Adaptive recognition method for multi-component rotation speed of rolling bearing

Abstract

A rolling bearing multi-component rotation speed adaptive identification method of the invention belongs to the bearing detection field, and comprises the following steps: obtaining weak magnetic signals during bearing operation, segmenting the obtained weak magnetic signals, and carrying out Fourier transform on each segmented signal to obtain a weak magnetic signal; constructing a time-frequency matrix according to all the segmented signals after Fourier transform; setting an extraction lower limit and an extraction upper limit, and extracting the time-frequency matrix according to the set extraction lower limit and the extraction upper limit to obtain time-frequency matrix subintervals; identifying first M local maximum values at each moment in the time-frequency matrix subinterval, identifying a frequency corresponding to each local maximum value, and accumulating the frequencies corresponding to the first M local maximum values at each moment and the frequency corresponding to the extraction lower limit to obtain a multi-component rotation frequency corresponding to each moment; calculating the corresponding multi-component bearing rotating speed according to the obtained multi-component rotating frequency corresponding to each moment; the method is used for solving the problem that the accuracy of distinguishing and extracting the multi-component running speed of the bearing is low.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of bearing detection, in particular to a rolling bearing multi-component rotating speed adaptive identification method. BACKGROUND

[0002] Rolling bearings are important components of mechanical rotation, and are related to the safe operation of rotating machinery. The application range covers many fields such as aviation, aerospace, vehicles, coal mines, etc. Especially for bearings used in aero-engines, the working environment is harsh, mainly manifested as high temperature, high speed, heavy load and oil mist, etc. The aero-engine bearing works in such a harsh environment, and the bearing bears great challenges. Once the aero-engine bearing fails, it may cause the engine to vibrate, the rotor to collide and grind, etc. The most direct reflection of rolling bearing failure is the change of the rotating speed of the bearing assembly. For example, when the inner ring of the bearing fails, the rotating speed of the inner ring and the rolling element will change periodically.

[0003] Currently, bearing fault diagnosis mainly focuses on the discrimination of fault frequency, which is mainly based on numerical calculation of bearing rotating speed, rolling element number, rolling element diameter, pitch circle diameter and contact angle, etc. Among them, except for the bearing rotating speed, the rest are inherent parameters of the bearing. The accuracy of the bearing inner ring rotating speed test and extraction is inevitably related to the fault discrimination. In addition, there is a phenomenon of rolling element and ring sliding in the bearing fault mode, which is mainly judged by the ratio of rolling element rotating speed to inner ring rotating speed. Therefore, the detection and extraction of rolling element and inner ring rotating speed during the operation of the bearing are involved. As can be seen, the discrimination of the rotating speed of the inner ring and the rolling element of the bearing is related to the effective diagnosis of the bearing fault, and it is of great significance to monitor the multi-component rotating speed of the bearing. SUMMARY

[0004] In order to solve the problem of low accuracy of bearing multi-component rotating speed discrimination and extraction in the prior art, a rolling bearing multi-component rotating speed adaptive identification method is proposed.

[0005] Step one, obtain the weak magnetic signal of the bearing during operation, segment the obtained weak magnetic signal, and perform Fourier transform on each segmented signal. A time-frequency matrix is constructed according to all the Fourier transformed segmented signals.

[0006] Step two: set the lower limit and upper limit of extraction, and extract the time-frequency matrix according to the set lower limit and upper limit of extraction to obtain a time-frequency matrix sub-interval.

[0007] Step three: identify the first M local maximum values at each time in the time-frequency matrix sub-interval, identify the frequency corresponding to each local maximum value, and add the frequency corresponding to the first M local maximum values at each time to the frequency corresponding to the lower limit of extraction to obtain the multi-component rotating frequency corresponding to each time.

[0008] Step four: according to the obtained multi-component bearing rotating speed corresponding to each time.

[0009] The beneficial effects of the present application are:

[0010] The rolling bearing multi-component rotating speed adaptive identification method of the present application first acquires the weak magnetic signal of the bearing operation, performs Fourier transform on the collected signal, identifies the local maximum value of the Fourier transformed signal, and reverses the frequency information in the original time-frequency spectrum to obtain the rolling bearing multi-component rotating frequency, and further obtain the rolling bearing multi-component rotating speed; Through the method proposed in the present application, the multiple operating speed components of the bearing operating assembly and the corresponding multiple frequency components can be adaptively extracted, and the bearing operating state can be more quickly judged, and the discrimination ability of the bearing fault diagnosis is improved. BRIEF DESCRIPTION OF DRAWINGS

[0011] Figure 1 A rolling bearing multi-component rotating speed adaptive identification method flow chart for the specific embodiment of the present application;

[0012] Figure 2 A weak magnetic signal time domain graph of the bearing operation for the specific embodiment of the present application;

[0013] Figure 3 A segmented signal after Fourier transform for the specific embodiment of the present application;

[0014] Figure 4 A time-frequency matrix time-frequency domain graph for the specific embodiment of the present application;

[0015] Figure 5 A pre-extracted multi-component time-frequency graph for the specific embodiment of the present application;

[0016] Figure 6 A multi-component rotating frequency frequency domain graph extracted for the specific embodiment of the present application;

[0017] Figure 7 A rolling bearing structure schematic diagram for the specific embodiment of the present application;

[0018] Figure 8 A bearing operating speed spectrum for the specific embodiment of the present application;

[0019] Figure 7 1 is an outer ring, 2 is a rolling body, and 3 is an inner ring. DETAILED DESCRIPTION

[0020] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described, obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of the present application.

[0021] Specific implementation method one: a rolling bearing multi-component rotating speed adaptive identification method, comprising:

[0022] Step one, obtaining the weak magnetic signal of the bearing operation, segmenting the obtained weak magnetic signal to obtain a plurality of segmented signals, performing Fourier transform on each segmented signal, and constructing a time-frequency matrix according to all the Fourier transformed segmented signals;

[0023] Step two: setting an extraction lower limit and an extraction upper limit, extracting the time-frequency matrix according to the set extraction lower limit and extraction upper limit to obtain a time-frequency matrix sub-interval;

[0024] Step three: identifying the first M local maximum values at each time in the time-frequency matrix sub-interval, identifying the frequency corresponding to each local maximum value, and respectively accumulating the frequency corresponding to the first M local maximum values at each time and the frequency corresponding to the extraction lower limit to obtain the multi-component rotating frequency corresponding to each time;

[0025] Step four: calculating the corresponding multi-component bearing rotating speed according to the obtained multi-component rotating frequency corresponding to each time.

[0026] Specifically, the rolling bearing multi-component rotating speed adaptive identification method proposed by the present application can adaptively extract various operating speed components of the bearing operating assembly and the corresponding frequency multiplication components, and can more quickly judge the bearing operating state, thereby improving the discrimination ability of the bearing fault diagnosis.

[0027] Further, the method for segmenting the obtained weak magnetic signal to obtain a plurality of segmented signals and performing Fourier transform on each segmented signal comprises:

[0028] The obtained weak magnetic signal is segmented into N equal parts to obtain a plurality of segmented signals:

[0029] ;

[0030] Wherein, the first segmented signal is f1(t), the second segmented signal is f2(t), the third segmented signal is f3(t), and the Nth segmented signal is fN(t);

[0031] performing Fourier transform on each segmented signal:​

[0032] ;

[0033] wherein, is the i-th segmented signal after Fourier transform, , is the Fourier transform, is the frequency of the i-th segmented signal, is the sampling frequency of the weak magnetic signal.

[0034] Specifically, the acquired weak magnetic signal of the bearing in operation is as shown in FIG. 1, and the signal sampling rate is 25600 Hz; in the specific embodiment of the present application, the weak magnetic signal is evenly segmented into 1828 segments, and the corresponding frequency domain transform of the 1828-th segmented signal, i.e., the Fourier transform of the 1828-th segmented signal, is as shown in FIG. 2. Figure 2 Figure 3

[0035] Further, the method for constructing the time-frequency matrix according to all the segmented signals after Fourier transform comprises:

[0036] arranging the time sequence according to the length of each segmented signal and the number of segmented signals of the weak magnetic signal;

[0037] ;

[0038] wherein, is the signal length of each segmented weak magnetic signal;

[0039] arranging all the segmented signals after Fourier transform according to the time sequence to form a two-dimensional time-frequency matrix.

[0040] Specifically, in the embodiment, the length of the segmented signal segment is 1828, and the two-dimensional time-frequency matrix is as shown in FIG. 3. Figure 4

[0041] Further, the method for extracting the sub-interval of the time-frequency matrix according to the set lower limit and upper limit comprises:

[0042] acquiring the highest rotating speed of the bearing test spectrum sequence, the lowest rotating speed of the bearing test spectrum sequence, and the number M of rotating speed components of the bearing component in operation;

[0043] calculating the theoretical rotating speed of the retainer according to the lowest rotating speed of the bearing test spectrum sequence

[0044] ;

[0045] ​​​​​​Wherein, d is the bearing rolling body diameter, D is the pitch diameter of the bearing, and a is the bearing contact angle;

[0046] Setting the highest rotational speed of the bearing test spectrum sequence The corresponding frequency is The theoretical rotational speed of the retainer The corresponding frequency is , and is taken as the lower limit of extraction, is taken as the upper limit of extraction to extract the time-frequency matrix sub-interval .

[0047] Specifically, in the present embodiment, the time-frequency of multiple components is pre-extracted, and the actual rotational speed of the bearing inner ring and the rolling body rotation is mainly extracted, as shown in Figure 5 In the present embodiment, the highest rotational speed of the bearing test spectrum sequence and the lowest rotational speed of the bearing test spectrum sequence are the same, both being 32460rpm, the number of rotational speed components of the bearing component is 2, the bearing rolling body diameter is 12.7mm, the pitch diameter of the bearing is 87.5mm, and the bearing contact angle is set to 33°, the value is set to 0.4;

[0048] Further, the method for accumulating the frequency corresponding to each time point and the lower limit of extraction is as follows:

[0049] ;

[0050] Wherein, is the time-frequency matrix sub-interval , is the frequency corresponding to the M component rotational frequency of each time point, is the frequency corresponding to the M local maximum of each time point in the time-frequency matrix sub-interval , is the lower limit of the corresponding frequency. Specifically, in the present embodiment

[0051] the value is 243hz, and the extracted multiple component rotational frequency information is as shown in . Figure 6

[0052] Further, the method for calculating the corresponding multiple component bearing rotational speed according to the obtained multiple component rotational frequency of each time point is as follows:

[0053] ;

[0054] Wherein, is the time-frequency matrix sub-interval​​ inner at the moment of

[0055] Embodiment II: a computer readable storage device, the storage device storing a computer program, the computer program being executed by a processor to implement the steps of the adaptive multi-component rotating speed identification method of rolling bearings as described in Embodiment I.

[0056] Embodiment III: an adaptive multi-component rotating speed identification device of rolling bearings, comprising a storage device, a processor, and a computer program stored in the storage device and executable on the processor, the processor executing the computer program to implement the steps of the adaptive multi-component rotating speed identification method of rolling bearings as described in Embodiment I.

[0057] Embodiment IV: a computer program product, comprising a computer program, the computer program being executed by a processor to implement the steps of the adaptive multi-component rotating speed identification method of rolling bearings as described in Embodiment I.

[0058] While the application has been described with reference to particular embodiments, it will be understood that the examples are merely for illustration and that many modifications can be made by persons of ordinary skill in the art. Other arrangements can be utilized and other methods can be implemented without departing from the spirit and scope of the application. Those skilled in the art will readily recognize a variety of ways in which the disclosed examples can be implemented. It is intended that the examples be limited only by the claims that follow.

Claims

1. A multi-component rotational speed self-adapting recognition method for rolling bearings, characterized in that, The method comprises the following steps: Step one, obtaining the weak magnetic signal of the bearing in operation, segmenting the obtained weak magnetic signal to obtain a plurality of segmented signals, performing Fourier transform on each segmented signal, and constructing a time-frequency matrix according to all the segmented signals after Fourier transform; Step two, setting an extraction lower limit and an extraction upper limit, and extracting the time-frequency matrix according to the set extraction lower limit and extraction upper limit to obtain a time-frequency matrix sub-interval; Step three, identifying the first M local maximum values at each time in the time-frequency matrix sub-interval, identifying the frequency corresponding to each local maximum value, and respectively accumulating the frequency corresponding to the first M local maximum values at each time and the frequency corresponding to the extraction lower limit to obtain a plurality of groups of component rotating frequencies corresponding to each time; Step four, calculating a plurality of groups of component bearing rotating speeds corresponding to each time according to the obtained plurality of groups of component rotating frequencies corresponding to each time.

2. The multi-component rotating speed self-adaptive identification method of a rolling bearing according to claim 1, characterized in that: The method for segmenting the obtained weak magnetic signal to obtain a plurality of segmented signals and performing Fourier transform on each segmented signal comprises the following steps: The acquired flux-weakening signal The average is divided into N equal parts to obtain a plurality of divided signals: ； wherein, is a first split signal, is a second split signal, is a third split signal, is an Nth split signal; The method for performing Fourier transform on each segmented signal comprises the following steps: ； wherein, is the i-th divided signal after Fourier transform, , is the Fourier transform, is the frequency of the i-th divided signal, is the sampling frequency of the weak magnetic signal.

3. The multi-component rotating speed self-adaptive identification method of a rolling bearing according to claim 2, characterized in that: The method for constructing a time-frequency matrix according to all the segmented signals after Fourier transform comprises the following steps: According to each split signal length and the number of field weakening signal split segments, a time sequence is constituted : ； wherein, is the signal length of each divided field-weakening signal; All the segmented signals after Fourier transform are arranged in time sequence to form a two-dimensional time-frequency matrix.

4. The multi-component rotating speed self-adaptive identification method of a rolling bearing according to claim 1, characterized in that: The method for setting an extraction lower limit and an extraction upper limit and extracting the time-frequency matrix according to the set extraction lower limit and extraction upper limit to obtain a time-frequency matrix sub-interval comprises the following steps: the highest rotational speed of the bearing test sequence the lowest rotational speed of the bearing test sequence and the number M of rotational speed components of the bearing component operation According to the lowest rotational speed of the bearing test spectrum sequence Calculate the theoretical rotational speed of the cage : ； Wherein, d is the diameter of the bearing rolling body, D is the pitch diameter of the bearing, and a is the contact angle of the bearing; Setting the highest rotational speed of the bearing test spectrum sequence The corresponding frequency is The theoretical rotational speed of the cage The corresponding frequency is Taking as the lower limit of extraction, Taking as the upper limit of extraction, the time-frequency matrix sub-interval is obtained by extracting the two-dimensional time-frequency matrix .

5. The multi-component rotating speed self-adaptive identification method of a rolling bearing according to claim 1, characterized in that: The method for respectively accumulating the frequency corresponding to the first M local maximum values at each time and the frequency corresponding to the extraction lower limit to obtain a plurality of groups of component rotating frequencies corresponding to each time comprises the following steps: ； wherein, is a sub-interval of the time-frequency matrix within M components of the time instant, is a sub-interval of the time-frequency matrix within M local maxima of the time instant, is the lower limit corresponding frequency.

6. The multi-component rotating speed self-adaptive identification method of a rolling bearing according to claim 1, characterized in that: The method for calculating a plurality of groups of component bearing rotating speeds corresponding to each time according to the obtained plurality of groups of component rotating frequencies corresponding to each time comprises the following steps: ； wherein sub-interval of a time-frequency matrix within at a time instant 7. A computer-readable storage device storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1-6. The computer program is executed by the processor to realize the steps of the rolling bearing multi-component rotating speed adaptive identification method according to any one of claims 1 to 6.

8. A rolling bearing multi-component rotational speed self-adapting recognition device comprising a storage device, a processor and a computer program stored in the storage device and executable on the processor, characterized in that, The processor executes the computer program to realize the steps of the rolling bearing multi-component rotating speed adaptive identification method according to any one of claims 1 to 6.

9. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the rolling bearing multi-component rotating speed adaptive identification method according to any one of claims 1 to 6.

Images