Motor bearing fault real-time detection method based on adaptive data acquisition

Abstract

The invention belongs to the field of fault detection, and discloses a motor bearing fault real-time detection method based on adaptive data acquisition. The method comprises the following steps: determining mounting positions of a vibration sensor and a rotating speed sensor, and setting detection parameters; acquiring N groups of periodic synchronous single-point sampling data Y; standard deviations of N groups of periodic synchronous single-point sampling data in the data Y are calculated respectively; whether a fault is generated or not is judged; and adaptive data acquisition is realized. According to the invention, through adaptive data acquisition, under the condition that the fault detection reliability is ensured, the data acquisition amount is reduced, and the data acquisition, data transmission and data processing costs are reduced.

Description

Technical Field

[0001] The invention belongs to the field of fault detection, and in particular relates to a real-time detection method for motor bearing faults based on adaptive data acquisition. Background Art

[0002] Motors are devices that realize the mutual conversion between electrical energy and mechanical energy, and are widely used in industry and daily life. Bearings are one of the key components of motors. They support the motor rotor and ensure the stability and accuracy of the motor during operation. The performance of bearings directly affects the working efficiency, noise level, temperature rise and life of the motor. Bearings may suffer various forms of damage during long-term use, such as pitting, plastic deformation, wear, fatigue peeling, fracture, etc. Bearing damage can cause abnormal vibration of the motor, thereby affecting the performance of the motor and even causing motor failure. Therefore, it is necessary to perform fault detection on the bearings during the operation of the motor to ensure the normal operation of the bearings.

[0003] At present, bearing fault detection often adopts fixed detection interval or variable detection interval. Fixed detection interval means to detect the health status of the bearing once at a certain time interval. For example, the health status of the bearing is detected once every other day. The duration of bearing fault detection is generally a few seconds or minutes. Variable detection interval means to determine the detection interval according to the bearing degradation state. Variable detection interval generally adopts a larger detection interval when the bearing degradation is relatively minor, such as fault detection once a week, and adopts a smaller detection interval when the bearing degradation is relatively serious, such as fault detection once an hour. Bearing fault may occur at any time during the use of the bearing, not at the fault detection period. In this case, the bearing fault detection methods based on fixed detection interval and variable detection interval cannot overcome the missed detection rate of fault detection.

[0004] In order to overcome the shortcomings of the above methods, some researchers have proposed the use of continuous fault detection methods, that is, continuous detection of bearing faults. From the perspective of data acquisition, this method assumes that the probability of bearing failure is the same in any working period, so the same sampling frequency is used to continuously detect bearing faults. This method also ignores the regularity of bearing degradation, which will seriously increase the detection cost and is not suitable for general industrial fields.

[0005] In order to overcome the shortcomings of the above two types of fault detection methods, the patent of the present invention discloses a real-time detection method for motor bearing faults based on adaptive data acquisition. The method disclosed in the patent of the present invention takes into account the degradation law of the bearing and adaptively determines the number of sampling data points in each rotation period of the motor, that is, when the bearing degradation is relatively mild, a larger sampling interval is used, and when the bearing degradation is relatively serious, a smaller sampling interval is used. It should be noted here that the sampling interval is different from the detection interval mentioned above. The sampling interval refers to the collection of a data point at a certain interval, for example, a data point is collected every time the shaft rotates one circle. Taking into account the degradation law of the bearing, the patent of the present invention designs a fault detection method that adaptively determines the sampling interval. The patent of the present invention reduces the amount of data collection while ensuring the reliability of bearing fault detection, effectively reduces the cost of data collection, transmission and processing, and is more suitable for general industrial fields. Summary of the invention

[0006] Purpose of the invention: In response to the problems pointed out in the background technology, the present invention discloses a real-time detection method for motor bearing faults based on adaptive data acquisition, which adaptively determines the number of data points collected within each rotation period of the motor shaft, reduces the amount of data collection while ensuring the reliability of bearing fault detection, and thus effectively reduces the cost of data collection, transmission and processing.

[0007] Technical solution: The present invention discloses a real-time detection method for motor bearing faults based on adaptive data acquisition, which specifically includes the following steps:

[0008] Step 1: Determine the installation location of the vibration sensor and the speed sensor, and set the detection parameters;

[0009] Step 2: Obtain N groups of periodic synchronous single-point sampling data Y;

[0010] Step 3: Calculate the standard deviation Z of N groups of periodic synchronous single-point sampling data in data Y respectively;

[0011] Step 4: Determine whether a fault occurs and implement adaptive data acquisition.

[0012] Furthermore, determining the installation positions of the vibration sensor and the rotation speed sensor in step 1 includes: the vibration sensor is installed at the pre-detection bearing to detect the vibration signal of the bearing, and the rotation speed sensor is installed near the motor shaft to measure the shaft speed.

[0013] Furthermore, the detection parameters set in step 1 include the initial number N0 of synchronously sampled data points per rotation period of the motor shaft, the minimum number N0 of synchronously sampled data points per rotation period of the motor shaft, and the minimum number N1 of synchronously sampled data points per rotation period of the motor shaft. min , the maximum number of synchronous sampling data points N in each rotation period of the motor shaft max, the motor shaft speed L for data collection, the thresholds X0 and X1; the initial number N0 of synchronously sampled data points per motor shaft rotation period is determined according to the health status of the detected bearing; when the detected bearing is a new bearing, the initial number N0 of synchronously sampled data points per motor shaft rotation period is set to N min ; When the health status of the detected bearing is unknown, the initial number N0 of synchronous sampling data points per rotation period of the motor shaft is set to (N min +N max ) / 2; the initial number N0 and the minimum number N of synchronous sampling data points per rotation period of the motor shaft min , the maximum number N max are all integers; when the initial number N0 of synchronous sampling data points per rotation period of the motor shaft is set to (N min +N max ) / 2, and (N min +N max ) / 2 is not an integer, then (N min +N max ) / 2 is rounded to an integer; the threshold values ​​X0 and X1 are determined according to the statistical values ​​of the standard deviation of the periodic synchronous single-point sampling data under the conditions of no fault and fault of the bearing; the threshold values ​​X0=μ0+σ0 and X1=μ1-σ1 are set, where μ0 and σ0 represent the mean and standard deviation of the standard deviation of the periodic synchronous single-point sampling data under the condition of no fault, respectively, and μ1 and σ1 represent the mean and standard deviation of the standard deviation of the periodic synchronous single-point sampling data under the condition of fault, respectively.

[0014] Furthermore, in step 2, according to the number N of synchronous sampling data points in each rotation period of the motor shaft, N groups of periodic synchronous single-point sampling data Y=[Y1, Y2, Y3, ..., Y i ,…,Y N ], where the i-th group of periodic synchronous single-point sampling data Y i =[y i ,y i+N ,y i+2N ,…,y i+(L-1)N ]; The periodic synchronous single-point sampling data is to synchronously collect a single data within each rotation period of the motor shaft; similarly, the periodic synchronous multi-point sampling data is to synchronously collect multiple data within each rotation period of the motor shaft; therefore, N groups of periodic synchronous single-point sampling data can also be regarded as a group of periodic synchronous multi-point sampling data.

[0015] Furthermore, in step 3, the standard deviation Z=[s(1), s(2), s(3), ..., s(i), ..., s(N)] of N groups of periodic synchronous single-point sampling data in data Y is calculated respectively, wherein the standard deviation of the i-th group of periodic synchronous single-point sampling data is Among them, M iis the mean of the ith group of periodically synchronized single-point sampling data.

[0016] Furthermore, in step 4, the average value of N standard deviations Z is calculated. The mean value M of the standard deviation is used as the fault indicator and compared with the threshold value; if the fault indicator M is less than or equal to the threshold value X0, the motor bearing is judged to be in a fault-free state, and the number of synchronous sampling data points N per rotation period of the motor shaft is adaptively adjusted to N min , and return to step 2 to continue fault detection; if the fault index M is greater than the threshold X0 and less than the threshold X1, it is impossible to reliably determine whether a fault has occurred. At this time, the number of synchronous sampling data points N per motor shaft rotation period is set to Among them, round() is a rounding function, the result is an integer, and returns to step 2 to continue fault detection; if the fault index M is greater than or equal to the threshold X1, it is considered that the motor bearing has a fault.

[0017] Beneficial effects:

[0018] 1. The method of the present invention discloses a real-time detection method for motor bearing faults based on adaptive data acquisition, which overcomes the shortcomings of existing real-time detection methods for motor bearing faults and can be applied to the general field of motor fault detection.

[0019] 2. The method of the present invention can adaptively determine the number of synchronously sampled data points in each rotation period of the motor shaft, thereby reducing the cost of data acquisition, data transmission and data processing while ensuring the reliability of fault detection.

[0020] 3. The computational complexity of the method of the present invention is low, and a control unit with a lower processing speed can be used, thereby reducing the cost of the control unit.

[0021] 4. The method of the present invention is not affected by the motor speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is a fault detection flow chart of the method of the present invention;

[0023] Figure 2 The probability density function and threshold diagram of the fault index when there is no fault and when there is a fault;

[0024] Figure 3 This is a schematic diagram of the motor bearing fault-free detection result;

[0025] Figure 4 This is a schematic diagram of the number of synchronously sampled data points per revolution period for fault-free detection of motor bearings;

[0026] Figure 5 This is a schematic diagram of the motor bearing fault detection result;

[0027] Figure 6 This is a schematic diagram of the number of synchronously sampled data points per revolution for motor bearing fault detection. DETAILED DESCRIPTION

[0028] The implementation process of the method of the present invention is described below with reference to the accompanying drawings.

[0029] refer to Figure 1 The present invention discloses a real-time detection method for motor bearing fault based on adaptive decision making, comprising the following steps:

[0030] Step 1: Determine the installation positions of the vibration sensor and the speed sensor, determine the initial number of synchronous sampling data points N0=25 per motor shaft rotation period, and the minimum number of synchronous sampling data points N per motor shaft rotation period min =20, the maximum number of synchronous sampling data points per motor shaft rotation cycle N max =30, the motor shaft speed L for data collection = 20, the threshold X0 = μ0 + σ0 = 0.344, X1 = μ1 - σ1 = 0.362. The probability density function of the fault index under the condition of no fault and fault is as follows: Figure 2 The bearing health status is unknown, so the initial number of synchronous sampling data points N0 per motor shaft rotation period is set to (N min +N max ) / 2=25.

[0031] Step 2: Obtain N groups of periodically synchronized single-point sampling data Y.

[0032] According to the number N of synchronous sampling data points in each rotation period of the motor shaft, N groups of periodic synchronous single-point sampling data Y are obtained, that is, a group of periodic synchronous multi-point sampling data.

[0033] According to the information collected by the vibration sensor, N=25 groups of periodic synchronous single-point sampling data Y are obtained, that is, a group of periodic synchronous multi-point sampling data.

[0034] Each row of data in data Y represents a group of periodically synchronized single-point sampling data.

[0035] Step 3: Calculate the standard deviation Z of 25 groups of periodic synchronous single-point sampling data in data Y respectively.

[0036] Z=[s(1),s(2),s(3),…,s(i),…,s(25)],

[0037] Among them, the standard deviation of the i-th group of periodic synchronous single-point sampling data is μ i is the mean of the ith group of periodically synchronized single-point sampling data.

[0038] Step 4: Determine whether a fault occurs and implement adaptive data acquisition.

[0039] Calculate the mean of N standard deviations Z The mean value M of the standard deviation is used as a fault indicator and compared with the threshold value X0 and the threshold value X1.

[0040] Figure 3 The detection results when the bearing is fault-free are given. In the figure, in the first detection, the fault indicator is identified to be greater than the threshold value X0 but not greater than the threshold value X1. At this time, the health status of the bearing cannot be reliably determined. The number of synchronous sampling data points N per rotation period of the motor shaft is adaptively adjusted to Return to step 2 to continue fault detection. If the fault index is lower than the threshold value X0 in the subsequent detection, the bearing is considered to be in a fault-free state. At this time, the number of synchronous sampling data points N per rotation period of the motor shaft is adaptively adjusted to N min , return to step 2 to continue fault detection. In the entire detection process, in the same way, the number N of synchronous sampling data points per rotation period of the motor shaft is adaptively adjusted according to the detected fault index, and the detection process is continued until the motor bearing detects a fault. Figure 4 The number of synchronously sampled data points N in each rotation period of the motor shaft when the bearing is fault-free is given. Since the bearing is in a fault-free state, the N collected in each detection is less than the maximum value of 30, reducing the amount of data collection.

[0041] Figure 5 This is the detection result when the bearing fails. As can be seen from the figure, the fault index is lower than the threshold value X1 in the 1st to 4th detections. According to step 4 of the present invention, it is determined that the fault has not occurred. At this time, data is adaptively collected based on the fault index, and the fault detection is continued in step 2. In the 5th detection, the fault index exceeds the threshold value X1, which means that the fault is detected, and the fault detection program is stopped at this time. Figure 6 The number of synchronously sampled data points N per motor shaft revolution period when the bearing fails is given. Since the bearing is in an early fault state, the collected N is less than the maximum value of 30 during some detection processes, reducing the amount of data collection. When the fault index is found to be greater than the threshold value X1 during the 5th detection, the number of synchronously sampled data points N per motor shaft revolution period is set to the maximum value of 30, so that fault detection can be performed reliably.

[0046] Compared with the traditional fault detection method, the method of the present invention has a lower computational complexity. The fault threshold is determined based on the statistical results of the fault indicators when the bearing has or does not have a fault, and compared with the fault indicators calculated from the test data to determine whether the fault has occurred and to achieve adaptive data acquisition. When the fault indicator M calculated from the test data is less than or equal to the threshold value X0, the bearing is determined to be in a fault-free state, and the number of synchronously sampled data points N in each rotation period of the motor shaft is set to the minimum value N min , and return to step 2 to continue fault detection; when M is greater than the threshold value X0 and less than the threshold value X1, it is impossible to reliably determine whether the bearing has a fault. At this time, N is set to And return to step 2 to continue fault detection; when M is greater than or equal to the threshold value X1, it is determined that the bearing is faulty. The method of the present invention adaptively determines the number N of synchronously sampled data points per rotation period of the motor shaft according to the bearing fault index, increases or decreases the amount of data collected in real time, and optimizes the amount of collected data. While ensuring the reliability of fault detection, the method of the present invention reduces the cost of data collection, transmission and processing.

[0047] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable people familiar with the technology to understand the content of the present invention and implement it accordingly, and they cannot be used to limit the protection scope of the present invention. Any equivalent transformation or modification made according to the spirit of the present invention should be included in the protection scope of the present invention.

Claims

1. A real-time detection method for motor bearing faults based on adaptive data acquisition, characterized in that: The following steps are involved: Step 1: Determine the installation location of the vibration sensor and the speed sensor, and set the detection parameters; Step 2: Obtain N groups of periodic synchronous single-point sampling data Y; Step 3: Calculate the standard deviation Z of N groups of periodic synchronous single-point sampling data in data Y respectively; Step 4: Determine whether a fault occurs and implement adaptive data acquisition.

2. The method for real-time detection of motor bearing faults based on adaptive data acquisition according to claim 1, characterized in that: Determining the installation positions of the vibration sensor and the rotation speed sensor in step 1 includes: the vibration sensor is installed at the pre-detection bearing to detect the vibration signal of the bearing, and the rotation speed sensor is installed near the motor shaft to measure the shaft speed.

3. The real-time detection method for motor bearing fault based on adaptive data acquisition according to claim 2 is characterized in that: The detection parameters set in step 1 include the initial number N0 of synchronously sampled data points per rotation period of the motor shaft, the minimum number N0 of synchronously sampled data points per rotation period of the motor shaft, min , the maximum number of synchronous sampling data points N in each rotation period of the motor shaft max , the motor shaft speed L for data collection, the thresholds X0 and X1; the initial number N0 of synchronously sampled data points per motor shaft rotation period is determined according to the health status of the detected bearing; when the detected bearing is a new bearing, the initial number N0 of synchronously sampled data points per motor shaft rotation period is set to N min ; When the health status of the detected bearing is unknown, the initial number N0 of synchronous sampling data points per rotation period of the motor shaft is set to (N min +N max ) / 2; the initial number N0 and the minimum number N of synchronous sampling data points per rotation period of the motor shaft min , the maximum number N max are all integers; when the initial number N0 of synchronous sampling data points per rotation period of the motor shaft is set to (N min +N max ) / 2, and (N min +N max ) / 2 is not an integer, then (N min +N max ) / 2 is rounded to an integer; the threshold values ​​X0 and X1 are determined according to the statistical values ​​of the standard deviation of the periodic synchronous single-point sampling data under the conditions of no fault and fault of the bearing; the threshold values ​​X0=μ0+σ0 and X1=μ1-σ1 are set, where μ0 and σ0 represent the mean and standard deviation of the standard deviation of the periodic synchronous single-point sampling data under the condition of no fault, respectively, and μ1 and σ1 represent the mean and standard deviation of the standard deviation of the periodic synchronous single-point sampling data under the condition of fault, respectively.

4. The method for real-time detection of motor bearing faults based on adaptive data acquisition according to claim 3 is characterized in that: In step 2, according to the number of synchronous sampling data points N in each rotation period of the motor shaft, N groups of periodic synchronous single-point sampling data Y=[Y1, Y2, Y3, ..., Y i ,…,Y N ], where the i-th group of periodic synchronous single-point sampling data Y i =[y i ,y i+N ,y i+2N ,…,y i+(L-1)N ].

5. The method for real-time detection of motor bearing faults based on adaptive data acquisition according to claim 4, characterized in that: In step 3, the standard deviation Z=[s(1), s(2), s(3), ..., s(i), ..., s(N)] of N groups of periodic synchronous single-point sampling data in data Y is calculated respectively, where the standard deviation of the i-th group of periodic synchronous single-point sampling data is Among them, M i is the mean of the ith group of periodically synchronized single-point sampling data.

6. The method for real-time detection of motor bearing faults based on adaptive data acquisition according to claim 5, characterized in that: In step 4, the average value of N standard deviations Z is calculated The mean value M of the standard deviation is used as the fault indicator and compared with the threshold value; If the fault index M is less than or equal to the threshold value X0, the motor bearing is judged to be in a fault-free state, and the number of synchronous sampling data points N per rotation period of the motor shaft is adaptively adjusted to N min , and return to step 2 to continue fault detection; If the fault index M is greater than the threshold X0 and less than the threshold X1, it is impossible to reliably determine whether a fault has occurred. In this case, the number of synchronously sampled data points N per motor shaft rotation period is set to And return to step 2 to continue fault detection; if the fault indicator M is greater than or equal to the threshold value X1, it is considered that the motor bearing has failed.

Images