Campus intelligent life terminal comprehensive management and control system based on big data

Abstract

The application relates to the technical field of Internet of Things, and discloses a campus intelligent life terminal comprehensive management and control system based on big data, which comprises a data acquisition module, a big data analysis module, a comprehensive management and control module and an execution feedback module. Through the data fusion center module, different system operation data is integrated in real time during the comprehensive management and control of the campus life terminal, cross-platform data sharing and consistency of the canteen, the access control and the energy terminal are ensured, the information island problem caused by independent operation of terminal data is avoided, the collected environmental factors and user behavior information are analyzed in association, terminal operation abnormity and resource conflict risks can be found in real time, the efficiency and reliability of terminal collaborative management are ensured, fault response delay is further reduced, a shunting scheme is automatically generated when sudden demand increases, the situation that the waiting time of students is prolonged is prevented, and the dynamic adaptability and service quality of the campus life terminal are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things, and in particular to a comprehensive management and control system for campus smart life terminals based on big data. Background Art

[0002] The Internet of Things refers to the connection of any object to the network through information sensing devices and agreed protocols. Objects exchange and communicate information through information dissemination media to achieve intelligent identification, positioning, tracking, and supervision functions.

[0003] The campus smart life terminal comprehensive management and control system is a key platform that integrates IoT perception and big data decision-making, aiming to improve campus management efficiency and life service quality. It is the core infrastructure supporting the operation of the smart campus, carrying the important function of optimizing resource allocation to ensure the collaborative management effect of campus terminals.

[0004] Currently, due to the existence of multiple types of independently deployed systems in campus life terminals, the operating data collected by different systems cannot be shared and integrated across terminals when conducting comprehensive management and control of canteens, access control, and energy terminals. When abnormal equipment operation occurs, the fault response will be delayed by several hours, and real-time linkage analysis of terminal status cannot be achieved.

[0005] Therefore, a comprehensive management and control system for campus smart life terminals based on big data is proposed to solve the above problems. Summary of the Invention

[0006] (1) Technical problems solved

[0007] In response to the shortcomings of the existing technology, the present invention provides a comprehensive management and control system for campus smart life terminals based on big data, which solves the problems raised in the above background technology.

[0008] (2) Technical solution

[0009] To achieve the above objectives, the present invention provides the following technical solutions: a comprehensive management and control system for campus smart life terminals based on big data, comprising:

[0010] Data collection module: This module collects operational data from campus life terminals in real time, including terminal type identification, operational parameters, environmental factors, and user behavior data. Terminal types include cafeteria ordering terminals, access control terminals, energy management terminals, and health monitoring terminals. Operational parameters include device status values, energy consumption values, and usage frequency. Environmental factors include temperature, humidity, and foot traffic. User behavior data includes student card swipe records, reservation requests, and feedback scores.

[0011] Big data analysis module: When terminal operating parameters deviate from the preset normal range, it determines whether the big data model activation threshold has been reached. If so, the big data model is called to build a terminal virtual map based on historical operating data, real-time collected data, and environmental factors, generate a state evolution trajectory, and predict the risk of terminal state abnormality;

[0012] Comprehensive management and control module: Based on prediction results, it matches the historical management and control database, assesses potential failures and optimizes paths. It then combines campus schedules with terminal health tolerance to determine whether intervention is needed. If necessary, it generates dynamic management and control strategies, including adjusting terminal operating modes, resource allocation, and alarm notifications.

[0013] Execution feedback module: After implementing the control strategy, it monitors the execution effect in real time, collects feedback data, optimizes big data model parameters, and forms a closed-loop control;

[0014] Among them, the system integrates Internet of Things devices through a cloud platform to achieve data sharing and collaborative management and control.

[0015] Preferably, the process of determining whether the big data model activation threshold is reached is:

[0016] The key parameters are normalized to dimensionless relative values:

[0017] ,

[0018] in is the benchmark value;

[0019] Calculate the relative deviation:

[0020]

[0021] The activation conditions are:

[0022] and

[0023] Character meaning: Real-time parameter physical values, : System preset benchmark physical value, : Normalized real-time parameters, : Normalized warning threshold, : The system allows the maximum normalization threshold, fixed value 1.2, : Model activation trigger threshold, fixed value 0.05.

[0024] Preferably, the process of generating the state evolution trajectory is:

[0025] With the time axis as the X-axis and the key operating parameters as the Y-axis, a multi-dimensional timing curve of the terminal status is constructed;

[0026] Generate standard state reference curve based on historical data;

[0027] Sampling at the same time point to obtain the current state data set and the reference state data set;

[0028] Calculate the cosine normalized vectors of two sets of data:

[0029]

[0030] when When , the curve type is linear, otherwise it is nonlinear;

[0031] in is the module length, is the current parameter vector component, is the standardized component, is the abnormality judgment threshold, set to 0.97, is the cosine similarity value, ranging from [-1, 1], which measures the degree of deviation between the current state and the standard state.

[0032] Preferably, the process of predicting the risk of abnormal terminal status includes:

[0033] When the curve type is linear, the least squares method is used to fit the trend line equation ,The critical fault threshold is input into the equation to predict the critical time point;

[0034] When the curve type is nonlinear, analyze the growth rate of key parameters: calculate the historical change rate through differential operation and determine the maximum growth rate , calculate the parameter close to the value , critical time , critical time point ;

[0035] in is the parameter change in the kth period, is the time interval, is the maximum normalized rate of change, is the normalized fault threshold, the threshold is 0.98, is the estimated time for the parameter to reach the fault threshold, is the current timestamp, is the predicted time point of failure occurrence.

[0036] Preferably, the process of determining whether intervention is needed by combining the campus schedule and the terminal health tolerance is as follows:

[0037] Obtain production plan information and terminal health tolerance, the latter defined as the difference between the current health score and the lower tolerance limit;

[0038] Construct equipment health data group and production task data group;

[0039] Calculate the Pearson correlation coefficient:

[0040]

[0041] in Health data, original value, range 0-100, such as comprehensive health index calculated based on vibration and temperature, For task data, is the normalized health score, is the mean of the normalized health score, is the mean of task occupancy, is the standard deviation of the normalized health score, is the standard deviation of task occupancy;

[0042] when When the health status is close to the upper limit of tolerance, it is determined that intervention is needed. , upper limit of health tolerance .

[0043] Preferably, the process of generating a dynamic control strategy is:

[0044] Control parameters based on feedback from big data models, combined with current operating conditions;

[0045] Match the intervention strategy library and screen the adaptive measures, including load adjustment, downtime maintenance, and resource reallocation;

[0046] Generate a plan: including intervention content, time window, resource requirements and implementation steps;

[0047] Example: When congestion is predicted at the cafeteria terminal during peak hours, a diversion strategy is generated to dynamically adjust the order queue.

[0048] Preferably, the big data model is trained using a machine learning algorithm:

[0049] Input layer: integrates real-time sensor data, historical database and environmental data;

[0050] Hidden layer: Feature extraction using deep neural networks and support vector machines;

[0051] Output layer: predicts state probability distribution and combines reinforcement learning to optimize strategy generation.

[0052] Preferably, it also includes a user interaction module:

[0053] Provide a mobile APP interface so students can submit personalized needs;

[0054] The system generates customized strategies based on user data: predicting students’ cafeteria preferences and optimizing menu allocation.

[0055] Preferably, the terminal health tolerance assessment method is:

[0056] Health score range 0-100, tolerance ;

[0057] When forecasting , then the intervention is triggered.

[0058] Preferably, the closed-loop optimization process of executing the feedback module includes:

[0059] Collect strategy execution data and calculate effect indicators;

[0060] Update model weights via gradient descent;

[0061] When the effect does not meet expectations, the analysis module is reactivated for iterative optimization.

[0062] (3) Beneficial effects

[0063] Compared with the existing technology, the present invention provides a comprehensive management and control system for campus smart life terminals based on big data, which has the following beneficial effects:

[0064] 1. In the present invention, by setting up a data fusion center module, during the comprehensive management and control of campus life terminals, the operating data of different systems are integrated in real time to ensure cross-platform data sharing and consistency of canteen, access control, and energy terminals, avoid the information island problem caused by independent operation of terminal data, and at the same time, correlate and analyze the collected environmental factors with user behavior information, so as to detect the risks of terminal operation anomalies and resource conflicts in real time, ensure the efficiency and reliability of terminal collaborative management, and further reduce fault response delays.

[0065] 2. In the present invention, by setting up an intelligent prediction engine module, the historical operation trajectory and real-time usage frequency are analyzed in the scene of dense crowds during peak hours in the cafeteria, and the potential trend of queue congestion at the ordering terminal is identified, so that the system can predict the risk of equipment overload and adjust the strategy in time. At the same time, it automatically generates a diversion plan when sudden demand increases, preventing students from waiting for a longer time, and ensuring the dynamic adaptability and service quality of campus life terminals.

[0066] 3. In the present invention, by setting up a strategy optimization and coordination module, in the process of campus energy scheduling and personalized services, the course plan and terminal health tolerance are associated, and resource allocation and service strategies are dynamically matched. The operation mode can be automatically adjusted according to the health status and actual needs of different terminals, reducing energy waste and service loss problems. At the same time, in user interaction, the ordering menu and access control authorization are optimized according to students' preferences, ensuring the accuracy of terminal resource utilization and the improvement of life experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 It is a schematic diagram of the overall system architecture of the present invention. DETAILED DESCRIPTION

[0068] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0069] See also Figure 1 The campus smart life terminal comprehensive management and control system based on big data includes:

[0070] Data collection module: This module collects operational data from campus life terminals in real time, including terminal type identification, operational parameters, environmental factors, and user behavior data. Terminal types include cafeteria ordering terminals, access control terminals, energy management terminals, and health monitoring terminals. Operational parameters include device status values, energy consumption values, and usage frequency. Environmental factors include temperature, humidity, and foot traffic. User behavior data includes student card swipe records, reservation requests, and feedback scores.

[0071] Big data analysis module: When terminal operating parameters deviate from the preset normal range, it determines whether the big data model activation threshold has been reached. If so, the big data model is called to build a terminal virtual map based on historical operating data, real-time collected data, and environmental factors, generate a state evolution trajectory, and predict the risk of terminal state abnormality;

[0072] Comprehensive management and control module: Based on prediction results, it matches the historical management and control database, assesses potential failures and optimizes paths. It then combines campus schedules with terminal health tolerance to determine whether intervention is needed. If necessary, it generates dynamic management and control strategies, including adjusting terminal operating modes, resource allocation, and alarm notifications.

[0073] Execution feedback module: After implementing the control strategy, it monitors the execution effect in real time, collects feedback data, optimizes big data model parameters, and forms a closed-loop control;

[0074] Among them, the system integrates IoT devices through the cloud platform to achieve data sharing and collaborative management and control;

[0075] The process of determining whether the big data model activation threshold has been reached is as follows:

[0076] The key parameters are normalized to dimensionless relative values:

[0077] ,

[0078] in is the benchmark value;

[0079] Calculate the relative deviation:

[0080]

[0081] The activation conditions are:

[0082] and

[0083] Character meaning: Real-time parameter physical values, : System preset benchmark physical value, : Normalized real-time parameters, : Normalized warning threshold, : The system allows the maximum normalization threshold, fixed value 1.2, : Model activation trigger threshold, fixed value 0.05;

[0084] The process of generating state evolution trajectory is:

[0085] With the time axis as the X-axis and the key operating parameters as the Y-axis, a multi-dimensional timing curve of the terminal status is constructed;

[0086] Generate standard state reference curve based on historical data;

[0087] Sampling at the same time point to obtain the current state data set and the reference state data set;

[0088] Calculate the cosine normalized vectors of two sets of data:

[0089]

[0090] when When , the curve type is linear, otherwise it is nonlinear;

[0091] in is the module length, is the current parameter vector component, is the standardized component, is the abnormality judgment threshold, set to 0.97, is the cosine similarity value, ranging from [-1, 1], which measures the degree of deviation between the current state and the standard state;

[0092] The process of predicting the risk of abnormal terminal status includes:

[0093] When the curve type is linear, the least squares method is used to fit the trend line equation ,The critical fault threshold is input into the equation to predict the critical time point;

[0094] When the curve type is nonlinear, analyze the growth rate of key parameters: calculate the historical change rate through differential operation and determine the maximum growth rate , calculate the parameter close to the value , critical time , critical time point ;

[0095] in is the parameter change in the kth period, is the time interval, is the maximum normalized rate of change, is the normalized fault threshold, the threshold is 0.98, is the estimated time for the parameter to reach the fault threshold, is the current timestamp, is the predicted time point of failure occurrence;

[0096] The process of combining campus schedule and terminal health tolerance to determine whether intervention is needed is as follows:

[0097] Obtain production plan information and terminal health tolerance, the latter defined as the difference between the current health score and the lower tolerance limit;

[0098] Construct equipment health data group and production task data group;

[0099] Calculate the Pearson correlation coefficient:

[0100]

[0101] in Health data, original value, range 0-100, such as comprehensive health index calculated based on vibration and temperature, For task data, is the normalized health score, is the mean of the normalized health score, is the mean of task occupancy, is the standard deviation of the normalized health score, is the standard deviation of task occupancy;

[0102] when When the health status is close to the upper limit of tolerance, it is determined that intervention is needed. , upper limit of health tolerance ;

[0103] The process of generating a dynamic control strategy is as follows:

[0104] Control parameters based on feedback from big data models, combined with current operating conditions;

[0105] Match the intervention strategy library and screen the adaptive measures, including load adjustment, downtime maintenance, and resource reallocation;

[0106] Generate a plan: including intervention content, time window, resource requirements and implementation steps;

[0107] Example: When congestion is predicted at the cafeteria terminal during peak hours, a diversion strategy is generated to dynamically adjust the order queue;

[0108] Big data models are trained using machine learning algorithms:

[0109] Input layer: integrates real-time sensor data, historical database and environmental data;

[0110] Hidden layer: Feature extraction using deep neural networks and support vector machines;

[0111] Output layer: predicts state probability distribution and generates optimization strategies based on reinforcement learning;

[0112] Also includes user interaction modules:

[0113] Provide a mobile APP interface so students can submit personalized needs;

[0114] The system generates customized strategies based on user data: predicting students’ cafeteria preferences and optimizing menu allocation;

[0115] The terminal health tolerance assessment method is:

[0116] Health score range 0-100, tolerance ;

[0117] When forecasting , then the intervention is triggered;

[0118] The closed-loop optimization process of the execution feedback module includes:

[0119] Collect strategy execution data and calculate effect indicators;

[0120] Update model weights via gradient descent;

[0121] When the effect does not meet expectations, the analysis module is reactivated for iterative optimization.

[0122] Example 1: Actual campus environment

[0123] The data acquisition module obtains real-time operational data through IoT sensors deployed at cafeteria ordering terminals, access control terminals, energy management terminals, and health monitoring terminals. Specifically, terminal type identification is used to distinguish device categories. Operating parameters include device status values, energy consumption values, and usage frequency. Environmental factors cover dynamic variables such as temperature, humidity, and traffic flow. User behavior data integrates student card swiping records, appointment requests, and feedback scores. These data are uniformly transmitted to the central server through the cloud platform to ensure data sharing and consistency across terminals. In a pilot project at a certain university, the system successfully reduced the problem of information islands. When the load of the cafeteria ordering terminal was abnormal, environmental factors were correlated and analyzed in real time, triggering an early warning mechanism to reduce response delays.

[0124] The big data analysis module starts the processing flow when it detects that the operating parameters deviate from the normal range. The system first normalizes the parameters to reduce unit differences and calculates the relative deviation to determine whether the activation threshold has been reached. When the threshold is reached, the built-in big data model is called to construct a terminal virtual mapping and simulate the state evolution trajectory. For access control terminals, the system analyzes historical access data to generate a reference curve, compares the cosine similarity with the current state, and predicts potential failure risks. In the energy management terminal scenario, the system identifies nonlinear growth trends, calculates critical time points, and warns operation and maintenance personnel in advance.

[0125] The comprehensive management and control module matches the historical management and control database based on the prediction results, evaluates the optimization path and fault risk, and combines the campus schedule and terminal health tolerance to determine the need for intervention. When intervention is required, it dynamically generates a management and control strategy. During peak hours in the cafeteria, the system identifies the trend of queue congestion at the ordering terminal, automatically generates a diversion strategy, and sends an alarm notification to the administrator's mobile terminal to ensure that service quality does not decline.

[0126] The execution feedback module monitors the effect after the strategy is implemented and optimizes the model parameters through a closed-loop mechanism. After a certain energy scheduling intervention, the system detects that the lighting energy consumption of empty classrooms has decreased and then updates the model weights. When the effect does not meet expectations, the analysis module is reactivated for iterative optimization. The entire implementation process emphasizes the matching of resource scheduling and production tasks to enhance the dynamic adaptability of terminals.

[0127] The system has been successfully verified in multiple campus scenarios. A university has reduced cafeteria congestion, shortened fault response time, and optimized energy utilization through the above implementation method. Formulaic calculations were avoided during implementation to ensure ease of operation and scalability.

[0128] Example 2: Campus access control scenario

[0129] The data collection module uses access control terminals deployed at various campus entrances to collect real-time operational data, including terminal type identification, operating parameters, environmental factors, and user behavior data. During a university sports meet, the system detected that the opening and closing frequency of the East Gate gate increased to 120 times per minute, while the pedestrian flow reached 150 people per minute and the ambient temperature exceeded 30°C. This data is integrated in real time through the cloud platform, reducing information silos and ensuring cross-terminal data sharing. When data deviates from the normal range, the system immediately initiates the processing process.

[0130] The big data analysis module determines whether to activate the model when parameters deviate. The system first normalizes key parameters and calculates the relative deviation. When the deviation exceeds the threshold, it calls the built-in model to build a virtual mapping of the access control terminal and generate a state evolution trajectory. Through cosine similarity analysis, the system predicts potential risks. During the peak period of the sports meeting, the similarity analysis showed that the device status deviated from the normal mode, predicted the critical time point, and warned the operation and maintenance personnel in advance.

[0131] The comprehensive management and control module matches the historical fault database based on the prediction results, evaluates the optimized path, and combines it with the campus schedule. The system calculates the Pearson correlation coefficient, determines whether intervention is needed, and then generates a dynamic strategy and sends an alert to the administrator. The system automatically reallocates resources to ensure traffic efficiency.

[0132] The execution feedback module monitors the effect after the strategy is implemented and optimizes the model parameters through a closed-loop mechanism. When the effect does not meet expectations, the analysis module is reactivated for iterative optimization. The entire implementation process improves the reliability of the access control system and optimizes resource utilization. The system emphasizes matching with production tasks and enhancing the dynamic adaptability of terminals.

[0133] Example 3: Teaching Building Energy Management Scenario

[0134] The data acquisition module collects real-time operational data from smart meters, air conditioning controllers, and lighting sensors deployed in classrooms, including terminal type identification, operating parameters, environmental factors, and user behavior data. During the final exam week at a certain university, the system monitored that the air conditioning load rate in Building A of the teaching building continued to exceed 85%. At the same time, light sensor data showed that rainy weather caused lighting energy consumption to increase by 40%.

[0135] These multi-source data are integrated in real time through the cloud platform. When the system detects that the energy consumption value deviates from the preset range, the analysis process is immediately triggered. The big data analysis module first normalizes the key parameters. When the calculation finds that the relative deviation exceeds the activation threshold of 0.05, the historical database is called to build a virtual mapping of the energy terminal. Through cosine similarity analysis, the system determines it as a linear abnormal trend and predicts the critical time point.

[0136] The comprehensive management and control module combines the campus schedule with the equipment health tolerance, calculates the Pearson correlation coefficient to confirm a strong correlation, and generates a dynamic management and control strategy: adjusting the operating mode, reallocating resources, sending alarm notifications to the logistics management center, and automatically implementing a time-sharing load reduction strategy between exams to avoid affecting the power supply stability of the examination room.

[0137] The execution feedback module monitors the effectiveness of the strategy in real time and optimizes the model through a closed-loop mechanism. When it detects that a classroom needs to extend its power supply due to a special activity, the system reactivates the analysis module to iterate the strategy. The entire implementation process achieves precise energy consumption control and maintains a zero-fault operation record for teaching equipment.

[0138] Data group parameters and weight definitions:

[0139] The current state data set contains the following parameters and weights, the sum of which is 1:

[0140] Parameter Type Specific parameters Weight Data Source Operating parameters (0.5) Real-time energy consumption value (kW) 0.3 Smart Meters Equipment load rate (%) 0.2 Air conditioning controller built-in sensor Environmental factors (0.3) Indoor and outdoor temperature difference (℃) 0.15 Temperature and humidity sensors Light intensity (Lux) 0.15 Light sensor User behavior (0.2) Frequency of classroom reservation (times / h) 0.1 Campus Reservation System Mobile terminal adjustment requests (times / h) 0.1 Student APP operation log

[0141] Reference status data group:

[0142] Parameter Type Baseline value Tolerance range Real-time energy consumption value 50kW ±10% Equipment load rate 70% ±15% Indoor and outdoor temperature difference 5℃ ±3℃

[0143] Health data and comprehensive health index calculation:

[0144] Health data sources and processing procedures:

[0145] Vibration data:

[0146] Source: triaxial accelerometer at the bottom of the device, sampling frequency 10kHz;

[0147] Processing: Calculate effective value ;

[0148] Health Score: , is the safety threshold;

[0149] Temperature data:

[0150] Source: Thermocouple embedded in the equipment heat sink;

[0151] Processing: Calculate the temperature rise rate , unit ℃ / min;

[0152] Health Score: , exponential decay model;

[0153] Comprehensive health index formula: ;

[0154] Solutions for data interactions:

[0155] To solve the coupling effect between parameters, the system adopts a hierarchical decision-making mechanism:

[0156] First-level filtering: eliminate strongly correlated parameters through the covariance matrix;

[0157] Secondary weighting: perform principal component analysis on the remaining parameters and redistribute the weights based on the proportion of eigenvalues;

[0158] Dynamic correction: Based on real-time feedback data, the weights are updated according to the following rules:

[0159]

[0160] ΔE: deviation between actual energy consumption and predicted value;

[0161] γ=0.1: Empirical correction factor, determined by fitting 50 sets of historical data.

[0162] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply the existence of any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0163] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A comprehensive management and control system for campus smart life terminals based on big data, characterized by: include: Data collection module: This module collects operational data from campus life terminals in real time, including terminal type identification, operational parameters, environmental factors, and user behavior data. Terminal types include cafeteria ordering terminals, access control terminals, energy management terminals, and health monitoring terminals. Operational parameters include device status values, energy consumption values, and usage frequency. Environmental factors include temperature, humidity, and foot traffic. User behavior data includes student card swipe records, reservation requests, and feedback scores. Big data analysis module: When terminal operating parameters deviate from the preset normal range, it determines whether the big data model activation threshold has been reached. If so, the big data model is called to build a terminal virtual map based on historical operating data, real-time collected data, and environmental factors, generate a state evolution trajectory, and predict the risk of terminal state abnormality; Comprehensive management and control module: Based on prediction results, it matches the historical management and control database, assesses potential failures and optimizes paths. It then combines campus schedules with terminal health tolerance to determine whether intervention is needed. If necessary, it generates dynamic management and control strategies, including adjusting terminal operating modes, resource allocation, and alarm notifications. Execution feedback module: After implementing the control strategy, it monitors the execution effect in real time, collects feedback data, optimizes big data model parameters, and forms a closed-loop control; Among them, the system integrates Internet of Things devices through a cloud platform to achieve data sharing and collaborative management and control.

2. The campus smart life terminal comprehensive management and control system based on big data according to claim 1 is characterized by: The process of determining whether the big data model activation threshold has been reached is as follows: The key parameters are normalized to dimensionless relative values: , in is the benchmark value; Calculate the relative deviation: The activation conditions are: and Character meaning: Real-time parameter physical values, : System preset benchmark physical value, : Normalized real-time parameters, : Normalized warning threshold, : The system allows the maximum normalization threshold, fixed value 1.2, : Model activation trigger threshold, fixed value 0.

05.

3. The campus smart life terminal comprehensive management and control system based on big data according to claim 1 is characterized by: The process of generating state evolution trajectory is: With the time axis as the X-axis and the key operating parameters as the Y-axis, a multi-dimensional timing curve of the terminal status is constructed; Generate standard state reference curve based on historical data; Sampling at the same time point to obtain the current state data set and the reference state data set; Calculate the cosine normalized vectors of two sets of data: when When , the curve type is linear, otherwise it is nonlinear; in is the module length, is the current parameter vector component, is the standardized component, is the abnormality judgment threshold, set to 0.97, For data group No. A component with a range of [-1,1], which measures the degree of deviation between the current state and the standard state.

4. The campus smart life terminal comprehensive management and control system based on big data according to claim 1 is characterized by: The process of predicting the risk of abnormal terminal status includes: When the curve type is linear, the least squares method is used to fit the trend line equation ,The critical fault threshold is input into the equation to predict the critical time point; When the curve type is nonlinear, analyze the growth rate of key parameters: calculate the historical change rate through differential operation and determine the maximum growth rate , calculate the parameter close to the value , critical time , critical time point ; in is the parameter change in the kth period, is the time interval, is the maximum normalized rate of change, is the normalized fault threshold, the threshold is 0.98, is the estimated time for the parameter to reach the fault threshold, is the current timestamp, is the predicted time point of failure occurrence.

5. The campus smart life terminal comprehensive management and control system based on big data according to claim 1 is characterized by: The process of combining campus schedule and terminal health tolerance to determine whether intervention is needed is as follows: Obtain production plan information and terminal health tolerance, the latter defined as the difference between the current health score and the lower tolerance limit; Construct equipment health data group and production task data group; Calculate the Pearson correlation coefficient: in Health data, original value, range 0-100, such as comprehensive health index calculated based on vibration and temperature, For task data, is the normalized health score, is the mean of the normalized health score, is the mean of task occupancy, is the standard deviation of the normalized health score, is the standard deviation of task occupancy; when When the health status is close to the upper limit of tolerance, it is determined that intervention is needed. , upper limit of health tolerance .

6. The campus smart life terminal comprehensive management and control system based on big data according to claim 1 is characterized by: The process of generating a dynamic control strategy is as follows: Control parameters based on feedback from big data models, combined with current operating conditions; Match the intervention strategy library and screen the adaptive measures, including load adjustment, downtime maintenance, and resource reallocation; Generate a plan: including intervention content, time window, resource requirements and implementation steps.

7. The campus smart life terminal comprehensive management and control system based on big data according to claim 1 is characterized by: The big data model is trained using a machine learning algorithm: Input layer: integrates real-time sensor data, historical database and environmental data; Hidden layer: Feature extraction using deep neural networks and support vector machines; Output layer: predicts state probability distribution and combines reinforcement learning to optimize strategy generation.

8. The campus smart life terminal comprehensive management and control system based on big data according to claim 1 is characterized by: Also includes user interaction modules: Provide a mobile APP interface so students can submit personalized needs; The system generates customized strategies based on user data: predicting students’ cafeteria preferences and optimizing menu allocation.

9. The campus smart life terminal comprehensive management and control system based on big data according to claim 1 is characterized by: The terminal health tolerance evaluation method is: Health score range 0-100, tolerance ; When forecasting , then the intervention is triggered.

10. The campus smart life terminal comprehensive management and control system based on big data according to claim 1 is characterized by: The closed-loop optimization process of the execution feedback module includes: Collect strategy execution data and calculate effect indicators; Update model weights via gradient descent; When the effect does not meet expectations, the analysis module is reactivated for iterative optimization.

Images