Bearing fault diagnosis method of vibration resonance auxiliary enhanced random resonance coupling system

Abstract

The application discloses a bearing fault diagnosis method of a vibration resonance auxiliary enhanced random resonance coupling system and belongs to the technical field of mechanical fault diagnosis, and comprises the following steps: collecting rolling bearing vibration signals by using a sensor, performing Hilbert envelope demodulation processing on the collected rolling bearing vibration signals to obtain amplitude envelopes of the vibration signals; a matching steady-state random resonance system is constructed, and the vibration resonance system and the matching steady-state random resonance system are combined into a coupling system through a nonlinear coupling mode; the obtained amplitude envelopes are taken as input signals of the constructed coupling system, and an optimization algorithm is adopted to realize optimization selection of multiple parameters of the system; the obtained optimal parameters are substituted into the coupling system, the amplitude envelopes are processed, spectrum analysis is performed on system outputs, and effective identification and diagnosis of rolling bearing faults are completed. The application can realize effective extraction of weak fault characteristics in rolling bearing vibration signals, and provides an effective solution for early weak fault diagnosis of rolling bearings.

Description

Technical Field

[0001] This invention belongs to the field of mechanical fault diagnosis technology, specifically relating to a bearing fault diagnosis method for a vibration resonance-assisted enhanced random resonance coupling system, which can effectively extract weak fault features from the vibration signal of rolling bearings. Background Technology

[0002] Rolling bearings are among the most widely used general-purpose mechanical components. Their health directly affects the operational quality of the entire machine. Failure of a rolling bearing often results in significant economic losses and even personal injury. Therefore, health monitoring and fault diagnosis of rolling bearings are of great importance. Vibration signal analysis is commonly used in bearing fault diagnosis. Analyzing the vibration signals of rolling bearings can yield their time-domain, frequency-domain, and time-frequency-domain characteristics, allowing for the assessment of the bearing's operating state. However, under actual operating conditions, the acquired vibration signals are easily affected by environmental background noise, machine operating noise, etc., resulting in a low signal-to-noise ratio and impacting the accuracy of fault diagnosis.

[0003] Traditional methods enhance and extract fault feature information by suppressing noise. However, these methods also destroy the fault feature information in the original signal, greatly affecting fault diagnosis. Stochastic resonance, on the other hand, reflects the positive role of noise in nonlinear systems. Specifically, under certain nonlinear conditions, the synergistic effect of weak periodic signals and noise (random interference) enhances the output of periodic signals in a nonlinear system. Stochastic resonance has the unique advantage of utilizing noise and is widely used in mechanical fault diagnosis. However, it still has some drawbacks. For example, it is greatly affected by noise intensity and type, and noise can only increase in one direction and is difficult to reduce; it is also subject to output saturation due to the influence of the potential function. Therefore, it is necessary to study a stochastic resonance system that can more effectively extract weak fault signals under strong noise backgrounds and apply it to the fault diagnosis technology of rolling bearings. Summary of the Invention

[0004] In order to overcome the shortcomings of the existing technology, the purpose of this invention is to provide a bearing fault diagnosis method for a vibration resonance-assisted enhanced random resonance coupling system, so as to achieve accurate identification and extraction of early weak fault characteristics of rolling bearings.

[0005] First, a matched steady-state potential function model is designed and constructed, which can match different potential well shapes according to the characteristics of the input signal, thus mitigating the output saturation phenomenon present in traditional bistable models. Then, a stochastic resonant coupling system is constructed based on the proposed model, and vibration resonance control is introduced to improve the system's detection performance for weak signals. The constructed vibration resonance-assisted enhanced stochastic resonant coupling system exhibits good bandpass filtering characteristics, effectively suppressing low-frequency and high-frequency interference components. Finally, particle swarm optimization is used to optimize the selection of multiple system parameters.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] A bearing fault diagnosis method for a vibration resonance-assisted enhanced stochastic resonance coupling system includes the following steps:

[0008] Step S1 Vibration signal preprocessing: Use a sensor to collect the vibration signal of the rolling bearing, and perform Hilbert envelope demodulation on the collected rolling bearing vibration signal to obtain the amplitude envelope of the vibration signal;

[0009] Step S2: Construct a coupled system, build a matched steady-state stochastic resonance system, and form a coupled system with the vibration resonance system through nonlinear coupling;

[0010] Step S3 System parameter optimization: The amplitude envelope obtained in step S1 is used as the input signal of the constructed coupled system, and the output signal-to-noise ratio is used as the objective function of the intelligent optimization algorithm. The number of iterations is used as the termination condition. The optimization selection of multiple system parameters is achieved by seeking the maximum output signal-to-noise ratio in multiple iterations.

[0011] Step S4, fault identification, involves substituting the optimal parameters obtained in step S3 into the coupled system, processing the amplitude envelope obtained in step S1, and performing spectrum analysis on the system output to complete the effective identification and diagnosis of rolling bearing faults.

[0012] A further improvement to the technical solution of the present invention is that the potential function of the matched steady-state stochastic resonance system in step S2 is:

[0013]

[0014] Where m and n are system parameters, and both are positive real numbers; by adjusting the system parameters m and n, the potential function U(x) can exhibit monostable or bistable characteristics.

[0015] A further improvement to the technical solution of the present invention is that the coupled system in step S2 consists of a matched steady-state stochastic resonance system and a bistable vibrational resonance system, and the specific mathematical model is as follows:

[0016]

[0017] Where, U1(y)=-|y| m / (2m)+n 2 y 2 It is the matching potential function, U2(y)=-ay 2 / 2+by 4 / 4 is the bistable state function, and a and b are system parameters; the output y1(t) of the matched steady-state random resonance and the output y2(t) of the bistable vibration resonance are linked together by the coupling coefficient γ; s(t) is the system input signal, and h(t) is the added high-frequency signal, i.e., h(t) = Hcos(2πf1t), where H and f1 are the amplitude and frequency, respectively.

[0018] A further improvement of the technical solution of the present invention lies in that: the output y2(t) of the bistable vibration resonance system passes through a passband cutoff frequency of f. p =0.75f1, stopband cutoff frequency is f s The result after applying a low-pass filter of 0.9f1 is that the final output of the coupled system is y(t) = y1(t) + y2(t).

[0019] A further improvement of the technical solution of the present invention is that the optimization algorithm in step S3 is a particle swarm optimization algorithm, and the signal-to-noise ratio of the coupled system output is used as the fitness function to optimize and select multiple parameters of the system.

[0020] A further improvement to the technical solution of this invention lies in the following: the system's multiple parameters mainly include: for a stochastic resonance system, adjusting parameters m and n to achieve stochastic resonance control; for a vibration resonance system, fixing system parameters a = b = 1 and adjusting amplitude H and frequency f1 to achieve vibration resonance control; for a coupled system, adjusting the coupling coefficient γ to achieve joint control of stochastic resonance and vibration resonance; when the frequency of the system input signal s(t) is much greater than 1Hz, the scale transformation factor R needs to be adjusted; the system parameters m, n, coupling coefficient γ, amplitude H and frequency f1 of the high-frequency signal, and scale factor R are adaptively adjusted using a particle swarm optimization algorithm, with the system output signal-to-noise ratio as the objective function, and the optimal parameter combination is obtained by seeking the maximum output signal-to-noise ratio in multiple iterations.

[0021] A further improvement of the technical solution of the present invention is that: the search range of multiple system parameters is set as follows: the search range of system parameters m and n is [0.01, 5], the search range of high frequency signal amplitude H and frequency f1 is [1, 5] and [200, 1000] respectively, the search range of coupling coefficient γ is [-1, 1], and the search range of scale transformation factor R is [500, 1500].

[0022] By employing the above technical solution, the bearing fault diagnosis method for a vibration resonance-assisted enhanced stochastic resonance coupling system provided by this invention has the following advantages compared with existing technologies:

[0023] This invention constructs a matched steady-state stochastic resonance model, which can match the optimal potential well shape according to the characteristics of the input signal, thereby weakening the output saturation phenomenon of the traditional bistable potential function and enhancing the stochastic resonance effect.

[0024] The vibration resonance-assisted enhanced random resonance coupling system constructed in this invention can utilize both the noise energy contained in the signal and the controllable high-frequency signal energy to enhance the characteristics of weak fault signals, thereby achieving a higher output signal-to-noise ratio and providing an effective solution for early weak fault diagnosis of rolling bearings. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly described below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0026] Figure 1 This invention provides a flow chart for a bearing fault diagnosis method in a vibration resonance-assisted enhanced random resonance coupling system.

[0027] Figure 2 The coupling system structure provided by this invention;

[0028] Figure 3 The time-domain waveform of the rolling bearing fault signal provided by the present invention;

[0029] Figure 4 The spectrum of rolling bearing fault signals provided by this invention;

[0030] Figure 5 The envelope spectrum of the rolling bearing fault signal provided by this invention;

[0031] Figure 6 The time-domain waveform is obtained after the rolling bearing vibration signal is processed by the method proposed in this invention.

[0032] Figure 7 The spectrum is obtained after the rolling bearing vibration signal is processed by the method proposed in this invention.

[0033] Figure 8 The time-domain waveform obtained after processing the rolling bearing vibration signal by classical bistable random resonance;

[0034] Figure 9This is the spectrum obtained after processing the vibration signal of a rolling bearing using classical bistable random resonance. Detailed Implementation

[0035] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The specific implementation process of the present invention will be further described below with reference to the accompanying drawings:

[0036] like Figure 1 As shown, the present invention provides a bearing fault diagnosis method for a vibration resonance-assisted enhanced stochastic resonance coupling system, the diagnosis method comprising the following steps:

[0037] (1) Vibration signal preprocessing: First, the Hilbert transform is used to perform envelope demodulation on the collected rolling bearing vibration signal; then, the obtained amplitude envelope is used as the input of the coupled system.

[0038] (2) Construct a matched steady-state stochastic resonance system and form a coupled system with the vibration resonance system through nonlinear coupling; the potential function of the matched steady-state stochastic resonance system is:

[0039]

[0040] Where m and n are system parameters, and both are positive real numbers; by adjusting the system parameters m and n, the potential function U(x) can exhibit monostable or bistable characteristics.

[0041] Set up Figure 2 The coupled system shown consists of a matched steady-state stochastic resonance system and a bistable vibrational resonance system. The specific mathematical model is as follows:

[0042]

[0043] Where, U1(y)=-|y| m / (2m)+n 2 y 2 It is the matching potential function, U2(y)=-ay 2 / 2+by 4 / 4 is the bistable state function, and a and b are system parameters; the output y1(t) of the matched steady-state stochastic resonance and the output y2(t) of the bistable vibration resonance are linked together by the coupling coefficient γ; s(t) is the system input signal, and h(t) is the added high-frequency signal, i.e., h(t) = Hcos(2πf1t), where H and f1 are the amplitude and frequency, respectively, and the parameters a and b of the vibration resonance system are set to 1.

[0044] The output y2(t) of the bistable resonant system is obtained through a passband cutoff frequency of f. p=0.75f1, stopband cutoff frequency is f s The result after applying a low-pass filter of 0.9f1 is that the final output of the coupled system is y(t) = y1(t) + y2(t).

[0045] (3) System parameter optimization: The particle swarm optimization algorithm is used to optimize and select multiple system parameters. First, the output signal-to-noise ratio of the coupled system is used as the fitness function of the particle swarm optimization algorithm. The main system parameters include: for the stochastic resonance system, parameters m and n are adjusted to achieve stochastic resonance control; for the vibration resonance system, the system parameters a = b = 1 are fixed, and the amplitude H and frequency f1 are adjusted to achieve vibration resonance control; for the coupled system, the coupling coefficient γ is adjusted to achieve joint control of stochastic resonance and vibration resonance; when the frequency of the system input signal s(t) is much greater than 1Hz, the scaling factor R is adjusted; the system parameters m, n, coupling coefficient γ, amplitude H and frequency f1 of the high-frequency signal, and scaling factor R are adaptively adjusted by the particle swarm optimization algorithm, with the system output signal-to-noise ratio as the objective function. The optimal parameter combination is obtained by seeking the maximum output signal-to-noise ratio in multiple iterations.

[0046] The search ranges for multiple system parameters are set as follows: the search ranges for system parameters m and n are [0.01, 5], the search ranges for high-frequency signal amplitude H and frequency f1 are [1, 5] and [200, 1000] respectively, the search range for coupling coefficient γ is [-1, 1], and the search range for scaling factor R is [500, 1500]. The number of iterations is set to 1000, and the optimal parameter combination is obtained by seeking the maximum output signal-to-noise ratio in multiple iterations.

[0047] (4) Fault identification: Input the amplitude envelope obtained in step (1) into the optimized coupling system to obtain the final output y(t) of the coupling system. By performing spectrum analysis on the output signal y(t), the rolling bearing fault can be effectively identified and diagnosed.

[0048] Effect analysis of the embodiments of the present invention:

[0049] The following explanation of the invention is based on vibration data from a rolling bearing of model SKF6203, with a sampling frequency of f. s =6400Hz, data length N=6400, drive motor speed is 1433r / min. The rolling bearing has an inner race fault; the inner race fault frequency f is calculated theoretically. i =117.52Hz, frequency conversion is f r =23.88Hz. Figures 3-5 The time-domain waveform, spectrum, and envelope spectrum of the rolling bearing vibration signal are shown. It can be seen that... Figure 3 and Figure 4 In the process, characteristic information related to rolling bearing failure is drowned out by noise and cannot be effectively identified; Figure 5 In the envelope spectrum shown, except for the transition frequency f r Besides the spectral peak at a certain point, there can be a fault frequency f in the inner ring. i A subtle amplitude was found at f; however, due to noise interference, i The fault characteristics at this location are difficult to identify accurately. The present invention will be used to process this signal.

[0050] First, the rolling bearing vibration signal is demodulated using Hilbert envelope, and the resulting amplitude envelope is input into the coupling system. Then, a particle swarm optimization algorithm is used to optimize the system's multiple parameters, yielding the optimal parameter combination: m = 1.0047, n = 0.01, γ = 1, H = 1.9061, f1 = 907.6301, R = 918.9437, corresponding to a system output signal-to-noise ratio (SNR) of -16.2203 dB. Next, the optimized coupling system is used to process the obtained amplitude envelope, and the result is as follows: Figure 6 and Figure 7 As shown in the figure. It can be seen from the figure that in Figure 6 In the time-domain waveform shown, the oscillation period of the signal is quite obvious. Figure 7 In the spectrum shown, the fault characteristic frequency f i The amplitude at this point is prominent and has good discernibility, with low-frequency noise and high-frequency interference being largely suppressed. In contrast, the amplitude envelope obtained was processed using a classical bistable stochastic resonance system. The optimal parameters obtained using the particle swarm optimization algorithm were a = 2.504, b = 3.908, and R = 1500, with an optimal system output signal-to-noise ratio (SNR) of -22.1455 dB. The results are as follows... Figure 8 and Figure 9 As shown. It can be seen that, except for the rotation frequency f r In addition to the prominent spectral peak at the point, the characteristic frequency f of the bearing inner ring fault is... i The amplitude at that point is not significant, and the system output signal-to-noise ratio is also significantly lower than the result obtained in this invention. The above comparative analysis results demonstrate that the method of this invention can achieve enhanced extraction of weak fault features, which is helpful for the accurate detection and diagnosis of rolling bearing faults.

[0051] This invention proposes a bearing fault diagnosis method for a vibration resonance-assisted enhanced stochastic resonance coupled system. The method constructs a matched steady-state stochastic resonance model, which mitigates the output saturation phenomenon present in traditional models. The coupled system constructed using the proposed matched steady-state stochastic resonance model and vibration resonance model exhibits good bandpass filtering characteristics, effectively filtering out high-frequency and low-frequency components and increasing the accuracy of fault feature extraction.

[0052] This article uses specific examples to illustrate the principles and implementation methods of the invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. The described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

Claims

1. A bearing fault diagnosis method of a vibration resonance-assisted enhanced stochastic resonance coupling system, characterized by The method comprises the following steps: Step S1: vibration signal preprocessing, collecting rolling bearing vibration signals by using a sensor, and performing Hilbert envelope demodulation processing on the collected rolling bearing vibration signals to obtain an amplitude envelope of the vibration signals; Step S2: coupling system construction, constructing a matching steady-state stochastic resonance system, and combining the vibration resonance system with the matching steady-state stochastic resonance system through nonlinear coupling to form a coupling system; The potential function of the matching steady-state stochastic resonance system in the step S2 is as follows: Wherein, m and n are system parameters, and are both positive real numbers; by adjusting system parameters and n, the potential function can exhibit monostable or bistable characteristics. The coupling system in the step S2 is composed of a matching steady-state stochastic resonance system and a bistable vibration resonance system, and the specific mathematical model is as follows: where is the matching potential function, is the bistable potential function, a and b are system parameters; the output of the matching stochastic resonance and the output of the bistable stochastic resonance are coupled together through the coupling coefficient ; is the system input signal, is the added high-frequency signal, i.e. , H and are the amplitude and frequency, respectively; Step S3: system parameter optimization, taking the amplitude envelope obtained in the step S1 as an input signal of the constructed coupling system, taking an output signal-to-noise ratio as an objective function of an intelligent optimization algorithm, taking an iteration number as a termination condition, and realizing optimization selection of system multiple parameters by seeking a maximum output signal-to-noise ratio in multiple iterations; Step S4: fault identification, substituting the optimal parameters obtained in the step S3 into the coupling system, processing the amplitude envelope obtained in the step S1, performing spectrum analysis on system output, and completing effective identification and diagnosis of rolling bearing faults.

2. The bearing fault diagnosis method of a vibration resonance-assisted enhanced stochastic resonance coupling system according to claim 1, characterized in that: Output of a bistable vibrational resonance system is the result of a low pass filter with a passband cutoff frequency of and a stopband cutoff frequency of The final output of the coupled system is ​ 3. The bearing fault diagnosis method of a vibration resonance-assisted enhanced stochastic resonance coupling system according to claim 1, characterized in that: The optimization algorithm in the step S3 is a particle swarm optimization algorithm, and a signal-to-noise ratio of the coupling system output is taken as a fitness function to optimize selection of system multiple parameters.

4. The bearing fault diagnosis method of a vibration resonance-assisted enhanced stochastic resonance coupling system according to claim 3, characterized in that: The system multi-parameters mainly include: for the stochastic resonance system, adjusting parameters m and n to realize stochastic resonance control; for the vibration resonance system, fixing system parameters , adjusting amplitude H and frequency to realize vibration resonance control; for the coupling system, adjusting coupling coefficient to realize joint control of stochastic resonance and vibration resonance; when the frequency of the system input signal is much greater than 1 Hz, adjusting scale transformation factor R; through the particle swarm optimization algorithm, the system parameters m, n, coupling coefficient , amplitude H and frequency of the high-frequency signal, and scale factor R are adaptively adjusted, the output signal-to-noise ratio of the system is taken as the objective function, the optimal parameter combination is obtained by seeking the maximum output signal-to-noise ratio in multiple iterations.

5. The bearing fault diagnosis method of a vibration resonance-assisted enhanced stochastic resonance coupling system according to claim 4, characterized in that: The search ranges of the system parameters are set as follows: the search ranges of the system parameters m and n are , the search ranges of the high-frequency signal amplitude H and the frequency are and , the search range of the coupling coefficient is , and the search range of the scale transformation factor R is .

Images