Data analysis-based teenager physical training method and system

Abstract

The invention relates to the technical field of physical health data analysis, in particular to a teenager physical training method and system based on data analysis, and the method comprises the following steps: dividing a load signal into a low-frequency band, a bandwidth frequency band and a high-frequency band based on the load intensity data in teenager training, analyzing the long-term change trend and fluctuation range of the load of the low-frequency band, and calculating the long-term change trend of the load of the high-frequency band; stability parameters are calculated in combination with bandwidth frequency band signal fluctuation, the high-frequency-band short-time load change rate and change rate distribution are analyzed, and a frequency band distribution parameter set is obtained. According to the invention, through load signal multi-band decomposition and parameterization processing, independent calculation and comprehensive analysis of load characteristics are realized, the dynamic change capture capability is enhanced, and through association of cumulative load and time distribution characteristics, dynamic interaction data support is provided, and load risk points and distribution areas are accurately identified. The intensity, time and frequency adjustment strategies are optimized, load balance and action coordination are achieved, overload and deviation risks are reduced, and the scientificity and controllability of teenager training are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of sports health data analysis, and in particular to a method and system for physical training of teenagers based on data analysis. Background Art

[0002] The field of sports and health data analysis technology includes technical methods for studying the physical functions, physical performance and related health indicators of individuals or groups during exercise by collecting, processing, analyzing and modeling various data related to sports and health. The core content of this technical field includes data acquisition, organization, analysis and data-based optimization decision-making and application. The field mainly covers data collection technology and data processing algorithms related to disciplines such as exercise physiology and sports mechanics, which are used to analyze the physical load, exercise intensity, physical fitness level and training effect during exercise. The field of sports and health data analysis also combines sensing technology, data modeling and artificial intelligence technology to provide reliable technical support for the formulation and optimization of sports training methods, and is the basis of physical monitoring and scientific training.

[0003] Among them, the physical training method for teenagers refers to a training method developed through data collection and analysis technology based on the physical characteristics and physical fitness characteristics of teenagers. The technical matters targeted by the patent subject cover the collection of related physical data of teenagers during training, including indicators such as heart rate, speed, strength, sensitivity, etc. Specifically, data during exercise is collected through wearable sensing equipment, physical fitness monitoring devices, etc., and the data is processed and analyzed in real time to realize the monitoring of the physical fitness status of teenagers. Based on the analysis results, the method further formulates scientific and personalized physical training plans, including training project selection, training load setting and training intensity grading, to complete the scientific management and improvement of physical training for teenagers.

[0004] Existing technologies in youth physical training methods mostly rely on direct data collection and static analysis, and have weak capabilities in capturing dynamic changes in load during training. They find it difficult to independently process the characteristics of different load frequency bands, resulting in the difficulty in timely identifying the risks of short-term load overload in high-frequency bands and long-term load accumulation in low-frequency bands. Existing technologies mostly analyze the distribution of load and training time as a whole, ignoring the analysis of load fluctuations and interval characteristics within the bandwidth frequency band. This rough processing method leads to uneven distribution of training intensity and increases the risk of fatigue accumulation. Existing technologies pay insufficient attention to the coordination of youth movements and deviation correction, and fail to effectively analyze the relationship between movement load distribution and deviation data, resulting in weak targeted movement correction, further amplification of movement deviations during training, and increased potential risks of sports injuries, limiting the existing technologies' ability to finely manage dynamic load and movement optimization during youth physical training. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the existing technology and propose a physical training method and system for teenagers based on data analysis.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a physical training method for teenagers based on data analysis, comprising the following steps:

[0007] S1: Based on the load intensity data of adolescent training, the load signal is divided into low-frequency band, broadband band and high-frequency band. The long-term change trend and fluctuation range of the low-frequency band load are analyzed. The stability parameters are calculated based on the signal fluctuation of the broadband band. The short-term load change rate and change rate distribution of the high-frequency band are analyzed to obtain the frequency band distribution parameter set.

[0008] S2: Based on the frequency band distribution parameter set and combined with the low-frequency band load trend value, the relationship between training time and cumulative load is analyzed, the bandwidth frequency band load stability and time interval characteristics are extracted to obtain the dynamic load interaction coefficient, the instantaneous overload frequency and fluctuation amplitude are identified, and the dynamic characteristics between differentiated loads are integrated and analyzed to obtain the training load risk point dataset;

[0009] S3: Based on the training load risk point dataset, fatigue accumulation areas in the low-frequency band are identified, the intensity fluctuation range and training time distribution balance of the bandwidth frequency band are analyzed, the overload concentrated distribution area in the high-frequency band is extracted, and the high-frequency overload risk range is obtained. The intensity, frequency, and time adjustment of the training load are optimized to obtain the training load optimization adjustment result;

[0010] S4: Based on the training load optimization adjustment results, adjust the intensity and frequency of the training action load, identify the training interval time, analyze the movement coordination abnormalities in combination with the movement stability parameters of the bandwidth frequency band, extract the movement deviation data and calculate the correction amplitude, and generate a youth physical training adjustment execution plan.

[0011] As a further solution of the present invention, the step of acquiring the frequency band distribution parameter set is specifically as follows:

[0012] S111: Based on the load intensity data from youth training, perform spectrum analysis to decompose the load signal into low-frequency band, broadband band, and high-frequency band. Extract the energy components and time series characteristics of the frequency band signals, perform denoising, analyze the energy distribution ratio and mean fluctuation range of the frequency bands, and generate a load intensity frequency division feature parameter set.

[0013] S112: Based on the load intensity frequency division characteristic parameter set, statistical analysis is performed on the time series characteristic values of the low-frequency load signal to identify the linear fitting parameters and fluctuation range of the long-term change trend, using the formula:

[0014]

[0015] Calculate the trend distribution parameters of low-frequency signals and generate the long-term change trend and fluctuation range of low-frequency bands;

[0016] Among them, P l Represents the low-frequency signal trend distribution parameter, v i Represents the time series value of the low-frequency load signal, w i represents the weight factor of the time point, α is the fluctuation adjustment coefficient, β is the smoothing weight adjustment parameter, and n is the total number of sample points;

[0017] S113: Based on the long-term change trend and fluctuation range of the low-frequency band, combined with the fluctuation characteristic parameters and stability distribution of the broadband frequency band signal, a differential operation of the short-term load change rate of the high-frequency band signal is performed to obtain the distribution density and change rate distribution range of the high-frequency band load change rate, identify the deviation index of the frequency band signal amplitude, and generate a frequency band distribution parameter set.

[0018] As a further solution of the present invention, the step of obtaining the dynamic load interaction coefficient is specifically as follows:

[0019] S211: Based on the frequency band distribution parameter set, extract the load value and time information of each time period from the training data, group the time period load values of the low frequency band data by time node, identify the load accumulation value of the time group, filter the time nodes with load change fluctuations, and obtain the low frequency band bandwidth load trend parameter set;

[0020] S212: Based on the low-band bandwidth load trend parameter set, and referring to the low-band load stability and time interval characteristics, the load stability and time interval characteristics are combined, and the formula is used:

[0021]

[0022] The influence of time interval correction on load stability is introduced into the adjustment coefficient, and the dynamic load interaction coefficient is calculated;

[0023] Among them, C t represents the dynamic load interaction coefficient, d j Represents the load change within the time interval, T j Represents the corresponding time interval, M1 represents the load stability weight coefficient, L j Represents the associated parameter of load stability, S j represents the characteristic value of the time interval, and k represents the total number of time periods.

[0024] As a further solution of the present invention, the steps for obtaining the training load risk point data set are specifically as follows:

[0025] S221: Based on the dynamic load interaction coefficient, decompose the dynamic numerical data by time period, extract the instantaneous load data at each time point, filter the values exceeding the threshold, perform partition marking and classification, classify the load intervals, and obtain the instantaneous overload load interval data set;

[0026] S222: Based on the instantaneous overload load interval data set, calculate the difference between the peak and the valley of the load value within the interval, record the quantitative data of the fluctuation amplitude, match the time series, analyze the starting point and end point of the load change, perform time marking, and obtain a load fluctuation characteristic distribution data table;

[0027] S223: Based on the load fluctuation characteristic distribution data table, filter the intervals where the fluctuation amplitude exceeds the set range, perform cross-analysis in combination with the time series and overload load data, classify and organize the risk intervals according to the fluctuation amplitude and overload frequency, integrate the risk load point data, and generate a load risk point data set.

[0028] As a further solution of the present invention, the step of obtaining the high-frequency overload risk interval is specifically as follows:

[0029] S311: Based on the training load risk point dataset, analyzing the fluctuation intensity and corresponding time intervals of the low-frequency training load, marking the parts where the fluctuation intensity is greater than the time stability, performing segmented analysis of the load change trend, and identifying the low-frequency fatigue accumulation interval;

[0030] S312: Analyze the distribution characteristics of the low-frequency fatigue accumulation interval and the high-frequency intensity fluctuation interval, perform time distribution statistics and intensity evaluation of the high-frequency training load, mark intervals where the training load intensity is higher than the low-frequency time average, identify differences in the time and intensity distribution of the training load, optimize the intensity fluctuation interval, and obtain a concentrated distribution area of overload in the high-frequency band;

[0031] S313: Based on the fatigue accumulation interval in the low frequency band and the concentrated distribution area of overload in the high frequency band, a matching analysis of training time and intensity is performed. Combined with the load risk point data, the overload area is analyzed using the formula:

[0032]

[0033] Calculate the high-frequency overload risk value and obtain the high-frequency overload risk range;

[0034] Among them, Q represents the high-frequency overload risk value, F H (t) represents the load intensity in the high frequency band, F L (t) represents the low-frequency load intensity, y represents the frequency intensity weight adjustment coefficient, T H (t) and T L(t) represents the training time distribution of high frequency band and low frequency band respectively, and T represents the length of the training time period.

[0035] As a further solution of the present invention, the steps for obtaining the training load optimization adjustment result are specifically as follows:

[0036] S321: Based on the high-frequency overload risk interval, extract load intensity, frequency, and time data, analyze load intensity changes in each time period, group and process and filter intensity mutation points, associate time period and frequency data, calculate frequency changes and intensity distribution, and obtain high-frequency overload load distribution data;

[0037] S322: Based on the high-frequency overload load distribution data, calculate the intensity change difference within the interval, perform group analysis according to time period, determine the interval that exceeds the change range, mark the time series, organize the load characteristic distribution characteristics, and obtain the load distribution characteristic adjustment benchmark;

[0038] S323: Based on the load distribution characteristic adjustment benchmark, adjust the intensity, frequency and time distribution of the training load, re-divide the time interval, balance the intensity data, smooth the frequency data, match and integrate with the time series, and generate the training load optimization adjustment result.

[0039] As a further solution of the present invention, the steps for obtaining the youth physical training adjustment execution plan are specifically as follows:

[0040] S411: Based on the training load optimization and adjustment result, extract the intensity and frequency parameters of the training action load and divide them into intervals, analyze the load variation amplitude of each action within the training cycle, identify the load coverage interval, and obtain the intensity and frequency adjustment intervals of the training action load;

[0041] S412: Based on the intensity and frequency adjustment intervals of the training action load, combined with the changing trend of the action frequency, extract the corresponding time period, identify the stability and volatility of the training action intervals, and calculate the training action stability parameter using the formula:

[0042]

[0043] Obtain the abnormal distribution area of movement coordination;

[0044] Among them, R represents the training action stability parameter, f h and f l are the load intensities of high-frequency and low-frequency training movements, t h and t l are the time intervals of high-frequency and low-frequency actions, respectively, and β and γ are adjustment coefficients;

[0045] S413: Calculate the deviation of the movement load in the area according to the movement coordination abnormality distribution area, adjust the movement frequency and intensity through movement intensity and coordination parameters, and generate a youth physical training adjustment execution plan.

[0046] A youth physical training system based on data analysis, wherein the youth physical training system based on data analysis is used to execute the above-mentioned youth physical training method based on data analysis, and the system comprises:

[0047] The load frequency band parameter analysis module divides the frequency bands based on the load intensity data of youth training, extracts the trend change value and fluctuation range value of the low frequency band, identifies the short-term change rate and distribution parameters of the high frequency band, and generates a frequency band distribution parameter set;

[0048] The dynamic training load identification module performs parameter calculation on the low-frequency band trend change value and the cumulative load based on the frequency band distribution parameter set, and generates a dynamic load characteristic data set by combining the bandwidth frequency band stability characteristics and the high-frequency band overload frequency distribution data;

[0049] The training risk area optimization module extracts the regional distribution value of the cumulative load change in the low-frequency band based on the dynamic load characteristic data set, performs a balance analysis based on the bandwidth frequency band fluctuation range and time interval parameters, integrates the high-frequency band overload concentration distribution parameters, and generates a high-frequency overload risk area;

[0050] The training adjustment execution module adjusts the intensity and frequency of the load signal based on the high-frequency overload risk area, combines the bandwidth frequency band stability characteristics and the training interval time distribution parameters, analyzes the movement offset and corrects the amplitude, and generates a youth physical training adjustment execution plan.

[0051] Compared with the prior art, the advantages and positive effects of the present invention are:

[0052] In the present invention, through multi-band decomposition and parameterization of load signals, the long-term load trend of the low-frequency band, the load stability characteristics of the broadband frequency band, and the short-term change rate of the high-frequency band are distinguished, so that the load characteristics of different frequency bands during training can be calculated and comprehensively analyzed separately, solving the problem of difficulty in subdividing load characteristics during training and improving the ability to capture dynamic changes in training load. By analyzing the correlation between the cumulative load and training time in the low-frequency band and the time interval characteristics of the broadband frequency band load, the coordination between the cumulative load and the interval distribution is enhanced. The overload frequency and fluctuation amplitude of the high-frequency band are further extracted, providing complete data support for the dynamic interaction between training loads. The training load risk points are accurately identified and the distribution areas are delineated, so that fatigue accumulation areas and high-frequency overload areas can be separately extracted, and the adjustment strategy of load intensity, time and frequency is optimized. The real-time adjustment of movement load achieves load balance and coordination of movement execution through comprehensive optimization of intensity, frequency and deviation correction, thereby significantly reducing the overload risk and movement deviation risk during training in youth physical training and enhancing the scientific nature and controllability of the overall training process. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a schematic diagram of the workflow of the present invention;

[0054] Figure 2 This is a flow chart of the frequency band distribution parameter set of the present invention;

[0055] Figure 3 It is a flow chart of the dynamic load interaction coefficient in the present invention;

[0056] Figure 4 This is a flow chart of the training load risk point data set in the present invention;

[0057] Figure 5 This is a flow chart of the high-frequency overload risk interval in the present invention;

[0058] Figure 6 This is a flow chart of the training load optimization adjustment results in the present invention;

[0059] Figure 7 The figure is a flow chart of the implementation scheme for adjusting the physical training of teenagers in the present invention. DETAILED DESCRIPTION

[0060] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0061] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0062] Example 1

[0063] See also Figure 1 The present invention provides a technical solution: a method for physical training of teenagers based on data analysis, comprising the following steps:

[0064] S1: Based on the load intensity data of adolescent training, the load signal is divided into low-frequency band, broadband band and high-frequency band. The long-term change trend and fluctuation range of the low-frequency band load are analyzed. The stability parameters are calculated based on the signal fluctuation of the broadband band. The short-term load change rate and change rate distribution of the high-frequency band are analyzed to obtain the frequency band distribution parameter set.

[0065] S2: Based on the frequency band distribution parameter set and combined with the low-frequency band load trend value, the relationship between training time and cumulative load is analyzed. The bandwidth frequency band load stability and time interval characteristics are extracted to obtain the dynamic load interaction coefficient. The instantaneous overload frequency and fluctuation amplitude are identified. The dynamic characteristics of differentiated loads are integrated and analyzed to obtain the training load risk point dataset.

[0066] S3: Based on the training load risk point dataset, fatigue accumulation areas in the low-frequency band are identified. The intensity fluctuation range and training time distribution balance of the bandwidth frequency band are analyzed. The high-frequency overload concentrated distribution area is extracted to obtain the high-frequency overload risk range. The intensity, frequency, and time adjustments of the training load are optimized to obtain the training load optimization adjustment results.

[0067] S4: Based on the training load optimization adjustment results, adjust the intensity and frequency of the training action load, identify the training interval time, analyze the movement coordination abnormalities in combination with the movement stability parameters of the bandwidth frequency band, extract the movement deviation data and calculate the correction amplitude, and generate the youth physical training adjustment implementation plan.

[0068] The frequency band distribution parameter set includes the low-frequency band load trend value, the broadband frequency band load stability parameter, and the high-frequency band load change rate distribution. The dynamic load interaction coefficient includes the load correlation parameter, the load stability characteristic, and the overload fluctuation characteristic. The training load risk point data set includes the fatigue accumulation area, the intensity fluctuation range, and the high-frequency overload distribution. The high-frequency overload risk range includes the overload concentration, the overload range, and the high-frequency load distribution characteristics. The training load optimization adjustment results include the load intensity adjustment value, the frequency optimization value, and the time distribution adjustment value. The youth physical training adjustment execution plan includes the action load adjustment parameters, the action frequency correction value, and the action deviation correction amplitude.

[0069] See also Figure 2 , the specific steps for obtaining the frequency band distribution parameter set are:

[0070] S111: Based on the load intensity data from youth training, perform spectrum analysis to decompose the load signal into low-frequency band, broadband band, and high-frequency band. Extract the energy components and time series characteristics of the frequency band signals, perform denoising, analyze the energy distribution ratio and mean fluctuation range of the frequency bands, and generate a load intensity frequency division feature parameter set.

[0071] The continuous signal is converted into a frequency domain signal through Fourier transform, and the low frequency band, broadband frequency band and high frequency band are separated by calculating the power spectral density function. The specific frequency range of each frequency band is determined. The energy density of the load signal is normalized and calculated to extract the energy characteristics of the signal in each frequency band. The frequency concentration parameter of the signal is calculated based on the peak amplitude corresponding to the frequency in the frequency domain. The time series data of the signal in each frequency band is denoised by the sliding average method. The signal energy proportion, mean fluctuation range and amplitude change rate of each frequency band are calculated. By comparing the energy distribution and amplitude fluctuation range of each time period in each frequency band, the time period with a low noise proportion is screened. The mean, standard deviation and peak frequency of the characteristic signal are further extracted to generate a load intensity frequency division characteristic parameter set.

[0072] S112: Based on the load intensity frequency division characteristic parameter set, statistical analysis is performed on the time series characteristic values of the low-frequency load signal to identify the linear fitting parameters and fluctuation range of the long-term change trend using the formula:

[0073]

[0074] Calculate the trend distribution parameters of low-frequency signals and generate the long-term change trend and fluctuation range of low-frequency bands;

[0075] Among them, P l Represents the low-frequency signal trend distribution parameter, v i Represents the time series value of the low-frequency load signal, w irepresents the weight factor of the time point, α is the fluctuation adjustment coefficient, β is the smoothing weight adjustment parameter, and n is the total number of sample points;

[0076] v i : represents the time series value of the low-frequency load signal at the i-th time point. The unit is "load unit". In the context of youth physical training, it is expressed in Newtons (N) or training load index (a dimensionless scoring standard). In this example, it is the training intensity sampling value (e.g., 12, 14, 15, 13, 14);

[0077] w i : is the time weight factor corresponding to the i-th time point, unitless. This value is the relative weight determined by time importance analysis, reflecting the degree of influence of each time point on the trend. By analyzing the importance of the time point in the training plan, such as whether it is at the training peak or recovery period, the weight is set accordingly. In this example, the values are 1.2, 1.3, 1.1, 1.2, and 1.4, which are proportional factors.

[0078] α: Volatility adjustment coefficient, used to amplify the weight of the volatility factor and enhance the ability to identify high-volatility time points. It is an empirically set parameter. In this example, α=0.8 is used. It has no unit.

[0079] β: is a smoothing weight adjustment parameter used to suppress fluctuation interference caused by extreme values and enhance the stability of trend judgment. The unit is a dimensionless constant. In this example, β = 0.5;

[0080] n: sample size, which is the total number of time series data points, in this case 5;

[0081] The formula is beneficial in that it enhances the ability to fit high-volatility signals by introducing a volatility adjustment coefficient, while reducing reliance on extreme signal values by adjusting the smoothing weight parameters, making the formula more stable in measuring long-term trends and volatility ranges.

[0082] Collect time series data of low frequency load signal v i , get the sequence value v i is 12, 14, 15, 13, 14, and the weight factor w i The weights are determined by time importance analysis, and are 1.2, 1.3, 1.1, 1.2, and 1.4, respectively. The fluctuation adjustment coefficient α = 0.8 and the adjustment parameter β = 0.5 are calculated based on the stability of the overall fluctuation range of the data;

[0083] Calculate the first part

[0084] Calculate the second part

[0085]

[0086] Combined formula for overall calculation:

[0087] The results show that the trend distribution parameter of the low-frequency signal is 15.14, which indicates a high level of long-term stability in the training load. Combining this with fluctuation range analysis can help optimize physical training plans more accurately, ultimately generating a long-term trend and fluctuation range for the low-frequency band.

[0088] The parameters in the formula are obtained by sampling and analyzing the low-frequency load time series in adolescent training. Combined with set factors such as time weight, fluctuation adjustment and smoothing coefficient, the indicator P1 reflecting the long-term trend and fluctuation of training load is calculated. The formula is unified in dimension as "training intensity unit" to facilitate comparison and adjustment with the original load data, and has good engineering practicality and interpretability.

[0089] S113: Based on the long-term change trend and fluctuation range of the low-frequency band, combined with the fluctuation characteristic parameters and stability distribution of the broadband frequency band signal, a differential operation of the short-term load change rate of the high-frequency band signal is performed to obtain the distribution density and change rate distribution range of the high-frequency band load change rate, identify the deviation index of the frequency band signal amplitude, and generate a frequency band distribution parameter set;

[0090] By performing differential operations on signal time series data, the instantaneous rate of change of load intensity is calculated, and the part of the rate of change with more significant short-term fluctuations is identified through the high-order difference method. By statistically analyzing the distribution density of the rate of change, the overall amplitude range of the short-term changes of high-frequency signals is calculated. By comparing the deviation values of the rate of change distribution density, the deviation index of the short-term rate of change distribution is calculated. The strength of the short-term fluctuations in the high-frequency band is judged by the level of the deviation index, and the time points with significant fluctuations are screened. The short-term rate of change characteristic data and the stability characteristics of the bandwidth frequency band are combined for analysis, and the frequency band distribution parameters are extracted, classified and organized into a frequency band distribution parameter set.

[0091] See also Figure 3 , the steps for obtaining the dynamic load interaction coefficient are as follows:

[0092] S211: Based on the frequency band distribution parameter set, extract the load value and time information of each time period from the training data, group the time period load values of the low-frequency band data by time node, identify the load accumulation value of the time group, filter the time nodes with load fluctuation, and obtain the low-frequency band bandwidth load trend parameter set;

[0093] In the process of extracting load values, the time series must first be segmented, and the time interval length is set as the basis for grouping. The cumulative load value in each period is calculated for each group, and the change trend is determined by the absolute value difference of the load change. The sum of the load values in each time period is used as the cumulative value, and the difference change is calculated based on the start value and end value of each time period. Then, the cumulative value of all time periods is compared with the change trend. The threshold of the differential amplitude is used to determine whether the load change in each period fluctuates greatly. The time period with large fluctuations is recorded as an abnormal point and serves as the key reference point for subsequent data analysis. When screening time points with large fluctuations, the specific fluctuation situation is determined based on the change amplitude and the standard deviation threshold of the time period. Those below the standard deviation are eliminated. The above steps complete the extraction of the load trend parameter set for each time period and obtain the low-frequency bandwidth load trend parameter set.

[0094] S212: Based on the low-band bandwidth load trend parameter set, and referring to the low-band load stability and time interval characteristics, the load stability and time interval characteristics are combined and the formula is used:

[0095]

[0096] The influence of time interval correction on load stability is introduced into the adjustment coefficient, and the dynamic load interaction coefficient is calculated;

[0097] Among them, C t represents the dynamic load interaction coefficient, d j Represents the load change within the time interval, T j Represents the corresponding time interval, M1 represents the load stability weight coefficient, L j Represents the associated parameter of load stability, S j represents the characteristic value of the time interval, and k represents the total number of time periods;

[0098] The benefit of this formula is that by combining time intervals and load stability and introducing weight correction parameters, the dynamic load interaction coefficient can be evaluated more accurately. At the same time, the time stability analysis capability of low-frequency loads is enhanced, and the calculation of the dynamic load interaction coefficient is more accurate.

[0099] C t The dynamic load interaction coefficient measures the degree of load fluctuation in each period and its impact on training time.

[0100] d j Represents the load change during a certain period of time, which is obtained by subtracting the initial load value from the final load value. The unit is training load unit (such as N, kg or index score);

[0101] T jRepresents the length of the corresponding time interval, in units of time (such as seconds, minutes);

[0102] M1 is the load stability weight coefficient, which is an empirical parameter used to adjust the stability impact weight. It has no unit and is set to 0.6, for example.

[0103] L j Represents the load stability parameter within a certain period of time, which is calculated based on the ratio of the load fluctuation range to the standard value. j It is obtained by dividing the load fluctuation range by the load standard value, and the unit is a dimensionless ratio;

[0104] S j Represents the time interval characteristic value, which is the average time interval of load changes in a certain period of time;

[0105] k is the total number of time periods, unitless;

[0106] Assume that the total load value of the collected data in a certain period of time is 500 units. The period is divided into 5 sub-intervals. The load values of each sub-interval are 80, 100, 120, 90, and 110 respectively. The time intervals are 10, 8, 12, 9, and 11 units of time respectively:

[0107] Calculate the load change in each subinterval: d1 = 100 - 80 = 20, d2 = 120 - 100 = 20, and so on;

[0108] Calculate the load stability parameter L j and the time interval characteristic value S j , for example, if the load fluctuation range is 40 units and the standard value is 100 units,

[0109] Substitute the formula and sum it step by step:

[0110] Set M1 = 0.6, S j =2, the final result is C t =12.5;

[0111] The results show that the value of the dynamic load interaction coefficient reflects the combined effect of load fluctuation and time interval characteristics in each period. The higher the value, the greater the fluctuation effect, which will be used in the adjustment of the dynamic load model;

[0112] The dynamic load interaction coefficient C t In the calculation of , the formula converts the load change rate d in each sub-interval into j / T j The corresponding load stability factor M1·L j / S jCombined with the above, it comprehensively reflects the coupling effect of short-term fluctuation intensity and time interval inhomogeneity during training. j With training load units (such as N), T j 、S j is the time unit (such as minutes), and the other parameters are dimensionless. Through dimensional unification, the final calculation result C t It has the dimension of load unit per time squared (such as N / min2), which is used to accurately identify and regulate potential overload risks and rhythm instability factors in training. This indicator has important application value in the optimization of physical training programs.

[0113] See also Figure 4 ,The steps for obtaining the training load risk point dataset are as follows:

[0114] S221: Based on the dynamic load interaction coefficient, the dynamic numerical data is decomposed by time period, the instantaneous load data at each time point is extracted, the values exceeding the threshold are filtered, and the partitions are marked and classified, and the load intervals are classified and processed to obtain the instantaneous overload load interval data set;

[0115] First, the training data of a day is decomposed by hour and time period, and the training load value of each time period is extracted into an independent data set to ensure that the instantaneous load data at each time point can clearly identify the time period characteristics. For each instantaneous load data, a dynamic load threshold is set, such as the 200-400 unit interval in youth physical training. Data points exceeding the set threshold are filtered, and the time and training items corresponding to the data are recorded. The instantaneous load data exceeding the threshold are divided into peak and non-peak areas according to the time period. The overload situation is marked in each interval, for example, by training type (such as sprint running, powerlifting, etc.), and classified and sorted. By analyzing the duration and load intensity of each set of overload data, the load intervals are further classified and processed, and the dynamic interaction coefficient of each load data is corrected to improve the analysis accuracy, finally forming an instantaneous overload load interval data set that includes partition marking, classification and time nodes.

[0116] S222: Based on the instantaneous overload interval data set, calculate the difference between the peak and the valley of the load value within the interval, record the quantitative data of the fluctuation amplitude, match the time series, analyze the starting and ending points of the load change, perform time marking, and obtain a load fluctuation characteristic distribution data table;

[0117] The load data within each interval is calculated point by point, extracting the peak and trough values of the load within the interval to calculate the quantitative data of the fluctuation amplitude. For example, for youth physical training, if the peak load of sprint running in a certain interval is 480 units and the trough value is 210 units, the calculated fluctuation amplitude is 270 units. This is recorded in a quantitative table, and the above fluctuation amplitude data is matched to the time series to ensure that each fluctuation amplitude can be accurately mapped to the training period. The starting and end points of the training load changes are further analyzed. For example, through analysis, a load fluctuation was found to start at the start time of sprint running training at 15:05 and end at the end time of the strength recovery phase at 15:35. Through this analysis, each interval is time-stamped, and the fluctuation amplitude and load change node at each time point are recorded in the data table. Ultimately, a load fluctuation characteristic distribution data table is formed to provide a basis for the next step of risk analysis.

[0118] S223: Based on the load fluctuation characteristic distribution data table, filter the intervals where the fluctuation amplitude exceeds the set range, perform cross-analysis based on the time series and overload data, classify and organize the risk intervals according to the fluctuation amplitude and overload frequency, integrate the risk load point data, and generate a load risk point data set;

[0119] Through the load fluctuation characteristic distribution data table, the fluctuation amplitude of each interval is screened one by one, and the intervals with fluctuation amplitudes exceeding the set range are extracted. For example, the fluctuation amplitude threshold is set to 250 units, and the training intervals with fluctuation amplitudes exceeding 250 units are screened. The screened time series data and instantaneous overload load data are combined to perform cross-analysis on the screened intervals. Through cross-analysis, the key intervals where the training load fluctuation amplitude and overload frequency overlap are identified. For example, in sprint training, the overload load frequency is 3 times / hour and the fluctuation amplitude reaches 300 units. The training intervals are classified and sorted according to the fluctuation amplitude and overload frequency, and high-risk intervals and medium-risk intervals are marked. The key load point data of the corresponding intervals are extracted. For example, a load point in the interval between 15:10 and 15:30 has three overloads, and the risk coefficient is a high value. After integrating all the marked risk load point data, a load risk point data set is generated, providing a reliable basis for optimizing the design and implementation of physical training for teenagers.

[0120] See also Figure 5 ,The specific steps for obtaining the high-frequency overload risk interval are:

[0121] S311: Based on the training load risk point dataset, analyze the fluctuation intensity and corresponding time intervals of the low-frequency training load, mark the parts where the fluctuation intensity is greater than the time stability, perform segmented analysis of the load change trend, and identify the low-frequency fatigue accumulation interval;

[0122] Frequency band associated data is extracted, and the fluctuation intensity is calculated by decomposing the low-frequency band load characteristics. The time series in the low-frequency band is discretized based on the sampling period of the training load point to obtain the fluctuation amplitude distribution in the time series. The load fluctuation intensity of each sampling point in the time series is calculated using the frequency distribution function, and the fluctuation range of the low-frequency band is determined. The local extreme point analysis method is used to mark the intervals with significant load intensity changes. Each interval is classified according to the significance of the load intensity fluctuation, and the significant fluctuation area and the stable area are distinguished. The area with significant load changes is identified to obtain the low-frequency band fatigue accumulation interval.

[0123] S312: Analyze the distribution characteristics of the low-frequency fatigue accumulation interval and the high-frequency intensity fluctuation interval, perform time distribution statistics and intensity evaluation of the high-frequency training load, mark intervals where the training load intensity is higher than the low-frequency time average, identify differences in the time and intensity distribution of the training load, optimize the intensity fluctuation interval, and obtain the concentrated distribution area of overload in the high-frequency band;

[0124] The intensity distribution characteristics and time coverage of the high-frequency training load were extracted, and the time distribution statistics of the high-frequency training load were performed according to the load fluctuation intensity in the fatigue accumulation interval. The frequency distribution characteristic function was used to calculate the load change rate at each time point. The variation range between the mean and maximum values of the training load intensity was calculated based on the discrete sampling values of the high-frequency time series. The areas with load intensity higher than the time mean of the low-frequency band were screened and marked as high-frequency overload areas. The time-intensity distribution difference coefficient of the training load was calculated using the fluctuation characteristics of the frequency band. The load distribution intervals with a difference coefficient greater than a specific threshold were optimized, and the areas with significant fluctuations were divided into independent high-frequency distribution areas to obtain the concentrated distribution areas of high-frequency overload.

[0125] S313: Based on the fatigue accumulation interval in the low-frequency band and the concentrated distribution area of overload in the high-frequency band, the matching analysis of training time and intensity is carried out. Combined with the load risk point data, the overload area is analyzed using the formula:

[0126]

[0127] Calculate the high-frequency overload risk value and obtain the high-frequency overload risk range;

[0128] Among them, Q represents the high-frequency overload risk value, F H (t) represents the load intensity in the high frequency band, F L (t) represents the low-frequency load intensity, y represents the frequency intensity weight adjustment coefficient, T H (t) and T L (t) represents the training time distribution of high frequency band and low frequency band respectively, and T represents the length of the training time period;

[0129] The formula is beneficial in that, by introducing the absolute value calculation of the difference in load intensity between the high-frequency band and the low-frequency band, combined with the relative volatility evaluation of the training time distribution, it can accurately identify the high-frequency band overload risk interval and quantify its impact;

[0130] F H (t) is obtained by statistically calculating the training load intensity of the high-frequency band sampling points. The specific calculation method is: take the maximum value of the training intensity value of each high-frequency band minus the minimum value and take the average. Set the high-frequency band training load intensity value to [320, 280, 340], then F H (t) = (340 - 280) / 3 = 20, where the unit is the training intensity unit (e.g., Newton N, kg, or dimensionless training index);

[0131] F L (t) is the training load intensity in the low frequency band. The difference range is calculated using the same method and the setting value is [240, 220, 260]. Then F L (t) = (260-220) / 3 = 13.3, the unit is the same as F H (t) consistency, maintaining uniformity of dimensions;

[0132] The adjustment coefficient y depends on the imbalance of the intensity distribution between the high-frequency band and the low-frequency band. It is calculated through the frequency-intensity difference rate function and is set to y = 1.2, which is an empirically set dimensionless parameter.

[0133] T H (t) and T L (t) are the time distribution density of high frequency band and low frequency band respectively. By calculating the uniformity of time distribution within the statistical time period, set T H (t) = 15 and T L (t) = 10, where the unit is the time unit (e.g., minutes or seconds);

[0134] T: The total length of the training cycle, which is the total number of time periods, in units of time (such as minutes)

[0135] Substituting into the formula:

[0136] The results show that the calculated high-frequency overload risk value is 0.808, which is in the lower range, indicating that the current high-frequency load distribution has little impact on the balance of training time distribution. This value can be further used as an important indicator for screening high-risk load intervals.

[0137] In the calculation of the high-frequency overload risk value Q, the dynamic matching coordination degree is measured by comparing the difference in training intensity between the high-frequency and low-frequency bands (corrected by the adjustment coefficient) and the corresponding difference in training time density. Among the various parameters, F H (t), F L(t) The unit is load intensity unit, T H (t), T L The unit of (t) is time, and the adjustment coefficient γ has no unit. The result of the formula, Q, has the dimension of load intensity per unit time (such as N / min), which is used to reflect the degree of uneven distribution of load and time in the high-frequency bands of training. Higher values indicate potential overload risks, and lower values indicate a more balanced training arrangement. This formula not only provides guidance for the adjustment of training programs, but also provides a quantitative reference for the screening of high-frequency risk areas.

[0138] See also Figure 6 The specific steps for obtaining the training load optimization adjustment results are as follows:

[0139] S321: Based on the high-frequency overload risk interval, extract load intensity, frequency, and time data, analyze load intensity changes in each time period, group and process, and filter intensity mutation points, associate time period and frequency data, calculate frequency changes and intensity distribution, and obtain high-frequency overload load distribution data;

[0140] The load intensity data for each time period is sliced and archived according to the time series. To divide the hourly data into several 10-minute intervals, the peak and average load intensity values for each interval are recorded. The rate of change of the load intensity within each time period is calculated. By comparing the load intensity gradient changes between the previous and next intervals, intensity mutation points are screened. For the mutation point data, the frequency and time information of the same interval are correlated. The frequency change trends before and after each mutation point are analyzed. The frequency values are then refined into fluctuation ranges and change rates based on the time period distribution, ultimately forming a corresponding relationship table between frequency changes and load intensity distribution. The calculation process specifically includes setting a mutation intensity threshold, marking a mutation point based on whether the load intensity change rate within the interval exceeds the set value, using the statistical distribution of frequency fluctuations to detect the significance of the frequency change before and after the mutation point, and summarizing the significance results with the time period information to obtain high-frequency overload load distribution data.

[0141] S322: Based on the high-frequency overload load distribution data, calculate the intensity change difference within the interval, perform group analysis according to time period, determine the interval that exceeds the change range, mark the time series, organize the load characteristic distribution characteristics, and obtain the load distribution characteristic adjustment benchmark;

[0142] Group by time period, record the load intensity extreme points in each time period, calculate the intensity change difference between each two consecutive time periods in turn, and mark the intervals that exceed the change range according to the size of the difference. The operation includes setting a baseline value for the intensity change range, identifying the intervals with abnormal fluctuations by comparing with the baseline value, marking the time series that exceeds the change range, and associating the marked time series with the frequency data to form a frequency-intensity correspondence curve. During the sorting process, further judge the load characteristic distribution characteristics of each interval through the trend of load intensity and frequency changes, extract the correlation points between the frequency change fluctuation range and the intensity mutation distribution of each time period, and form a load distribution characteristic adjustment benchmark to guide the design of load adjustment strategy.

[0143] S323: Based on the load distribution characteristics, the benchmark is adjusted to adjust the intensity, frequency, and time distribution of the training load, the time interval is re-divided, the intensity data is balanced, the frequency data is smoothed, and the data is matched and integrated with the time series to generate the training load optimization adjustment result;

[0144] First, the training time is divided into several detailed intervals, and the load intensity baseline value of each interval is reset. The adjustment method includes balancing the intensity data. By calculating the average value of the intensity distribution in each interval, the data that exceeds the interval average value is normalized to avoid local excessive load from disrupting the overall training rhythm. At the same time, the frequency data is smoothed to eliminate abnormal peak points in the frequency fluctuation, and the frequency change trend of the previous and next intervals is used for interpolation and completion. The data with large fluctuations is adjusted to a continuous curve, the processed intensity data is matched with the time series, and the adjusted frequency curve is used to correct the load distribution characteristics of each time period. Finally, the training load optimization adjustment result is generated to ensure that the distribution of load intensity and frequency is more stable in the training plan.

[0145] See also Figure 7 The specific steps for obtaining the youth physical training adjustment implementation plan are as follows:

[0146] S411: Based on the training load optimization adjustment results, extract the intensity and frequency parameters of the training action load and divide them into intervals, analyze the load variation amplitude of each action within the training cycle, identify the load coverage interval, and obtain the intensity and frequency adjustment intervals of the training action load;

[0147] By analyzing the cycle data of training movements, the interval range of training intensity is established, and the amplitude of changes in the strength of training movements is measured. The historical data of the motion sampler is used to extract the maximum and minimum training intensity and movement frequency interval of each movement. The key change points of the training movements are marked by the training load intensity change curve. According to the training frequency distribution within the cycle and the standard that the movement load value is higher than the average of the whole cycle, the time periods with high or low movement load are screened. The load coverage rate is calculated by matching the load frequency with the training intensity. The time periods with low coverage rate are adjusted in a complementary manner. The training load changes are standardized to generate the intensity and frequency adjustment intervals of the training movement load.

[0148] S412: Based on the intensity and frequency adjustment intervals of the training action load, combined with the changing trend of the action frequency, the corresponding time period is extracted, the stability and volatility of the training action intervals are identified, and the training action stability parameter is calculated using the formula:

[0149]

[0150] Obtain the abnormal distribution area of movement coordination;

[0151] Among them, R represents the training action stability parameter, f h and f l are the load intensities of high-frequency and low-frequency training movements, t h and t l are the time intervals of high-frequency and low-frequency actions, respectively, and β and γ are adjustment coefficients;

[0152] The benefit of this formula is that, by introducing the intensity and time distribution parameters of high-frequency and low-frequency training movements, as well as two weight adjustment coefficients, it can more accurately capture anomalies in movement stability caused by intensity changes and uneven time distribution, thereby accurately identifying areas of coordination abnormalities.

[0153] f h The load intensity of high-frequency action is calculated based on the monitoring value of action load intensity. For example, the historical load value of high-frequency training action [310, 280, 300] is extracted and its average value is calculated to obtain f h =296.7;

[0154] f l is the load intensity of low-frequency action. By the same method, the historical data of low-frequency training load is [250, 240, 260]. The average value is calculated to get f l =250;

[0155] t h and t l The time distribution of high-frequency and low-frequency actions respectively. By analyzing the action time records, the interval average value of high-frequency actions is obtained as th =5, the average value of low-frequency action interval is t l =8;

[0156] The adjustment coefficients β and γ are used to control the matching sensitivity between intensity and time interval, β = 1.2 and γ = 0.8, both of which are determined based on the sensitivity curve of the rate of change of movement coordination;

[0157] Substituting the above values into the formula:

[0158]

[0159] The results show that the stability index of the current training movement is 916.6, which is higher than the normal range of coordination (for example, between 500-800), indicating that there are obvious abnormal coordination areas in the movement. The cause can be further analyzed and the abnormal areas can be marked.

[0160] S413: Calculate the deviation of the movement load in the area based on the movement coordination abnormality distribution area, adjust the movement frequency and intensity based on the movement intensity and coordination parameters, and generate an adjustment execution plan for the youth physical training;

[0161] First, the time range and load deviation data of the abnormal distribution area are extracted, and the deviation value is calculated through the fluctuation rate of the movement load intensity within the abnormal time period. Combined with the historical data records of the movement frequency, the matching relationship between the load deviation and the frequency adjustment within the time range is evaluated. The time period with the deviation value higher than the fixed threshold is used to calculate the correction amplitude. The correction amplitude in each time period is summarized, and the training movement frequency and intensity are adjusted to generate an adjustment implementation plan for youth physical training.

[0162] The youth physical training system based on data analysis is used to implement the above-mentioned youth physical training method based on data analysis. The system includes:

[0163] The load frequency band parameter analysis module divides the frequency bands based on the load intensity data of youth training, extracts the trend change value and fluctuation range value of the low frequency band, identifies the short-term change rate and distribution parameters of the high frequency band, and generates a frequency band distribution parameter set;

[0164] The dynamic training load identification module performs parameter calculations on the low-frequency band trend change value and the accumulated load based on the frequency band distribution parameter set, and combines the bandwidth band stability characteristics with the high-frequency band overload frequency distribution data to generate a dynamic load characteristic data set.

[0165] The training risk area optimization module extracts the regional distribution value of the cumulative load change in the low-frequency band based on the dynamic load characteristic data set, performs a balance analysis based on the bandwidth frequency band fluctuation range and time interval parameters, and integrates the concentrated distribution parameters of the high-frequency band overload to generate the high-frequency overload risk area;

[0166] The training adjustment execution module adjusts the intensity and frequency of the load signal based on the high-frequency overload risk area, combines the bandwidth frequency band stability characteristics and the training interval time distribution parameters, analyzes the movement offset and corrects the amplitude, and generates a youth physical training adjustment execution plan.

[0167] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A youth physical training method based on data analysis, characterized in that: The following steps are involved: S1: Based on the load intensity data of adolescent training, the load signal is divided into low-frequency band, broadband band and high-frequency band. The long-term change trend and fluctuation range of the low-frequency band load are analyzed. The stability parameters are calculated based on the signal fluctuation of the broadband band. The short-term load change rate and change rate distribution of the high-frequency band are analyzed to obtain the frequency band distribution parameter set. S2: Based on the frequency band distribution parameter set and combined with the low-frequency band load trend value, the relationship between training time and cumulative load is analyzed, the bandwidth frequency band load stability and time interval characteristics are extracted to obtain the dynamic load interaction coefficient, the instantaneous overload frequency and fluctuation amplitude are identified, and the dynamic characteristics between differentiated loads are integrated and analyzed to obtain the training load risk point dataset; S3: Based on the training load risk point dataset, fatigue accumulation areas in the low-frequency band are identified, the intensity fluctuation range and training time distribution balance of the bandwidth frequency band are analyzed, the overload concentrated distribution area in the high-frequency band is extracted, and the high-frequency overload risk range is obtained. The intensity, frequency, and time adjustment of the training load are optimized to obtain the training load optimization adjustment result; S4: Based on the training load optimization adjustment results, adjust the intensity and frequency of the training action load, identify the training interval time, analyze the movement coordination abnormalities in combination with the movement stability parameters of the bandwidth frequency band, extract the movement deviation data and calculate the correction amplitude, and generate a youth physical training adjustment execution plan.

2. The method for physical training of teenagers based on data analysis according to claim 1, characterized in that: The steps for obtaining the frequency band distribution parameter set are specifically as follows: S111: Based on the load intensity data from youth training, perform spectrum analysis to decompose the load signal into low-frequency band, broadband band, and high-frequency band. Extract the energy components and time series characteristics of the frequency band signals, perform denoising, analyze the energy distribution ratio and mean fluctuation range of the frequency bands, and generate a load intensity frequency division feature parameter set. S112: Based on the load intensity frequency division characteristic parameter set, statistical analysis is performed on the time series characteristic values of the low-frequency load signal to identify the linear fitting parameters and fluctuation range of the long-term change trend, using the formula: Calculate the trend distribution parameters of low-frequency signals and generate the long-term change trend and fluctuation range of low-frequency bands; Among them, P l Represents the low-frequency signal trend distribution parameter, v i Represents the time series value of the low-frequency load signal, w i represents the weight factor of the time point, α is the fluctuation adjustment coefficient, β is the smoothing weight adjustment parameter, and n is the total number of sample points; S113: Based on the long-term change trend and fluctuation range of the low-frequency band, combined with the fluctuation characteristic parameters and stability distribution of the broadband frequency band signal, a differential operation of the short-term load change rate of the high-frequency band signal is performed to obtain the distribution density and change rate distribution range of the high-frequency band load change rate, identify the deviation index of the frequency band signal amplitude, and generate a frequency band distribution parameter set.

3. The method for physical training of teenagers based on data analysis according to claim 2, characterized in that: The steps for obtaining the dynamic load interaction coefficient are specifically as follows: S211: Based on the frequency band distribution parameter set, extract the load value and time information of each time period from the training data, group the time period load values of the low frequency band data by time node, identify the load accumulation value of the time group, filter the time nodes with load change fluctuations, and obtain the low frequency band bandwidth load trend parameter set; S212: Based on the low-band bandwidth load trend parameter set, and referring to the low-band load stability and time interval characteristics, the load stability and time interval characteristics are combined, and the formula is used: The influence of time interval correction on load stability is introduced into the adjustment coefficient, and the dynamic load interaction coefficient is calculated; Among them, C t represents the dynamic load interaction coefficient, d j Represents the load change within the time interval, T j Represents the corresponding time interval, M1 represents the load stability weight coefficient, L j Represents the associated parameter of load stability, S j represents the characteristic value of the time interval, and k represents the total number of time periods.

4. The method for physical training of teenagers based on data analysis according to claim 3, characterized in that: The steps for obtaining the training load risk point dataset are specifically as follows: S221: Based on the dynamic load interaction coefficient, decompose the dynamic numerical data by time period, extract the instantaneous load data at each time point, filter the values exceeding the threshold, perform partition marking and classification, classify the load intervals, and obtain the instantaneous overload load interval data set; S222: Based on the instantaneous overload load interval data set, calculate the difference between the peak and the valley of the load value within the interval, record the quantitative data of the fluctuation amplitude, match the time series, analyze the starting point and end point of the load change, perform time marking, and obtain a load fluctuation characteristic distribution data table; S223: Based on the load fluctuation characteristic distribution data table, filter the intervals where the fluctuation amplitude exceeds the set range, perform cross-analysis in combination with the time series and overload load data, classify and organize the risk intervals according to the fluctuation amplitude and overload frequency, integrate the risk load point data, and generate a load risk point data set.

5. The method for physical training of teenagers based on data analysis according to claim 4, characterized in that: The steps for obtaining the high-frequency overload risk interval are specifically as follows: S311: Based on the training load risk point dataset, analyzing the fluctuation intensity and corresponding time intervals of the low-frequency training load, marking the parts where the fluctuation intensity is greater than the time stability, performing segmented analysis of the load change trend, and identifying the low-frequency fatigue accumulation interval; S312: Analyze the distribution characteristics of the low-frequency fatigue accumulation interval and the high-frequency intensity fluctuation interval, perform time distribution statistics and intensity evaluation of the high-frequency training load, mark intervals where the training load intensity is higher than the low-frequency time average, identify differences in the time and intensity distribution of the training load, optimize the intensity fluctuation interval, and obtain a concentrated distribution area of overload in the high-frequency band; S313: Based on the fatigue accumulation interval in the low frequency band and the concentrated distribution area of overload in the high frequency band, a matching analysis of training time and intensity is performed. Combined with the load risk point data, the overload area is analyzed using the formula: Calculate the high-frequency overload risk value and obtain the high-frequency overload risk range; Among them, Q represents the high-frequency overload risk value, F H (t) represents the load intensity in the high frequency band, F L (t) represents the low-frequency load intensity, y represents the frequency intensity weight adjustment coefficient, T H (t) and T L (t) represents the training time distribution of high frequency band and low frequency band respectively, and T represents the length of the training time period.

6. The method for physical training of teenagers based on data analysis according to claim 5, characterized in that: The steps for obtaining the training load optimization adjustment result are specifically as follows: S321: Based on the high-frequency overload risk interval, extract load intensity, frequency, and time data, analyze load intensity changes in each time period, group and process and filter intensity mutation points, associate time period and frequency data, calculate frequency changes and intensity distribution, and obtain high-frequency overload load distribution data; S322: Based on the high-frequency overload load distribution data, calculate the intensity change difference within the interval, perform group analysis according to time period, determine the interval that exceeds the change range, mark the time series, organize the load characteristic distribution characteristics, and obtain the load distribution characteristic adjustment benchmark; S323: Based on the load distribution characteristic adjustment benchmark, adjust the intensity, frequency and time distribution of the training load, re-divide the time interval, balance the intensity data, smooth the frequency data, match and integrate with the time series, and generate the training load optimization adjustment result.

7. The method for physical training of teenagers based on data analysis according to claim 6, characterized in that: The steps for obtaining the youth physical training adjustment implementation plan are as follows: S411: Based on the training load optimization and adjustment result, extract the intensity and frequency parameters of the training action load and divide them into intervals, analyze the load variation amplitude of each action within the training cycle, identify the load coverage interval, and obtain the intensity and frequency adjustment intervals of the training action load; S412: Based on the intensity and frequency adjustment intervals of the training action load, combined with the changing trend of the action frequency, extract the corresponding time period, identify the stability and volatility of the training action intervals, and calculate the training action stability parameter using the formula: Obtain the abnormal distribution area of movement coordination; Among them, R represents the training action stability parameter, f h and f l are the load intensities of high-frequency and low-frequency training movements, t h and t l are the time intervals of high-frequency and low-frequency actions, respectively, and β and γ are adjustment coefficients; S413: Calculate the deviation of the movement load in the area according to the movement coordination abnormality distribution area, adjust the movement frequency and intensity through movement intensity and coordination parameters, and generate a youth physical training adjustment execution plan.

8. A youth physical training system based on data analysis, characterized in that: The method for physical training of teenagers based on data analysis according to any one of claims 1 to 7, wherein the system comprises: The load frequency band parameter analysis module divides the frequency bands based on the load intensity data of youth training, extracts the trend change value and fluctuation range value of the low frequency band, identifies the short-term change rate and distribution parameters of the high frequency band, and generates a frequency band distribution parameter set; The dynamic training load identification module performs parameter calculation on the low-frequency band trend change value and the cumulative load based on the frequency band distribution parameter set, and generates a dynamic load characteristic data set by combining the bandwidth frequency band stability characteristics and the high-frequency band overload frequency distribution data; The training risk area optimization module extracts the regional distribution value of the cumulative load change in the low-frequency band based on the dynamic load characteristic data set, performs a balance analysis based on the bandwidth frequency band fluctuation range and time interval parameters, integrates the high-frequency band overload concentration distribution parameters, and generates a high-frequency overload risk area; The training adjustment execution module adjusts the intensity and frequency of the load signal based on the high-frequency overload risk area, combines the bandwidth frequency band stability characteristics and the training interval time distribution parameters, analyzes the movement offset and corrects the amplitude, and generates a youth physical training adjustment execution plan.

Images