Equipment health management method and device based on Internet of Things

Through the combination of Internet of Things technology, intelligent early warning and machine learning models, the shortcomings of real-time and intelligent perception in existing equipment fault diagnosis and health management are solved, real-time monitoring and predictive maintenance of equipment health status are realized, and maintenance costs are reduced And improves equipment operation safety.

CN119066550APending Publication Date: 2024-12-03ROCKET FORCE UNIV OF ENG
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202411066970.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-05
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

The existing equipment fault diagnosis and health management methods lack real-time and intelligent perception, relying on manual judgment and offline data, resulting in high maintenance costs and lack of system learning capabilities.

Method used

The equipment health management method based on the Internet of Things is adopted to monitor the equipment status in real time through sensor networks, and fault warning, diagnosis and predictive maintenance are used to use intelligent early warning and machine learning models to achieve visual monitoring and predictive maintenance decisions of equipment health.

Benefits of technology

Real-time monitoring and fault warning of equipment health status is realized, maintenance costs are reduced, equipment operation safety is improved, and future failure rates are reduced through predictive maintenance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119066550A_ABST
    Figure CN119066550A_ABST
Patent Text Reader

Abstract

The invention discloses an equipment health management method and device based on the Internet of Things, and relates to the technical field of equipment health management, and the method specifically comprises the following steps: S1, equipment health monitoring and sensing; s2, preprocessing equipment health state monitoring data; s3, equipment fault intelligent early warning is carried out; s4, carrying out equipment fault diagnosis; s5, equipment fault analysis and optimization: positioning a fault part through S4, judging a fault type, providing an emergency scheme and a corresponding subsequent maintenance scheme, analyzing fault key factors, and summarizing a fault feature rule; s6, equipment predictive maintenance is carried out; according to the invention, a machine learning modeling method is utilized, real-time operation data and historical data of the equipment are fused, the overall monitoring of the health degree of the equipment can be realized in a visual mode, the unhealthy state of the equipment can be found in time, and the situation that the abnormal condition of the equipment needs to be manually checked and judged in the past is changed, so that an equipment prediction maintenance strategy is given; the maintenance cost is reduced, and the operation safety of the equipment is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of equipment health management, and in particular to an equipment health management method and device based on the Internet of Things. Background Art

[0002] With the rapid development of modern science and technology, modern engineering systems are becoming increasingly complex, and the integration and intelligence of systems are constantly improving. The maintenance and support costs of modern engineering complex systems are also becoming increasingly high. At present, internationally, the fault prediction and health management of large and complex equipment are increasingly becoming an important basis for autonomous logistics support. Through the Internet of Things technology, the online operation data of equipment can be obtained in real time, and the data flow and information flow can be analyzed and processed.

[0003] At present, the fault diagnosis and health management methods of equipment are mainly determined based on the experience of technicians and offline data. Although condition monitoring technology is used in the operation process, some information related to the equipment status is collected, and the corresponding maintenance strategy is determined based on the evaluation of the equipment status, this model has poor real-time performance and lacks online collection of intelligent perception information on the health status of the equipment, as well as the comprehensive application of real-time perception status. The equipment maintenance work does not have the ability of systematic learning. For this reason, we propose an equipment health management method and device based on the Internet of Things. Summary of the invention

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] An equipment health management method based on the Internet of Things specifically includes the following steps:

[0006] S1. Equipment health status monitoring and perception: By deploying sensors and establishing a sensor network, real-time monitoring and perception of equipment health status based on the Internet of Things is achieved, and equipment health status data is collected and stored;

[0007] S2. Equipment health status monitoring data preprocessing: Use sensors to monitor equipment online, collect and analyze equipment real-time status parameters and upload them in a timely manner, understand the equipment's operating conditions, filter and organize data, identify working condition information in the data, and eliminate non-important variables;

[0008] S3. Intelligent warning of equipment failure: The abnormal critical value of equipment health is set through the intelligent warning monitoring formula. When the health reaches the abnormal critical point, the system issues a health alarm, and when the health reaches the critical point of failure, a fault warning is issued. The health curve of the equipment's operating status is displayed in real time in a visual way through screen projection, realizing three levels of alarm: normal health, health alarm and equipment failure warning;

[0009] S4. Equipment fault diagnosis: Through the established equipment fault diagnosis model, abnormal parts of the equipment are located and diagnosed, and the equipment faults are attributed and analyzed. The factors causing the faults are discovered, and the characteristic variables of the faults are summarized. By comparing and monitoring the real-time operation data and historical data of each key component, abnormal fluctuations in the status of the components are discovered, so as to determine which part of the fault has caused the decline in equipment health;

[0010] S5, Equipment Fault Analysis and Optimization: Through S4, locate the fault part, determine the fault type, provide emergency plan and corresponding follow-up maintenance plan, analyze the key factors of the fault, summarize the fault characteristics, and propose preventive measures to reduce the occurrence rate of future faults;

[0011] S6. Predictive maintenance of equipment: Based on S4 fault diagnosis and S5 fault analysis, maintenance decision measures are proposed, and a comprehensive assessment of the equipment structure recovery status is conducted after the maintenance task is completed to confirm the effectiveness of the maintenance task. The equipment maintenance department can use machine learning prediction models to predict the performance and remaining service life of equipment or components. Before a serious equipment downtime accident occurs, the use of effective prediction functions can ensure that there is enough time to formulate and implement maintenance plans.

[0012] The machine learning prediction model is to learn the characteristics of equipment historical data and fit its nonlinear evolution law. The process is as follows:

[0013] Step 1: Collect historical fault data and performance degradation data of multiple groups of equipment, as well as intelligent perception data based on the Internet of Things;

[0014] Step 2: Standardize the data.

[0015] Step 3: Create a time series data set for the standardized equipment data and divide it into a training set, a validation set, and a test set;

[0016] Step 4: Build a machine learning prediction model, set parameters and hyperparameters, train the model using the training set, and verify the model training effect using the validation set;

[0017] Step 5: Take a certain number of frames of historical fault data samples in the test set and input them into the machine learning prediction model. The model will predict the performance evolution process and remaining life of the equipment for a certain period of time.

[0018] Preferably, in S3, normal health, health warning and equipment failure warning are indicated by a green light, a yellow light and a red light respectively.

[0019] Preferably, in S3, the critical values ​​of the yellow light and the red light may be set to 30% and 50% respectively according to the equipment health requirement.

[0020] Preferably, the intelligent early warning monitoring formula in S3 is as follows: Assuming that the equipment has m state points in operation, and each state point is composed of n observation points, the observation vector at a certain moment is defined as

[0021] Y i =[X1(H i ),X2(H i ),...,X n (H i )],

[0022] Where Y i is the observation vector of the equipment in the i-th operating state, i = 1,…,m, X j (H i ) is the observation value of the equipment at the jth observation point in the ith operating state, j = 1, 2, …, n.

[0023] Preferably, the comprehensive evaluation method of equipment performance in S6 is as follows: T i Set as the current time, T i +1 is the starting time for the endpoint or equipment to perform the next task in a given time period, E n is the performance threshold of the equipment in normal state, Et is the performance threshold of the equipment in available state, and i The previous performance characteristic parameter observations were analyzed to obtain T i +1 predicted value, if E(T i +1)≥E n , the equipment structure is currently in normal state; if E(T i )>Et and E(T i +1)<E n , the equipment performance is in a declining state; E(T i )≤Et, the equipment structure has failed or the equipment performance has reached the lower limit and is unusable.

[0024] A device for equipment health management based on the Internet of Things comprises a memory and at least one processor, wherein the memory stores instructions, and the at least one processor calls the instructions in the memory so that the device for equipment health management based on the Internet of Things executes any one of the methods for equipment health management based on the Internet of Things.

[0025] Compared with the prior art, the present invention has the following beneficial effects:

[0026] 1. The present invention is based on the Internet of Things technology. By rationally arranging sensors, the health status data of equipment can be intelligently sensed. The real-time operation data of equipment and historical data can be integrated by using the machine learning modeling method. The overall monitoring of the health of equipment can be realized in a visual way, and the unhealthy state of equipment can be discovered in time. This changes the previous method of manually checking and judging abnormal conditions of equipment or traditional preventive maintenance, thereby providing equipment predictive maintenance strategies, which is conducive to reducing maintenance costs and improving the operation safety of equipment.

[0027] The present invention detects abnormal fluctuations in component status through comparative monitoring of real-time operating data and historical data of various key components, determines which part of the equipment health is reduced due to failure, and assists in positioning and diagnosing abnormal parts of the equipment. By locating the fault part and determining the fault type, the module prompts emergency plans and corresponding follow-up maintenance plans. The module analyzes the attribution of the fault, screens the characteristic variables of the fault, analyzes the key factors of the fault, summarizes the fault characteristic laws, and thus proposes predictive maintenance measures to reduce the future failure rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a flow chart of an equipment health management method based on the Internet of Things proposed by the present invention. DETAILED DESCRIPTION

[0029] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0030] Example:

[0031] Reference Figure 1 , this solution provides an embodiment: a device health management method based on the Internet of Things, specifically comprising the following steps:

[0032] S1. Equipment health status monitoring and perception: By deploying sensors and establishing a sensor network, real-time monitoring and perception of equipment health status based on the Internet of Things is achieved, and equipment health status data is collected and stored;

[0033] S2. Equipment health status monitoring data preprocessing: Use sensors to monitor equipment online, collect and analyze equipment real-time status parameters and upload them in a timely manner, understand the equipment's operating conditions, filter and organize data, identify working condition information in the data, and eliminate non-important variables;

[0034] S3. Intelligent warning of equipment failure: The abnormal critical value of equipment health is set through the intelligent warning monitoring formula. When the health reaches the abnormal critical point, the system issues a health alarm, and when the health reaches the critical point of failure, a fault warning is issued. The health curve of the equipment's operating status is displayed in real time in a visual way through screen projection, realizing three levels of alarm: normal health, health alarm and equipment failure warning;

[0035] S4. Equipment fault diagnosis: Through the established equipment fault diagnosis model, abnormal parts of the equipment are located and diagnosed, and the equipment faults are attributed and analyzed. The factors causing the faults are discovered, and the characteristic variables of the faults are summarized. By comparing and monitoring the real-time operation data and historical data of each key component, abnormal fluctuations in the status of the components are discovered, so as to determine which part of the fault has caused the decline in equipment health;

[0036] S5, Equipment Fault Analysis and Optimization: Through S4, locate the fault part, determine the fault type, provide emergency plan and corresponding follow-up maintenance plan, analyze the key factors of the fault, summarize the fault characteristics, and propose preventive measures to reduce the occurrence rate of future faults;

[0037] S6. Predictive maintenance of equipment: Based on S4 fault diagnosis and S5 fault analysis, maintenance decision measures are proposed, and a comprehensive assessment of the equipment structure recovery status is conducted after the maintenance task is completed to confirm the effectiveness of the maintenance task. The equipment maintenance department can use machine learning prediction models to predict the performance and remaining service life of equipment or components. Before a serious equipment downtime accident occurs, the use of effective prediction functions can ensure that there is enough time to formulate and implement maintenance plans.

[0038] The machine learning prediction model is to learn the characteristics of equipment historical data and fit its nonlinear evolution law. The process is as follows:

[0039] Step 1: Collect historical fault data and performance degradation data of multiple groups of equipment, as well as intelligent perception data based on the Internet of Things;

[0040] Step 2: Standardize the data.

[0041] Step 3: Create a time series data set for the standardized equipment data and divide it into a training set, a validation set, and a test set;

[0042] Step 4: Build a machine learning prediction model, set parameters and hyperparameters, train the model using the training set, and verify the model training effect using the validation set;

[0043] Step 5: Take a certain number of frames of historical fault data samples in the test set and input them into the machine learning prediction model. The model will predict the performance evolution process and remaining life of the equipment for a certain period of time.

[0044] In this embodiment, normal health, health warning and equipment failure warning in S3 are indicated by a green light, a yellow light and a red light respectively.

[0045] In this embodiment, in S3, according to the equipment health requirement, the critical values ​​of the yellow light and the red light can be set to 30% and 50% respectively.

[0046] In this embodiment, the intelligent early warning monitoring formula in S3 is as follows: Assuming that there are m state points in the equipment operation state, and each state point is composed of n observation points, the observation vector at a certain moment is defined as

[0047] Y i =[X1(H i ),X2(H i ),...,X n (H i )],

[0048] Where Y i is the observation vector of the equipment in the i-th operating state, i = 1,…,m, X j (H i ) is the observation value of the equipment at the jth observation point in the ith operating state, j = 1, 2, …, n. The real-time data of the equipment has a similarity relationship with the predicted data of the state point. A similarity value is obtained by distance similarity calculation. As the equipment runs, the real-time data is continuously input into the model, thereby obtaining the similarity curve of the equipment's operating state.

[0049] In this embodiment, the comprehensive evaluation method of equipment performance in S6 is as follows: T i Set as the current time, T i +1 is the starting time for the endpoint or equipment to perform the next task in a given time period, E n is the performance threshold of the equipment in normal state, Et is the performance threshold of the equipment in available state, and i The previous performance characteristic parameter observations were analyzed to obtain T i +1 predicted value, if E(T i +1)≥E n , the equipment structure is currently in normal state; if E(T i )>Et and E(T i +1)<E n , the equipment performance is in a declining state; E(T i )≤Et, the equipment structure has failed or the equipment performance has reached the lower limit and is unusable. The remaining life of the equipment can be predicted by combining the comprehensive evaluation of the equipment performance.

[0050] An equipment health management device based on the Internet of Things includes a memory and at least one processor, wherein instructions are stored in the memory, and the at least one processor calls the instructions in the memory to enable the equipment health management device based on the Internet of Things to execute any one of the equipment health management methods based on the Internet of Things.

[0051] The above are only preferred specific implementation modes of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical solutions and inventive concepts of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for equipment health management based on the Internet of Things, characterized in that: The specific steps include: S1. Equipment health status monitoring and perception: By deploying sensors and establishing a sensor network, real-time monitoring and perception of equipment health status based on the Internet of Things is achieved, and equipment health status data is collected and stored; S2. Equipment health status monitoring data preprocessing: Use sensors to monitor equipment online, collect and analyze equipment real-time status parameters and upload them in a timely manner, understand the equipment's operating conditions, filter and organize data, identify working condition information in the data, and eliminate non-important variables; S3. Intelligent warning of equipment failure: The abnormal critical value of equipment health is set through the intelligent warning monitoring formula. When the health reaches the abnormal critical point, the system issues a health alarm, and when the health reaches the critical point of failure, a fault warning is issued. The health curve of the equipment's operating status is displayed in real time in a visual way through screen projection, realizing three levels of alarm: normal health, health alarm and equipment failure warning; S4. Equipment fault diagnosis: Through the established equipment fault diagnosis model, abnormal parts of the equipment are located and diagnosed, and the equipment faults are attributed and analyzed. The factors causing the faults are discovered, and the characteristic variables of the faults are summarized. By comparing and monitoring the real-time operation data and historical data of each key component, abnormal fluctuations in the status of the components are discovered, so as to determine which part of the fault has caused the decline in equipment health; S5, Equipment Fault Analysis and Optimization: Through S4, locate the fault part, determine the fault type, provide emergency plan and corresponding follow-up maintenance plan, analyze the key factors of the fault, summarize the fault characteristics, and propose preventive measures to reduce the occurrence rate of future faults; S6. Predictive maintenance of equipment: Based on S4 fault diagnosis and S5 fault analysis, maintenance decision measures are proposed, and a comprehensive assessment of the equipment structure recovery status is conducted after the maintenance task is completed to confirm the effectiveness of the maintenance task. The equipment maintenance department can use machine learning prediction models to predict the performance and remaining service life of equipment or components. Before a serious equipment downtime accident occurs, the use of effective prediction functions can ensure that there is enough time to formulate and implement maintenance plans.

2. According to the equipment health management method based on the Internet of Things according to claim 1, it is characterized in that: In S3, normal health, health warning and equipment failure warning are indicated by a green light, a yellow light and a red light respectively.

3. According to the equipment health management method based on the Internet of Things according to claim 1, it is characterized in that: In S3, according to the equipment health requirement, the critical values ​​of the yellow light and the red light can be set to 30% and 50% respectively.

4. The equipment health management method based on the Internet of Things according to claim 1, characterized in that: The intelligent early warning monitoring formula in S3 is as follows: Assuming that the equipment has m state points in operation, and each state point is composed of n observation points, the observation vector at a certain moment is defined as Y i =[X1(H i ),X2(H i ),...,X n (H i )], Where Y i is the observation vector of the equipment in the i-th operating state, i = 1,…,m, X j (H i ) is the observation value of the equipment at the jth observation point in the ith operating state, j = 1, 2, …, n.

5. The equipment health management method based on the Internet of Things according to claim 1, characterized in that: The comprehensive evaluation method of equipment performance in S6 is as follows: i Set as the current time, T i +1 is the starting time for the endpoint or equipment to perform the next task in a given time period, E n is the performance threshold of the equipment in normal state, Et is the performance threshold of the equipment in available state, and i The previous performance characteristic parameter observations were analyzed to obtain T i +1 predicted value, if E(T i +1)≥E n , the equipment structure is currently in normal state; if E(T i )>Et and E(T i +1)<E n , the equipment performance is in a declining state; E(T i )≤Et, the equipment structure has failed or the equipment performance has reached the lower limit and is unusable.

6. An equipment health management device based on the Internet of Things, characterized in that: The system includes a memory and at least one processor, wherein the memory stores instructions, and the at least one processor calls the instructions in the memory so that the equipment health management device based on the Internet of Things executes the equipment health management method based on the Internet of Things as described in any one of claims 1-5.

Citation Information

Cited By

  • Intelligent storage equipment health state evaluation and predictive maintenance method and system

    CN120338757A

  • Management system based on Internet of Things equipment health monitoring platform

    CN120528955A

  • Equipment health management system and method based on multi-channel parameter diagram

    CN120632681A