Method, device, storage medium and machine equipment for bearing condition monitoring

Abstract

The present invention relates to a method, device, storage medium and machine equipment for bearing status monitoring. The method comprises: obtaining the temperature relationship between different bearings in the machine equipment (a); selecting a reference bearing in real time and determining the real-time temperature of the reference bearing as the reference temperature (b); determining the real-time normal temperature range of different bearings of the machine equipment in combination with the temperature relationship and the reference temperature (c); measuring the temperature of the bearing in the machine equipment (d); comparing the measured bearing temperature with the corresponding real-time normal temperature range (e). The device is used to execute the above method. The computer-readable medium carries instructions, and when the instructions are executed, the device executes the above method. The machine equipment comprises a bearing and a temperature sensor. When the bearing is monitored in condition by the above method, the temperature sensor measures the temperature of the bearing in the machine equipment.

Description

Technical Field

[0001] The present invention relates to a method, a device, a storage medium and the machine equipment itself for monitoring the condition of a bearing in the machine equipment. Background Art

[0002] In machinery and equipment such as gearboxes in wind turbines, it is necessary to monitor the working conditions of rotating components such as bearings, shafts, gears, etc., to avoid service failure of the wind power station due to failure of these rotating components, especially bearings.

[0003] However, current gearbox condition monitoring systems typically use vibration and temperature sensors mounted on the housing. These sensors can only monitor the condition of the support bearings. There is currently no effective condition monitoring solution for many other bearings in the gearbox, such as the planetary gear bearings.

[0004] also, Figure 1 The figure shows the lead time for identifying bearing failure based on existing bearing monitoring methods. Figure 1 As shown in Figure 1, current bearing condition detection methods based on bearing temperature can only identify bearing failures in their late stages. Furthermore, while vibration-based bearing condition detection methods can identify bearing failures in their early stages, as described above, they can only be used to monitor a subset of bearings in a gearbox and are very expensive to implement. Summary of the Invention

[0005] Therefore, the main purpose of the present invention is to propose a method, device, storage medium and machine equipment itself for condition monitoring of bearings in machine equipment, which can particularly identify bearing failure in the early failure stage of the bearing based on the bearing temperature, and can preferably be implemented at a lower cost.

[0006] To achieve the above objectives, a method for monitoring the condition of bearings in machine equipment is proposed, the method comprising:

[0007] Obtain the temperature relationship between different bearings in machinery and equipment;

[0008] selecting a reference bearing in real time and determining the real-time temperature of the reference bearing as a reference temperature;

[0009] Determine the real-time normal temperature range of different bearings of the machine equipment by combining the obtained temperature relationship and the determined reference temperature;

[0010] Measuring the temperature of bearings in machinery and equipment;

[0011] Compare the measured bearing temperature with the corresponding real-time normal temperature range.

[0012] This method is based on the fact that multiple components in the same machine equipment, such as bearings, have a correlation in temperature. Specifically, since there is a certain relationship between the temperature of each bearing and the condition parameters of the bearing, and there is a certain relationship between the corresponding condition parameters of different bearings, the temperature relationship between different bearings can be established through the above-mentioned condition parameters. Here, the temperature relationship between different bearings in the same machine equipment can be determined in real time in a reasonable manner, especially during the condition monitoring of the bearings. Preferably, the temperature relationship includes the temperature information of all bearings in the machine equipment that need to be condition monitored. Preferably, the temperature relationship includes the temperature information of each bearing under most working conditions of the machine equipment, especially under all working conditions. Preferably, the temperature relationship is expressed as the temperature range information of each bearing.

[0013] Here, a bearing that will not fail, or whose probability of failure is so low as to be negligible, is selected as a reference bearing. The existence of the reference bearing is based on the assumption that the probability of simultaneous failure of multiple bearings in the same machine equipment is extremely low and negligible. Preferably, the selection of the reference bearing and the determination of the corresponding reference temperature are performed in real time while the machine equipment is operating in an operational mode. Therefore, the reference temperature is associated with the specific operational mode of the machine equipment and with a specific monitoring moment.

[0014] Since the temperature relationship between different bearings in the same machine equipment is actually a relative relationship, by substituting a reference bearing and its corresponding reference temperature into the above temperature relationship, the temperature range of other bearings in the specific machine equipment under normal operating conditions at a specific monitoring moment can be obtained, that is, the above-mentioned real-time normal temperature range. Since the real-time normal temperature range of each bearing is provided, it is not necessary to set the alarm temperature value for identifying bearing failure based on bearing temperature to the maximum value under all possible operating conditions as in existing solutions. Therefore, by comparing the bearing temperature measured in real time with the corresponding real-time normal temperature range, it is possible to promptly identify whether the bearing temperature is abnormal based on the corresponding failure judgment rules, thereby determining whether the corresponding bearing has failed. In this case, it is possible to identify bearing failure in the early stage of bearing failure based on the bearing temperature. This can reduce gearbox downtime caused by bearing failure.

[0015] In a preferred embodiment, the temperature relationship between different bearings in the machine equipment is obtained by a machine learning method.

[0016] This method allows the temperature relationship between different bearings in a machine to be determined without requiring precise knowledge of various conditional parameters that may affect bearing temperature, particularly those that are difficult to calculate or measure. Furthermore, complex formulas describing the relationship between bearing temperature and these parameters are unnecessary. Consequently, the apparatus for performing the bearing condition monitoring method can be simplified, particularly without requiring the construction of multiple sensor units for measuring various conditional parameters. Furthermore, the amount of data computation required during the bearing condition monitoring method is also reduced.

[0017] Particularly preferably, the machine learning method includes: predefining a temperature relationship model between different bearings in the machine equipment; obtaining bearing temperature data of the bearings in the machine equipment under different working conditions of the machine equipment; and training the temperature relationship model with the help of the bearing temperature data.

[0018] The predefined temperature relationship model can be easily established, particularly by avoiding direct correlation with difficult-to-measure and calculable parameters such as clearance and lubricity. By measuring the actual bearing temperatures of bearings in actual machinery and equipment under different operating conditions and training the predefined temperature relationship model with this measured bearing temperature data, the temperature relationship model can be optimized for the specific machinery and equipment structure and operating conditions, thereby determining the temperature relationships between different bearings in the actual machinery and equipment.

[0019] Advantageously, the different operating states of the machine device include operating states of the machine device under different load conditions and / or under different speed conditions.

[0020] Thus, the machine equipment can be subjected to different load conditions and / or speed conditions, and the temperature data of the bearings in the machine equipment can be measured and recorded accordingly. Particularly advantageously, the different load conditions include the load conditions involved in all possible operating states of the machine equipment. Particularly advantageously, the different speed conditions include the speed conditions involved in all possible operating states of the machine equipment.

[0021] Advantageously, the steps in the machine learning method are performed during factory testing of the machine equipment and / or after the machine equipment is installed at the work site: obtaining bearing temperature data of bearings in the machine equipment under different working conditions of the machine equipment; and training the temperature relationship model with the help of the bearing temperature data.

[0022] Thus, the temperature relationship between different bearings in a machine can be obtained after the machine is assembled at the factory and / or after it is installed at the work site. The temperature relationship obtained in this manner indirectly relates to parameters that are difficult to calculate and measure, such as possible errors in the assembly of the machine itself and its installation at the work site, the actual clearance of each bearing, and the lubricant flow rate within the machine. Therefore, the temperature relationship obtained in this manner is highly adaptable to the actual machine, and the steps for obtaining the temperature relationship are easy to perform.

[0023] In a preferred embodiment, the bearing condition monitoring method includes: measuring the temperature of bearings used to support multiple components in a machine device that are approximately identical in structural design and operating conditions; and selecting a reference bearing from the bearings of these components by comparing the measured component bearing temperatures.

[0024] This is based on the assumption that the probability of simultaneous failure of bearings used in multiple components with nearly identical structural designs and operating conditions is extremely low and negligible. Therefore, the bearings used in these multiple components are preferably configured as identically as possible. Furthermore, the temperatures of the bearings in the multiple components should be similar or even identical at the same time. Therefore, by comparing these bearing temperatures, failed bearings can be promptly identified. Consequently, a non-failed component bearing can be selected in real time as a reference bearing, and its temperature used as the reference temperature.

[0025] In particular, when a machine includes a planetary gear transmission, planetary gears of the same stage in the planetary gear transmission can be considered components with the same structural design and operating conditions. Therefore, a reference bearing can be selected from the planetary gear bearings supporting the planetary gears of the same stage. The bearing condition monitoring method includes measuring the temperature of bearings of planetary gears of the same stage in the planetary gear transmission; and selecting a reference bearing from the bearings of these planetary gears by comparing the measured planetary gear bearing temperatures.

[0026] Advantageously, the bearing condition monitoring method includes selecting a reference bearing based on a temperature difference between bearings of components having the same structural design and operating conditions and / or a variation trend of the temperature difference.

[0027] In particular, the bearing condition monitoring method includes: selecting one of the bearings of components having the same structural design and working conditions, which has the lowest bearing temperature, as a reference bearing.

[0028] Particularly advantageously, the bearing condition monitoring method includes determining whether component bearings have failed based on the temperature difference and / or the changing trend of the temperature difference between bearings of components with identical structural design and operating conditions. In this case, failure identification of component bearings, such as planetary gear bearings, no longer requires comparison of their temperatures with corresponding real-time normal temperature ranges; instead, failure is already detected during the selection of a reference bearing.

[0029] In a preferred embodiment, in particular, with respect to the determination of the comparison result between the measured bearing temperature and the corresponding real-time normal temperature range, the bearing condition monitoring method includes: if the measured bearing temperature of the bearing is within the corresponding real-time normal temperature range, it is determined that the bearing is working well; if the measured bearing temperature of the bearing is outside the corresponding real-time normal temperature range and / or the degree of departure from the corresponding real-time normal temperature range increases over time, it is determined that the bearing has failed.

[0030] Advantageously, the bearing condition monitoring method is at least partially performed in the machine.

[0031] Advantageously, the bearing condition monitoring method is at least partially executed in a cloud server or in a remote physical server.

[0032] Accordingly, a device for monitoring the condition of a bearing in a machine device is also proposed, wherein the device is used to perform the above-mentioned bearing condition monitoring method. Here, the bearing condition monitoring device particularly includes a temperature sensing unit and a data processing unit.

[0033] When the bearing condition monitoring device executes the bearing condition monitoring method, the temperature sensing unit measures the temperature of the bearings in the machine equipment in at least one of the following steps: obtaining the temperature relationship between different bearings in the machine equipment; selecting a reference bearing in real time and determining the real-time temperature of the reference bearing as the reference temperature; determining the real-time normal temperature range of different bearings of the machine equipment in combination with the obtained temperature relationship and the determined reference temperature; measuring the temperature of the bearings in the machine equipment; and comparing the measured bearing temperature with the corresponding real-time normal temperature range.

[0034] When the bearing condition monitoring device executes the bearing condition monitoring method, the data processing unit performs data processing and calculation on the bearing temperature measured by the temperature sensing unit in at least one of the following steps: obtaining the temperature relationship between different bearings in the machine equipment; selecting a reference bearing in real time and determining the real-time temperature of the reference bearing as the reference temperature; determining the real-time normal temperature range of different bearings of the machine equipment in combination with the obtained temperature relationship and the determined reference temperature; measuring the temperature of the bearings in the machine equipment; and comparing the measured bearing temperature with the corresponding real-time normal temperature range.

[0035] In an advantageous embodiment, the bearing condition monitoring device is integrated into the machine.

[0036] In an alternative embodiment, the temperature sensor unit in the bearing condition monitoring device is integrated into the machine equipment, and the data processing unit in the bearing condition monitoring device is partially integrated into the machine equipment.

[0037] Accordingly, a computer-readable storage medium is also provided, which carries instructions that, when executed, cause at least one device, particularly the aforementioned bearing condition monitoring device, to perform the aforementioned bearing condition monitoring method. The computer-readable storage medium may be tangible and non-transitory. The instructions may be one or more executable programs, or portions thereof, that cause the bearing condition monitoring device to perform the bearing condition monitoring method.

[0038] Accordingly, a machine equipment is also proposed, which includes a bearing and a temperature sensor assembly, wherein when the bearing in the machine equipment is monitored by the above-mentioned bearing condition monitoring method, the temperature sensor assembly measures the temperature of the bearing in the machine equipment in at least one of the following steps: obtaining the temperature relationship between different bearings in the machine equipment; selecting a reference bearing in real time and determining the real-time temperature of the reference bearing as the reference temperature; determining the real-time normal temperature range of different bearings of the machine equipment in combination with the obtained temperature relationship and the determined reference temperature; measuring the temperature of the bearing in the machine equipment; and comparing the measured bearing temperature with the corresponding real-time normal temperature range.

[0039] In a preferred embodiment, the machine equipment also includes at least one data processor, wherein, when the bearing condition monitoring method is used to monitor the condition of the bearings in the machine equipment, the at least one data processor performs data processing and calculations on the bearing temperature measured by the temperature sensor assembly in at least one of the following steps: obtaining the temperature relationship between different bearings in the machine equipment; selecting a reference bearing in real time and determining the real-time temperature of the reference bearing as the reference temperature; determining the real-time normal temperature range of different bearings of the machine equipment in combination with the obtained temperature relationship and the determined reference temperature; and comparing the measured bearing temperature with the corresponding real-time normal temperature range.

[0040] In one advantageous embodiment, the machine comprises multiple components that are approximately identical in terms of structural design and operating conditions. Advantageously, the machine can be a gearbox including a planetary gear train, wherein the planetary gears of the same stage in the planetary gear train are approximately identical in terms of structural design and operating conditions. It is particularly advantageous that the bearings of the planetary gears of the same stage in the planetary gear train are all equipped with wireless temperature sensors, preferably wireless passive temperature sensors. Alternatively, the machine can be a roller press having multiple rollers that are approximately identical in terms of structural design and operating conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The preferred embodiments of the present invention are schematically described below with reference to the accompanying drawings.

[0042] Figure 1 shows the lead time for identifying bearing failure according to existing bearing monitoring methods;

[0043] Figure 2 A flow chart of a bearing condition monitoring method according to a preferred embodiment is shown;

[0044] Figure 3 An embodiment of the step of selecting a reference bearing in the above preferred embodiment is shown;

[0045] Figure 4 An example of the step of identifying planetary gear bearing failure in the above preferred embodiment is shown. DETAILED DESCRIPTION

[0046] Figure 2 A flowchart of a bearing condition monitoring method according to a preferred embodiment is shown. In this embodiment, a bearing condition monitoring device executes the bearing condition monitoring method, described in detail below, to monitor the operating condition of all bearings in a gearbox for a wind turbine based on bearing temperature. In other embodiments, the operating condition of only a portion of the bearings in the gearbox may be monitored.

[0047] like Figure 2 As shown, in the bearing condition monitoring method according to this embodiment, it is first necessary to obtain the temperature relationship (a) between different bearings in the gearbox. Here, the temperature relationship between different bearings in the gearbox is obtained in particular by using a machine learning method.

[0048] To obtain the temperature relationships between the various bearings in a gearbox using a machine learning approach, a temperature relationship model (a1) is first predefined between the various bearings in the gearbox. Next, bearing temperature data (a2) is obtained for all bearings in the gearbox under various load and speed conditions of the gearbox or wind turbine, for example, using temperature sensors installed at each bearing, particularly during product testing after the gearbox is assembled at the factory or after the gearbox is installed in a wind turbine. This bearing temperature data is then used to train the temperature relationship model predefined in step (a1), for example, using a data processor integrated into the wind turbine, particularly the gearbox, or a data processing unit in a cloud server or a remote physical server, to optimize the temperature relationship model and thereby obtain bearing temperature relationships specifically tailored to the actual gearbox.

[0049] The bearing condition monitoring method of this embodiment also requires, during operation of the bearing condition monitoring device (i.e., monitoring the bearing conditions of all bearings in the gearbox), to identify, in real time, a reference bearing serving as a benchmark for bearing temperature relationships and its corresponding reference temperature. To this end, a reference bearing is selected in real time, and the real-time temperature of the reference bearing is determined as reference temperature (b).

[0050] Figure 3 The example of selecting a reference bearing in this embodiment is schematically shown. Since the probability of simultaneous failure of bearings used for multiple components that are approximately identical in terms of structural design and working conditions is extremely low, and the planetary gears of the same stage in the planetary gear transmission mechanism of the gearbox are the aforementioned approximately identical components, it is possible to select the bearings used to support the planetary gears of the same stage, for example, Figure 3 A reference bearing is selected from the four planetary gear bearings 1, 2, 3, and 4 shown in FIG. To this end, the temperature of each planetary gear bearing can be measured using wireless temperature sensors, particularly wireless passive temperature sensors, installed at each planetary gear bearing (b1). The measured planetary gear bearing temperatures can then be compared using a data processor integrated into the wind turbine, particularly the gearbox, or a data processing unit in a cloud server or a remote physical server. In this embodiment, the planetary gear bearing with the lowest temperature, namely the third planetary gear bearing 3, can be selected as the reference bearing, and the temperature of the third planetary gear bearing 3, 39°C, can be used as the corresponding reference temperature.

[0051] Then, if Figure 2As shown, the gearbox bearing temperature relationship obtained in step (a) is combined with the reference bearing and corresponding reference temperature determined in step (b) to comprehensively determine the real-time normal temperature range of each bearing in the gearbox (c). This step can also be performed by a data processor integrated in the wind turbine, especially the gearbox, or a data processing unit in a cloud server or a remote physical server.

[0052] Here, if Figure 2 As shown, it is also necessary to measure the temperature (d) of the corresponding gearbox bearing in real time by means of temperature sensors respectively installed at each gearbox bearing in the gearbox.

[0053] Then, if Figure 2 As shown, for each gearbox bearing in the gearbox, or for each gearbox bearing in the gearbox other than the planetary gear bearing that served as the alternative reference bearing in step (b), the gearbox bearing temperature measured in step (d) is compared (e) with the corresponding real-time normal temperature range at the monitoring time obtained in step (c), thereby determining whether the corresponding gearbox bearing has failed. This bearing failure determination step, performed through temperature comparison, can be performed by a data processor integrated into the wind turbine, particularly the gearbox, or a data processing unit in a cloud server or a remote physical server. Because the benchmark for determining bearing failure based on bearing temperature, i.e., the real-time normal temperature range, is optimized for the actual gearbox and is a judgment benchmark for the actual monitoring time, bearing temperature anomalies can be identified promptly, thereby enabling bearing failure identification based on bearing temperature at the early stages of bearing failure. This can reduce gearbox downtime caused by bearing failure. Furthermore, the method provided by this embodiment is cost-effective and easy to implement.

[0054] Here, the corresponding bearing failure judgment rule is, for example: if the temperature of the gearbox bearing measured in step (d) is within the corresponding real-time normal temperature range obtained in step (c), it is judged that the bearing is working well; if the temperature of the gearbox bearing measured in step (d) is outside the corresponding real-time normal temperature range obtained in step (c) and / or the degree of departure from the corresponding real-time normal temperature range increases over time, it is judged that the bearing has failed.

[0055] In one embodiment, the planetary gear bearings 1, 2, 3, 4 (see Figure 3 ) Whether the planetary gear has failed can be determined by the temperature difference of the planetary gear bearing temperature and / or the changing trend of the temperature difference. Figure 4 The temperature T of each planetary gear bearing 1, 2, 3, 4 is shown i (i=1, 2, 3, 4) and the lowest bearing temperature Tmin The temperature difference changes over time. Figure 4 It can be seen that the second planetary gear bearing 2 has the lowest bearing temperature T min The temperature difference of the third planetary gear bearing 3 (T2-T min Here, it can be determined that the second planetary gear bearing 2 has failed.

[0056] While the foregoing descriptions illustrate possible embodiments, it should be understood that numerous variations exist through combinations of all known and other technical features and implementations readily conceivable to a skilled artisan. Furthermore, it should be understood that the exemplary embodiments serve merely as examples and in no way limit the scope, application, or configuration of the present invention. The foregoing descriptions are intended primarily to provide a skilled artisan with technical guidance for implementing at least one exemplary embodiment. Various modifications, particularly regarding the functionality and structure of the components described, may be made without departing from the scope of the claims.

Claims

1. A method for monitoring the condition of a bearing in a machine device, the method comprising: Obtaining a temperature relationship between different bearings in the machine equipment (a); selecting a reference bearing in real time and determining the real-time temperature of the reference bearing as a reference temperature (b); Determining the real-time normal temperature range of different bearings of the machine equipment in combination with the temperature relationship and the reference temperature (c); Measuring the temperature of a bearing in the machine equipment (d); The measured bearing temperature is compared with the corresponding real-time normal temperature range (e).

2. The method according to claim 1, wherein: The temperature relationship is obtained by a machine learning method.

3. The method according to claim 2, wherein: The machine learning method comprises: Predefine a temperature relationship model (a1) between different bearings in the machine equipment; Acquiring bearing temperature data (a2) of a bearing in the machine equipment under different working conditions of the machine equipment; The temperature relationship model (a3) ​​is trained with the aid of the bearing temperature data.

4. The method according to claim 3, wherein: The different working states of the machine equipment include the operating states of the machine equipment under different load conditions and / or under different speed conditions.

5. The method according to claim 3, wherein: The following steps in the machine learning method are performed during factory testing of the machine equipment and / or after the machine equipment is installed at the work site: Acquiring bearing temperature data (a2) of a bearing in the machine equipment under different working conditions of the machine equipment; The temperature relationship model (a3) ​​is trained with the aid of the bearing temperature data.

6. The method according to claim 1, wherein: The method comprises: Measuring the temperature of bearings in the machine equipment for supporting a plurality of components having the same structural design and working conditions (b1); The reference bearing is selected from the bearings of the component by comparing the measured component bearing temperatures.

7. The method according to claim 6, wherein: In the case where the machine device includes a planetary gear transmission mechanism, the method includes: Measuring the temperature of the bearings of the same stage planetary gears in the planetary gear transmission mechanism; The reference bearing is selected from the bearings of the planetary gear by comparing the measured planetary gear bearing temperature.

8. The method according to claim 6, wherein: The method comprises: The reference bearing is selected based on the temperature difference of the component bearing temperature and / or the variation trend of the temperature difference.

9. The method according to claim 6, wherein: The method comprises: Select the bearing with the lowest bearing temperature (T min ) as the reference bearing.

10. The method according to claim 6, wherein: The method comprises: Whether the component bearing fails is determined based on the temperature difference of the component bearing temperature and / or the change trend of the temperature difference.

11. The method according to claim 1, wherein: The method comprises: If the measured bearing temperature of the bearing is within the corresponding real-time normal temperature range, it is judged that the bearing is working well; If the measured bearing temperature of the bearing is outside the corresponding real-time normal temperature range and / or the degree of departure from the real-time normal temperature range increases over time, it is determined that the bearing has failed.

12. The method according to any one of claims 1 to 11, wherein: The method is at least partially performed in the machine device.

13. The method according to any one of claims 1 to 11, wherein: The method is at least partially performed in a cloud server or in a remote physical server.

14. A device for condition monitoring of a bearing in a machine device, the device being configured to execute the method according to any one of claims 1 to 13.

15. A computer-readable storage medium carrying instructions which, when executed, cause at least one device to perform the method of any one of claims 1 to 13.

16. A machine device, comprising a bearing and a temperature sensor assembly, wherein: When the condition of the bearing is monitored by the method according to any one of claims 1 to 13, the temperature sensor assembly measures the temperature of the bearing in the machine equipment in at least one of the following steps: Obtaining a temperature relationship between different bearings in the machine equipment (a); selecting a reference bearing in real time and determining the real-time temperature of the reference bearing as a reference temperature (b); Determining the real-time normal temperature range of different bearings of the machine equipment in combination with the temperature relationship and the reference temperature (c); Measuring the temperature of a bearing in the machine equipment (d); The measured bearing temperature is compared with the corresponding real-time normal temperature range (e).

17. The machine device according to claim 16, wherein: The machine equipment further comprises at least one data processor, wherein, when the condition of the bearing is monitored by the method according to any one of claims 1 to 13, the at least one data processor performs data processing and calculation on the bearing temperature measured by the temperature sensor assembly in at least one of the following steps: Obtaining a temperature relationship between different bearings in the machine equipment (a); selecting a reference bearing in real time and determining the real-time temperature of the reference bearing as a reference temperature (b); Determining the real-time normal temperature range of different bearings of the machine equipment in combination with the temperature relationship and the reference temperature (c); Measuring the temperature of a bearing in the machine equipment (d); The measured bearing temperature is compared with the corresponding real-time normal temperature range (e).

18. The machine device according to claim 16, wherein: The machine structure comprises a plurality of components which are identical in terms of structural design and operating conditions.

Images