this specification, multiple instances may implement components, operations, or structures described as single instances. Although individual operations of one or more methods are shown and described as separate operations, one or more of the separate operations may be performed simultaneously and do not need to be performed in the order shown. Structures and functions presented as separate components in the example configuration may be implemented as composite structures or components. Similarly, structures and functions presented as single components may be implemented as separate components. These and other variations, modifications, additions, and improvements fall within the scope of this document.

[0085] While an overview of the subject matter has been described with reference to specific example embodiments, various modifications and changes can be made to these embodiments without departing from the broader scope of embodiments of this disclosure. Such embodiments of the subject matter are referred to herein, individually or collectively, by the term "invention," and are used for convenience only and are not intended to limit the scope of this application to any single disclosure or concept, should more than one disclosure or concept be disclosed in fact.

[0086] The embodiments described herein have been described in sufficient detail to enable those skilled in the art to practice the disclosed teachings. Other embodiments may be used and derived therefrom, such that structural and logical substitutions and changes may be made without departing from the scope of this disclosure. Therefore, the detailed description should not be construed as limiting, and the scope of the various embodiments is defined only by the appended claims and the full scope of their equivalents.

Claims

1. A method for multi-source data analysis and regulation of physical fitness status for university sports teams, characterized in that, Includes the following steps: The collected multi-source physical fitness data were subjected to integrity verification and time synchronization processing to screen out the first relevant candidate indicators characterizing the athlete's physical fitness status. Based on the candidate indicators of the athlete's first correlation as the display dimension of physical fitness status, adaptive processing is performed on the axis scale of the display dimension of physical fitness status visualization. The statistical feature parameters corresponding to the current stage of each physical fitness indicator in the final set of physical fitness indicators are subjected to axis-scale adaptive processing to generate an axis-scale adaptive configuration set of statistical feature parameters for the current stage. The system acquires and analyzes the athlete's current motion parameters and physical condition data, triggers real-time rendering updates of the athlete's 3D model, and performs consistency checks on the display results of the athlete's 3D model.

2. The method for multi-source data analysis and regulation of physical fitness status for college sports teams as described in claim 1, characterized in that: The process of performing integrity verification and time synchronization on the collected multi-source physical fitness data is as follows: Acquire multi-source physical fitness data of athletes, construct a time series graph of multi-source physical fitness data of athletes, count the instantaneous fluctuation amplitude of each physical fitness data in the time series graph of multi-source physical fitness data of athletes, and at the same time, preset a time window, and count the difference between the maximum and minimum timestamp of each sensor data point at the same time in the time window, which is recorded as the timestamp deviation value of each sensor data point. If the timestamp deviation of a certain sensor data point and the instantaneous fluctuation amplitude of each physical fitness data are both higher than or equal to the corresponding deviation threshold and instantaneous fluctuation amplitude, then the fluctuation of the athlete's multi-source physical fitness data within the time window is determined to be an unreliable abnormal fluctuation, and time alignment correction processing or resampling interpolation processing is performed on the data within the time window.

3. The method for multi-source data analysis and regulation of physical fitness status for college sports teams as described in claim 1, characterized in that: The screening process yields the first relevant candidate indicators characterizing the athlete's physical condition. The specific process is as follows: After completing the time alignment and numerical correction of multi-source physical fitness data, the athlete's current training content and training phase identifier information are read and mapped to a set of training phase parameters. Statistical analysis of the time series of various physical fitness indicators in multi-source physical fitness data of athletes within the corresponding time window yields the statistical results of each physical fitness indicator, including the mean, standard deviation rate of change, percentage of valid data, and continuous valid duration of each physical fitness indicator during the training phase. The statistical results of each physical fitness indicator of the athlete are matched with the preset stage association rules. When at least one of the statistical results of a certain physical fitness indicator in the training stage matches the stage association rule, it is determined as the first correlation candidate indicator, thereby obtaining the first correlation candidate indicators of the athlete.

4. The method for multi-source data analysis and regulation of physical fitness status for college sports teams as described in claim 1, characterized in that: The specific process of using the athlete's primary correlation candidate indicators as the display dimensions for visualizing physical fitness is as follows: The candidate indicators with the highest relevance are sorted, and the candidate indicators are truncated and filtered according to the readability constraints of the radar chart stored in the database to obtain the final set of physical fitness indicators. The final set of physical fitness indicators will be used as the display dimension of the radar chart, and the corresponding statistical parameters and filtering results will be recorded.

5. The method for multi-source data analysis and regulation of physical fitness status for college sports teams as described in claim 1, characterized in that: The axis scale of the visualization dimension for physical fitness is adaptively processed, specifically as follows: Read the statistical parameter sequence formed by the final set of physical fitness indicators in the current training phase, and calculate the maximum and minimum values ​​of the physical fitness indicators of each first-correlation candidate indicator in this phase. Based on the maximum and minimum values ​​of the physical fitness indicators of each first-relevance candidate indicator in this stage, the initial axis range of the corresponding physical fitness indicator is constructed, and the interquartile range and concentration of the final displayed physical fitness indicator set in the current training stage are statistically analyzed. The axis range is adaptively adjusted according to the interquartile range and concentration of the final displayed physical fitness indicator set in the current training stage, and all axis configuration results and statistical constraint results formed in this stage are uniformly encapsulated into a structured rendering parameter set of the radar chart.

6. The method for multi-source data analysis and regulation of physical fitness status for college sports teams as described in claim 1, characterized in that: The adaptive processing of the axis scale of the display dimension for visualizing physical fitness also includes: Read the structured rendering parameter set of the athlete's statistical chart, and extract the sequence length of the statistical parameter of each physical fitness indicator, the coverage ratio of the statistical parameter value of each physical fitness indicator in its corresponding final axis range, and the proportion of the effective change range of each physical fitness indicator in its final axis range. If the length of any physical fitness indicator statistical parameter sequence is less than the number of dimension nodes configured in the statistical chart stored in the database, the coverage ratio of the statistical parameter value in its corresponding final axis range is less than the preset coverage ratio threshold in the final axis range, and the proportion of the effective change range of the physical fitness indicator in its final axis range is less than the effective change range proportion threshold in the final axis range stored in the database, then the display dimension is marked as an abnormal state in the structured rendering parameter set, and the abnormal handling process is triggered.

7. The method for multi-source data analysis and regulation of physical fitness status for college sports teams as described in claim 1, characterized in that: The specific process of acquiring and analyzing the athlete's current movement parameters and physical condition data is as follows: Extract the athlete's current training content and training stage identifier information, preset several monitoring times, and obtain the statistical feature parameter sequence vector generated by the athlete under the current training stage and training content at each monitoring time. Compare the statistical feature parameter sequence vector generated by the athlete under the current training stage and training content with the statistical feature parameter sequence vector generated by the athlete under the current training stage and training content at the previous time point element by element to obtain the average rate of change of the components of the statistical feature parameter sequence vector generated by the athlete under the current training stage and training content.

8. The method for multi-source data analysis and control of physical fitness status for college sports team training as described in claim 7, characterized in that: The specific process for triggering the real-time rendering update of the athlete's 3D model is as follows: The average rate of change of the statistical feature parameter sequence vector components generated by the athlete under the current training stage and training content is compared with the fluctuation allowable threshold stored in the database. If the average rate of change of the statistical feature parameter sequence vector components generated by the athlete under the current training stage and training content is higher than or equal to the fluctuation allowable threshold stored in the database, the statistical feature parameter sequence vector generated by the athlete under the current training stage and training content is mapped to the set of rendering nodes of the 3D model to obtain the updated rendering area of ​​the 3D model of the athlete under the current training stage, and the real-time rendering update of the 3D model is automatically triggered in the updated rendering area of ​​the 3D model of the athlete under the current training stage.

9. The method for multi-source data analysis and regulation of physical fitness status for college sports teams as described in claim 1, characterized in that: The consistency verification process for the display results of the athlete's 3D model is as follows: The system reads the rendered values ​​of the three-dimensional models of each athlete's physical fitness indicators within the same time window and calculates the difference between them and the actual values ​​to obtain the error value of the rendered values ​​of the three-dimensional models of each athlete's physical fitness indicators. It then determines whether the rendered values ​​of each athlete's physical fitness indicators are within the allowable error range.

10. The method for multi-source data analysis and regulation of physical fitness status for college sports teams as described in claim 9, characterized in that: The specific process for determining whether the rendered values ​​of each physical fitness indicator of the athlete are within the allowable error range is as follows: The error values ​​of the 3D model rendering values ​​of the athlete's various physical fitness indicators are compared with the allowable deviation range stored in the configuration file. If the error value of the 3D model rendering value of a certain physical fitness indicator of an athlete exceeds the allowable deviation range stored in the configuration file, the physical fitness indicator will be marked as abnormal, and rendering correction and data abnormality warnings will be triggered.

Images