Exercise prescription recommendation system and method for health data management

Abstract

The invention provides an exercise prescription recommendation system and method for health data management. The system comprises a search space construction module used for obtaining exercise prescription index data, performing exhaustion combination on the exercise prescription index data, and constructing a search space based on the exercise prescription index data after exhaustion combination; the fitness function acquisition module is used for acquiring the real-time physical index data from the intelligent terminal and designing a fitness function of a particle swarm algorithm according to the real-time physical index data; and the exercise prescription recommendation module is used for searching an optimal solution in the search space by using a particle swarm algorithm based on the fitness function, and recommending an exercise prescription corresponding to the searched optimal solution. According to the method, the search space is constructed through exhaustion combination, potential effective exercise prescriptions cannot be missed when the particle swarm optimization is used for searching in the search space, and the exercise prescription found based on the search space and the fitness function is the optimal exercise prescription obtained according to real-time physical indexes and exercise prescription index data.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of sports rehabilitation, and particularly relates to a sports prescription recommendation system and method for health data management. BACKGROUND

[0002] Sports prescription combined with physical medicine is a sports program combining sports science and medical principles, aiming to help individuals improve their physical condition and prevent or treat certain diseases. It is a systematized and personalized fitness program formulated by a doctor or sports worker for an exerciser according to age, gender, health status, exercise experience, and function level of the heart-lung or motor organs in the form of a prescription. Its characteristic is to treat the disease according to the individual, and the prescription is different for different people. Sports prescription is different from ordinary physical exercise and general treatment method. Sports prescription is a targeted, purposeful, selective and controlled exercise therapy. A sports prescription generally includes exercise type, exercise intensity, exercise time, exercise frequency, exercise progress and matters needing attention. The formulation of sports prescription should be strictly in accordance with the system, and generally divided into four stages: testing, formulation, execution and feedback. The most critical thing in formulating a sports prescription is to conduct a systematic test on the exerciser or patient to obtain comprehensive data required for formulating a sports prescription. The test generally includes examination of the motor system, examination of the cardiovascular system, examination of the respiratory system, examination of the nervous system, etc. After testing, relevant personnel formulate different sports prescriptions for different objects. The types of sports prescriptions can be roughly divided into therapeutic sports prescriptions, which are mainly aimed at treating diseases and improving rehabilitation effects; preventive sports prescriptions, which are mainly aimed at enhancing physical fitness, preventing diseases and improving health level; fitness and beauty sports prescriptions, which are mainly aimed at improving physical fitness, sports ability and beauty. At the same time, the formulation of sports prescription should be based on the specific situation of each exerciser or patient to formulate a sports prescription that meets the individual's objective conditions and requirements. In the execution stage of sports prescription, in addition to sticking to exercise, it is necessary to fully communicate with the prescription formulator to let the formulator understand the body feedback so as to make subsequent adjustments. After the execution of a cycle, the executor should timely feedback the information to the prescription formulator, and the feedback purpose is to adjust. That is, according to the changes and physical conditions of the executor, the sports prescription is adjusted according to the new data through retesting and other ways. In order to facilitate information management and updating, a health data management platform can be established to obtain the personal information, health status, exercise habits and other data of the user. The sports prescription formulator can complete the formulation of sports prescription by setting and modifying the above data indicators.

[0003] Currently, many platforms and companies have developed data management platforms to simplify the management of users' exercise prescription data. After acquiring users' personal information, health status, and exercise habits, these platforms can set and modify these elements to create exercise prescriptions. Furthermore, the platform should record users' exercise data and health indicators, presenting them in charts and other formats to help users understand their health status and recovery progress.

[0004] However, when existing health data management platforms create exercise prescriptions, once a patient selects an exercise prescription, the patient will continue to follow that prescription. However, a patient's health status is constantly changing. If the exercise prescription provided by the platform cannot be adjusted in a timely and targeted manner, it will be difficult to achieve the desired effect on the patient's rehabilitation. In other words, traditional fitness platform systems cannot automatically adjust exercise programs based on the exerciser's exercise feedback, nor can they regularly evaluate the exercise effects produced by the exercise program. Summary of the Invention

[0005] This invention provides an exercise prescription recommendation system and method for health data management, which can solve the technical problem in the prior art that the system cannot automatically adjust the exercise plan based on the exerciser's exercise feedback.

[0006] This invention provides an exercise prescription recommendation system for health data management, comprising a smart terminal and a central controller, wherein the central controller includes:

[0007] The search space construction module is used to obtain exercise prescription index data, exhaustively combine the exercise prescription index data, and construct a search space based on the exhaustively combined exercise prescription index data.

[0008] A fitness function acquisition module is used to acquire real-time physical fitness index data from a smart terminal and design the fitness function of the particle swarm algorithm based on the real-time physical fitness index data.

[0009] The exercise prescription recommendation module is used to search for the optimal solution in the search space based on the fitness function using the particle swarm optimization algorithm, and then recommend the exercise prescription corresponding to the optimal solution.

[0010] Furthermore, the search space construction module constructs the search space, including:

[0011] The search space construction module retrieves exercise prescription indicator data:

[0012] S = {s1,…s} i ,…,s n}

[0013]

[0014] Among them, s1~s n These are different exercise prescription indicators, where n is the total number of exercise prescription indicators; It is an exercise prescription indicator. i All options, m i It is an exercise prescription indicator. i The number of options;

[0015] The search space building module exhaustively combines options for different exercise prescription indicators:

[0016]

[0017] ...

[0018]

[0019] in, It is all exercise prescriptions obtained by exhaustively combining options of different exercise prescription indicators;

[0020] The search space construction module treats each motion prescription as the position of a particle in the particle swarm optimization algorithm, forming the search space.

[0021] Furthermore, the fitness function of the particle swarm optimization algorithm designed in the fitness function acquisition module is:

[0022]

[0023] Among them, B E B T These are the target physical fitness index data and the real-time physical fitness index data obtained by the smart terminal at time T; ω i It is the weighting coefficient of the i-th exercise prescription indicator.

[0024] Furthermore, the exercise prescription recommendation module recommends the exercise prescription corresponding to the optimal solution found in the search, including:

[0025] The exercise prescription recommendation module uses a particle swarm optimization algorithm to search for the optimal solution in a constructed search space. The particle swarm optimization algorithm includes:

[0026] S1. Initialize the position and velocity of each particle in the particle swarm;

[0027] S2. Calculate the fitness value of each particle based on the fitness function;

[0028] S3. For each particle, compare its current fitness value with the fitness value corresponding to its individual historical best position pbest. If the current fitness value is higher, then update the historical best position pbest with the current position.

[0029] S4. For each particle, compare its current fitness value with the fitness value corresponding to the global best position gbest. If the current fitness value is higher, update the position of the current particle to the global best position gbest.

[0030] S5. Update the velocity and position of each particle;

[0031] S6. If the set termination condition is not met, return to S2. Stop iterating when the number of iterations reaches the maximum number of iterations or the increment of the best fitness value is less than the given threshold.

[0032] The exercise prescription recommendation module recommends the exercise prescription corresponding to the optimal solution.

[0033] Furthermore, the update formula for the velocity and position of each particle in the exercise prescription recommendation module is as follows:

[0034]

[0035]

[0036] in,

[0037] These are the d-th component of the velocity vector of particle i in the k-th and (k-1)-th iterations, respectively.

[0038] These are the d-th component of the velocity vector at position i in the k-th and (k-1)-th iterations, respectively.

[0039] c1 and c2 are acceleration constants used to adjust the maximum learning step size;

[0040] r1 and r2 are random functions used to increase the randomness of the search, with values ​​ranging from [0,1].

[0041] w is the inertial weight used to adjust the search range of the solution space.

[0042] A method for recommending exercise prescriptions for health data management includes:

[0043] Obtain exercise prescription indicator data, exhaustively combine the exercise prescription indicator data, and construct a search space based on the exhaustively combined exercise prescription indicator data.

[0044] Obtain real-time physical fitness data, and design the fitness function of the particle swarm optimization algorithm based on the real-time physical fitness data;

[0045] The particle swarm optimization algorithm is used based on the fitness function to search for the optimal solution in the search space, and the exercise prescription corresponding to the optimal solution is recommended.

[0046] Furthermore, the construction of the search space includes:

[0047] Obtain exercise prescription indicator data:

[0048] S = {s1,…s} i ,…,s n}

[0049]

[0050] Among them, s1~s n These are different exercise prescription indicators, where n is the total number of exercise prescription indicators; It is an exercise prescription indicator. i All options, m i It is an exercise prescription indicator. i The number of options;

[0051] Exhaustive combinations of different exercise prescription indicators:

[0052]

[0053] ...

[0054]

[0055] in, It is all exercise prescriptions obtained by exhaustively combining options of different exercise prescription indicators;

[0056] Each motion prescription is used as the position of a particle in the particle swarm optimization algorithm, forming a search space.

[0057] Furthermore, the fitness function of the particle swarm optimization algorithm is:

[0058]

[0059] Among them, B E B T These are the target physical fitness index data and the real-time physical fitness index data obtained by the smart terminal at time T; ω i It is the weighting coefficient of the i-th exercise prescription indicator.

[0060] Furthermore, the step of recommending the exercise prescription corresponding to the optimal solution found in the search includes:

[0061] The optimal solution is searched in a constructed search space using a particle swarm optimization algorithm, which includes:

[0062] S1. Initialize the position and velocity of each particle in the particle swarm;

[0063] S2. Calculate the fitness value of each particle based on the fitness function;

[0064] S3. For each particle, compare its current fitness value with the fitness value corresponding to its individual historical best position pbest. If the current fitness value is higher, then update the historical best position pbest with the current position.

[0065] S4. For each particle, compare its current fitness value with the fitness value corresponding to the global best position gbest. If the current fitness value is higher, update the position of the current particle to the global best position gbest.

[0066] S5. Update the velocity and position of each particle;

[0067] S6. If the set termination condition is not met, return to S2. Stop iterating when the number of iterations reaches the maximum number of iterations or the increment of the best fitness value is less than the given threshold.

[0068] Recommend the exercise prescription corresponding to the optimal solution.

[0069] Furthermore, the update formula for the velocity and position of each particle is as follows:

[0070]

[0071]

[0072] in,

[0073] These are the d-th component of the velocity vector of particle i in the k-th and (k-1)-th iterations, respectively.

[0074] These are the d-th component of the velocity vector at position i in the k-th and (k-1)-th iterations, respectively.

[0075] c1 and c2 are acceleration constants used to adjust the maximum learning step size;

[0076] r1 and r2 are random functions used to increase the randomness of the search, with values ​​ranging from [0,1].

[0077] w is the inertial weight used to adjust the search range of the solution space.

[0078] This invention provides an exercise prescription recommendation system and method for health data management, which has the following advantages compared with the prior art:

[0079] The particle swarm optimization (PSO) algorithm designed based on real-time physical fitness data in this invention can intelligently adjust exercise prescriptions to better suit individual conditions. Furthermore, the exhaustive search space constructed in this invention ensures the comprehensiveness of the exercise prescriptions; the PSO algorithm will not miss any potential effective exercise prescriptions when searching this space. The exercise prescriptions found based on this search space and the fitness function are the optimal exercise prescriptions obtained from real-time physical fitness data and exercise prescription indicator data. Attached Figure Description

[0080] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0081] In the attached diagram:

[0082] Figure 1 This is the flowchart provided in this manual;

[0083] Figure 2 This is a schematic diagram of the search space construction provided in this manual;

[0084] Figure 3 This is the particle swarm optimization algorithm flowchart provided in this manual. Detailed Implementation

[0085] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. However, it should be understood that the scope of protection of the present invention is not limited to the specific implementation.

[0086] Example

[0087] An exercise prescription recommendation system for health data management includes a smart terminal and a central controller, wherein the central controller includes:

[0088] Search space building module

[0089] The search space construction module obtains exercise prescription indicator data, exhaustively combines these indicators, and constructs the search space based on this exhaustive combination. The module dynamically builds the search space using the exercise prescription indicator data. This means the system can more flexibly consider different exhaustive combinations of exercise to adapt to the user's current needs. As users' health goals and physical conditions may change over time, real-time updates to the search space help the system remain sensitive to these changes. This includes:

[0090] Exercise prescription index data acquisition unit

[0091] The exercise prescription index data acquisition unit acquires a set of exercise prescription indexes S:

[0092] S = {s1,…s} i ,…,s n}

[0093]

[0094] Among them, s1~s n These are different exercise prescription indicators, which may involve the type, intensity, frequency, duration, etc. of exercise; n is the total number of exercise prescription indicators. It is an exercise prescription indicator. i All options, such as sports type options, include running, swimming, Tai Chi, fitness, badminton, etc.; m i It is a dynamic prescription index s i The number of options.

[0095] Search space building unit

[0096] like Figure 2 As shown, the search space construction unit exhaustively combines options for different exercise prescription indicators to generate multiple exhaustive combinations of exercise prescriptions. Each possible exhaustive combination represents a potential exercise prescription.

[0097]

[0098] ...

[0099]

[0100] in, It is all exercise prescriptions obtained by exhaustively combining options of different exercise prescription indicators. One of the exercise prescriptions may be: {swimming, high intensity, once a week, 2 hours each time, ...}.

[0101] Each motion prescription obtained after exhaustive combination is used as the position of a particle in the particle swarm optimization algorithm, forming the search space.

[0102] Fitness function acquisition module

[0103] The fitness function acquisition module is used to design the fitness function for the particle swarm optimization algorithm based on physical fitness data obtained from the smart terminal. The fitness function should be able to evaluate the quality of each exercise prescription, i.e., whether the exercise prescription helps the user achieve their health goals. Through the fitness function acquisition module, the system can better adapt to the user's current health condition and goals. Different people may need different exercise prescriptions at different times, so real-time updates can provide more personalized suggestions. A user's physical condition and health status may change over time. Through real-time monitoring and updates, the system can reflect these changes promptly and adjust the exercise prescription accordingly. This helps ensure that the user is consistently engaging in exercise that is most beneficial to them.

[0104] Unit for obtaining real-time physical fitness data

[0105] Collect users' real-time physical fitness data B T This may include physiological indicators such as height, weight, blood pressure, and heart rate. Obtain the user's target physical fitness data B. E Physiological indicators such as height, weight, blood pressure, and heart rate after weight loss, increased muscle mass, and lowered blood pressure.

[0106] The fitness function for the particle swarm optimization algorithm includes:

[0107]

[0108] Among them, B E B T These are the target physical fitness index data and the real-time physical fitness index data obtained by the smart terminal at time T. ω i ω is the weight coefficient of the i-th exercise prescription indicator, n is the total number of exercise prescription indicators, and ω i is the weighting coefficient of the i-th exercise prescription indicator data. This weighting coefficient can be obtained by fitting historical data using a linear regression method, i.e.:

[0109] Historical data of each exercise prescription indicator and the corresponding physical fitness indicator data are collected. Each exercise prescription indicator is used as input, and the regression coefficient is obtained by minimizing the difference between the predicted value of the physical fitness indicator data in the regression model and the actual value of the corresponding physical fitness indicator data. The average of these regression coefficients is used as the weight coefficient of the corresponding exercise prescription indicator data.

[0110] The fitness function formula aims to minimize the difference between the target physical fitness data and the current physical fitness data, while also taking into account the influence of a set of exercise indicators. Specifically:

[0111] Target fitness data refers to the ideal or desired fitness level that is desired to be achieved, potentially representing a desired level of health or physical condition. Current fitness data refers to the current fitness level, representing the current health or physical condition. Exercise index 1... Exercise index n is a set of exercise indicators, which may include different types of exercise, exercise intensity, duration, etc., used to describe exercise behavior. ω i ......ω n These are the weights corresponding to the exercise indicators, used to determine the importance of each indicator in the overall assessment. The numerator of the formula represents the desired minimization of the difference between the target physical fitness data and the current physical fitness data. This can be understood as the desire to approach or achieve the target physical fitness state by changing exercise behavior. The denominator represents a weighted sum of the exercise indicators, with the weights reflecting the relative importance of each indicator. The entire denominator can be seen as a comprehensive measure of the effect or contribution of exercise. Dividing these two parts, the entire formula can be understood as the desire to minimize the difference between the target physical fitness data and the current physical fitness data by rationally selecting exercise behavior, taking into account the influence of the weights of different exercise indicators. During the optimization process, the system will attempt to adjust exercise behavior to minimize this ratio, thereby bringing the physical fitness data closer to the target.

[0112] Exercise prescription recommendation module

[0113] The exercise prescription recommendation module uses a particle swarm optimization (PSO) algorithm based on a fitness function to search for the optimal solution within the search space and recommends the corresponding exercise prescription to the user. PSO is a heuristic optimization algorithm inspired by simulating the behavior of social groups such as flocks of birds or schools of fish. It was first proposed by James Kennedy and Russell Eberhart in 1995. In PSO, candidate solutions are called "particles," which search for the optimal solution by moving within the solution space. Each particle has a position and velocity, adjusted based on its individual and group experience. The behavior of the entire swarm is simulated by mimicking the movement of each particle in the search space. The basic idea is as follows:

[0114] Initialization: A certain number of particles are randomly generated in the search space, each particle having a random initial position and velocity.

[0115] Evaluation: Calculate the fitness of each particle, i.e., the value of the objective function.

[0116] Update individual optimal position: For each particle, update its individual optimal position based on its individual experience and current position.

[0117] Update the swarm optimal position: Update the global optimal position of the entire swarm based on the fitness of all particles.

[0118] Update velocity and position: Update the velocity and position of each particle according to certain rules.

[0119] Repeat: Repeat the above steps until the stopping condition is met, such as reaching the maximum number of iterations or finding a satisfactory solution.

[0120] In updating their velocity and position, particles are influenced by two factors: individual experience and swarm experience. Individual experience makes particles tend to maintain their historically optimal positions, while swarm experience makes them tend to move to better positions within the swarm. In this way, the entire swarm can collaboratively search the solution space to find the global optimum. Particle swarm optimization (PSO) is widely used in function optimization, neural network training, image processing, machine learning, and other fields, and is particularly suitable for high-dimensional, nonlinear, and complex optimization problems.

[0121] Particle Swarm Optimization Search Unit:

[0122] Each exhaustive combination of exercise prescriptions is treated as a particle, and a swarm of particles is randomly initialized in the search space. The particle swarm optimization (PSO) algorithm is used to update the position and velocity of each particle to find the optimal solution in the search space. Through continuous iteration, the PSO algorithm finds the exercise prescription with the optimal fitness function value. Based on the optimal solution found by the PSO algorithm, a corresponding exercise prescription is recommended to the user. The optimal exercise prescription should be an exercise program with the best fitness while meeting the user's physical and health goals. The exercise prescription recommendation module uses the PSO algorithm to search for the optimal solution in a constantly updated search space. In this way, the system can find the most beneficial exercise prescription for the user based on the latest information. Real-time updates help avoid using outdated or no longer applicable exercise prescriptions. Specifically, this includes:

[0123] In D-dimensional space, there are N particles (corresponding to n motion indices). (One exercise prescription):

[0124] Position of particle i: x i ={x i1 ,…x i2 ,…,x iD}, x i Substitute the values ​​into the fitness function to find the fitness value;

[0125] The velocity of particle i: v i ={v i1 ,…v i2 ,…,v iD};

[0126] The best position for particle i: pbesti ={p i1 ,…p i2 ,…,p iD};

[0127] The best location visited by the population: gbest={g1,…g2,…,g D};

[0128] Typically, the range of positional variation in the d-th dimension (1≤d≤D) is limited to [X]. min,d ,X max,d Within; the range of velocity variation is limited to [-V] min,d V max,d Within, that is, if v during the iteration process id X id If the boundary value is exceeded, the velocity or position of that dimension is restricted to the maximum velocity or boundary position of that dimension.

[0129] like Figure 2 As shown, the iterative steps for finding the particle swarm optimization algorithm include:

[0130] S1. Initialize the position and velocity of each particle in the particle swarm;

[0131] S2. Calculate the fitness value of each particle based on the fitness function;

[0132] S3. For each particle, compare its current fitness value with the fitness value corresponding to its historical best position pbest. If the current fitness value is higher, then update the historical best position pbest with the current position.

[0133] S4. For each particle, compare its current fitness value with the fitness value corresponding to the global best position gbest. If the current fitness value is higher, update the current particle's position to the global best position gbest.

[0134] S5. Update the velocity and position of each particle according to the formula:

[0135] The formula for updating the d-th dimension velocity of particle i is:

[0136]

[0137] The formula for updating the d-th dimension position of particle i is:

[0138]

[0139] in, These are the d-th component of the velocity vector of particle i in the k-th and (k-1)-th iterations, respectively.

[0140] These are the d-th component of the velocity vector at position i in the k-th and (k-1)-th iterations, respectively.

[0141] c1 and c2 are acceleration constants used to adjust the maximum learning step size;

[0142] r1 and r2 are random functions used to increase the randomness of the search, with values ​​ranging from [0,1].

[0143] w is the inertial weight used to adjust the search range of the solution space.

[0144] S6. If the termination condition is not met, return to S2. Stop iterating when the number of iterations reaches the maximum number of iterations or the increment of the best fitness value is less than the given threshold.

[0145] like Figure 1 As shown, the specific limitations of the exercise prescription recommendation method for health data management can be found in the limitations of exercise prescription recommendation for health data management mentioned above, and will not be repeated here.

[0146] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. An exercise prescription recommendation system for health data management, comprising a smart terminal and a central controller, characterized in that, The central control includes: The search space construction module is used to obtain exercise prescription index data, exhaustively combine the exercise prescription index data, and construct a search space based on the exhaustively combined exercise prescription index data. A fitness function acquisition module is used to acquire real-time physical fitness index data from a smart terminal and design the fitness function of the particle swarm algorithm based on the real-time physical fitness index data. The exercise prescription recommendation module is used to search for the optimal solution in the search space based on the fitness function using the particle swarm optimization algorithm, and then recommend the exercise prescription corresponding to the optimal solution.

2. The exercise prescription recommendation system for health data management according to claim 1, characterized in that, The search space construction module constructs the search space, including: The search space construction module retrieves exercise prescription indicator data: S={s1,…s i ,…,s n } Among them, s1~s n These are different exercise prescription indicators, where n is the total number of exercise prescription indicators; It is an exercise prescription indicator. i All options, m i It is an exercise prescription indicator. i The number of options; The search space building module exhaustively combines options for different exercise prescription indicators: …… in, It is all exercise prescriptions obtained by exhaustively combining options of different exercise prescription indicators; The search space construction module treats each motion prescription as the position of a particle in the particle swarm optimization algorithm, forming the search space.

3. The exercise prescription recommendation system for health data management according to claim 2, characterized in that, The fitness function of the particle swarm optimization algorithm designed in the fitness function acquisition module is: Among them, B E B T These are the target physical fitness index data and the real-time physical fitness index data obtained by the smart terminal at time T; ω i It is the weighting coefficient of the i-th exercise prescription indicator.

4. The exercise prescription recommendation system for health data management according to claim 3, characterized in that, The exercise prescription recommendation module recommends the exercise prescription corresponding to the optimal solution found in the search, including: The exercise prescription recommendation module uses a particle swarm optimization algorithm to search for the optimal solution in a constructed search space. The particle swarm optimization algorithm includes: S1. Initialize the position and velocity of each particle in the particle swarm; S2. Calculate the fitness value of each particle based on the fitness function; S3. For each particle, compare its current fitness value with the fitness value corresponding to its individual historical best position pbest. If the current fitness value is higher, then update the historical best position pbest with the current position. S4. For each particle, compare its current fitness value with the fitness value corresponding to the global best position gbest. If the current fitness value is higher, update the position of the current particle to the global best position gbest. S5. Update the velocity and position of each particle; S6. If the set termination condition is not met, return to S2. Stop iterating when the number of iterations reaches the maximum number of iterations or the increment of the best fitness value is less than the given threshold. The exercise prescription recommendation module recommends the exercise prescription corresponding to the optimal solution.

5. The exercise prescription recommendation system for health data management according to claim 4, characterized in that, The update formula for the velocity and position of each particle in the exercise prescription recommendation module is as follows: in, These are the d-th component of the velocity vector of particle i in the k-th and (k-1)-th iterations, respectively. These are the d-th component of the velocity vector at position i in the k-th and (k-1)-th iterations, respectively. c1 and c2 are acceleration constants used to adjust the maximum learning step size; r1 and r2 are random functions used to increase the randomness of the search, with values ​​ranging from [0,1]. w is the inertial weight used to adjust the search range of the solution space.

6. A method for recommending exercise prescriptions for health data management, characterized in that, include: Obtain exercise prescription indicator data, exhaustively combine the exercise prescription indicator data, and construct a search space based on the exhaustively combined exercise prescription indicator data. Obtain real-time physical fitness data, and design the fitness function of the particle swarm optimization algorithm based on the real-time physical fitness data; The particle swarm optimization algorithm is used based on the fitness function to search for the optimal solution in the search space, and the exercise prescription corresponding to the optimal solution is recommended.

7. The method for recommending exercise prescriptions for health data management according to claim 6, characterized in that, The construction of the search space includes: Obtain exercise prescription indicator data: S={s1,…s i ,…,s n } Among them, s1~s n These are different exercise prescription indicators, where n is the total number of exercise prescription indicators; It is an exercise prescription indicator. i All options, m i It is an exercise prescription indicator. i The number of options; Exhaustive combinations of different exercise prescription indicators: …… in, It is all exercise prescriptions obtained by exhaustively combining options of different exercise prescription indicators; Each motion prescription is used as the position of a particle in the particle swarm optimization algorithm, forming a search space.

8. The method for recommending exercise prescriptions for health data management according to claim 7, characterized in that, The fitness function of the particle swarm optimization algorithm is: Among them, B E B T These are the target physical fitness index data and the real-time physical fitness index data obtained by the smart terminal at time T; ω i It is the weighting coefficient of the i-th exercise prescription indicator.

9. The method for recommending exercise prescriptions for health data management according to claim 8, characterized in that, The step of recommending the exercise prescription corresponding to the optimal solution found in the search includes: The optimal solution is searched in a constructed search space using a particle swarm optimization algorithm, which includes: S1. Initialize the position and velocity of each particle in the particle swarm; S2. Calculate the fitness value of each particle based on the fitness function; S3. For each particle, compare its current fitness value with the fitness value corresponding to its individual historical best position pbest. If the current fitness value is higher, then update the historical best position pbest with the current position. S4. For each particle, compare its current fitness value with the fitness value corresponding to the global best position gbest. If the current fitness value is higher, update the position of the current particle to the global best position gbest. S5. Update the velocity and position of each particle; S6. If the set termination condition is not met, return to S2. Stop iterating when the number of iterations reaches the maximum number of iterations or the increment of the best fitness value is less than the given threshold. Recommend the exercise prescription corresponding to the optimal solution.

10. The method for recommending exercise prescriptions for health data management according to claim 9, characterized in that, The update formulas for the velocity and position of each particle are as follows: in, These are the d-th component of the velocity vector of particle i in the k-th and (k-1)-th iterations, respectively. These are the d-th component of the velocity vector at position i in the k-th and (k-1)-th iterations, respectively. c1 and c2 are acceleration constants used to adjust the maximum learning step size; r1 and r2 are random functions used to increase the randomness of the search, with values ​​ranging from [0,1]; w is an inertial weight used to adjust the search range of the solution space.

Images