WO2022174441A1

WO2022174441A1 - Method and apparatus for monitoring health state of mechanical apparatus or mechanical component

Info

Publication number: WO2022174441A1
Application number: PCT/CN2021/077222
Authority: WO
Inventors: 王民刚; 董婷婷
Original assignee: 罗伯特·博世有限公司; 王民刚; 董婷婷
Priority date: 2021-02-22
Filing date: 2021-02-22
Publication date: 2022-08-25
Also published as: CN116888450A

Abstract

The present invention relates to a method (100) for monitoring the health state of a mechanical apparatus or a mechanical component. The method comprises at least the following steps: i) acquiring actual vibration data collected in or on a mechanical apparatus or a mechanical component; and ii) on the basis of a health diagnosis model, determining, from the actual vibration data, whether an anomaly is present in the mechanical apparatus or the mechanical component, and the type and/or grade of the anomaly, wherein the health diagnosis model is constructed on the basis of calculated vibration data calculated by means of a big data corrected kinetic model of the mechanical apparatus or the mechanical component. The present invention further relates to an apparatus for monitoring the health state of a mechanical apparatus or a mechanical component, and a corresponding computer program product and a corresponding vehicle. According to the present invention, real-time and reliable health state monitoring of an electric motor and a gear drive apparatus are realized.

Description

A method and apparatus for monitoring the health status of a mechanical device or mechanical component

technical field

The present invention relates to a method for monitoring the state of health of a mechanical device or mechanical component. Furthermore, the invention relates to a device for monitoring the state of health of a machine or machine component, a corresponding computer program product and a corresponding vehicle.

Background technique

With the enhancement of people's awareness of environmental protection and the promotion of relevant policies, the electrification of commercial vehicles has shown a great development trend. For commercial vehicle electrification, the main task is to select the optimum powertrain system for the vehicle. Today, long-axis motors are still a major trend in the market. Nevertheless, integrated electric drive axle and coaxial electric drive axle have received more and more attention and research and development because they can improve the efficiency of powertrain and save more battery space. However, an important disadvantage of these two types of electric drive axles is that they are subjected to high vibrations during the driving process of the vehicle as unsprung structures. This high vibration will bring great risks and challenges to the health of the electric drive axle, especially the motor and gear transmission.

Therefore, it is desirable to be able to monitor the health of electric motors and gearing in real time to make decisions about fault repair or predictive maintenance.

SUMMARY OF THE INVENTION

The object of the present invention is achieved by providing a method for monitoring the health state of a mechanical device or mechanical component in real time, which at least includes the following steps:

i) obtaining actual vibration data collected in or on said mechanical device or mechanical component; and

ii) Determining from actual vibration data whether an abnormality exists in a mechanical device or mechanical component and the type and/or level of abnormality present A corrected dynamic model is constructed from the calculated vibration data.

According to an optional embodiment of the present invention, the computational vibration data is acquired by the following method: based on the finite element method, a dynamic model is established for mechanical devices or mechanical components with different types and/or different levels of anomalies, respectively, and learning and correction through big data Each dynamic model; in turn, calculated vibration data of the mechanical device or mechanical component under different types and/or different levels of anomalies are calculated from the corrected dynamic model.

According to an optional embodiment of the present invention, the health diagnosis model is constructed in the following manner: extracting characteristic values of the calculated vibration data and constructing the health diagnosis model based on each characteristic value.

According to an optional embodiment of the present invention, a health diagnosis model is constructed in the following manner: based on each feature value for each type of abnormality, corresponding symptom feature families are respectively determined, wherein each symptom feature family includes at least one feature, and based on Each eigenvalue determines a corresponding range of abnormal eigenvalues for each feature in each symptom feature family, wherein step ii) is performed in the following manner: when it is identified that the actual vibration data for at least one feature of a symptom feature family is located in the corresponding When the eigenvalue is within the range of abnormal eigenvalues, it is concluded that the mechanical device or mechanical component has the abnormal type corresponding to the symptom feature family.

According to an optional embodiment of the present invention, the health diagnosis model is constructed in the following manner: based on the abnormality of each characteristic value at each level, the corresponding characteristic value interval is respectively determined, wherein step ii) is performed in the following manner: when the mechanical When the eigenvalues of the actual vibration data of the device or mechanical component are within a eigenvalue interval, it is concluded that the mechanical device or mechanical component has an abnormality of a level corresponding to the eigenvalue interval.

According to an optional embodiment of the present invention, the actual vibration data and the calculated vibration data relate to the same location in or on the mechanical device or mechanical component, wherein it is determined based on a dynamic model that the external surface of the mechanical device or mechanical component has At least one location that is sensitive to vibrations induced by an abnormality of a mechanical device or mechanical component is used as the site.

According to an optional embodiment of the present invention, the arrangement of the vibration measuring points used to collect actual vibration data satisfies any one, any multiple or all of the following conditions:

1) There are at least two vibration measuring points arranged at intervals from each other in the axial direction;

2) At least two vibration measuring points are arranged at intervals from each other in the circumferential direction;

3) There are at least two vibration measuring points arranged at intervals from each other in the radial direction;

4) There is at least one vibration measuring point on each of the two axial end faces of the shell;

5) At least one vibration measuring point is arranged close to the power input part and/or the power output part of the mechanical device or mechanical part.

According to an optional embodiment of the present invention, the kinetic model is corrected with the aid of test data from the laboratory before the kinetic model is corrected with the aid of big data.

In another aspect, the object of the present invention is also achieved by an apparatus for monitoring the state of health of a mechanical device or a mechanical component, the apparatus comprising a processor and a computer-readable storage medium communicatively connected to the processor, a computer The readable storage medium stores computer instructions which, when executed by the processor, implement the steps according to the method described above, wherein the apparatus is configured as a vehicle-side device or a server.

In the case where the device is configured as a server, the vehicle side transmits the vibration data collected by the vibration sensor arranged in the vehicle to the server in real time, periodically or in response to a data acquisition request.

In yet another aspect, the objects of the invention are also achieved by a computer program product comprising computer instructions which, when executed by a processor, implement the steps according to the method described above.

In a further aspect, the object of the invention is also achieved by a vehicle comprising an electric machine, a gearing and at least one vibration sensor arranged in and/or on the electric machine and/or the gearing, said vibration sensor communicatively connected with a device according to the above description, wherein, in particular, the at least one vibration sensor is at a position determined based on a dynamic model or a solution that satisfies any one or more of the above conditions 1)-5) Arranged in and/on the motor and/or gear transmission.

The present invention has the following advantages:

- Real-time and reliable monitoring of the health status of machinery or machinery components and the ability to predict impending failures in machinery or machinery components;

- Flexibility to adapt to mechanical devices or mechanical components with any possible vibration sensor arrangement;

-It is only necessary to arrange a vibration sensor on the casing of the mechanical device or mechanical component to grasp the health status of the mechanical device or mechanical component, with low modification difficulty and low cost.

Other advantages and advantageous embodiments of the inventive subject matter will be apparent from the description, drawings and claims.

Description of drawings

Further features and advantages of the present invention can be further elucidated by the following detailed description of specific embodiments with reference to the accompanying drawings. The attached drawings are:

FIG. 1 shows a structural block diagram of an apparatus for monitoring the health status of a mechanical device or a mechanical component according to an exemplary embodiment of the present invention;

FIG. 2 shows a flowchart of a method for monitoring the state of health of a mechanical device or mechanical component according to an exemplary embodiment of the present invention;

3A and 3B illustrate an arrangement scheme of vibration measuring points for a motor according to an exemplary embodiment of the present invention;

Fig. 4 shows the photograph of the test bench set up in order to obtain test data;

5A and 5B respectively show the vibration time domain data and vibration frequency domain data of the motor under different shaft eccentricities calculated by the dynamic model of the motor;

Figure 6 shows a flow chart of a step of the method according to the invention;

Figure 7 shows a flowchart of a sub-step of the steps shown in Figure 6;

Figure 8 shows a flow chart of another step of the method according to the invention;

Figure 9 shows a flow chart of a further step of the method according to the invention;

Figure 10 shows a flowchart of a sub-step of the steps shown in Figure 9;

Figure 11 shows a flow chart of a further step of the method according to the invention; and

FIG. 12 shows a flowchart of a sub-step of the steps shown in FIG. 11 .

Detailed ways

In order to make the technical problems, technical solutions and beneficial technical effects to be solved by the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and multiple exemplary embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the protection scope of the present invention. In the drawings, the same or similar reference numbers refer to the same or equivalent parts.

FIG. 1 shows a structural block diagram of an apparatus 1 for monitoring the health status of a mechanical device or mechanical component according to an exemplary embodiment of the present invention. The mechanical devices may broadly include various types of mechanical devices, such as electric motors and gear transmissions (such as transmissions) in drive trains, especially vehicle drive trains (especially integrated and coaxial transaxles). gearbox, reducer, differential). The mechanical components may broadly include various types of mechanical components, such as rotating parts (such as rotors, drive shafts, gears, or the like) used in motors and gear transmissions, as well as bearings (such as rolling bearings), flanges, flanges, etc. Blue, shell and bolts, etc.

When the device 1 is applied to a mechanical device, it is capable of monitoring the state of health of the mechanical device as a whole or at least one of its components. When the device 1 is applied to a mechanical component, it is capable of monitoring the state of health of the mechanical component as a whole or at least a part thereof.

The apparatus 1 includes a processor 10 and a computer-readable storage medium 20 communicatively connected to the processor 10. The computer-readable storage medium 20 stores computer instructions, which, when executed by the processor 10, implement the method according to the present invention. The steps of the method 100 of the invention will be described in detail below.

Further, the vibration sensor 30 is communicatively connected to the device 1 or the processor 10, and is disposed in or on the mechanical device or mechanical component for collecting vibration signals of the mechanical device or mechanical component. The vibration signal collected by the vibration sensor 30 can be acquired by the device 1 as test data to correct the dynamic model and the health diagnosis model of the mechanical device or mechanical component, which will be described in detail below, or as measurement data to analyze and evaluate the mechanical device or machine. The health status of the component.

The vibration sensor 30 may be any suitable type of sensor known in the art capable of capturing vibration signals, such as a vibration acceleration sensor.

In one example, at least one vibration sensor 30 can be arranged on the housing of the motor or gear when monitoring the state of health of the motor or gear by means of the device 1 .

In one example, the device 1 may be configured as a server, while the vibration sensor 30 is provided in or on the vehicle. In another example, both the device 1 and the vibration sensor 30 are provided in or on the vehicle. In this case, the health status of vehicle devices or vehicle components can be monitored in real time at the vehicle end and relevant warnings can be given to the vehicle user.

Figure 2 shows a flow diagram of a method 100 for monitoring the state of health of a mechanical device or mechanical component according to an exemplary embodiment of the present invention.

In the method 100, in step S110, a dynamic model of the mechanical device or mechanical component is established. Illustratively, the dynamic model can be built based on a finite element method.

In one example, a dynamic model of a healthy machine or machine component is established.

Additionally or alternatively, a dynamic model of the mechanical device or mechanical component where the anomaly exists is established. In this case, preferably, corresponding dynamic models are respectively established for mechanical devices or mechanical components having different types and/or different levels (ie, severity) of anomalies.

As used herein, the term "abnormality" should be broadly understood as any abnormal phenomenon occurring in a mechanical device or part of a mechanical device that reduces or degrades the function and/or efficiency of itself or the equipment in which it is located, which includes not only mechanical devices or Failure or defect in a mechanical part that has caused the function and/or characteristics of itself or the equipment in which it is located to deviate from the normal range, and also includes a mechanical device or mechanical part that causes the function and/or characteristics of itself or the equipment in which it is located. / or "sub-health" problems with decreased efficiency but not yet deviating from the normal range.

Exemplarily, in the case where the monitored mechanical device is a motor, a dynamic model of the motor can be established for different types of abnormality such as bearing abnormality, transmission shaft abnormality, rotor abnormality, etc. The abnormality of the outer ring, the abnormality of the inner ring of the bearing, the abnormality of the bearing cage, and the abnormality of the bearing rolling element are different types of abnormality to establish the dynamic model of the motor. Additionally or alternatively, a dynamic model of the motor can also be established separately for different levels of bearing outer ring anomalies or different levels of shaft eccentricity.

5A and 5B respectively show the vibration time domain data and vibration frequency domain data of the motor under different shaft eccentricity calculated by the dynamic model of the motor, wherein the time domain curves 21a, 22a, 23a and 24a are respectively The frequency domain curves 21b, 22b, 23b and 24b are obtained when the motor has shaft eccentricity of 0mm, 1mm, 3mm and 5mm, correspondingly, the frequency domain curves 21b, 22b, 23b and 24b are obtained when the motor has shaft eccentricity of 0mm, 1mm, 3mm and 5mm, respectively. acquired.

Next, in step S120, vibration measurement data of the mechanical device or mechanical component is acquired.

In an example, the acquired vibration measurement data may include test data acquired from a lab bench. Figure 4 shows a photo of the experimental bench set up to acquire test data. In the test bench, a plurality of vibration measuring points are arranged on the motor casing, and the vibration sensors located at the corresponding vibration measuring points can capture the corresponding vibration data.

Additionally or alternatively, the acquired vibration measurement data may include big data from the vehicle side, the OEM side, and/or the OEM side.

According to an embodiment of the present invention, step S120 further includes (see FIG. 6 ):

In step S121, at least one part in or on the mechanical device or mechanical component that is sensitive to vibration induced by the abnormality of the mechanical device or mechanical component is selected as a vibration measuring point.

In one example, vibration measuring points are selected on the outer surface of a mechanical device or mechanical component. In the case where the mechanical device is a motor or a gear transmission, the vibration measuring points are selected on the outer surface of the motor housing or the housing of the gear transmission.

In one example, locations in or on a mechanical device or mechanical component that are susceptible to abnormally induced vibrations can be determined by means of a dynamic model. In this case, step S121 further includes (see FIG. 7 ):

In step S1211, the vibration data of the mechanical device or mechanical component at M locations are calculated with the aid of the dynamic model, and the M locations are approximately uniformly distributed in the outer surface of the mechanical device or mechanical component at appropriate intervals, in particular , the number of these M parts is large enough to roughly outline the outer contour of the mechanical device or mechanical component;

Then, in step S1212, the M sets of vibration data are compared to filter out one or more sets of data with the most significant vibration response, and the parts corresponding to the filtered data are regarded as parts sensitive to vibration.

Additionally or alternatively, vibration measuring points can also be selected by general rule or experience. Generally speaking, in the case of a motor or gear transmission, the selected vibration measuring point can satisfy any one or more or all of the following conditions:

3) There are at least two vibration measuring points arranged at a distance from each other in the radial direction;

4) At least one vibration measuring point is provided on each of the two axial end faces of the housing.

Additionally, at least one vibration measuring point is arranged close to the input part (eg the input shaft) and/or the output part (eg the output shaft) of the electric motor or gear transmission.

Additionally or alternatively, any two of the vibration measuring points disposed on the housing may be made to differ in any one or more of axial, circumferential and radial directions.

3A and 3B illustrate schematic diagrams of a set of vibration measuring points A1 , A2 , A3 and A4 for the motor 50 according to an exemplary embodiment of the present invention. In this embodiment, the first vibration measuring point A1 is located at the radially outermost position of the motor casing, the second vibration measuring point A2 is located at the approximate axial center of the motor casing, and the third vibration measuring point A3 is located at the outermost position in the radial direction of the motor casing. The axial end face 51 on the output side of the motor housing is positioned close to the output shaft 52 , and the fourth vibration measuring point A4 is positioned on the other axial end face 53 of the motor housing opposite to the axial end face 51 .

The arrangement scheme of the vibration measuring points shown in FIGS. 3A and 3B complies with all of the above-mentioned conditions 1) to 4). Specifically, for condition 1), the vibration measuring points A1, A2, A3, and A4 are distributed with an interval from each other in the axial direction; for condition 2), the vibration measuring point A4 and the remaining vibration measuring points A1, A2, and A3 are circumferentially distributed. Upwards with a spaced arrangement; for condition 3), the vibration measuring points A1, A2, A3 and A4 are arranged with a distance from each other in the radial direction; for condition 4), the vibration measuring point A3 is arranged on the shaft of the output side of the motor housing On the end face 51, the vibration measuring point A4 is arranged on the other axial end face 53 of the motor housing.

Additionally, the selected vibration measuring points are arranged such that the overall vibration measuring points can comprehensively reflect the health status of all components in the mechanical device or mechanical components that are prone to vibration induced by their own defects.

Next, in step S122, the vibration sensor is arranged on the selected vibration measuring point.

Then, in step S123, the vibration measurement data of the mechanical device or the mechanical component is collected by means of the vibration sensor.

Subsequently, in step S130, each dynamic model established in step S110 is corrected using the vibration measurement data acquired in step S120.

In an example, step S130 further includes (see FIG. 8 ):

In step S131, the kinetic model is corrected using the test data collected from the bench; and

Then, in step S132, the kinetic model is further corrected through big data learning using the acquired big data.

Exemplarily, the dynamic model can be modified in at least one of the following ways: modifying material properties; modifying boundary conditions; modifying the positional deviation of the vibration data output part of the model relative to the vibration measuring points.

Illustratively, a dynamic model of a healthy machine or machine component is calibrated using vibration measurements collected from a healthy machine or machine component, utilizing vibration collected from a machine or machine component that has various types or levels of anomalies The measurement data corrects a kinetic model with corresponding deficiencies.

Next, in step S140, calculated vibration data of the mechanical device or mechanical component at the location corresponding to the arrangement position of the vibration sensor of the monitored mechanical device or mechanical component is calculated from the corrected dynamic model.

In an embodiment of the present invention, in step S140, calculated vibration data of a healthy mechanical device or mechanical component, ie, so-called baseline data, is calculated by means of a dynamic model established based on the healthy mechanical device or mechanical component.

Additionally or alternatively, in step S140, the mechanical devices or mechanical components in different types and/or mechanical components are respectively calculated by means of the respectively established dynamic models based on the mechanical devices or mechanical components with different types and/or different levels of abnormality. Calculated vibration data for different levels of anomaly.

Then, in step S150, a health diagnosis model is constructed based on the calculated vibration data obtained in step S140.

In an example, step S150 further includes (see FIG. 9 ):

In step S151, feature extraction is performed on the calculated vibration data to obtain the feature value of each feature;

Next, in step S152, a health diagnosis model is constructed based on the feature extraction result.

In an example, step S152 further includes (see FIG. 10 ):

In step S1521, based on the extracted feature values for each type of abnormality, respectively determine corresponding symptom feature families, wherein each symptom feature family includes at least one feature, and is determined for each feature in each symptom feature family. The corresponding range of abnormal eigenvalues;

Then, in step S1522, the corresponding feature value intervals are respectively determined based on the extracted feature values for each level of abnormality.

According to the present invention, the health diagnosis model is constructed such that when it is identified that the actual vibration data of the mechanical device or mechanical component has an eigenvalue within the corresponding abnormal eigenvalue range for at least one feature of a symptom feature family, then the mechanical device is obtained. Or the conclusion that a mechanical component has an abnormal type corresponding to that symptom family. In particular, the health diagnosis model is constructed such that the mechanical device is only obtained when it is identified that the actual vibration data of the mechanical device or mechanical component has eigenvalues within the corresponding abnormal eigenvalue range for all the features of a symptom feature family Or the conclusion that a mechanical component has an abnormal type corresponding to that symptom family.

Additionally or alternatively, the health diagnosis model is constructed such that when it is identified that the eigenvalues of the actual vibration data of the mechanical device or mechanical component are located within a eigenvalue interval, it is concluded that the mechanical device or mechanical component exists corresponding to the eigenvalue interval. The level of abnormal conclusion.

Optionally, step S152 may further include determining a corresponding weight for each feature in the symptom feature family. Further, the health diagnostic model is constructed to take into account the weighting of features when determining the type and/or level of anomaly from actual vibration data of the machine or machine component.

In an example, in step S1521, the following features may be selected as features in the symptom feature family corresponding to a certain abnormality type: the relative health state of the feature value calculated by the dynamic model of the abnormality of this type The baseline eigenvalues below have clearly identifiable deviations.

In an example, in step S1521, the range of abnormal feature values for each abnormal type may be determined based on the corresponding feature values calculated by the dynamic model with various types of abnormalities and optionally the baseline feature values, respectively.

In one example, in step S1522, the feature value interval of each abnormality level may be determined respectively based on the corresponding feature value and optionally the baseline feature value calculated by the dynamic model of the abnormality with various levels.

In one example, the baseline eigenvalues may be extracted from calculated vibration data calculated by a dynamic model of a healthy mechanism or machine component, and may additionally or alternatively be from actual vibrations collected from a healthy mechanism or machine component Data Extraction.

In one example, different anomaly types may be assigned the same or different symptom feature families, wherein the different symptom feature families have at least one different feature, but it is not excluded that some of the same features may be included. Additionally or alternatively, a feature that is shared by multiple symptom feature families may have a variable range of outlier feature values. That is, the feature may have different or the same range of abnormal feature values for different abnormal types.

In one example, the extracted features include any suitable time or frequency domain features, such as:

i. Frequency domain amplitude;

ii. frequencies with significant magnitudes;

iii. the magnitude of said significant magnitude; and

iv. The dispersion index of the magnitude in the time domain, its first and second derivatives, including the dispersion index and/or the first or second derivative of the

1. Maximum, average and minimum values

2. The occurrence of a peak value greater than a certain threshold within a fixed time, wherein the certain threshold is determined by analytical model and finite element analysis;

3. Variance and Standard Deviation

4. Peak-to-peak ratio and peak-to-average ratio

5. Statistical distribution kurtosis and skewness; and

v. Significant time periods in the cepstrum, this is especially true for gearbox failures and some bearing failures.

Using the computational vibration data of the dynamic model to construct a health diagnosis model has the following advantages: on the one hand, the dynamic model that is double-corrected by the test data and big data has high confidence and the calculated vibration data output from this is highly close to the real value; On the one hand, the dynamic model can output the vibration response data of the mechanical device or mechanical component at any position, so that a health diagnosis model can be constructed that is suitable for the mechanical device or mechanical component with various sensor arrangements.

In an example, when using the test data and big data to correct the dynamics model in step S130, the test data and big data are additionally used to correct the health diagnosis model, especially the abnormal feature value range of the symptom feature family and its features can be corrected. and the eigenvalue interval for the ranks.

In one example, symptom feature families and their abnormal feature value ranges for their features and feature value intervals for grades are stored in a database for the device 1 to recall when performing a health diagnosis of a mechanical device or mechanical component. Optionally, the database can be updated as the kinetic model and the health diagnostic model are corrected.

Next, in step S160, the health status of the mechanical device or the mechanical component is monitored by means of the health diagnosis model.

In an example, step S160 further includes (see FIG. 11 ):

In step S161, obtain the actual vibration data of the monitored mechanical device or mechanical component;

Next, in step S162, whether there is an abnormality in the mechanical device or mechanical component and the type and/or level of the abnormality present are identified from the actual vibration data based on the health diagnosis model.

In an example, step S162 further includes (see FIG. 12 ):

In step S1621, feature extraction is performed on the actual vibration data to obtain the feature value of each feature;

Next, in step S1622, it is determined whether any symptom feature family is hit based on each feature value, wherein "hit" refers to the feature obtained in step S1621 for at least one, especially all features in the corresponding symptom feature family The value falls within its abnormal feature value range; if no symptom family is hit, in step S1625 the user is informed that the mechanical device or mechanical component is currently healthy and returns to continue monitoring; if there is a hit symptom family, then Draw the conclusion that the mechanical device or mechanical component has the abnormal type corresponding to the hit symptom feature family and go to step S1623;

In step S1623, the hit feature value interval is determined based on each feature value, and the mechanical device or mechanical component has an abnormal level corresponding to the hit feature value interval based on this, and then in step S1624, the user is notified of the mechanical device or mechanical component. Currently, there may be the abnormality type determined in step S1622 and the abnormality level determined in step S1623. Next, go back to continue monitoring.

In an example, step S1624 is performed in the following manner: audible or visual warning information is output to the user.

The method for selecting vibration measurement points used for acquiring vibration measurement data described above in the explanation of step S120 is also applicable to the collection of actual vibration data in step S160.

The method according to the present invention can not only identify faults that have already occurred in mechanical devices or mechanical components and require corresponding repairs, but also can identify abnormality in mechanical devices or mechanical components in advance before they have not evolved into actual faults. Existence, which can help relevant personnel to carry out relevant inspections and predictive maintenance, so as to avoid the further deterioration of abnormality or the occurrence of accidents.

While some embodiments have been described, these embodiments have been presented by way of example only, and are not intended to limit the scope of the inventions. The appended claims and their equivalents are intended to cover all modifications, substitutions and changes as fall within the scope and spirit of the invention.

Claims

A method (100) for monitoring the state of health of a mechanical device or mechanical component, the method comprising at least the steps of:

i) obtaining actual vibration data collected in or on said mechanical device or mechanical component; and

ii) Determining from actual vibration data whether there is an abnormality in a mechanical device or mechanical component and the type and/or level of the abnormality present A data-corrected dynamic model is constructed from the calculated vibration data.
The method (100) of claim 1, wherein:

The computational vibration data is obtained in the following ways: based on the finite element method, dynamic models are established for mechanical devices or mechanical components with different types and/or levels of anomalies, respectively, and each dynamic model is corrected through big data learning; The dynamic model calculates the calculated vibration data of a mechanical device or mechanical component under different types and/or different levels of anomalies.
The method (100) of claim 2, wherein:

The health diagnosis model is constructed in the following manner: extracting characteristic values from the calculated vibration data and constructing a health diagnosis model based on each characteristic value.
The method (100) of claim 3, wherein:

A health diagnosis model is constructed in the following manner: corresponding symptom feature families are respectively determined for various types of abnormalities based on each feature value, wherein each symptom feature family includes at least one feature, and based on each feature value, a Each feature determines a corresponding range of abnormal feature values, wherein step ii) is performed in the following manner: when it is identified that the actual vibration data has feature values that lie within the corresponding abnormal feature value range for at least one feature of a symptom feature family , then it is concluded that the mechanical device or mechanical component has the abnormal type corresponding to the symptom feature family.
The method (100) according to claim 3 or 4, characterized in that,

The health diagnosis model is constructed in the following way: based on the abnormality of each eigenvalue at each level, the corresponding eigenvalue interval is respectively determined, wherein, step ii) is performed in the following way: when the characteristics of the actual vibration data of the mechanical device or mechanical component are identified When the value is within a eigenvalue interval, it is concluded that the mechanical device or mechanical component has an abnormality of a level corresponding to the eigenvalue interval.
The method (100) according to any of the preceding claims, characterized in that,

The actual vibration data and the calculated vibration data relate to the same location in or on the mechanical device or mechanical part, wherein the external surface of the mechanical device or mechanical part is determined based on the dynamic model to be induced by the abnormality of the mechanical device or mechanical part The vibration-sensitive at least one location serves as the site.
The method (100) according to any one of the preceding claims, characterized in that the arrangement of the vibration measuring points for collecting actual vibration data satisfies any one, any multiple or all of the following conditions:

1) There are at least two vibration measuring points arranged at intervals from each other in the axial direction;

2) At least two vibration measuring points are arranged at intervals from each other in the circumferential direction;

3) There are at least two vibration measuring points arranged at intervals from each other in the radial direction;

4) There is at least one vibration measuring point on each of the two axial end faces of the housing;

5) At least one vibration measuring point is arranged close to the power input part and/or the power output part of the mechanical device or mechanical part.
The method (100) according to any of the preceding claims, characterized in that,

The dynamic model is corrected by means of test data from the laboratory before the dynamic model is corrected by means of big data, wherein the mechanical device or mechanical component is a power plant or power component, such as an electric motor or a gear transmission for a vehicle.
An apparatus (1) for monitoring the state of health of a mechanical device or mechanical component, comprising a processor (10) and a computer-readable storage medium (20) communicatively connected to the processor (10), the computer-readable The storage medium (20) stores computer instructions which, when executed by the processor (10), implement the steps of the method (100) according to any one of the preceding claims, wherein the device (1) Configured as a vehicle-side device or server.
A vehicle comprising an electric machine, a gear transmission and at least one vibration sensor (30) arranged in and/or on the electric machine and/or the gear transmission, the vibration sensor (30) being the same as the one according to claim 9 The device (1) is communicatively connected, wherein, in particular, the at least one vibration sensor (30) is arranged in and/on the motor and/or gearing in the manner defined in claim 6 or 7.