Bearing life prediction system constructed based on health indexes

Abstract

The invention relates to the technical field of bearing life prediction, and discloses a health index-based bearing life prediction system, which comprises a data acquisition module, an analysis processing module and an evaluation early warning module. According to the bearing service life prediction system constructed based on the health indexes, a sensing data unit and a feature data unit are arranged through a data acquisition module, a prediction data set is acquired, a constant prediction unit calculates a constant wear index # imgabs0 #, a variable prediction unit calculates a variable wear index # imgabs1 #, an evaluation early warning module calculates a health index # imgabs2 #, and the health index # imgabs2 # is calculated. According to the method, the abrasion degree of the bearing is rapidly calculated, manual overhaul is not needed, mistakes are not prone to being missed, time and labor are saved, comprehensive evaluation stability is high, the evaluation early warning module calculates the remaining life # imgabs4 # according to the health index # imgabs3 #, a reference data set is compared, the operation state of the bearing is judged, a signal and the health index are generated, and the index measurement detection result is more accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of bearing life prediction, and specifically to a bearing life prediction system constructed based on health indicators. Background Art

[0002] A bearing is a precision component used to support a rotating shaft of a machine. A bearing consists of an inner ring, an outer ring, rolling elements, a cage, etc. The main function of a bearing is to bear axial and radial loads, and at the same time reduce friction and resistance to ensure the smooth operation of mechanical equipment. Bearings play a crucial role in various industrial fields, such as automotive, aviation, aerospace, construction, manufacturing, etc. As a key basic component for mechanical operation, bearings are of great significance for ensuring the smooth operation of mechanical equipment. With the continuous development of technology, the types and performance of bearings are also constantly improving, providing strong support for the technological progress of various industrial fields. According to the structural characteristics and working principles of bearings, bearings are divided into several categories, such as sliding bearings, rolling bearings, spherical roller bearings, angular contact ball bearings, tapered roller bearings, spherical plain bearings, and magnetic bearings. Wear will occur when there is friction in a bearing, and the performance state of the bearing directly affects the operating efficiency and safety of the entire equipment. Therefore, fault detection and life prediction of bearings are of great significance for improving the operating reliability of equipment, reducing maintenance costs, and avoiding potential economic losses and personnel safety hazards. With the proposal of the concept of Prognostics and Health Management (PHM), the operation and maintenance strategy of rotating equipment has gradually changed from passive maintenance to active prediction. It can not only give early warnings of potential fault risks, but also optimize the use and maintenance strategies of equipment according to the prediction results, further improving the use efficiency and life of the equipment. Bearing fault detection and life prediction play a crucial role in modern industrial production. They can not only ensure the normal operation of equipment, but also save a large amount of maintenance and replacement costs for enterprises. During the operation of a bearing, vibrations will be generated, and these vibration signals contain a large amount of information about the bearing state. A widely used detection method is to measure these vibration parameters, such as vibration acceleration values and envelope values, to judge whether the working state of the bearing is normal. The vibration signals of the bearing are collected by vibration sensors. After the signals are denoised, feature extraction and analysis are performed on them to diagnose potential problems of the bearing, thereby predicting its remaining life and providing data support for further improving the bearing life.

[0003] Currently, due to factors such as bearing type and working environment, the accuracy of the predicted life results of traditional bearing life prediction systems is not stable enough. In actual use, in high-altitude areas, the atmospheric pressure is low, which causes the evaporation rate of lubricating oil to accelerate, thereby affecting the service life of the bearing. It is necessary to manually inspect a large number of bearings in sequence to judge the degree of wear and replace them in time to ensure the normal operation of the machine and extend the service life of the bearing. The labor intensity is high and it is easy to miss or make mistakes. Summary of the Invention

[0004] In view of the deficiencies of the prior art, the present invention provides a bearing life prediction system constructed based on health indicators, which has the advantages of high comprehensive evaluation stability and more accurate index measurement and detection results, and solves the problem of low accuracy of the traditional bearing life prediction system for predicting the bearing life in high altitude areas.

[0005] To achieve the above object, the present invention provides the following technical solution: A bearing life prediction system constructed based on health indicators, including a data acquisition module, an analysis and processing module, and an evaluation and warning module; The data acquisition module includes a sensing data unit and a feature data unit. The sensing data unit collects a sensing data set in real time by electrically connecting a vibration sensor, a thermometer, and a barometer, and transmits it to the data acquisition module through a network. The feature data unit collects a feature data set by connecting to a database through a network and transmits it to the data acquisition module through a network. The prediction data set is composed of the sensing data set and the feature data set. The data acquisition module is connected to the analysis and processing module through a network; The analysis and processing module numbers the prediction data set according to its characteristics. The analysis and processing module collects a reference data set by connecting to a database through a network and numbers it. The reference data set consists of the reference vibration frequency of bearing operation, the reference temperature of bearing operation, the reference air pressure of bearing operation, the reference characteristics of bearing operation, the reference constant wear index, the reference variable wear index, the reference health index, and the reference bearing life duration. The reference data set numbers correspond one by one to the prediction data set numbers. The analysis and processing module includes a constant prediction unit and a variable prediction unit. The constant prediction unit calculates the constant wear index according to the prediction data set and the reference data set and transmits it to the evaluation and warning module through a network. The variable prediction unit calculates the variable wear index according to the prediction data set and the reference data set and transmits it to the evaluation and warning module through a network. The analysis and processing module is connected to the evaluation and warning module through a network; The evaluation and warning module calculates the health index according to the constant wear index , the variable wear index and the reference data set, and saves it to the database through a network. The evaluation and warning module judges the bearing operation state according to the health index , compares it with the reference data set, generates corresponding signals and health indicators. The prompt alarm module includes a prompt unit and an alarm unit. The prompt unit transmits data by connecting to the database through a network according to the signal. The alarm unit calculates the remaining life according to the signal by comparing with the reference data set and outputs the data by connecting to the console through a network.

[0006] ​Preferably, the analysis and processing module numbers the sensor data set and the feature data set according to the characteristics of the prediction data set, and the sensor data set number is , and , the feature data set number is , , ,... .

[0007] Preferably, the analysis and processing module numbers the reference vibration frequency of the bearing operation, the reference temperature of the bearing operation, the reference air pressure of the bearing operation, the reference characteristics of the bearing operation, the reference constant wear index, the reference variable wear index, the reference health index and the reference life span of the bearing according to the characteristics of the reference data set, and the reference data set number corresponds to the prediction data set number one by one, and the reference data set number is , , , , , , and .

[0008] Preferably, the constant prediction unit calculates the constant wear index according to the prediction data set and the reference data set. , and its calculation formula is as follows: , represents the constant wear index, represents the tolerance parameter of the constant wear calculation formula, Indicates the vibration frequency generated by the current bearing operation. Indicates the actual temperature value of the current bearing operating environment. Indicates the actual air pressure value of the current bearing operating environment. It indicates the constant wear index calculated by substituting the actual vibration frequency, temperature and air pressure of the current bearing into the constant wear calculation formula.

[0009] Preferably, the variable prediction unit calculates the variable wear index according to the prediction data set and the reference data set. , and its calculation formula is as follows: In the formula, represents the variable wear index, It shows the intersection of the frequency value range under bearing variable vibration and the reference vibration frequency range under normal operation of the bearing. It represents the intersection of the bearing variable temperature range and the reference temperature range under normal operation of the bearing. It represents the intersection of the bearing variable air pressure value range and the reference air pressure range under normal operation of the bearing.

[0010] Preferably, the evaluation and early warning module calculates the health index according to the constant wear index , variable wear index and the reference data set, and the calculation formula is as follows: In the formula, represents the health index, represents the constant wear index, represents the proportion of the actual variable wear index within the range of the reference variable wear index, represents the reference characteristic maximum value of a specific type of bearing, represents the fixed average parameter of the health index formula.

[0011] Preferably, the evaluation and early warning module judges the operating state of the bearing by comparing the health index with the reference data set. When the health index is less than or equal to the reference health index , it is judged that the bearing is in a healthy state, and a prompt signal and a health index of 0 are generated accordingly. When the health index is greater than the reference health index , it is judged that the bearing is in a worn state, and an alarm signal and a health index of 1 are generated accordingly.

[0012] Preferably, the prompt unit connects to the database according to the prompt signal through the network, and transmits the constant wear index , variable wear index , variable wear index, health index and the bearing data with a health index of 0 to the database.

[0013] Preferably, the alarm unit counts the bearings with a health index of 1 according to the alarm signal, compares them with the reference data set, and calculates the remaining life , and the calculation formula is as follows: In the formula, represents the remaining life, represents the reference service life of the bearing, represents the used time of the bearing with a health index of 1.

[0014] Preferably, the alarm unit counts the number of bearings with a health index of 1 according to the alarm signal, and statistically counts their types, models and position data through the network connection to the database, and integrally outputs the constant wear index , variable wear index , health index and the remaining life of the bearings with a health index of 1and transmits the data to the console.

[0015] Compared with the prior art, the present invention provides a bearing life prediction system constructed based on health indicators, having the following beneficial effects: The present invention sets a sensing data unit and a feature data unit through a data acquisition module to collect a prediction data set. The analysis and processing module numbers the prediction data set according to the characteristics of the prediction data set, establishes a reference data set, and numbers it. The constant prediction unit calculates a constant wear index , normalizes the actual vibration frequency, temperature, and air pressure sensing data of the bearing operation, and the variable prediction unit calculates a variable wear index , timely senses the influence of variable factors on the bearing life. The evaluation and warning module calculates a health index according to the constant wear index , variable wear index and the reference data set . The system quickly calculates the wear degree of the bearing, does not require manual inspection, is not prone to omission and error, saves time and effort, evaluates and collects data more comprehensively, and has high overall evaluation stability; The present invention judges the bearing operation state by the evaluation and warning module according to the health index , compares with the reference data set. The health index reference health index , judges that the bearing is in a healthy state, and correspondingly generates a prompt signal and a health index of 0. The health index reference health index is greater than the reference health index , judges that the bearing is in a worn state, and correspondingly generates an alarm signal and a health index of 1. The prompt unit connects to the database according to the prompt signal through the network, and the alarm unit calculates the remaining life by comparing the alarm signal with the reference data set , counts the number, type, model, and position data of the bearings with a health index of 1, integrates and outputs them to the console, assisting the operator to intuitively understand the operation status and remaining life of the bearings, guiding the adoption of corresponding maintenance measures, reducing the failure rate and downtime of the equipment, and making the index measurement of the detection result more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a schematic diagram of the structure system of the present invention. EMBODIMENTS

[0017] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0018] Please refer toFigure 1 , a bearing life prediction system constructed based on health indicators, including a data acquisition module, an analysis and processing module, and an evaluation and warning module; The data acquisition module includes a sensing data unit and a feature data unit. The sensing data unit collects a sensing data set in real time through electrical connection with a vibration sensor, a thermometer, and a barometer. The vibration sensor, thermometer, and barometer are fixedly installed near the bearing, monitor and collect signals in real time, and transmit them to the data acquisition module through a network. The feature data unit collects a feature data set through a network connection to a database, collects feature data related to the bearing life, including the operating frequency domain feature, envelope feature, etc. of the bearing in the time domain, and transmits it to the data acquisition module through a network. The prediction data set is composed of the sensing data set and the feature data set. The data acquisition module is connected to the analysis and processing module through a network; The analysis and processing module numbers the prediction data set according to its characteristics. The sensing data set is numbered , and , and the numbers correspond to the actual vibration frequency, actual temperature, and actual air pressure of the bearing operation. The feature data set is numbered , , ,... , and the numbers correspond to the time domain feature, frequency domain feature, envelope feature, type feature, and life feature of the bearing operation. The analysis and processing module collects a reference data set through a network connection to a database and numbers it. The reference data set consists of the reference vibration frequency of the bearing operation, the reference temperature of the bearing operation, the reference air pressure of the bearing operation, the reference feature of the bearing operation, the reference constant wear index, the reference variable wear index, the reference health index, and the reference bearing life duration. The numbers of the reference data set and the prediction data set correspond one by one. The numbers of the reference data set are , , , , , , and . The analysis and processing module includes a constant prediction unit and a variable prediction unit. The constant prediction unit calculates the constant wear index according to the prediction data set and the reference data set, and transmits it to the evaluation and warning module through a network. Its calculation formula is as follows: In the formula, represents the constant wear index, represents the tolerance parameter of the constant wear calculation formula, represents the vibration frequency generated by the current bearing operation, represents the actual temperature value of the current bearing operation environment, represents the actual air pressure value of the current bearing operation environment, It represents the constant wear index calculated by substituting the actual vibration frequency, temperature value, and air pressure value of the current bearing operation into the constant wear calculation formula. According to the constant wear index , the actual vibration frequency, temperature, and air pressure sensing data of the bearing operation are normalized, so as to evaluate the collected data more comprehensively and have higher stability in subsequent predictions; The variable prediction unit calculates the variable wear index according to the prediction data set and the reference data set , and transmits it to the evaluation and warning module through the network. Its calculation formula is as follows: In the formula, represents the variable wear index, represents the intersection part of the frequency value range under variable vibration of the bearing and the reference vibration frequency range under the normal operation state of the bearing, represents the intersection part of the variable temperature value range of the bearing and the reference temperature range under the normal operation state of the bearing, represents the intersection part of the variable air pressure value range of the bearing and the reference air pressure range under the normal operation state of the bearing. According to the variable wear index , it can timely perceive the influence of variable factors on the bearing life, which is convenient for more accurate measurement of the bearing health index detection results in the future; The evaluation and warning module is based on the constant wear index. A bearing life prediction system constructed based on health indicators according to claim 3, wherein: the evaluation and warning module is based on the constant wear index , variable wear index and the reference data set, calculates the health index , and its calculation formula is as follows: In the formula, represents the health index, represents the constant wear index, represents the proportion of the actual variable wear index within the reference variable wear index range, represents the reference characteristic maximum value of a specific type of bearing, represents the fixed average parameter of the health index formula. According to the health index , it comprehensively analyzes the influence of current constant factors and periodic variable factors on the bearing life. The system can quickly calculate the wear degree of the bearing, without the need for manual inspection and evaluation, is not easy to miss or make mistakes, and saves time and effort; The evaluation and warning module judges the bearing operation state according to the health index , compares it with the reference data set, generates corresponding signals and health indicators. The health indicators, as a measurement tool, provide an objective and quantitative method to evaluate the bearing wear degree. By setting specific indicators, accurate measurement and evaluation make the results more valuable for reference, so as to make more reasonable decisions. The health index Reference health index , it is determined that the bearing is in a healthy state, and a prompt signal and a health index of 0 are generated correspondingly. The health index, Reference health index > Reference health index , it is determined that the bearing is in a worn state, and an alarm signal and a health index of 1 are generated correspondingly. The prompt alarm module includes a prompt unit and an alarm unit. The prompt unit is network-connected to the database according to the prompt signal and will measure the constant wear index , variable wear index , variable wear index, health index and the bearing data with a health index of 0 are transmitted to the database. The alarm unit compares the alarm signal with the reference data set and calculates the remaining life , and its calculation formula is as follows: In the formula, represents the remaining life, represents the reference life duration of the bearing, represents the used duration of the bearing with a health index of 1; The alarm unit counts the number of bearings with a health index of 1 according to the alarm signal, and statistically counts their types, models and location data through network connection to the database, and integrates and outputs the constant wear index , variable wear index , health index and the remaining life of the bearing with a health index of 1 data are transmitted to the console to assist the operator in intuitively understanding the operating conditions and remaining life of the bearing, guiding the adoption of corresponding maintenance measures, such as replacing the bearing, adjusting the operating parameters, etc., to reduce the failure rate and downtime of the equipment.

[0019] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A bearing life prediction system constructed based on health indicators, characterized in that: It includes a data acquisition module, an analysis and processing module, and an evaluation and early warning module; The data acquisition module includes a sensor data unit and a feature data unit. The sensor data unit collects a sensor data set in real time by electrically connecting a vibration sensor, a thermometer and a barometer, and transmits it to the data acquisition module through a network. The feature data unit collects a feature data set by connecting to a database through a network, and transmits it to the data acquisition module through a network. The predicted data set consists of a sensor data set and a feature data set. The data acquisition module is connected to an analysis and processing module through a network; the analysis and processing module numbers the predicted data set according to its features. The analysis and processing module collects a reference data set by connecting to a database through a network and numbers it. The reference data set consists of a reference vibration frequency of a bearing operation, a reference temperature of a bearing operation, a reference air pressure of a bearing operation, a reference feature of a bearing operation, a reference constant wear index, a reference variable wear index, a reference health index and a reference life span of a bearing. The reference data set number corresponds to the predicted data set number one by one. The analysis and processing module includes a constant prediction unit and a variable prediction unit. The constant prediction unit calculates the constant wear index based on the predicted data set and the reference data set. The variable prediction unit calculates the variable wear index based on the prediction data set and the reference data set. and transmitted to the evaluation and early warning module through the network, the analysis and processing module is connected to the evaluation and early warning module through the network; the evaluation and early warning module is based on the constant wear index , variable wear index and reference data sets to calculate the health index and is saved to the database through the network. The evaluation and early warning module is based on the health index , compare with the reference data set, judge the bearing operation status, generate corresponding signals and health indicators, the prompt alarm module includes a prompt unit and an alarm unit, the prompt unit connects to the database to transmit data according to the signal network, and the alarm unit compares the reference data set according to the signal to calculate the remaining life , and output data through the network connection console.

2. The bearing life prediction system based on health indicators according to claim 1, characterized in that: The analysis and processing module numbers the sensor data set and the feature data set according to the features of the prediction data set. The sensor data set number is , and , the feature data set number is , , ,... .

3. The bearing life prediction system based on health indicators according to claim 2, characterized in that: The analysis and processing module numbers the reference vibration frequency of bearing operation, the reference temperature of bearing operation, the reference air pressure of bearing operation, the reference feature of bearing operation, the reference constant wear index, the reference variable wear index, the reference health index, and the reference service life duration of the bearing according to the characteristics of the reference data set. The reference data set numbers correspond one by one to the prediction data set numbers, and the reference data set numbers are 、 、 、 、 、 、 and 。 4. The bearing life prediction system based on health indicators according to claim 3, characterized in that: The constant prediction unit calculates a constant wear index based on a prediction data set and a reference data set , and its calculation formula is as follows: In the formula, represents the constant wear index, represents the tolerance parameter of the constant wear calculation formula, represents the vibration frequency generated by the current bearing operation, represents the actual temperature value of the current bearing operation environment, represents the actual air pressure value of the current bearing operation environment, represents the constant wear index calculated after substituting the actual vibration frequency, temperature value and air pressure value of the current bearing operation into the constant wear calculation formula.

5. The bearing life prediction system based on health indicators according to claim 3, wherein: The variable prediction unit calculates a variable wear index based on a prediction data set and a reference data set , and its calculation formula is as follows: In the formula, represents the variable wear index, represents the intersection of the frequency value range under variable vibration of the bearing and the reference vibration frequency range under the normal operating state of the bearing, represents the intersection of the variable temperature value range of the bearing and the reference temperature range under the normal operating state of the bearing, represents the intersection of the variable air pressure value range of the bearing and the reference air pressure range under the normal operating state of the bearing.

6. The bearing life prediction system based on health indicators according to claim 3, wherein: The evaluation and warning module calculates the health index based on the constant wear index , the variable wear index , and the reference data set. The calculation formula is as follows: In the formula, represents the health index, represents the constant wear index, represents the proportion of the actual variable wear index within the range of the reference variable wear index, represents the maximum reference feature value of a specific type of bearing, represents the fixed average parameter of the health index formula.

7. The bearing life prediction system based on health indicators according to claim 6, characterized in that: The evaluation and warning module determines the operating state of the bearing by comparing the health index with the reference data set. When the health index is less than or equal to the reference health index , it is determined that the bearing is in a healthy state, and a prompt signal and health index 0 are generated accordingly. When the health index is greater than the reference health index , it is determined that the bearing is in a worn state, and an alarm signal and health index 1 are generated accordingly.

8. The bearing life prediction system based on the construction of health indicators according to claim 7, wherein: The said prompting unit is network-connected to the database according to the prompting signal, and transmits the constant wear index , variable wear index , variable wear index, health index and the bearing data with a health index of 0 to the database.

9. The bearing life prediction system based on health indicators as claimed in claim 7, wherein: The alarm unit counts the bearings with a health index of 1 based on the alarm signal, compares with the reference data set, and calculates the remaining life , and its calculation formula is as follows: In the formula, represents the remaining life, represents the reference life duration of the bearing, represents the used duration of the bearing with a health index of 1.

10. A bearing life prediction system constructed based on health indicators according to claim 9, characterized in that: The alarm unit counts the number of bearings with a health index of 1 based on the alarm signal, and statistically calculates their types, models, and location data through a network connection to the database, and integrates and outputs a constant wear index , a variable wear index , a health index , and the remaining life of the bearings with a health index of 1 The data is transmitted to the console.

Images