Full-automatic oil feeding system based on equipment diagnosis operation and maintenance big data model

Abstract

The invention discloses a full-automatic oil feeding system based on an equipment diagnosis operation and maintenance big data model, and relates to the technical field of equipment lubrication, and the system comprises a three-dimensional state monitoring sensing device which collects vibration, temperature and displacement data of an equipment bearing in real time; the big data analysis module is connected with the three-dimensional state monitoring sensing device, internally provided with a bearing database model, and used for calculating the running state of the bearing according to the collected data and generating a diagnosis report; the precise oil feeding control unit is used for receiving an oil feeding quantity instruction of the big data analysis module and executing quantitative oil feeding to a lubricating point; the feedback monitoring module monitors faults in real time and feeds back states; and the self-learning module is used for dynamically adjusting a follow-up oil feeding quantity instruction according to the bearing operation data after oil feeding. According to the method, the oil feeding amount and the oil feeding state required during equipment operation are calculated through equipment monitoring data and a bearing big data model, manual intervention is not needed, and therefore the full-automatic oil feeding mode is achieved, the equipment operation efficiency is improved, and the production continuity is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of equipment lubrication technology, specifically to a fully automatic oiling system based on a big data model for equipment diagnosis and maintenance. Background Technology

[0002] Currently, there are three main methods for lubricating equipment bearings: manual lubrication, single-point lubrication, and automatic lubrication systems. Regardless of the method chosen, manual setting of the lubrication parameters and quantity is unavoidable. This setting is usually based on guidance provided in the equipment manual or relies on the operator's experience and intuition. However, in modern equipment management and digital equipment management, the scientific rigor and precision of these methods are clearly insufficient to meet the demands of high-precision equipment maintenance.

[0003] Existing technologies, such as CN109685227A, disclose a big data-based industrial equipment lubrication system and method. This system includes multiple modules such as an uploading unit, a query unit, a diagnostic unit, a lubrication unit, and a processing unit. Its core is to establish a large lubrication database and compare the lubrication information sent by the diagnostic unit with the database to determine specific lubrication operations and solutions. While this technology has certain theoretical advantages, in practical applications, it does not perform online optimization for key data such as bearing vibration, temperature, and displacement, resulting in an inability to achieve optimal equipment maintenance. Therefore, the existing lubrication methods can no longer meet the oiling requirements after equipment upgrades.

[0004] In view of this, and in response to the numerous problems mentioned above, this case came into being. Summary of the Invention

[0005] The purpose of this invention is to provide a fully automatic lubrication system based on a big data model for equipment diagnosis and maintenance, aiming to provide a more scientific and precise solution for equipment bearing lubrication, thereby filling the gaps in existing technologies and improving the overall level of equipment maintenance.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a fully automatic lubrication system based on a big data model for equipment diagnosis and maintenance, comprising:

[0007] A three-dimensional condition monitoring sensor is used to collect vibration, temperature, and displacement data of equipment bearings in real time.

[0008] The big data analysis module is connected to the three-dimensional condition monitoring sensor device, has a built-in bearing database model, calculates the bearing operating status based on the collected data and generates a diagnostic report;

[0009] The precision oil supply control unit receives the oil supply command from the big data analysis module and performs quantitative oil supply to the lubrication points;

[0010] The feedback monitoring module monitors lubrication point blockage or system malfunction in real time and provides feedback on the status via touchscreen / MES system / fault signal.

[0011] The self-learning module dynamically adjusts subsequent oil supply commands based on bearing operating data after oiling.

[0012] Preferably, the three-dimensional condition monitoring sensing device includes a high-frequency vibration sensor, an infrared temperature sensor, and an axial displacement sensor.

[0013] Preferably, the big data analysis module generates the fuel injection command through the following steps:

[0014] Compare with historical fault data in the bearing database;

[0015] When the diagnostic report indicates "low grease", calculate the required grease volume based on the bearing model, speed, and load parameters;

[0016] Output encrypted commands to the precision fuel supply control unit.

[0017] Preferably, the precision oil supply control unit includes:

[0018] A multi-channel metering pump, with each channel corresponding to an independent lubrication point;

[0019] Solenoid valve assembly controls the on / off supply of lubricating grease;

[0020] Flow meter, real-time calibration of output grease volume.

[0021] Preferably, the feedback monitoring module triggers an alarm when it detects any of the following conditions:

[0022] The pressure at the lubrication point continues to exceed the threshold.

[0023] The deviation between the measured value and the commanded value of the flow meter is > ±5%;

[0024] System power / communication interruption.

[0025] Preferably, the self-learning module optimizes the model in the following ways:

[0026] Analyze the rate of change of bearing vibration data after lubrication;

[0027] If the vibration reduction does not meet expectations, increase the amount of fuel injected next time;

[0028] If the temperature rise exceeds the limit, reduce the amount of oil applied next time and mark the bearing as abnormal.

[0029] Preferably, the big data analysis module is deployed on an edge computing gateway or cloud server and communicates with industrial equipment via the OPC UA protocol.

[0030] Preferably, the system further includes a data storage unit for storing bearing database models, historical operating data, and lubrication records;

[0031] The bearing database model is a neural network model based on machine learning, and the training data includes bearing failure cases, material property maps, and operating condition simulation data.

[0032] Preferably, the system executes synchronously when a fault signal is triggered:

[0033] Lock the current fuel injection command;

[0034] Push fault codes and handling suggestions to the MES system;

[0035] Store abnormal data in the security log.

[0036] A fully automated lubrication method based on a big data model for equipment diagnosis and maintenance, for bearing lubrication scenarios, includes the following steps:

[0037] The operating data of the equipment bearings are collected through a three-dimensional condition monitoring sensor.

[0038] Based on the bearing database model and big data analysis, determine whether the bearing needs lubrication and the required amount of oil;

[0039] Send a quantitative oil supply command to the precision oil supply control unit;

[0040] Monitor the lubrication process and issue an alarm when abnormalities occur;

[0041] Based on the feedback from the equipment's operating data after lubrication, optimize the subsequent lubrication strategy;

[0042] The judgment step includes using machine learning algorithms to analyze historical data in order to predict the optimal fuel injection amount and timing.

[0043] The lubrication command includes the amount of oil supplied, the frequency of oil supply, and the information on the lubrication point identification.

[0044] Compared with the prior art, the beneficial effects of the present invention are: the fully automatic lubrication system based on the equipment diagnosis and maintenance big data model relies on the equipment diagnosis and maintenance big data model, uses the data obtained by online monitoring of equipment bearings, performs calculations through the big data model, and cooperates with the precision lubrication control system to achieve a fully automatic lubrication mode for grease without manual intervention.

[0045] By monitoring data from online equipment diagnostics, such as bearing vibration, temperature, and displacement, the operating status of the equipment and the causes affecting that status can be determined, and relevant equipment diagnostic reports can be generated. These reports enable equipment managers to perform predictive maintenance, thereby improving equipment uptime and ensuring sustained production continuity. Within these reports, the bearing database and operating data can be used to calculate whether a certain condition is caused by "lack of lubrication," and instructions can be automatically sent to the lubrication system to achieve unmanned, automated lubrication.

[0046] By combining equipment monitoring data with a bearing big data model, the required lubrication quantity and lubrication status during equipment operation can be calculated without manual intervention, thus achieving a fully automated lubrication mode. This eliminates the instability of manually set lubrication data, improves equipment operating efficiency, and ensures production continuity. Attached Figure Description

[0047] Figure 1 This is a schematic diagram of the system architecture of the present invention;

[0048] Figure 2 This is a schematic diagram of the optimization logic of the self-learning module of the present invention;

[0049] Figure 3 This is a schematic diagram of the verification data for this invention. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] Please see Figures 1-3 The present invention provides the following technical solutions:

[0052] A fully automated lubrication system based on a big data model for equipment diagnosis and maintenance includes:

[0053] A three-dimensional condition monitoring sensor is used to collect vibration, temperature, and displacement data of equipment bearings in real time.

[0054] The big data analysis module connects to the three-dimensional condition monitoring sensor device, has a built-in bearing database model, calculates the bearing operating status based on the collected data and generates a diagnostic report;

[0055] The precision lubrication control unit receives the lubrication quantity command from the big data analysis module and performs quantitative lubrication to the lubrication points;

[0056] The feedback monitoring module monitors lubrication point blockage or system malfunction in real time and provides feedback on the status via touchscreen / MES system / fault signal.

[0057] The self-learning module dynamically adjusts subsequent oil supply commands based on bearing operating data after oiling.

[0058] The three-dimensional condition monitoring sensing device includes a high-frequency vibration sensor (sampling rate ≥20kHz, range ±500g), an infrared temperature sensor (accuracy ±0.5℃, temperature range -20~150℃), and an axial displacement sensor (resolution 0.01mm).

[0059] The big data analytics module generates fuel injection commands through the following steps:

[0060] Compare with historical fault data in the bearing database;

[0061] When the diagnostic report indicates "low grease", calculate the required grease volume based on the bearing model, speed, and load parameters;

[0062] Output encrypted commands to the precision fuel supply control unit.

[0063] The precision fuel injection control unit includes:

[0064] Multi-channel piezoelectric ceramic metering pump (flow control accuracy ±1%), each channel has an independent lubrication point;

[0065] Solenoid valve assembly controls the on / off supply of lubricating grease;

[0066] Equipped with a HART protocol flow meter to verify the output grease volume in real time.

[0067] The feedback monitoring module triggers an alarm when it detects any of the following conditions:

[0068] When the lubrication point pressure continuously exceeds the threshold, P max =0.2×P rated (P rated (Rated working pressure);

[0069] The deviation between the measured value and the commanded value of the flow meter is > ±5%;

[0070] System power / communication interruption; failure code E201 is triggered if there is no signal for more than 500ms.

[0071] The self-learning module optimizes the model in the following ways:

[0072] Analyze the rate of change of bearing vibration data after lubrication;

[0073] If the vibration reduction does not meet expectations, increase the amount of fuel injected next time;

[0074] If the temperature rise exceeds the limit, reduce the amount of oil applied next time and mark the bearing as abnormal;

[0075] Implementation requires pre-setting such as Figure 2 The quantization thresholds shown are: K = bearing type coefficient (0.78 for ball bearings, 1.02 for roller bearings); P = bearing load (kN); RPM = rotational speed; T base =Ambient temperature.

[0076] The big data analytics module is deployed on an edge computing gateway (Intel i7 processor / 16GB RAM) and communicates with industrial equipment via the OPC UA (ISO / IEC 62541 standard) protocol, with a data transmission latency of <50ms.

[0077] The system also includes a data storage unit for storing bearing database models, historical operating data, and lubrication records;

[0078] The bearing database model is a machine learning-based neural network model. The training data includes bearing failure cases, material property maps, and operating condition simulation data. It is an LSTM neural network based on the TensorFlow framework, with 12 nodes in the input layer (6-dimensional vibration spectrum + temperature + displacement + rotational speed + load + ambient humidity).

[0079]

[0080]

[0081] The system executes synchronously when a fault signal is triggered:

[0082] Lock the current fuel injection command;

[0083] Push fault codes and handling suggestions to the MES system;

[0084] Store abnormal data in the security log.

[0085] A fully automated lubrication method based on a big data model for equipment diagnosis and maintenance, for bearing lubrication scenarios, includes the following steps:

[0086] The operating data of the equipment bearings are collected through a three-dimensional condition monitoring sensor.

[0087] Based on the bearing database model and big data analysis, determine whether the bearing needs lubrication and the required amount of oil;

[0088] Send a quantitative oil supply command to the precision oil supply control unit;

[0089] Monitor the lubrication process and issue an alarm when abnormalities occur;

[0090] Based on the feedback from the equipment's operating data after lubrication, optimize the subsequent lubrication strategy;

[0091] The judgment process includes using machine learning algorithms to analyze historical data in order to predict the optimal fuel injection rate and timing.

[0092] The lubrication command includes the amount of oil supplied, the frequency of oil supply, and information on the lubrication point.

[0093] Verification was performed using bearing operating data from multiple scenarios, and the results are as follows: Figure 3 As shown, this system improves the accuracy of vibration reduction control compared to traditional methods, increases the speed of temperature anomaly identification, and ensures a high fault identification rate, which helps to improve equipment operating efficiency and ensure production continuity.

[0094] Contents not described in detail in this specification are prior art known to those skilled in the art. Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A fully automated lubrication system based on a big data model for equipment diagnosis and maintenance, characterized in that, include: A three-dimensional condition monitoring sensor is used to collect vibration, temperature, and displacement data of equipment bearings in real time. The big data analysis module is connected to the three-dimensional condition monitoring sensor device, has a built-in bearing database model, calculates the bearing operating status based on the collected data and generates a diagnostic report; The precision oil supply control unit receives the oil supply command from the big data analysis module and performs quantitative oil supply to the lubrication points; The feedback monitoring module monitors lubrication point blockage or system malfunction in real time and provides feedback on the status via touchscreen / MES system / fault signal. The self-learning module dynamically adjusts subsequent oil supply commands based on bearing operating data after oiling.

2. The fully automatic lubrication system based on a big data model for equipment diagnosis and maintenance as described in claim 1, characterized in that: The three-dimensional condition monitoring sensing device includes a high-frequency vibration sensor, an infrared temperature sensor, and an axial displacement sensor.

3. The fully automatic lubrication system based on a big data model for equipment diagnosis and maintenance as described in claim 1, characterized in that: The big data analysis module generates fuel injection instructions through the following steps: Compare with historical fault data in the bearing database; When the diagnostic report indicates "low grease", calculate the required grease volume based on the bearing model, speed, and load parameters; Output encrypted commands to the precision fuel supply control unit.

4. The fully automatic lubrication system based on a big data model for equipment diagnosis and maintenance as described in claim 1, characterized in that: The precision oil supply control unit includes: A multi-channel metering pump, with each channel corresponding to an independent lubrication point; Solenoid valve assembly controls the on / off supply of lubricating grease; Flow meter, real-time calibration of output grease volume.

5. The fully automatic lubrication system based on a big data model for equipment diagnosis and maintenance as described in claim 1, characterized in that: The feedback monitoring module triggers an alarm when it detects any of the following conditions: The pressure at the lubrication point continues to exceed the threshold. The deviation between the measured value and the commanded value of the flow meter is > ±5%; System power / communication interruption.

6. The fully automatic lubrication system based on a big data model for equipment diagnosis and maintenance as described in claim 1, characterized in that: The self-learning module optimizes the model in the following ways: Analyze the rate of change of bearing vibration data after lubrication; If the vibration reduction does not meet expectations, increase the amount of fuel injected next time; If the temperature rise exceeds the limit, reduce the amount of oil applied next time and mark the bearing as abnormal.

7. The fully automatic lubrication system based on a big data model for equipment diagnosis and maintenance as described in claim 1, characterized in that: The big data analytics module is deployed on an edge computing gateway or cloud server and communicates with industrial equipment via the OPC UA protocol.

8. The fully automatic lubrication system based on a big data model for equipment diagnosis and maintenance as described in claim 1, characterized in that: The system also includes a data storage unit for storing bearing database models, historical operating data, and lubrication records; The bearing database model is a neural network model based on machine learning, and the training data includes bearing failure cases, material property maps, and operating condition simulation data.

9. The fully automatic lubrication system based on a big data model for equipment diagnosis and maintenance as described in claim 1, characterized in that: The system executes synchronously when a fault signal is triggered: Lock the current fuel injection command; Push fault codes and handling suggestions to the MES system; Store abnormal data in the security log.

10. A fully automated lubrication method based on a big data model for equipment diagnosis and maintenance, characterized in that, Using the system described in any of claims 1-9 for bearing lubrication scenarios, the following steps are included: The operating data of the equipment bearings are collected through a three-dimensional condition monitoring sensor. Based on the bearing database model and big data analysis, determine whether the bearing needs lubrication and the required amount of oil; Send a quantitative oil supply command to the precision oil supply control unit; Monitor the lubrication process and issue an alarm when abnormalities occur; Based on the feedback from the equipment's operating data after lubrication, optimize the subsequent lubrication strategy; The judgment step includes using machine learning algorithms to analyze historical data in order to predict the optimal fuel injection amount and timing. The lubrication command includes the amount of oil supplied, the frequency of oil supply, and the information on the lubrication point identification.

Images