Intelligent management system of scientific exercise gymnasium based on big data analysis
Through multi-source sensor networks and big data analysis, an intelligent gym management system is built to achieve multi-dimensional data fusion and user value stratification, provide differentiated services and preventive maintenance, solve the problems of low management efficiency and poor user experience in existing gyms, and improve management efficiency and user satisfaction.
Patent Information
- Application Number
- CN202510963897.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-14
- Publication Date
- 2025-10-17
AI Technical Summary
Existing gym management systems rely on a single sensor to collect data, lack multi-dimensional data fusion, have a single user management layer, and rely on manual inspections and post-fault repairs for equipment maintenance. It is difficult to reduce fault analysis costs through preventive warnings, resulting in low management efficiency and poor user experience.
A multi-source sensor network is used to collect multi-dimensional data, build a user movement feature model, conduct user value stratification and equipment health prediction through big data analysis, and combine visualization technology to optimize resource scheduling and achieve differentiated services and preventive maintenance.
Through multi-dimensional data collection and analysis, user experience is improved, equipment downtime is reduced, resource allocation is optimized, operating costs are reduced, and management efficiency and decision-making scientificity are improved.
Smart Images
Figure CN120804888A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application relates to the technical field of intelligent management of gyms, in particular to an intelligent management system of scientific gyms based on big data analysis. BACKGROUND
[0002] With the development of the Internet of Things technology, gym intelligent management, as a leading scheme integrating big data and the Internet of Things technology, is gradually promoting the digital transformation of the fitness industry; through a multi-source sensor network and intelligent analysis algorithms, user exercise data collection, equipment state monitoring and resource scheduling management are realized, thereby providing technical support for the improvement of venue operation efficiency; however, the prior art still has many bottlenecks. The prior art relies on a single sensor to collect data, lacks a multi-dimensional data fusion mechanism, user management is single-layered, lacks deep mining of behavior data, and the device maintenance mode is still manual inspection and post-failure maintenance, which is difficult to reduce the cost of failure analysis through preventive early warning; resulting in low gym management efficiency and poor user experience.
[0003] In order to solve the above-mentioned defects, the technical scheme is provided. SUMMARY
[0004] The purpose of the application is to solve the problem of low gym management efficiency and poor user experience, and the intelligent management system of scientific gyms based on big data analysis is proposed.
[0005] The purpose of the application can be achieved by the following technical scheme: The intelligent management system of scientific gyms based on big data analysis comprises: A data collection module: multi-dimensional data collection is performed through a multi-source sensor network, user exercise data, device parameters and environmental data are obtained and stored in a gym database; A data processing module: used for deep integration and feature processing of the collected multi-dimensional data, and a user exercise feature model is constructed; An intelligent analysis module: analyzes user value, device health and energy consumption, including a user value layering unit, a device health prediction unit and an energy consumption optimization unit; A resource scheduling module: presents the use of gym resources through visual technology to assist management decision-making.
[0006] Further, the specific construction process of the user exercise feature model is as follows: A unified clock source is used to time stamp align multi-dimensional data and convert it into standard international unit values; when communication is interrupted or sensors are temporarily disabled, data missing within 3 hours is filled by linear interpolation or based on the statistical value of the nearest sensor; more than 3 hours is marked as invalid; Bind physiological data and body data through user smart bracelet ID, associate running parameters to users and devices based on device number, call corresponding environment parameters to build complete motion context; Filter data through preset threshold range: When data exceeds preset threshold range , determine as critical abnormality, get abnormality degree through formula , wherein is observation value, is user historical same kind data mean, is standard deviation; when , replace with moving average of front and rear data; When data exceeds preset threshold range , determine as sensor failure or data transmission error, discard data point and record abnormality; The multi-layer architecture of real-time data layer, recent data layer and historical data layer of fitness database realizes efficient data management.
[0007] Further, the specific operation steps of the user value stratification unit include: Extract consumption frequency F, consumption amount M and recent active time R in user behavior data, introduce user duration Y and participation Two extended indicators, get participation through formula , wherein, is course completion rate, is social activity level; is interaction index; are influence weight factors of course completion rate, social activity level and interaction index respectively; After normalization processing of consumption frequency F, consumption amount M, recent active time R, user duration Y and participation , put them into formula: Get user value ; wherein, are influence weight factors of consumption frequency, consumption amount, recent active time, user duration and participation respectively; Based on user value, divide users into diamond, gold, silver and ordinary users, and match differentiated service strategy.
[0008] Further, the specific implementation process of the differentiated service strategy includes: Provide exclusive personalized training plan service for diamond users, get training intensity index through formula ; wherein, represents user historical maximum load; represents a user training target value, represents a current physical fitness level value, represents a fatigue recovery index; are weight factors of historical maximum load, training target value, physical fitness level value and fatigue recovery index, respectively; adjust the training plan based on the training intensity index change trend; Gold medal users provide sports injury risk warning services, collect user training actions by deploying RGB camera arrays, extract three-dimensional coordinates of key skeletal joints based on deep neural networks, and calculate dynamic time warping distances with standard actions to obtain action deviation ; Compare the number of training times in the user cycle time with the industry standard frequency to obtain the training frequency coefficient ; Through the formula obtain the user injury risk index ; wherein represents the load index of the joint; is the risk weight factor of the joint, is the total number of detected joints; are the influence weight factors of training frequency and action deviation, respectively; Take protective measures and adjust the training plan based on the user injury risk index.
[0009] Further, the specific implementation process of the differentiated service strategy also includes: Silver medal users provide intelligent group class recommendation and priority registration services, extract user preference weights by analyzing user behavior data to obtain user preference coefficients ; Through the formula obtain the course matching degree ; wherein, represents the user preference coefficient, represents the course corresponding feature value; is the feature dimension; represents the course popularity score, represents the course difficulty coefficient; are the influence weight factors of the course popularity score and the course difficulty coefficient, respectively; Based on the course matching degree, the top 5 group courses most suitable for the user are screened and recommended; when the course is open, silver medal users have priority over ordinary users to register; Ordinary users provide basic training guidance services, and obtain the basic training difficulty value through the formula ; wherein, represents the initial basic difficulty; represents the current training score; represents the baseline training score; represents the continuous completion rate; The difference between the current training score and the basic training score and the weight factor affecting the continuous completion rate; Adjust the difficulty, weight and number of sets of basic training movements through the basic training difficulty value; Users who were active for less than 3 days in the past week are identified as users at risk of churn and retention measures are triggered, including personalized coupons and event invitations.
[0010] Furthermore, the specific implementation steps of the device health prediction unit are as follows: The speed and movement distance are calculated by the collected equipment to obtain the Single running time of each component , obtain the standard working hours of the equipment based on the equipment instruction manual ; Count the number of equipment failures during the cycle time to get the failure frequency , after normalization, enter the formula Get device health index ;in, Represents the monitored components, Indicates the number of monitored components in the equipment; For the The importance weight of each component refers to the degree of impact of the component on the overall health of the equipment; is the impact weight factor of the fault frequency; when When the device is in a stable state, the existing operation and maintenance rhythm is maintained, the inspection cycle is extended, and a device health index trend chart is generated; when When a fault occurs, it is determined to be in a warning state, the contribution of each component is traced, preventive calibration is performed on high-risk components, and equipment parameters are adjusted; when When a fault occurs, it is determined to be in a high-risk state; the fault type is predicted by matching the historical fault database, triggering spare parts preparation and on-site maintenance; when When the device is detected as faulty, a standby fitness area is opened, the faulty device is marked as out of service for repair, and a full component diagnosis is performed. By formula Get the residual value of the equipment ;in, Indicates the amount of equipment purchased, Indicates the actual service life of the equipment from the time it was put into use to the occurrence of failure; Indicates the expected service life as specified in the instruction manual; Indicates the residual rate; If the repair cost exceeds 50% of the residual value of the equipment, the scrapping process is triggered and a new equipment is purchased synchronously.
[0011] Further, the specific implementation mode of the energy consumption optimization unit is as follows: Through the formula The zone energy consumption prediction value is obtained , wherein, Indicate the basic energy consumption of air conditioning, ventilation and lighting respectively; ; The energy consumption coefficient of PM2.5 triggering ventilation is ; The natural light intensity is The influence weight factor of the air conditioning basic energy consumption, the ventilation basic energy consumption and the lighting basic energy consumption is respectively The is compared with the historical same period data, when is greater than 20% of the historical same period data, it is determined that the abnormal state is triggered, and the abnormal state process analysis is triggered; Through the formula: The air conditioning energy consumption prediction value is obtained ; The ventilation energy consumption prediction value is obtained ; The lighting energy consumption prediction value is obtained ; By comparing , and with the historical same period data respectively, the abnormal source is judged.
[0012] Further, the specific implementation mode of the resource scheduling module is as follows: Based on the real-time summary of the use data of various resources in the gym, a resource state data model is constructed; a time series algorithm is used to predict the resource demand trend in the future 7 days; a hierarchical visual interface is used to display information, including a global view, a resource detail view and a time trend view; based on the demand trend, the human resource scheduling is optimized; based on the course reservation and the coach expertise, the work is arranged; based on the equipment health index, the maintenance plan is generated, and the maintenance time is coordinated to reduce the influence on operation.
[0013] Compared with the prior art, the beneficial effects of the present application are: The application collects user physiological data, equipment operation parameters and environmental data in real time through a multi-source sensor network, constructs a user motion feature model after high-precision timestamp alignment and abnormality processing, carries out user value stratification based on multi-dimensional indexes, and provides differentiated services for users in different levels, thereby significantly improving the experience of users. The application calculates a health index by collecting equipment operation parameters and failure frequency and combining component importance weight, realizes preventive maintenance of equipment failure, triggers component tracing and intervention measures when the health index is lower than a threshold, reduces equipment downtime compared with traditional after-maintenance, constructs an energy consumption prediction model based on environmental parameters, identifies abnormal energy consumption of air conditioning, ventilation and lighting systems by real-time comparison with historical data, reduces energy consumption through corresponding strategies, optimizes personnel scheduling, equipment maintenance plan and course arrangement through a visual interface and time series prediction, realizes efficient allocation of venue resources, reduces operation cost, and improves management efficiency and scientificity of decision-making. BRIEF DESCRIPTION OF DRAWINGS
[0014] In order to facilitate the understanding of those skilled in the art, the application will be further described below in conjunction with the drawings. Figure 1 The system general block diagram of the application. DETAILED DESCRIPTION
[0015] The technical solutions of the application will be described below in conjunction with the embodiments, obviously, the described embodiments are only part of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor belong to the protection scope of the application.
[0016] It should be understood that the terms "include" and "contain" used in the specification and claims of the present disclosure indicate the presence of the described features, whole, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, whole, steps, operations, elements, components and / or their sets.
[0017] It should also be understood that the terms used in the specification of the present disclosure are only for the purpose of describing specific embodiments and are not intended to limit the present disclosure. As used in the specification and claims of the present disclosure, the singular forms "a", "an" and "the" are intended to include plural forms unless the context clearly indicates otherwise. It should be further understood that the term "and / or" used in the specification and claims of the present disclosure means any combination of one or more of the associated listed items and all possible combinations, and includes these combinations.
[0018] As Figure 1As shown, the application is an intelligent management system of scientific sports gym based on big data analysis, which comprises a data acquisition module, a data processing module, an intelligent analysis module and a resource scheduling module.
[0019] The data acquisition module collects multi-dimensional data through the multi-source sensor network deployed in the equipment area, obtains corresponding user exercise data, equipment parameters and environmental data, and stores them in the pre-constructed fitness database. The multi-source sensor includes biological sensors, equipment sensors and environmental sensors; among them, at the biological sensor level, the user collects physiological data in real time through the smart bracelet and electromyographic arm band equipment, and the physiological data includes heart rate variability, electromyographic signal and blood oxygen saturation. The intelligent device terminal data interface is connected with the user's smart bracelet device to obtain the user's basic physical data and user behavior data, and the basic physical data includes age, gender, height and weight; the user behavior data includes course reservation, consumption frequency, consumption amount and recent active time. The equipment sensor network is deployed on the transmission components of aerobic equipment and the stress points of strength equipment to realize multi-dimensional equipment running parameter acquisition, among them, the speed and movement distance of the treadmill are measured in real time through the belt encoder; the resistance motor sensor of the elliptical machine is used to measure the resistance torque; the barbell rod pressure sensor of the strength equipment measures the load weight in real time, and at the same time, the equipment running state is monitored in real time through the motor current, voltage and temperature sensors, including current state, voltage state, temperature state and mechanical health state. The environmental sensor network is deployed every 50 The density of the fitness area is deployed, which is fixed by ceiling or wall, to realize comprehensive monitoring of environmental parameters, among them, the SHT30 sensor is used to monitor the temperature and humidity of the fitness area in real time, and the temperature and humidity changes of the fitness area are fed back in real time at a sampling rate of one minute; the The PM2.5 sensor monitors indoor air, and controls indoor air quality in real time; the light intensity is collected through the photoresistor sensor to provide data support for venue lighting adjustment.
[0020] The data processing module is used for deep integration and feature processing of the collected multi-dimensional data, and constructs the corresponding user exercise feature model in combination with the pre-constructed fitness database, and the specific construction process is as follows: The physiological data, equipment operation parameter data and environmental parameter data collected are time-stamped by using a unified high-precision clock source, and are converted into standard international unit values with clear physical meanings corresponding to the physiological data, equipment operation parameter data and environmental parameter data; high precision refers to a time precision error range of nanoseconds to microseconds; when data loss within 3 hours caused by communication interruption or transient sensor failure, the data loss is determined as short-time loss, and linear interpolation or filling based on the statistical value of the nearest sensor is used; for data loss exceeding 3 hours, the data loss is determined as long-time loss, and is marked as invalid segment data; Based on the user smart bracelet ID, the real-time collected physiological data is uniquely bound to the user's body data; the equipment number of the current aerobic equipment or strength equipment used by the user is associated with the collected equipment operation parameters to the current user and aerobic equipment or strength equipment; based on the specific position of the aerobic equipment or strength equipment where the user is located, the environmental parameters collected by the environmental sensors at the corresponding position are called to build a complete context environment of the user's exercise state.
[0021] The data is preliminarily screened by the preset threshold range of each type of data to realize the data anomaly detection and filtering mechanism; the data exceeding the threshold range is processed in the following two ways: When the data exceeds the preset threshold range , it is determined as critical abnormality, and the abnormality degree is calculated by the formula , wherein is the observation value, is the mean value of the user's historical data of the same type, is the standard deviation; when , the current data point is marked as abnormal and is replaced by the moving average of the previous and subsequent data; When the data exceeds the preset threshold range , it is determined as sensor failure or data transmission error, and the current data point is directly discarded and the abnormal event is recorded in the data log.
[0022] The fitness database adopts a multi-layer data storage architecture to realize efficient data management: Real-time data layer: a distributed in-memory database is used to store high-frequency sampling data in the last 24 hours, realize millisecond-level data access, and support real-time analysis and feedback; Recent data layer: a columnar storage database is used to save medium-frequency data in the last 30 days, support fast aggregation query and time series analysis; History data layer: a data lake is built based on the Hadoop ecosystem to realize low-cost storage and batch analysis of massive historical data, and the storage period can be up to 3 years; The intelligent analysis module analyzes user value, device health and energy consumption based on the structured data output by the data processing module, including a user value stratification unit, a device health prediction unit and an energy consumption optimization unit.
[0023] The user value stratification unit realizes user stratification and accurate marketing based on user behavior data, and is specifically implemented as follows: By extracting the consumption frequency F, consumption amount M and recent active time R in the user behavior data, while introducing the user duration Y and participation degree two extended indicators, the participation degree is calculated by the formula , wherein, is the course completion rate, is the social activity degree, such as the number of invited friends and the number of group course participation, is the interaction index, which refers to the App usage frequency and feedback times; are the influence weight factors of the course completion rate, the social activity degree and the interaction index, respectively; After normalization processing, the consumption frequency F, the consumption amount M, the recent active time R, the user duration Y and the participation degree are substituted into the formula: to calculate the user value ; wherein, are the influence weight factors of the consumption frequency, the consumption amount, the recent active time, the user duration and the participation degree, respectively; Based on the user value score, users are divided into four levels: diamond users (90-100 points), gold medal users (75-89 points), silver medal users (60-74 points) and ordinary users (0-59 points).
[0024] Differentiated service strategies are automatically matched based on different levels of users, including: Diamond users are provided with exclusive personalized training plan services, and the training intensity index is calculated by the formula ; wherein, represents the user's historical maximum load, which refers to the maximum intensity that can be tolerated in the past training; represents the user's training target value, represents the current physical fitness level value, represents the fatigue recovery index, which is calculated by a weighted formula based on training frequency, sleep quality and subjective fatigue; are the weight coefficients of the historical maximum load, the training target value, the physical fitness level value and the fatigue recovery index, respectively; Adjust the weight, number of sets, and rest time of each training session based on the training intensity index's changing trend to ensure that training skills promote progress without leading to overtraining. When the training intensity index increases or decreases, the training intensity is increased or decreased, allowing users to always adapt to the training intensity. Gold users are provided with sports injury risk warning services. By deploying RGB camera arrays in the fitness area, multi-angle video streams of user training movements are collected. The three-dimensional coordinates of the user's key bone joints are extracted in real time through a deep neural network to build a human skeletal motion model. The key bone points include the shoulder joint, elbow joint, wrist joint and knee joint. When the user performs training, the dynamic time regularization distance between the user's bone point trajectory and the standard movement in the preset professional coach's standard movement library is calculated to obtain the movement deviation. ; Count the number of training times within the user cycle time and compare it with the industry standard frequency to obtain the training frequency coefficient ; Through the formula Calculate the user injury risk index ;in Indicates the The load index of each joint is calculated based on the movement pattern combined with body weight and load; is the risk weight factor for the joint, The total number of joints detected; are the influencing weight factors of training frequency and action deviation respectively; when If it is determined to be a low-level injury risk, basic protective training suggestions will be pushed through the user app, including targeted joint stability exercise videos; a training safety report will be sent once a week to show the current load distribution of each joint; and the normal training plan progress will be maintained; when When the user is at a moderate injury risk, a warning is triggered and a joint protection instruction video produced by a professional coach is pushed. The training plan is adjusted to reduce the training volume of high-risk movements by 20%. High-risk movements are marked in the user's training log and alternative movement suggestions are provided. The injury risk is reassessed every 3 days and the risk status is updated in real time. The user's precautions are marked for the coach on the venue's smart screen. when If it is determined to be a high-level injury risk, training will be suspended immediately, high-risk training items will be frozen, and they will be forcibly replaced with rehabilitation training content; an emergency notification will be sent to the sports rehabilitation therapist terminal to arrange a one-on-one assessment within 48 hours; a detailed risk analysis report will be generated, including action video playback and joint force analysis; professional medical advice and medical referral information will be pushed when necessary; and the training plan will be re-formulated until the user's injury risk index drops below 60 before resuming normal training; Silver users are provided with intelligent group course recommendations and priority registration services. By analyzing user behavior data, the user's preference weights for different types of courses are extracted to obtain the user preference coefficient. ; Through the formula Calculate the course matching degree ;in, Indicates user Item preference coefficient, Indicates the characteristic value corresponding to the course, is the feature dimension, which refers to the number of course features involved in the calculation; represents the course popularity score, Indicates the course difficulty coefficient, which is calculated through a weighted formula using coach ratings, user post-class surveys, and physical fitness data; are the influencing weight factors of course popularity score and course difficulty coefficient respectively; Based on the course matching degree, the top five most suitable group courses for users are screened and recommended, and displayed in the promotion area on the user's app homepage. When the course matching degree exceeds the preset threshold, Silver users will have priority over regular users to register for the courses that are open for registration four hours in advance. Based on the course matching degree distribution of different Silver user groups, group course scheduling and instructor assignment are optimized. Provide basic training guidance services for ordinary users through formula Calculate and obtain the basic training difficulty value; among them, Indicates the initial basic difficulty; Indicates the current training score; represents the baseline training score; Represents the continuous completion rate, which refers to the percentage of times the training plan is completed continuously; are the difference between the current training score and the benchmark training score, and the influencing weight factors of the continuous completion rate; Adjust the difficulty, weight, and number of sets of basic training exercises through the basic training difficulty value to ensure that ordinary users are always in a comfortable training state; match the corresponding training videos and movement demonstrations based on the basic training difficulty value to match the teaching content with the user's current ability; Users who have been active for less than 3 days in the past week are identified as users at risk of churn and retention measures are triggered, including personalized coupons and event invitations; user value scores are automatically recalculated every week to achieve dynamic evaluation and stratification of user value, allowing gyms to implement refined operations for users of different values.
[0025] The equipment health prediction unit monitors equipment health and predicts faults by analyzing real-time data streams from equipment sensors. The specific implementation process is as follows: The real-time measurement speed and movement distance of aerobic equipment or strength equipment collected by the equipment sensor are calculated to obtain the first The actual single running time of each component , based on the instruction manual of aerobic equipment or strength equipment, obtain the standard working time designed for the equipment At the same time, the number of equipment failures during the cycle time is counted to obtain the failure frequency , after normalization, enter the formula Calculate the equipment health index ;in, Represents the monitored components, Indicates the number of monitored components in the equipment; For the The importance weight of each component refers to the degree of impact of the component on the overall health of the equipment; is the impact weight factor of the fault frequency; when When the device is in a stable state, the existing operation and maintenance rhythm is maintained, and the inspection cycle is extended, such as from every 15 days to every 30 days. A device health index trend chart is generated, and the current device is marked as a benchmark device for performance comparison during the acceptance of new equipment. when When a component is in a warning state, it is determined to be in a source-traceable state, the contribution of each component is calculated and ranked, and the original data collected by the equipment sensors is used to determine whether the component exceeds the rated design load. Intervention measures are then implemented, and preventive calibration is performed on high-risk components, including sensor calibration and transmission component lubrication. Equipment parameters are adjusted, including reducing the peak load of strength equipment and limiting the continuous high-speed operation time of treadmills. Monitoring efforts are also increased in the short term to verify the effectiveness of interventions and prevent the risk from worsening. when When it is determined to be a high-risk state; start the standardized process, call the historical fault library, and match the current The growth rate and the actual single operation time of the component change, predict the fault type, trigger the pre-preparation of component spare parts, and dispatch maintenance personnel for on-site repair. After the repair is completed, the equipment health index is recalculated. When the work order is closed, the equipment resumes normal use; when When the equipment fails, it is determined to be faulty; the spare fitness area is opened, the faulty equipment is marked as out of service for repair and the repair progress is announced, and the repair process is carried out by full component diagnosis; through the formula Calculate the residual value of the equipment ;in, Indicates the amount of equipment purchased, Indicates the actual service life of the equipment from the time it was put into use to the occurrence of failure; Indicates the expected service life based on the instruction manual; Residual value rate, which is set at 5% to 10% based on industry management and determined based on the actual equipment conditions; If the repair cost exceeds 50% of the equipment's residual value, the scrapping process will be directly triggered and new equipment will be purchased simultaneously.
[0026] The energy consumption optimization unit analyzes environmental sensor data to achieve refined management of the gym's energy consumption. The specific implementation process is as follows: Through the environmental sensor network every 50 The density deployment of a group of fitness areas is fixed on the ceiling or wall to achieve comprehensive monitoring of environmental parameters. The collected environmental parameter data is calculated through the formula Calculate and obtain regional energy consumption forecast ,in, Indicates the basic energy consumption of the air conditioner, which is calculated by the real-time power of the air conditioner and the corresponding operating time; Indicates the basic energy consumption of ventilation, which is calculated by the rated power of the fan and the operating time; Indicates the basic energy consumption of lighting; calculated by the total power of the lamp and the duration of its on-time; ; is the energy consumption coefficient of ventilation triggered by PM2.5, ; is the natural light intensity; are the influencing weight factors of basic energy consumption for air conditioning, basic energy consumption for ventilation and basic energy consumption for lighting respectively; Compare the regional energy consumption forecast value with the historical data for the same period. When the regional energy consumption forecast value is greater than 20% of the historical data for the same period, it is determined to be an abnormal state, triggering the abnormal state process analysis, calling the basic energy consumption of air conditioning, ventilation and lighting, and splitting the regional energy consumption forecast formula; By formula Calculate the predicted value of air conditioning energy consumption ; By formula Calculate the ventilation energy consumption forecast ; By formula Calculate the lighting energy consumption forecast ; By comparing the predicted values of air conditioning energy consumption, ventilation energy consumption, and lighting energy consumption with the historical data for the same period, the abnormal source can be determined; When the air conditioner energy consumption prediction value is greater than 10% of the historical same period data, it is determined that the air conditioner is abnormal; the air conditioner abnormal optimization strategy is executed, including adjusting the air conditioner set temperature, keeping ; detecting the door and window opening state through the door and window sensor, if there is a door and window that is not closed, the intelligent device sends a closing instruction; judging the air conditioner state based on the device health index, if , generating a maintenance work order and notifying the maintenance personnel to repair; if it cannot be repaired temporarily, adjusting the fitness area air conditioner running time, only starting in the peak period, and switching to the temporary area sharing in the flat peak period; calling the temperature and humidity data of the fitness area closest to the current abnormal fitness area, comparing the deviation value, when the deviation value of the two is greater than the preset threshold value, determining that the sensor is false, marking the sensor as an abnormal state, enabling the backup sensor data, and pushing the sensor abnormality to remind the maintenance personnel to manually review; When the ventilation energy consumption prediction value is greater than 10% of the historical same period data, it is determined that the ventilation is abnormal; the ventilation abnormal optimization strategy is executed, including judging the ventilation device state based on the device health index; if , generating a maintenance work order and notifying the maintenance personnel to repair; comparing the historical same period P and relationship, if the deviation between the current set threshold value and the historical optimal threshold value is greater than 15%, automatically triggering calibration and generating a threshold optimization suggestion, popping up a window on the administrator control end for confirmation, supporting one-key application or manual modification, and after modification, automatically synchronizing to the ventilation control system; calling the PM2.5 sensor data of the three areas closest to the current area, if the deviation value between the current area P and the three areas closest to the current area is greater than the preset threshold value, it is determined that the sensor is false, the sensor is marked as an abnormal state, the backup sensor data is enabled, and the sensor abnormality is pushed to remind the maintenance personnel to manually review; adjusting the ventilation port air volume distribution in combination with the human flow heat; When the lighting energy consumption prediction value is greater than 10% of the historical same period data, it is determined that the lighting is abnormal; the lighting abnormal optimization strategy is executed, including real-time comparison of the natural light intensity and the lamp power, if and the lamp power is not reduced by 50%, it is determined that the photosensitive control is invalid, the forced dimming mode is temporarily switched, the brightness is adjusted based on the natural light intensity, such as , the brightness is adjusted to 30%, and a photosensitive calibration work order is generated and pushed to the maintenance personnel for repair; calling the photosensitive resistance sensor data of the three areas closest to the current area, if the deviation value between the current area light intensity and the three areas closest to the current area is greater than the preset threshold value, it is determined that the sensor is false, the sensor is marked as an abnormal state, the backup sensor data is enabled, and the sensor abnormality is pushed to remind the maintenance personnel to manually review; optimizing the lighting scheme based on the area function and historical lighting data; If the air conditioner energy consumption prediction value, the ventilation energy consumption prediction value and the lighting energy consumption prediction value are greater than 10% of the historical same period data at the same time, all are determined as abnormal, then air conditioner abnormal optimization strategy, ventilation abnormal optimization strategy and lighting abnormal optimization strategy are respectively executed.
[0027] The resource scheduling module intuitively presents the resource usage of the gym through a visualization technique, and assists in management decision-making, and the specific implementation is as follows: Based on the real-time summary of the usage data of various resources of the gym, including equipment usage rate, site occupation, personnel distribution and trainer work load, a unified resource state data model is constructed. Based on the historical data layer data, a time series algorithm is used to predict the resource demand trend in each period in the next 7 days, including people flow prediction, equipment demand prediction and course reservation prediction. A hierarchical visualization interface is adopted, including a global view, which intuitively displays the real-time people flow density and equipment usage rate of each area of the venue through a heat map, and the color indicates the utilization rate from low to high. A resource detail view is adopted, which displays the current usage state, cumulative usage time and maintenance state of each type of equipment. A time trend view is adopted, which displays the predicted people flow and key resource demand in each period in the next 7 days in the form of a line chart. Based on the resource demand trend, the scheduling of human resources is optimized, the scheduling suggestion includes recommending the scheduling period of front desk and cleaning personnel based on people flow prediction; optimizing the work arrangement of trainers based on course reservation prediction and trainer expertise; generating equipment maintenance plans based on equipment health indexes, coordinating the work time of maintenance personnel, and optimizing the impact of maintenance time on normal operation.
[0028] The preferred embodiments disclosed above are only used to help explain the present application. The preferred embodiments do not describe all the details, nor limit the application to the specific implementation. Obviously, according to the content of the specification, many modifications and changes can be made. The specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the application, so that those skilled in the art can well understand and utilize the application. The application is limited by the claims and their entire scope and equivalents.
Claims
1. An intelligent management system for a scientific sports gym based on big data analysis, characterized by: include: Data acquisition module: collects multi-dimensional data through a multi-source sensor network, obtains user motion data, device parameters and environmental data and stores them in a fitness database; Data processing module: used to deeply integrate and process the collected multi-dimensional data and build a user movement feature model; Intelligent analysis module: Analyzes user value, equipment health, and energy consumption, including user value stratification unit, equipment health prediction unit, and energy consumption optimization unit; Resource scheduling module: uses visualization technology to present the gym's resource usage and assist management decisions.
2. The intelligent management system of a scientific sports gym based on big data analysis according to claim 1 is characterized in that: The specific construction process of the user motion feature model is as follows: A unified clock source is used to align the timestamps of multi-dimensional data and convert them into standard international unit values. When data is missing within three hours due to communication interruption or instantaneous sensor failure, linear interpolation or statistical values based on the most recent sensor are used to fill in the gaps. Segments exceeding three hours are marked as invalid. Bind physiological data and body data through the user's smart bracelet ID, associate running parameters with the user and the device based on the device number, and call the corresponding environmental parameters to build a complete sports context; Filter data by preset threshold range: When the data exceeds the preset threshold , determined to be a critical anomaly, through the formula Get the abnormality level ,in, is the observed value, is the average value of similar historical data of the user, is the standard deviation; when When , the moving average of the previous and next data is used instead; When the data exceeds the preset threshold range , it is determined to be a sensor failure or data transmission error, the data point is discarded and the anomaly is recorded; The multi-layer architecture of the fitness database, which consists of real-time data layer, recent data layer and historical data layer, enables efficient data management.
3. The intelligent management system of a scientific sports gym based on big data analysis according to claim 1 is characterized in that: The specific operation steps of the user value stratification unit include: Extract consumption frequency F, consumption amount M and recent active time R from user behavior data, and introduce user duration Y and engagement Two extended indicators, through the formula Get engagement ,in, is the course completion rate, For social activity; is the interaction index; These are the impact weight factors of course completion rate, social activity and interaction index; Consumption frequency F, consumption amount M, recent active time R, user duration Y and engagement After normalization, enter the formula: Get user value ;in, These are the influence weight factors of consumption frequency, consumption amount, recent active time, user duration and engagement; Based on user value, users are divided into diamond, gold, silver and ordinary users, matching differentiated service strategies.
4. The intelligent management system for scientific sports gym based on big data analysis according to claim 3 is characterized in that: The specific implementation process of the differentiated service strategy includes: Diamond users are provided with exclusive personalized training plan services through formula Get the training intensity index ;in, Indicates the historical maximum load of the user; Indicates the user training target value, Indicates the current physical fitness level. represents fatigue recovery index; They are the weight factors of historical maximum load, training target value, physical fitness level value and fatigue recovery index; Adjust the training plan based on the trend of changes in the training intensity index; Gold users are provided with sports injury risk warning services. By deploying an RGB camera array to collect user training movements, the 3D coordinates of key skeletal joints are extracted based on a deep neural network, and the dynamic time regularization distance from the standard movement is calculated to obtain the movement deviation. ; Compare the number of training times within the user cycle time with the industry standard frequency to obtain the training frequency coefficient ; Through the formula Get user injury risk index ;in Indicates the Load index of each joint; is the risk weight factor for the joint, The total number of joints detected; are the influencing weight factors of training frequency and action deviation respectively; Take protective measures and adjust training plans based on the user's injury risk index.
5. The intelligent management system of a scientific sports gym based on big data analysis according to claim 4 is characterized in that: The specific implementation process of the differentiated service strategy also includes: Silver users are provided with intelligent group course recommendations and priority registration services. By analyzing user behavior data, the user's preference weights for different courses are extracted to obtain the user preference coefficient. ; Through the formula Get course matching ;in, Indicates user Item preference coefficient, Indicates the corresponding characteristic value of the course; is the feature dimension; represents the course popularity score, Indicates the difficulty coefficient of the course; are the influencing weight factors of course popularity score and course difficulty coefficient respectively; Based on the course matching degree, we screen and recommend the top 5 most suitable group courses for users; when the courses are available, silver users have priority over ordinary users to register; Provide basic training guidance services for ordinary users through formula Get the basic training difficulty value; among them, Indicates the initial basic difficulty; Indicates the current training score; represents the baseline training score; represents the continuous completion rate; The difference between the current training score and the basic training score and the weight factor affecting the continuous completion rate; Adjust the difficulty, weight and number of sets of basic training movements through the basic training difficulty value; Users who were active for less than 3 days in the past week are identified as users at risk of churn and retention measures are triggered, including personalized coupons and event invitations.
6. The intelligent management system of a scientific sports gym based on big data analysis according to claim 1 is characterized in that: The specific implementation steps of the equipment health prediction unit are as follows: The speed and movement distance are calculated by the collected equipment to obtain the Single running time of each component , obtain the standard working hours of the equipment based on the equipment instruction manual ; Count the number of equipment failures during the cycle time to get the failure frequency , after normalization, enter the formula Get device health index ;in, Represents the monitored components, Indicates the number of monitored components in the equipment; For the The importance weight of each component refers to the degree of impact of the component on the overall health of the equipment; is the impact weight factor of the fault frequency; when When the device is in a stable state, the existing operation and maintenance rhythm is maintained, the inspection cycle is extended, and a device health index trend chart is generated; when When a fault occurs, it is determined to be in a warning state, the contribution of each component is traced, preventive calibration is performed on high-risk components, and equipment parameters are adjusted; when When a fault occurs, it is judged as a high-risk state; the fault type is predicted by matching the historical fault database, triggering spare parts preparation and on-site maintenance; when When the device is detected as faulty, a standby fitness area is opened, the faulty device is marked as out of service for repair, and a full component diagnosis is performed. By formula Get the residual value of the equipment ;in, Indicates the amount of equipment purchased, Indicates the actual service life of the equipment from the time it was put into use to the occurrence of failure; Indicates the expected service life specified in the instruction manual; represents the residual value rate; If the repair cost exceeds 50% of the equipment's residual value, the scrapping process is triggered and new equipment is purchased simultaneously.
7. The intelligent management system for scientific sports gym based on big data analysis according to claim 1 is characterized in that: The specific implementation of the energy consumption optimization unit is as follows: By formula Get regional energy consumption forecast ,in, Represent the basic energy consumption of air conditioning, ventilation and lighting respectively; ; is the energy consumption coefficient of ventilation triggered by PM2.5, ; is the natural light intensity; are the influencing weight factors of basic energy consumption for air conditioning, basic energy consumption for ventilation and basic energy consumption for lighting respectively; Will Comparing with historical data for the same period, When the value is greater than 20% of the historical data for the same period, it is determined to be an abnormal state and the abnormal state process analysis is triggered; By formula: Get the predicted value of air conditioning energy consumption ; Get the predicted value of ventilation energy consumption ; Get lighting energy consumption forecast ; By 、 and Compare with historical data for the same period to determine the source of the anomaly.
8. The intelligent management system for scientific sports gym based on big data analysis according to claim 1 is characterized in that: The specific implementation of the resource scheduling module is as follows: Based on real-time aggregated usage data of various gym resources, a resource status data model is constructed; a time series algorithm is used to predict resource demand trends for the next seven days; and a hierarchical visual interface is used to display information, including a global view, a resource detail view, and a time trend view. Optimize human resource scheduling based on demand trends; arrange work based on course reservations and coach expertise; Generate maintenance plans based on equipment health indexes and coordinate repair times to minimize operational impacts.