Intelligent health management system and method based on Internet of Things technology

Abstract

The invention discloses an intelligent health management system and method based on the Internet of Things technology, and belongs to the technical field of electronic digital data processing, and the method specifically comprises the steps: achieving the precise evaluation and abnormal early warning of the health state of equipment through real-time data collection and intelligent screening in combination with cloud analysis, effectively preventing the fault of the equipment, and improving the safety of the equipment. The production efficiency and the safety are improved; the intelligent recommendation algorithm is utilized, personalized management suggestions are provided according to the equipment state and use habits, and the suggestion quality is continuously optimized through intelligent pushing and interactive feedback mechanisms, so that the scientificity of equipment management and the user experience are enhanced, and long-term stable operation and efficient maintenance of the equipment are promoted.

Description

Technical Field

[0001] The present invention belongs to the technical field of electronic digital data processing, and more specifically, is an intelligent health management system and method based on Internet of Things technology. Background Art

[0002] The Internet of Things refers to the technology that connects any object to the Internet for information exchange and communication through information sensing devices such as RFID, infrared sensors, global positioning systems, laser scanners, etc. With the rapid development of Internet of Things technology, its application in the field of health management is becoming more and more extensive, and it can realize remote monitoring of human physiological parameters. With the continuous accumulation of health data, how to effectively store, process and analyze this data has become an important issue. Traditional health management methods have problems such as untimely data collection, inaccurate analysis, and non-personalized services.

[0003] For example, the patent with authorization announcement number CN116546111B discloses a smart ring health management method and system based on the Internet of Things technology, including: dividing the usage data collected by the smart ring, and obtaining the importance of important data according to the distribution characteristics of the usage data in different windows, and encoding and compressing the important data separately according to the importance to obtain the corresponding compressed data. This technical solution divides the usage data obtained by the smart ring, analyzes the division results separately, and encodes and compresses them, thereby improving the compression rate and transmission efficiency of important data in the usage data, and further improving the security of storing important data, thereby realizing intelligent management of health data of smart rings based on the Internet of Things technology.

[0004] The above existing technologies all have the following problems: although the compression rate and transmission efficiency of important data are improved through complex data division and encoding compression processes, the complexity and computational burden of the system are increased; they mainly focus on data division, compression and security improvement, and are relatively simple in function implementation; they lack real-time performance; they mainly process the health management data of smart rings, and the application scenarios of smart rings may be relatively limited. Summary of the invention

[0005] In view of the shortcomings of the existing technology, the present invention proposes an intelligent health management system and method based on the Internet of Things technology. Through real-time data collection and intelligent screening, combined with cloud analysis, accurate assessment of the health status of equipment and abnormal warning are achieved, which effectively prevents equipment failures and improves production efficiency and safety. By using an intelligent recommendation algorithm, personalized management suggestions are provided according to the equipment status and usage habits, and the quality of suggestions is continuously optimized through intelligent push and interactive feedback mechanisms, which enhances the scientific nature of equipment management and user experience and promotes the long-term stable operation and efficient maintenance of equipment.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] Intelligent health management methods based on IoT technology include:

[0008] Step S1: collecting the operating parameter data and environmental data of the equipment in real time through the sensor array, combining the collected operating parameter data and environmental data by using the head-to-tail splicing method to obtain a fused data set;

[0009] Step S2: Introduce an intelligent data screening strategy to screen the fused data set, obtain multi-dimensional operation data, and send the multi-dimensional operation data to the cloud server via a wireless network;

[0010] Step S3: The cloud server receives and stores the multi-dimensional operation data in a distributed manner, and uses the pre-trained dynamic equipment status evaluation model to analyze the multi-dimensional operation data to evaluate the health status of the equipment;

[0011] Step S4: Integrate an abnormality detection mechanism in the equipment status assessment process. When abnormal fluctuations in equipment operation data are found, the early warning system is immediately triggered, and an alarm message is sent to the equipment manager according to the preset alarm strategy;

[0012] Step S5: Based on the equipment status assessment results, combined with the equipment usage habits and maintenance goals, the intelligent recommendation algorithm is used to generate personalized equipment management suggestions, and the equipment management suggestions are pushed to the equipment manager through the mobile terminal at a preset time point through the intelligent push algorithm. At the same time, an interactive feedback mechanism is provided to optimize the intelligent recommendation algorithm based on user feedback.

[0013] Specifically, the specific steps of step S2 include:

[0014] S2.1: Get the fused dataset X = {x 1 ,...,x n}, where x n represents the nth fused data, and n represents the number of fused data;

[0015] S2.2: For X = {x 1 ,…,x n} to perform conditional screening. The conditional screening formula is:

[0016]

[0017] Among them, h 1 represents the fusion data feature threshold, s tong represents the statistical characteristics of the fused data, μ n represents the mean of the fused dataset X, σ n represents the standard deviation of the fused data set X, h 2Represents x n Relative to μ n The standard deviation of downward deviation σ n The limit of h 3 Represents x n Relative to μ n The standard deviation of upward deviation σ n The limit, s fu Indicates a composite filtering condition, con 1 represents the temperature condition in the fused data, con 2 represents the pressure condition in the fused data, con 3 represents the vibration conditions in the fused data;

[0018] S2.3: Integrate the selected fusion data to obtain multi-dimensional operation data And use symmetric encryption algorithm to send multi-dimensional operation data to the cloud server through wireless network, where x m Represents the m-th multidimensional running data.

[0019] Specifically, the specific steps of step S5 include:

[0020] S5.1: Obtain multi-dimensional operating data and equipment health status assessment results, and pre-process them using intelligent data cleaning and enhancement methods;

[0021] S5.2: Based on the pre-processed multi-dimensional operation data, use deep learning algorithms to analyze the usage habits of the equipment and set dynamic maintenance goals based on the type, importance and preset maintenance strategy of the equipment;

[0022] S5.3: Based on the health status assessment results and usage habit analysis results of the device, use the intelligent recommendation algorithm to generate personalized device management suggestions. The formula of the intelligent recommendation algorithm is:

[0023]

[0024] Among them, T ui Indicates the recommended health status assessment results and usage habit analysis results, r l represents the score of the lth health status assessment result, represents the score of the lth usage habit analysis result, α represents the regularization term coefficient, U and V both represent the low-rank matrices after the matrix consisting of the device health status assessment results and the usage habit analysis results is decomposed, and || || represents the 2-norm.

[0025] Specifically, the specific steps of step S5 also include:

[0026] S5.4: Based on the preset time point, real-time health status assessment results or user requests, the neural network model is used to evaluate the best push timing, and the device management suggestions are pushed to the device manager through personalized notifications;

[0027] S5.5: Establish an interactive feedback mechanism to allow equipment managers to provide feedback on received equipment management suggestions and collect feedback data;

[0028] S5.6: Analyze the collected feedback data, evaluate the effectiveness of the intelligent recommendation algorithm and the intelligent push algorithm, and iteratively optimize the intelligent recommendation algorithm and the intelligent push algorithm based on the evaluation results.

[0029] Specifically, the specific steps of the intelligent data cleaning and enhancement method of S5.1 include:

[0030] S5.11: Obtain multi-dimensional operating data and the health status assessment results of the equipment to generate equipment status data;

[0031] S5.12: Use deep learning algorithms to automatically identify outliers in device status data. Based on the identified outliers, use an improved regression prediction algorithm to correct or delete them. If they cannot be reasonably corrected, delete the data point. At the same time, use a variational autoencoder model to learn the distribution of device status data and generate new device status data samples to enhance sparse data areas. The formula of the improved regression prediction algorithm is:

[0032]

[0033] Among them, z represents the corrected equipment status data, β 0 and β b represents the regression coefficient, δ b Represents the coefficient of the kth data square term in the device status data, φ k represents the coefficient of the kth interaction term in the device status data, η represents the coefficient of the time variable, t represents the time variable, ε represents the random error term, N represents the number of any two data in the device status data as interaction terms, and

[0034] S5.13: Perform quality assessment on preprocessed and augmented data.

[0035] Specifically, the specific formula of the variational autoencoder model in S5.12 is:

[0036] L=λ 1 ×L jiao +λ 2 ×L tro+g×KL[q(v|θ)||p(v;θ)]+u×R;

[0037] Among them, L represents the total loss value when reconstructing the data, λ 1 and λ 2 represents the hyperparameter, L jiao represents the cross entropy reconstruction loss, L tro represents the L2 reconstruction loss, g represents the weight of the KL divergence, KL[q(v|θ)||p(v;θ)] represents the KL divergence between the posterior distribution q(v|θ) and the prior distribution p(v;θ), q(v|θ) represents the posterior distribution, p(v;θ) represents the prior distribution, u represents the weight of the regularization term, R represents the regularization term, θ represents the variational autoencoder model parameter, and v represents the latent variable.

[0038] Specifically, the maintenance strategies in S5.2 include preventive maintenance, predictive maintenance and corrective maintenance; and the maintenance goals include fault prevention, performance optimization, cost control and resource allocation.

[0039] The intelligent health management system based on the Internet of Things technology includes: data acquisition module, data transmission module, data analysis module, anomaly detection module, and intelligent recommendation module;

[0040] The data acquisition module collects the operating parameters and environmental data of the equipment in real time by integrating intelligent devices and sensors;

[0041] The data transmission module is used to encrypt and transmit the collected fusion data to the cloud server through wireless communication technology;

[0042] The data analysis module is used to perform preprocessing operations such as cleaning, denoising and labeling on the received physiological data, and use machine learning methods to analyze the preprocessed data to extract the health status information of the device, and at the same time, generate personalized device management suggestions based on the analysis results;

[0043] The anomaly detection module is used to detect anomalies in the equipment status evaluation process and find abnormal situations in time;

[0044] The intelligent recommendation module generates personalized equipment management suggestions based on the equipment status evaluation results, and provides an interactive feedback mechanism to optimize the recommendation algorithm.

[0045] Specifically, the intelligent recommendation module includes: an intelligent recommendation unit, an intelligent push unit, and a feedback unit;

[0046] The intelligent recommendation unit generates equipment management suggestions using an intelligent recommendation algorithm in combination with equipment status, usage habits and maintenance goals;

[0047] The intelligent push unit is used to push the device management suggestion to the device manager via the mobile terminal at a preset time point;

[0048] The feedback unit is used to collect user feedback and optimize the intelligent recommendation algorithm based on the collected user feedback.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] 1. The present invention proposes an intelligent health management method based on the Internet of Things technology. By real-time collection of multi-dimensional operation and environmental data of the equipment, combined with intelligent screening and cloud analysis, it realizes accurate assessment of the health status of the equipment and abnormal warning, improves the timeliness and efficiency of equipment maintenance, and reduces the risk of failure; based on the in-depth analysis of the equipment status and the personalized needs of users, the intelligent recommendation and push algorithm provides customized management suggestions for equipment managers, and continuously optimizes the recommendation accuracy through interactive feedback, realizing the intelligent, personalized and continuous optimization of equipment management, and improving management efficiency.

[0051] 2. The present invention proposes an intelligent health management system based on Internet of Things technology, and optimizes and improves the architecture, operation steps and processes. The system has the advantages of simple processes and low investment and operating costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a framework diagram of the intelligent health management method based on the Internet of Things technology of the present invention;

[0053] Figure 2 This is a workflow diagram of the intelligent health management method based on Internet of Things technology of the present invention;

[0054] Figure 3 This is an architecture diagram of the intelligent health management system based on Internet of Things technology of the present invention. DETAILED DESCRIPTION

[0055] Example 1

[0056] See also Figure 1-Figure 2 , an embodiment provided by the present invention: an intelligent health management method based on Internet of Things technology, comprising the following steps:

[0057] Step S1: collecting the operating parameter data and environmental data of the equipment in real time through the sensor array, combining the collected operating parameter data and environmental data by using the head-to-tail splicing method to obtain a fused data set;

[0058] Among them, the operating parameter data includes temperature, pressure and speed.

[0059] Furthermore, the specific steps of step S1 include:

[0060] (1) Deploy multiple sensors on the target device to collect the device's operating parameters and environmental data, where the operating parameter data includes temperature, pressure, and vibration, and the environmental data includes temperature, humidity, and light;

[0061] (2) Each sensor collects data in real time according to a preset sampling frequency and converts the collected data into a digital signal. It should be noted that during the data collection process, data synchronization needs to be considered to ensure that the data collected by different sensors are consistent in time;

[0062] (3) Preprocess the collected raw data, including data cleaning, data compression and data format unification;

[0063] (4) Combine the collected operating parameter data and environmental data using head-to-tail splicing. Here, “head-to-tail splicing” means arranging the two types of data together in chronological order or logical order to form a new data set. At the same time, during the splicing process, it is necessary to ensure the accuracy and consistency of the data to avoid data misalignment or loss;

[0064] (5) The spliced ​​data are fused to form a fused data set to improve the accuracy and reliability of the data. The fused data set will include both the operating parameters and environmental data of the equipment.

[0065] Step S2: Introduce an intelligent data screening strategy to screen the fused data set, obtain multi-dimensional operation data, and send the multi-dimensional operation data to the cloud server via a wireless network;

[0066] Step S3: The cloud server receives and stores the multi-dimensional operation data in a distributed manner, and uses the pre-trained dynamic equipment status evaluation model to analyze the multi-dimensional operation data to evaluate the health status of the equipment;

[0067] Furthermore, the specific steps of step S3 include:

[0068] (1) After receiving the encrypted data packet, the server performs decryption processing to restore the original multi-dimensional operation data, and uses the distributed storage system HDFS to store the restored multi-dimensional operation data in multiple nodes to improve the reliability and scalability of the multi-dimensional operation data;

[0069] (2) loading a pre-trained dynamic equipment status assessment model, and at the same time, pre-processing the multi-dimensional operation data, and inputting the pre-processed multi-dimensional operation data into the dynamic equipment status assessment model, executing the model reasoning process to assess the health status of the equipment, wherein the model reasoning process is the prior art content in this field, is not the inventive solution of the present application, and is not elaborated here;

[0070] (3) The dynamic equipment status assessment model outputs the equipment health status assessment results, including health score, fault warning, and remaining service life information. Based on the assessment results, a corresponding report or alarm is generated and sent to the user;

[0071] (4) Users take timely measures based on the feedback information they receive.

[0072] Step S4: Integrate an abnormality detection mechanism in the equipment status assessment process. When abnormal fluctuations in equipment operation data are found, the early warning system is immediately triggered, and an alarm message is sent to the equipment manager according to the preset alarm strategy;

[0073] Furthermore, the specific steps of step S4 include:

[0074] (1) Obtain historical multi-dimensional operation data and load the pre-built anomaly detection model;

[0075] (2) Based on the historical multi-dimensional operation data of the equipment, train the anomaly detection model so that it can identify abnormal fluctuations in the equipment operation data;

[0076] (3) Integrate the trained anomaly detection model into the equipment status assessment system, enabling it to analyze equipment operation data in real time and set anomaly thresholds to determine when to consider multi-dimensional equipment operation data as abnormal;

[0077] (4) Configure the early warning system, including setting alarm trigger conditions, alarm levels, and alarm recipients, to ensure that the early warning system can quickly and accurately trigger alarms and send alarm information to the corresponding equipment managers;

[0078] (5) Develop a preset alarm strategy based on the importance of the equipment and the consequences of failures. The alarm strategy should clearly define the response measures and urgency of different levels of alarms so that equipment managers can respond quickly;

[0079] (6) Real-time monitoring of the multi-dimensional operation data of the equipment and the output results of the anomaly detection model. When abnormal fluctuations in the equipment operation data are found, the early warning system is immediately triggered and an alarm message is sent to the equipment manager according to the preset alarm strategy;

[0080] (7) Collect feedback from equipment managers and continuously optimize and improve the anomaly detection mechanism and early warning system.

[0081] Step S5: Based on the equipment status assessment results, combined with the equipment usage habits and maintenance goals, the intelligent recommendation algorithm is used to generate personalized equipment management suggestions, and the equipment management suggestions are pushed to the equipment manager through the mobile terminal at a preset time point through the intelligent push algorithm. At the same time, an interactive feedback mechanism is provided to optimize the intelligent recommendation algorithm based on user feedback.

[0082] It should be noted that the intelligent recommendation algorithm should be able to comprehensively consider the current status of the equipment, historical maintenance records and preset maintenance goals. At the same time, equipment management suggestions include preventive maintenance plans, repair suggestions, and operation optimization suggestions; the preset time point refers to when the equipment is about to reach the maintenance cycle or the equipment status is abnormal.

[0083] The specific steps of step S2 include:

[0084] S2.1: Get the fused dataset X = {x 1 ,...,x n}, where x n represents the nth fused data, and n represents the number of fused data;

[0085] S2.2: For X = {x 1 ,...,x n} to perform conditional screening. The conditional screening formula is:

[0086]

[0087] Among them, h 1 represents the fusion data feature threshold, s tong represents the statistical characteristics of the fused data, μ n represents the mean of the fused dataset X, σ n represents the standard deviation of the fused data set X, h 2 Represents x n Relative to μ n The standard deviation of downward deviation σ n The limit of h 3 Represents x n Relative to μ n The standard deviation of upward deviation σ n The limit, s fu Indicates a composite filtering condition, con 1 represents the temperature condition in the fused data, con 2 represents the pressure condition in the fused data, con 3 represents the vibration conditions in the fused data;

[0088] For example, suppose there is a data set Y containing temperature y 1 , humidity 2 and light intensity y 3 Three features, if you want to filter out data points with a temperature between 20℃ and 30℃ and a humidity greater than 60%, the filtering condition can be expressed as: (20≤y 1 ≤30)∧(y 2 >60), if the light intensity is not too high, such as less than 1000 lux), the composite condition is: (20≤y1 ≤30)∧(y 2 >60)∧(y 3 <1000).

[0089] S2.3: Integrate the selected fusion data to obtain multi-dimensional operation data And use symmetric encryption algorithm to send multi-dimensional operation data to the cloud server through wireless network, where x m Represents the m-th multidimensional running data.

[0090] Furthermore, the specific steps of using the symmetric encryption algorithm for data transmission in S2.3 include:

[0091] (1) Obtaining multi-dimensional operation data

[0092] (2) Use the AES symmetric encryption algorithm and key to encrypt multi-dimensional data Perform encryption processing to ensure that the encryption process complies with relevant security standards and specifications;

[0093] (3) Encapsulating the encrypted multi-dimensional operation data, timestamp, data identifier, and identification information of the encryption algorithm into an encrypted data packet, and formatting the encrypted data packet according to the requirements of the network protocol so as to be transmitted on the wireless network. The specific process of data encapsulation is the prior art content in this field and is not the inventive solution of the present application, and is not described in detail here.

[0094] (4) Select Wi-Fi wireless network technology for data transmission, and use secure network protocols TLS / SSL to protect the security of the data transmission process and prevent data from being intercepted or tampered with, so as to send the encrypted data packet to the designated address and port of the cloud server through the wireless network;

[0095] (5) After receiving the encrypted data packet, the cloud server first performs an integrity check, then uses the same key and encryption algorithm as the sender to decrypt the encrypted data packet, restore the original multi-dimensional operation data, and store the decrypted data.

[0096] The specific steps of step S5 include:

[0097] S5.1: Obtain multi-dimensional operating data and equipment health status assessment results, and pre-process them using intelligent data cleaning and enhancement methods;

[0098] S5.2: Based on the pre-processed multi-dimensional operation data, use a deep learning algorithm to analyze the usage habits of the equipment, and set dynamic maintenance goals according to the type, importance and preset maintenance strategy of the equipment. The process of the deep learning algorithm performing the analysis is the prior art content in this field, and is not the inventive solution of the present application, and will not be described in detail here;

[0099] S5.3: Generate personalized device management suggestions using intelligent recommendation algorithms based on device health status assessment results and usage habit analysis results;

[0100] S5.4: Based on the preset time point, real-time health status assessment results or user requests, the neural network model is used to evaluate the best push timing, and the device management suggestions are pushed to the device manager through personalized notifications;

[0101] Furthermore, the specific steps of S5.4 include:

[0102] (1) Set the push time point, obtain real-time health status assessment results, receive user request data, and combine them to form push data;

[0103] (2) extracting push timing evaluation features from the collected push data, including time features, device health status features, and user behavior features, and encoding and scaling the push timing evaluation features;

[0104] (3) loading a pre-built convolutional neural network model, and using the push timing evaluation feature data to train the neural network model, adjusting the model parameters through a back propagation algorithm to minimize the loss function, and obtaining a trained convolutional neural network model, wherein the back propagation algorithm is a prior art content in this field, is not an inventive solution of the present application, and is not described in detail here, and the specific training process of the convolutional neural network model is also a prior art content in this field, is not an inventive solution of the present application, and is not described in detail here;

[0105] (4) Use the test set data in the push timing evaluation feature data to evaluate the trained convolutional neural network model, verify the generalization ability of the convolutional neural network model, and tune the convolutional neural network model based on the evaluation results, including adjusting hyperparameters and improving the model architecture;

[0106] (5) The real-time health status assessment results and user requests are input into the trained neural network model. The model evaluates the best push timing based on the input data and outputs the corresponding push decision;

[0107] (6) Generate personalized equipment management suggestions based on the push decision information, and push the equipment management suggestions to equipment managers through SMS, email, and APP push;

[0108] (7) Collect feedback from device managers on push notifications, such as click-through rate, conversion rate, and satisfaction, and iteratively optimize the neural network model based on the feedback results to improve push notification effects.

[0109] S5.5: Establish an interactive feedback mechanism to allow equipment managers to provide feedback on received equipment management suggestions and collect feedback data;

[0110] S5.6: Analyze the collected feedback data, evaluate the effects of the intelligent recommendation algorithm and the intelligent push algorithm, and iteratively optimize the intelligent recommendation algorithm and the intelligent push algorithm based on the evaluation results. The effects of the intelligent recommendation algorithm and the intelligent push algorithm are evaluated by calculating the accuracy index, and the calculation process of the accuracy index is the prior art content in this field, and is not the inventive solution of the present application, and will not be elaborated here.

[0111] The specific steps of the intelligent data cleaning and enhancement method of S5.1 include:

[0112] S5.11: Obtain multi-dimensional operating data and the health status assessment results of the equipment to generate equipment status data;

[0113] S5.12: Use deep learning algorithms to automatically identify outliers in device status data. Based on the identified outliers, use an improved regression prediction algorithm to correct or delete them. If they cannot be reasonably corrected, delete the data point. At the same time, use a variational autoencoder model to learn the distribution of device status data and generate new device status data samples to enhance sparse data areas. The formula of the improved regression prediction algorithm is:

[0114]

[0115] Among them, z represents the corrected equipment status data, β 0 and β b represents the regression coefficient, δ b Represents the coefficient of the kth data square term in the device status data, φ k represents the coefficient of the kth interaction term in the device status data, η represents the coefficient of the time variable, t represents the time variable, ε represents the random error term, which represents the unexplained part or noise in the data, N represents the number of any two data in the device status data as interaction terms, and

[0116] It should be noted that in the present invention, the modified formula introduces the linear term β b x b , square term and the interaction term φ k x i xj , and taking into account the time trend ηt and the error term ε, the flexibility and explanatory power of the regression prediction algorithm model are enhanced, and the linear and nonlinear relationships between variables and the interactions between variables can be captured, while considering the impact of time changes on the predicted values, thereby improving the accuracy and applicability of the prediction.

[0117] S5.13: Perform quality assessment on preprocessed and augmented data.

[0118] The specific formula of the variational autoencoder model in S5.12 is:

[0119] L=λ 1 ×L jiao +λ 2 ×L tro +g×KL[q(v|θ)||p(v;θ)]+u×R;

[0120] Among them, L represents the total loss value when reconstructing the data, λ 1 and λ 2 represents the hyperparameter, L jiao represents the cross entropy reconstruction loss, L tro represents the L2 reconstruction loss, g represents the weight of the KL divergence, KL[q(v|θ)||p(v;θ)] represents the KL divergence between the posterior distribution q(v|θ) and the prior distribution p(v;θ), q(v|θ) represents the posterior distribution, p(v;θ) represents the prior distribution, u represents the weight of the regularization term, R represents the regularization term, θ represents the variational autoencoder model parameter, and v represents the latent variable, that is, there is no direct observation of the variable in the variational autoencoder model, where || is used to separate the two probability distributions. The smaller the KL[q(z|θ)||p(z;θ)] value, the closer the two distributions are, and the more effective the model representation in the latent space is.

[0121] It should be noted that in the present invention, the modified formula improves the training effect and generalization ability of the variational autoencoder model by combining multiple loss functions and regularization terms. Specifically, the formula in the present invention not only takes into account the reconstruction loss to optimize the model's ability to reconstruct input data, but also introduces the KL divergence term to constrain the distribution of latent variables to make it close to the prior distribution, thereby enhancing the stability and generation ability of the model. In addition, the addition of the regularization term helps prevent the model from overfitting and improves the generalization performance of the model. This multi-objective optimization method enables the variational autoencoder model to comprehensively consider multiple performance indicators during the training process, thereby obtaining a more comprehensive and excellent performance.

[0122] The maintenance strategies in S5.2 include preventive maintenance, predictive maintenance and corrective maintenance; the maintenance goals include fault prevention, performance optimization, cost control and resource allocation.

[0123] Among them, preventive maintenance: regularly performed maintenance activities aimed at preventing equipment failures, including cleaning, lubrication, tightening, adjustment, and replacement of wearing parts;

[0124] Predictive maintenance: using monitoring technology and data analysis methods to monitor and predict the operating status of equipment in real time, detect potential faults in advance and take maintenance measures. It usually involves the collection, processing and analysis of equipment operation data, and the use of machine learning or deep learning algorithms to predict faults.

[0125] Corrective maintenance: Maintenance activities performed after equipment failure, aimed at restoring the normal operation of the equipment, including fault location, troubleshooting and repair.

[0126] Dynamic maintenance objectives include:

[0127] Failure prevention: Reduce the probability and frequency of equipment failures by optimizing preventive maintenance plans;

[0128] Performance optimization: Improve equipment operating efficiency and performance by adjusting equipment operating parameters, optimizing maintenance strategies, etc.

[0129] Cost control: Under the premise of ensuring the normal operation of the equipment, reasonably control the maintenance cost to avoid unnecessary waste;

[0130] Resource allocation: Rationally allocate maintenance resources and manpower based on equipment maintenance needs and priorities to ensure that key equipment is maintained in a timely and effective manner.

[0131] Example 2

[0132] See also Figure 3 Another embodiment provided by the present invention is an intelligent health management system based on Internet of Things technology, comprising:

[0133] Data collection module, data transmission module, data analysis module, anomaly detection module, intelligent recommendation module;

[0134] The data acquisition module collects the operating parameters and environmental data of the equipment in real time by integrating intelligent devices and sensors;

[0135] A data transmission module, used to encrypt and transmit the collected fusion data to a cloud server through wireless communication technology;

[0136] The data analysis module is used to perform preprocessing operations such as cleaning, denoising and labeling of the received physiological data, and use machine learning methods to analyze the preprocessed data to extract the health status information of the device. At the same time, based on the analysis results, it generates personalized device management suggestions;

[0137] The anomaly detection module is used to detect anomalies in the equipment status evaluation process and discover abnormal situations in a timely manner;

[0138] The intelligent recommendation module generates personalized equipment management suggestions based on the equipment status assessment results, and provides an interactive feedback mechanism to optimize the recommendation algorithm.

[0139] The data analysis module includes: a data storage unit and a status evaluation unit;

[0140] A data storage unit, used for distributed storage of received multi-dimensional operation data;

[0141] The status assessment unit uses a pre-trained dynamic equipment status assessment model to analyze multi-dimensional operating data, assess the health status of the equipment, and promptly identify potential problems or abnormal conditions of the equipment.

[0142] The anomaly detection module includes: an anomaly detection unit and an early warning unit;

[0143] Anomaly detection unit, integrated in the status assessment process, is used to monitor equipment operation data in real time and detect abnormal fluctuations to quickly identify abnormal conditions in the equipment operation status;

[0144] The early warning unit is used to send an alarm message to the equipment manager according to the preset alarm strategy when an abnormality is detected.

[0145] Function: Notify the equipment manager promptly so that appropriate measures can be taken.

[0146] The intelligent recommendation module includes: intelligent recommendation unit, intelligent push unit, and feedback unit;

[0147] The intelligent recommendation unit generates equipment management suggestions using intelligent recommendation algorithms based on equipment status, usage habits, and maintenance goals;

[0148] Intelligent push unit, used to push device management suggestions to device managers through mobile terminals at preset time points to ensure timely communication of suggestions;

[0149] The feedback unit is used to collect user feedback and optimize the intelligent recommendation algorithm based on the collected user feedback to improve the accuracy of recommendations.

[0150] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0151] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0152] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in the field may also change, modify, replace and modify the above-mentioned embodiments without departing from the purpose and scope of protection of the present invention, and all of these are within the protection of the present invention.

Claims

1. An intelligent health management method based on Internet of Things technology, characterized in that: include: Step S1: collecting the operating parameter data and environmental data of the equipment in real time through the sensor array, combining the collected operating parameter data and environmental data by using the head-to-tail splicing method to obtain a fused data set; Step S2: Introduce an intelligent data screening strategy to screen the fused data set, obtain multi-dimensional operation data, and send the multi-dimensional operation data to the cloud server via a wireless network; Step S3: The cloud server receives and stores the multi-dimensional operation data in a distributed manner, and uses the pre-trained dynamic equipment status evaluation model to analyze the multi-dimensional operation data to evaluate the health status of the equipment; Step S4: Integrate an abnormality detection mechanism in the equipment status assessment process. When abnormal fluctuations in equipment operation data are found, the early warning system is immediately triggered, and an alarm message is sent to the equipment manager according to the preset alarm strategy; Step S5: Based on the equipment status assessment results, combined with the equipment usage habits and maintenance goals, the intelligent recommendation algorithm is used to generate personalized equipment management suggestions, and the equipment management suggestions are pushed to the equipment manager through the mobile terminal at a preset time point through the intelligent push algorithm. At the same time, an interactive feedback mechanism is provided to optimize the intelligent recommendation algorithm based on user feedback.

2. The intelligent health management method based on Internet of Things technology according to claim 1, characterized in that: The specific steps of step S2 include: S2.1: Get the fused data set X = {x1,…,x n }, where x n represents the nth fused data, and n represents the number of fused data; S2.2: For X = {x1,…,x n } to perform conditional screening, the conditional screening formula is: Among them, h1 represents the fusion data feature threshold, s tong represents the statistical characteristics of the fused data, μ n represents the mean of the fused dataset X, σ n represents the standard deviation of the fused data set X, and h2 represents x n Relative to μ n The standard deviation of downward deviation σ n The limit of h3 represents x n Relative to μ n The standard deviation of upward deviation σ n The limit, s fu Represents the composite screening condition, con1 represents the temperature condition in the fused data, con2 represents the pressure condition in the fused data, and con3 represents the vibration condition in the fused data; S2.3: Integrate the selected fusion data to obtain multi-dimensional operation data And use symmetric encryption algorithm to send multi-dimensional operation data to the cloud server through wireless network, where x m Represents the mth multidimensional running data.

3. The intelligent health management method based on Internet of Things technology as claimed in claim 2, characterized in that: The specific steps of step S5 include: S5.1: Obtain multi-dimensional operating data and equipment health status assessment results, and pre-process them using intelligent data cleaning and enhancement methods; S5.2: Based on the pre-processed multi-dimensional operation data, use deep learning algorithms to analyze the usage habits of the equipment and set dynamic maintenance goals based on the type, importance and preset maintenance strategy of the equipment; S5.3: Based on the health status assessment results and usage habit analysis results of the device, use the intelligent recommendation algorithm to generate personalized device management suggestions. The formula of the intelligent recommendation algorithm is: Among them, T ui Indicates the recommended health status assessment results and usage habit analysis results, r l represents the score of the lth health status assessment result, represents the score of the lth usage habit analysis result, α represents the regularization term coefficient, U and V both represent the low-rank matrices after the matrix consisting of the device health status assessment results and the usage habit analysis results is decomposed, and || || represents the 2-norm.

4. The intelligent health management method based on Internet of Things technology as claimed in claim 3, characterized in that: The specific steps of step S5 also include: S5.4: Based on the preset time point, real-time health status assessment results or user requests, the neural network model is used to evaluate the best push timing, and the device management suggestions are pushed to the device manager through personalized notifications; S5.5: Establish an interactive feedback mechanism to allow equipment managers to provide feedback on received equipment management suggestions and collect feedback data; S5.6: Analyze the collected feedback data, evaluate the effectiveness of the intelligent recommendation algorithm and the intelligent push algorithm, and iteratively optimize the intelligent recommendation algorithm and the intelligent push algorithm based on the evaluation results.

5. The intelligent health management method based on Internet of Things technology according to claim 4, characterized in that: The specific steps of the intelligent data cleaning and enhancement method of S5.1 include: S5.11: Obtain multi-dimensional operating data and the health status assessment results of the equipment to generate equipment status data; S5.12: Use deep learning algorithms to automatically identify outliers in device status data. Based on the identified outliers, use an improved regression prediction algorithm to correct or delete them. If they cannot be reasonably corrected, delete the data point. At the same time, use a variational autoencoder model to learn the distribution of device status data and generate new device status data samples to enhance sparse data areas. The formula of the improved regression prediction algorithm is: Among them, z represents the corrected device status data, β0 and β b represents the regression coefficient, δ b Represents the coefficient of the kth data square term in the device status data, φ k represents the coefficient of the kth interaction term in the device status data, η represents the coefficient of the time variable, t represents the time variable, ε represents the random error term, N represents the number of any two data in the device status data as interaction terms, and S5.13: Perform quality assessment on preprocessed and augmented data.

6. The intelligent health management method based on Internet of Things technology according to claim 5, characterized in that: The specific formula of the variational autoencoder model in S5.12 is: L=λ1×L jiao +λ2×L tro +g×KL[q(v|θ)||p(v;θ)]+u×R; Among them, L represents the total loss value when reconstructing the data, λ1 and λ2 represent hyperparameters, and L jiao represents the cross entropy reconstruction loss, L tro represents the L2 reconstruction loss, g represents the weight of the KL divergence, KL[q(v|θ)||p(v;θ)] represents the KL divergence between the posterior distribution q(vθ) and the prior distribution p(v;θ), q(vθ) represents the posterior distribution, p(v;θ) represents the prior distribution, u represents the weight of the regularization term, R represents the regularization term, θ represents the variational autoencoder model parameter, and v represents the latent variable.

7. The intelligent health management method based on Internet of Things technology according to claim 6, characterized in that: The maintenance strategies in S5.2 include preventive maintenance, predictive maintenance and corrective maintenance; the maintenance goals include fault prevention, performance optimization, cost control and resource allocation.

8. An intelligent health management system based on Internet of Things technology, which is used to implement the intelligent health management method based on Internet of Things technology according to any one of claims 1 to 7, characterized in that: include: Data collection module, data transmission module, data analysis module, anomaly detection module, intelligent recommendation module; The data acquisition module collects the operating parameters and environmental data of the equipment in real time by integrating intelligent devices and sensors; The data transmission module is used to encrypt and transmit the collected fusion data to the cloud server through wireless communication technology; The data analysis module is used to perform preprocessing operations such as cleaning, denoising and labeling on the received physiological data, and use machine learning methods to analyze the preprocessed data to extract the health status information of the device, and at the same time, generate personalized device management suggestions based on the analysis results; The anomaly detection module is used to detect anomalies in the equipment status evaluation process and find abnormal situations in time; The intelligent recommendation module generates personalized equipment management suggestions based on the equipment status evaluation results, and provides an interactive feedback mechanism to optimize the recommendation algorithm.

9. The intelligent health management system based on Internet of Things technology as claimed in claim 8, characterized in that: The intelligent recommendation module includes: an intelligent recommendation unit, an intelligent push unit, and a feedback unit; The intelligent recommendation unit generates equipment management suggestions using an intelligent recommendation algorithm in combination with equipment status, usage habits and maintenance goals; The intelligent push unit is used to push the device management suggestion to the device manager via the mobile terminal at a preset time point; The feedback unit is used to collect user feedback and optimize the intelligent recommendation algorithm based on the collected user feedback.

Images