70 results about "Training load" patented technology

The invention provides a personalized fitness system and method based on multi-modal data and an adaptive large model. According to the method, multi-modal data are integrated, and heterogeneous data collaborative analysis is realized by using a space-time alignment algorithm and a confidence coefficient weighting mechanism. A multi-task learning framework is adopted, physiological prediction, exercise risk early warning and psychological incentive strategy generation tasks are synchronously processed, a user stage target is adapted through a dynamic attention mechanism, and federal learning and transfer learning technologies are combined. A nonlinear periodic planning engine based on reinforcement learning dynamically adjusts training load and action difficulty according to real-time biomechanical simulation results and recovery state evaluation, and meanwhile, a lightweight model compression technology and end-side reverse dynamics calculation are adopted. Based on the scheme, the exercise performance prediction precision and the early warning timeliness are improved, and the exercise loss risk caused by overtraining is reduced; the cold start data bottleneck is broken through; low-delay virtual coach interaction and privacy protection are realized; and dynamically adjusting the training intensity and the incentive strategy.

The invention relates to the field of swimming physical training analysis, and particularly discloses a swimming underwater physical training effect analysis system based on training data, which is characterized in that video data, motion data and physiological data of a swimmer are acquired by integrating an underwater camera, an inertial measurement unit, a myoelectricity sensor, a heart rate sensor and a blood oxygen sensor; space-time synchronous analysis of multi-modal data is realized; on the basis of joint angle change consistency and muscle activation synchronism, whether the action of the swimmer is standard or not is judged, AR real-time feedback is conducted, and real-time correction of the swimmer in training is facilitated; the training load of the swimmer is evaluated based on the underwater speed, the stroke, the stroke frequency, the swimming distance and the swimming duration, and overrun early warning is carried out, so that making of a personalized training plan is facilitated, and meanwhile, overtraining is avoided; by monitoring the heart rate, the oxyhemoglobin saturation, the muscle activation duration and the joint activity condition of the swimmer, whether the swimmer has the injury risk or not is judged, early warning is triggered, and the injury occurrence rate is reduced.

The invention discloses an adjustable handheld respiratory muscle training device which comprises a respiratory muscle training device body, a handle is detachably installed at the lower end of the respiratory muscle training device body through a bolt, an adjusting cavity is rotatably connected to the front end of the respiratory muscle training device body in a sealed mode, and a respiratory gas hood is installed at the front end of the adjusting cavity. An air hole is formed in the upper portion of the front end of the respiratory muscle training device body, a plurality of sets of adjusting holes are formed in the adjusting cavity in a surrounding mode, the air hole corresponds to one set of adjusting holes, the number of the adjusting holes is ten, and the ten sets of adjusting holes are sequentially formed in the adjusting cavity in a surrounding mode from large to small in aperture. In particular to an adjustable handheld respiratory muscle trainer convenient to use, the trainer has a multi-gear adjusting function and can provide different degrees of respiratory training loads for a user, and by means of the trainer, the user can effectively exercise respiratory muscles, improve vital capacity, increase athletic performance and improve the health level of the respiratory system.

The invention discloses a personalized physical training load intelligent regulation and control system based on multi-modal physiological data fusion, and relates to the technical field of intelligent physical training and health monitoring. The system comprises a multi-modal data acquisition module, an intelligent fusion analysis module, a personalized decision module and a real-time regulation and control module. The method comprises the following steps: acquiring physiological, motion and environmental parameters of a user, and performing multi-modal data fusion by using a time sequence-causal attention mechanism to generate a comprehensive physiological state feature vector; a dynamic personalized baseline is established based on a meta-learning framework, a load adjustment decision is generated by adopting a hierarchical reinforcement learning algorithm, and regulation and control safety is ensured through causal reasoning verification; and finally, load adjustment is executed through the real-time regulation and control module to form closed-loop control. According to the method, the problems that in the prior art, training load regulation depends on a single physiological parameter, the individuation degree is insufficient, and decision making is lack of scientificity are solved, and accurate, self-adaptive and safe individualized physical training management is achieved.

The invention discloses an off-road training plan generation method and device, electronic equipment and a computer storage medium. The method comprises the steps of obtaining feature information of a target off-road race and an initial long-term training load value of a user; according to the feature information and the initial long-term training load value, judging whether the user can successfully complete the target off-road race or not; and when it is judged that the user can successfully complete the target off-road running event, determining course arrangement of an off-road running training plan based on the training course library. In the whole off-road running training plan generation process, the off-road running training plan with pertinence, reasonability and scientificity can be generated, and therefore the off-road running ability of the runner can be improved in a targeted mode by combining the characteristics of the target off-road running event and the actual ability of the runner.

The invention discloses a physical training load regulation and control method and system based on self-adaptive physiological coupling, and the method comprises the steps: collecting an electrocardiosignal, a light volume pulse signal, a respiration signal, a surface electromyographic signal and a body surface temperature signal, and generating a multimode sequence with a time mark; self-adaptive time alignment is executed, a timestamp graph is constructed, clock drift estimation is carried out, segmented monotonic reversible time transformation is established, and a multimode sequence is mapped to a unified time axis; reconstructing the missing section and the artifact section, establishing a mask auto-encoder with physiological priori constraints, combining time domain consistency constraints and frequency domain consistency constraints, and outputting a continuous differentiable reconstructed sequence. According to the method, by introducing adaptive time alignment, cross-modal signal reconstruction and dual-time scale state modeling, coordinated recognition and dynamic load regulation and control of multi-source physiological signals are achieved, coupling analysis in the training process is more stable, and an individual model can be continuously and adaptively updated.

The invention discloses fitness equipment for high-strength recovery training, and belongs to the technical field of fitness equipment. Fitness equipment for high-strength recovery training comprises a rack and a balance weight swing rod arranged in the rack, the rack is provided with a control area and a balance weight area, one end of the balance weight swing rod is movably connected with the interior of the control area, the other end of the balance weight swing rod extends into the balance weight area, and a balance weight adjusting piece is hinged to the balance weight swing rod; a winding part used for winding the flexible connecting piece is rotationally arranged in the control area, and the end, away from the winding part, of the flexible connecting piece is connected between the movable end and the hinged end of the balance weight swing rod; the pressing rod is pressed down through an arm, the swing arm and the converter can be driven to swing, in the swing process, the flexible connecting piece can be rolled up to achieve lifting of the balance weight swing rod, on-load training is achieved, through the balance weight adjusting piece, the training load can be conveniently adjusted, autonomous exercise operation can be achieved, and the training efficiency is improved. The positive influence is generated on the conscious exercise of the patient.

The invention provides a multi-mode biofeedback-guided training intensity adjusting method in the pelma rehabilitation process, and belongs to the technical field of biofeedback guidance. A pressure sensor and a surface electromyographic electrode are fixed to the pelma surface of a patient, and a pelma pressure distribution signal and a lower limb muscle group surface electromyographic signal are synchronously collected; filtering, denoising and rectifying preprocessing are carried out on the collected signals, a real-time biological feedback adjustment value and a training load adaptability evaluation parameter are calculated through an adaptive intensity adjustment model of a liquid neural network structure, and the training intensity is dynamically adjusted by adopting a gating weighting function according to an adjustment value interval; establishing a training intensity optimization equation set comprising a load distribution equation and a fatigue monitoring equation to calculate an optimal training parameter, and feeding back the adjusted intensity parameter to rehabilitation equipment to realize closed-loop adaptive adjustment. The technical problem that real-time self-adaptive training intensity adjustment based on multi-mode biological signals cannot be achieved in the sole rehabilitation training process is solved.

The present invention provides a personalized rehabilitation training method and system based on state monitoring, which relates to the field of rehabilitation training technology. The method includes: obtaining electromyographic amplitude, root mean square, and spectrum median, calculating the rate of change, monitoring muscle fiber mobilization, blood oxygen, lactate metabolism, and heart rate variability, extracting the rate of change of joint angles and moment of inertia, and obtaining a training load reference value. In the present invention, by real-time monitoring and analysis of multiple physiological parameters including electromyographic signals, blood oxygen saturation, and heart rate variability, an accurate training load reference value is established to improve the accuracy of neuromuscular coordination assessment. By comparing the rate of change of electromyographic signals in short and long time windows, the training load is accurately adjusted to reduce fatigue risk. By real-time adjustment of the rehabilitation robot's motion trajectory and optimization of the training rhythm, it is ensured that the training intensity matches the individual's metabolic capacity, and dynamic matching and personalized adjustment of the training plan are achieved, effectively improving the safety and effectiveness of training.

The present application relates to the technical field of training equipment, and discloses a kettle lifting training device and a training method. According to the training load requirement, corresponding counterweights are assembled on the counterweight kettle body. The counterweight kettle body is internally provided with a damping mechanism that is close to or away from the guide column. The counterweight kettle body is slidably sleeved on the guide column and is located at the bottom of the guide column. While the counterweight kettle body is lifted, the damping force between the damping mechanism and the guide column is released, and the kettle lifting training is performed up and down along the guide column. After the training interval or the training is completed, the counterweight kettle body is loosened, and the damping force is generated between the damping mechanism and the guide column. The counterweight kettle body slowly slides down or remains in the position when the counterweight kettle body is loosened. The present application can assist in stabilizing the lifting direction of the trainer and controlling the falling range of the kettle body by the guide column. After the hand is released, the damping deceleration effect is achieved, the trainer is protected, and the trainer is not injured due to the rapid falling of the counterweight kettle body. The present application is suitable for home or outdoor training.

To provide a training device permitting efficient training in accordance with practical movements of sports.SOLUTION: A training device 1 includes: a cable 4 delivered with movements of a user; a reel 5 for winding the cable 4; a rotary encoder 8 for detecting a rotational state of the reel 5; a magnetic viscous fluid device 9 for imparting a brake force to the reel 5; and a control unit 10 for calculating a stroke quantity of the cable 4 based on an output value of the rotary encoder 8. The control unit 10 changes the size of a training load in the process of movements by controlling the brake force based on the stroke quantity.SELECTED DRAWING: Figure 1

The utility model provides a self-weight type canoeing action correction training device. The self-weight type canoeing action correction training device comprises a rack; the carrier is arranged on the top of the rack, the carrier is movably arranged in the set direction, and the set direction is obliquely arranged relative to the horizontal direction; the scribing rod assembly comprises a scribing rod and a flexible connecting piece, the flexible connecting piece is connected with the scribing rod and the rack, the connecting position between the flexible connecting piece and the rack is located on one side, in the width direction, of the carrier and is close to the moving path of the carrier, and the width direction of the carrier is perpendicular to the set direction. According to the self-weight type canoeing action correction trainer, a user can conduct single-side canoeing action training without external force interference through the rowing rod assembly, the trainer can well correct the canoeing action of the user, meanwhile, the weight of the user serves as a training load for canoeing action training, a resistance structure does not need to be additionally arranged, and the training efficiency is improved. And the structure of the trainer is simplified and the cost is low.

The invention discloses a sports event risk prediction method and system based on a risk level model, and belongs to the technical field of risk prevention and control. The method comprises the following steps: acquiring a historical athlete injury case data set; acquiring multiple data in each historical damage case, and generating a rule containing conditional statements and risk level adjustment values; acquiring multiple body data of the target athlete, and performing preprocessing; obtaining a total risk exposure degree integral based on the preprocessed multiple body data; and if the total risk exposure integral is not less than a preset integral threshold, generating a high risk level and sending an early warning instruction. According to the method, the body data and the training load of the athlete are comprehensively evaluated, the quantitative rule is generated in combination with the historical injury cases, the accurate and dynamic injury risk early warning effect is achieved, and the problems that in the prior art, risk evaluation is not accurate, and early warning is not timely are solved.

This invention discloses a method and system for constructing a data chain for a scientific fitness guidance platform, relating to the field of medical and health information technology. The method collects multi-source data, including users' basic physical attributes, medical and health data, physical fitness indicators, exercise intensity, and training completion, and performs timestamp alignment and unified access to construct a continuous and complete fitness data chain. It calculates the exercise load safety coefficient, exercise ability adaptation coefficient, and stage intervention effect coefficient, and sequentially constructs mechanisms for exercise load safety assessment, exercise ability adaptation analysis, and stage intervention effect assessment to dynamically determine exercise safety risks, training program matching degree, and stage intervention effectiveness. Furthermore, it automatically generates collaborative adjustment strategies for training load, training structure, and stage parameters, and uses the adjustment results as initial conditions for the next stage of training and data collection, achieving closed-loop feedback control, thereby improving the safety, adaptability, and continuous effectiveness of fitness guidance.

The invention provides intelligent resistance training equipment and method. The method comprises the following steps: synchronously collecting resistance signals of a neck and lower limbs when a user executes a collaborative resistance action; determining force application delay time between the neck and the lower limbs according to the resistance signals of the neck and the lower limbs; according to the angular velocity component of the neck of the user and the anti-resistance signal of the neck, determining a posture-force coupling coefficient when the user executes the collaborative anti-resistance action; determining a load adjustment coefficient of a lower limb target load based on the force application delay time and the posture-force coupling coefficient in combination with a preset load adjustment rule; and adjusting the lower limb load applied by the equipment according to the load adjustment coefficient, and generating prompt information containing posture coordination abnormity at the same time. By adopting the scheme of the invention, the training load conforming to the current coordination capability of the user can be dynamically allocated.

The utility model discloses an assembly capable of adjusting the hanging height of a barbell rod and fitness equipment. The assembly comprises a barbell guide rod, a sliding piece, a barbell rod, a mounting column and a hanging bolt assembly. The barbell rod is coupled with the barbell guide rod through a sliding piece, and the barbell rod is connected with two hooks; a plurality of positioning holes are formed in the mounting column; the hanging bolt assembly comprises a U-shaped supporting piece, a positioning part combined with the U-shaped supporting piece, an auxiliary supporting part and a hanging bolt. The U-shaped supporting piece is operably mounted at different heights of the mounting column through the matching of the positioning part and the positioning hole and the matching of the auxiliary supporting part and the mounting column; the barbell rod is rotated to enable the hook and the hanging bolt to realize hook-bolt connection, so that the barbell rod is hung at a fixed position. The fitness equipment is composed of a frame, the assembly and a balance weight assembly, and the balance weight assembly is used for flexibly adjusting training loads. According to the structure, the manufacturing cost of the fitness equipment is reduced, a user can conveniently and independently use the fitness equipment, and meanwhile use flexibility is kept.

The invention discloses an individualized intelligent reminding method and system for riding training management, and the method comprises the steps: determining an intraday individualized tolerance boundary which is used for limiting the relation between physiological load and training intensity based on user basic data, historical training records and multi-source state data obtained in a preset collection stage; real-time streams of multi-source state data are obtained in the riding process, normalization processing is conducted on the real-time streams according to context factors related to the environment and working conditions, and standardized state vectors are formed in a unified evaluation space; estimating the residual training load according to the standardized state vector; determining a target operation type and a trigger condition thereof according to a preset decision rule; and after the riding is finished, updating the parameters of the individualized tolerance boundary and the decision rule on the current day based on the response of the user to the prompt, and using an updating result for subsequent riding. According to the technical scheme, the safety and efficiency of riding training are improved.

1. Name of the design product: weight-bearing bracelet. 2. Use of the design product: for increasing the training load. 3. Design points of the design product: in shape. 4. Picture or photo best indicating the design points: perspective view.

The invention relates to a rehabilitation training scheme generation method and device, electronic equipment and a medium, and relates to the field of data processing.The method comprises the steps that the maximum exercise load power of a patient during load testing on a power vehicle and a heart rate change diagram and a respiratory rate change diagram of the patient in the testing process are obtained; fusing the heart rate change diagram and the respiratory rate change diagram to obtain a state change diagram of the patient, determining state quality corresponding to a plurality of preset power intervals of the patient from the state change diagram, and determining the number of training stages and the training load power of each training stage based on the state quality and the maximum exercise load power, and generating a rehabilitation training scheme of the patient based on the training load power of each training stage. The method has the effect of accurately formulating the rehabilitation training scheme more suitable for the self condition of the patient.

The present application relates to swimming training equipment technical field, disclose a kind of swimming training device with flow rate self-adapting regulation function and control method.The device includes pool body, partition plate, adjusting mechanism and propulsion module;The upper side of the inner wall of the front end of pool body is provided with water outlet, the lower side of the inner wall of the front end is provided with backwater port, the upper side of the inner wall of the rear end is provided with pool tail drain port;Partition plate is set in the inside of pool body, for dividing pool body into upper training water area and lower function water area, one end of partition plate close to pool tail drain port is provided with throttling window component;The water outlet end of propulsion module is communicated with water outlet, suction end is respectively communicated with backwater port and pool tail drain port, to constitute pool tail bypass backflow path and lower backflow path.Adjusting partition plate position and throttling window component opening, to adjust the tail discharge capacity of training main flow and training load.The present application can improve the problem that pool tail discharge is not timely under short pool body working condition, and it is easy to form stagnation and backflow interference.

The present invention discloses a sports fitness training cloud data collection system and method, which relates to the technical field of fitness data collection, and comprises the following steps: collecting sports fitness training data from multiple data sources; dividing participants with similar characteristics into the same group, calculating the change amount of each body part data of the participants in each group before and after training, and determining the body parts affected by each sports fitness training type; screening out a group with similar characteristics to a user from a sub-data set, and generating an initial training plan in combination with the gap between the user's training goal and current physical fitness level; calculating the total training load of each body part, adjusting the initial training plan, and generating a final training plan; collecting sports fitness training cloud data according to the body part corresponding to each sports fitness training in the final training plan; not only ensuring the personalized implementation of the training plan, but also improving the accuracy and effectiveness of the data.

The utility model discloses a kind of body-building equipment of high-strength recovery training belongs to body-building equipment technical field.A kind of body-building equipment of high-strength recovery training, including frame, counterweight swing lever being arranged in frame, frame has control area and counterweight area, counterweight swing lever one end is movably connected in control area, counterweight swing lever other end extends to inside counterweight area, counterweight adjusting part is hinged with counterweight swing lever;Flexible connecting piece, winding part for winding flexible connecting piece is rotatably arranged in control area, one end of flexible connecting piece away from winding part is connected between the movable end and hinged end of counterweight swing lever;The utility model can swing with swing arm and converter by arm pressing down pressing rod, flexible connecting piece can be wound to realize the lifting of counterweight swing lever in swing process, realize training with load, after counterweight adjusting part, it is convenient to adjust the size of training load, can realize independent exercise operation, and has positive influence on patient consciously exercising.

An adaptive sports analytics system for personalized running training and performance monitoring, wherein the system (100) comprises: a module (1) for capturing the runner profile and baseline values, configured to capture runner information, training history, targets, injury history, availability, access to equipment, and baseline values ​​from functional tests; a module (2) for generating adaptive training plans, configured to create a personalized running and functional training plan based on the runner information and baseline values ​​from functional tests; a module (3) for functional training content and progress, configured to store and provide training records, including exercises for core stability, balance, flexibility, mobility, strength, and power;a module (4) for performance monitoring using wearables and sensors, configured to receive running performance and compliance data from at least one wearable, mobile device, sensor, treadmill, or manual input interface; a fatigue, recovery, and injury risk analysis engine (5) designed to analyze signals related to training load, recovery, pain, compliance, and performance to generate an adjustment recommendation or risk indicator; a functional outcome and progress analysis module (6) designed to store baseline and follow-up assessment values ​​and generate progress reports; and a user interface, coach dashboard, and feedback management module (7) designed to display personalized training sessions, exercise guidance, progress reports, alerts, and updated training recommendations.

The invention discloses a health suggestion generation method and device, a medium and a program product, and relates to the technical field of sports health data calculation and intelligent evaluation. The health suggestion generation method comprises the following steps: calculating the physical ability of a user on that day according to the training load of the user on that day and the potential physical ability of the user, and calculating the fatigue of the user on that day according to the fatigue of the user on that day and the training load of the user on that day; determining a user training state value and a fusion coefficient according to the ratio of the user's physical ability on the day to the user's fatigue on the day and the user portrait; summing a product value of the user sleep influence value and the fusion coefficient with a user training state value to obtain a user physical state evaluation value; and displaying a health suggestion generated based on the user physical state evaluation value. According to the technical scheme provided by the embodiment of the invention, the health suggestions accurately fitting the actual body state of the user can be given, and the user satisfaction is greatly improved.

The utility model relates to a freestyle swimming land one-to-one simulation training device. The freestyle swimming land one-to-one simulation training device comprises a base, a supporting frame, a tension trainer, a trunk supporting pad and foot supporting points (foot fasteners). The tension training device and the trunk supporting pad are both installed on the base through a supporting frame. The tension trainer is arranged at the front end of the device; the trunk supporting pad is arranged in the middle of the device and can support the cross beam to rotate, so that the heeling action of the trunk part can be cooperatively trained while the strength and posture of the arm part are trained at the front part; the foot training mechanism at the rear part of the device is used for synchronously training leg beating actions by controlling and training feet, so that the whole body strength is balanced and coordinated; the foot fasteners provide training loads for the leg beating action, and all-around coordination training from the arms to the body and then to the legs and the feet can be achieved. The whole body is exercised at the same time, the overall balance degree is trained, and the up-and-down coordination ability of the whole body is improved.

The invention discloses a cerebral apoplexy lower limb intelligent rehabilitation training system, which relates to the technical field of rehabilitation instruments and comprises a walking module, a body fixing bracket, a foot fixer, a leg adjusting module, a waist and shoulder auxiliary fixing module and a training posture recognition and correction module. According to the training posture recognition and correction module, a binocular 3D camera is innovatively adopted to collect data, a Voxel CNN network and multi-constraint energy function optimization are combined, high-precision joint coordinate recognition, joint angle calculation and motion trajectory similarity evaluation are achieved, and a self-adaptive admittance control algorithm based on performance is matched, so that the precision of the robot is improved. The training load and the auxiliary force can be dynamically adjusted in real time according to the rehabilitation ability of a patient, customized rehabilitation training of one person and one strategy is achieved, the one-step training mode of a traditional instrument is thoroughly changed, rehabilitation training better meets the individual requirements of the patient, and the rehabilitation effect and the training efficiency are remarkably improved.

This application relates to the field of intelligent motion analysis technology, and discloses a multi-dimensional data-driven method and system for volleyball teaching effect feedback. The method includes: collecting real-time training motion and ball-hitting trajectory data to form a raw dataset; extracting features from the raw dataset and analyzing the training motion chain to obtain motion analysis results; if the motion deviation exceeds a threshold, classifying the data to determine a problem classification label; combining the problem classification label with subjective feelings and training load data to generate a personalized feedback report; monitoring subsequent training data according to the feedback report and updating the motion analysis results; comparing the motion analysis results before and after the update to calculate a continuity evaluation index; and generating a visualization chart based on the continuity evaluation index to determine a personalized training optimization path. This application solves the problems of delayed feedback, fragmented data, and lack of continuous evaluation in traditional volleyball teaching, improving the accuracy of volleyball motion diagnosis, the personalization level of training programs, and overall teaching efficiency.

This utility model discloses a tongue tip trainer for speech and oral motor rehabilitation, including a support and a tongue muscle elastic training component. The support has a training space; the tongue muscle elastic training component includes a tongue tip component and an elastic component. The tongue tip component is located within the training space and is connected to the support via the elastic component. When the user's tongue presses against the tongue tip component, the tongue tip component moves within the training space, forcing the elastic component to undergo elastic deformation, thereby providing motion resistance. The tongue tip trainer for speech and oral motor rehabilitation provided by this utility model utilizes the pressing action of the tongue against the tongue tip component during use, enabling the tongue tip component to move within the training space and simultaneously forcing the elastic component to undergo elastic deformation, thus providing motion resistance. This motion resistance can improve the tongue muscle training effect. Furthermore, it can provide variable training loads to meet the different oral motor rehabilitation needs of various users, thereby effectively improving training results and satisfying diverse training requirements.

The invention discloses a dance training load prediction method and system based on training data. The method comprises the following steps: S1, acquiring a basic heart rate of a dancer; s2, calculating the metabolic equivalent of the dancer for calculating a heart rate load value, and carrying out endurance training to obtain the maximum heart rate value in the training process; s3, calculating a heart rate load prediction index of the dancer according to the basic heart rate; s4, performing standardization processing on the obtained heart rate load prediction indexes; s5, predicting a heart rate load value through a linear regression equation; and S6, calculating energy consumption according to the heart rate data in the exercise process. According to the dance training load prediction method and system based on the training data, the heart rate data of the dancer can be calculated for subsequent prediction of the heart rate load of the dancer and prediction of the energy consumption of the dancer, and personalized training plans can be made for different dancers.

The present application belongs to the field of medical care information technology, and particularly relates to a movement risk assessment method based on machine learning, wherein historical data and injury records are arranged to obtain a data set, high correlation indexes related to various movement injuries are obtained through feature screening, and independent regression models are established between the high correlation indexes and the risk of various injuries; the physical quality and physiological index system data of a new training personnel group are obtained through testing, the regression models are used to predict the injury risk indexes of each part, the training personnel are grouped and trained through K-means clustering algorithm according to the injury risk indexes; and the historical training load data, functional action screening scores and the next day training plan content are fused to construct a model to evaluate the injury risk index of the training personnel on the next day, so that the present method is suitable for large-scale training groups, the index dimensions are comprehensive and targeted, and the accuracy and efficiency of risk assessment are improved.