A rotating machinery vibration order analysis method based on angle domain quasi-synchronous iteration

Abstract

This invention discloses a method for analyzing the order of vibration in rotating machinery based on quasi-synchronous iteration in the angle domain, relating to the field of vibration signal order monitoring and analysis technology in rotating machinery. The method involves acquiring angle domain signals at equal angles to form an initial A / D angle domain acquisition sequence. Iterative coefficients are calculated on this initial A / D angle domain acquisition sequence to obtain a quasi-synchronous reconstruction sequence of the vibration signal after quasi-synchronous iteration. Based on this quasi-synchronous reconstruction sequence, a Fourier transform is performed to obtain the order spectrum of the rotating machinery vibration signal. Further analysis of the order spectrum yields the amplitude of each order of the vibration signal. This invention can reduce the impact of synchronization errors caused by sampling at non-integer multiples of rotational speed on accuracy, even when the number of angle domain sampling points is not an integer multiple of the rotational speed.

Description

Technical Field

[0001] This invention relates to the field of vibration signal order monitoring and analysis technology for rotating machinery, specifically a method for vibration order analysis of rotating machinery based on quasi-synchronous iteration in the angle domain. Background Technology

[0002] Rotating machinery is a type of mechanical equipment widely used in industries such as automobiles, aviation, petrochemicals, and power. Vibration problems have always been one of the main factors restricting the safety, stability, and lifespan of such machinery. Traditional time / frequency analysis methods are not ideal for analyzing non-stationary vibration signals from rotating machinery with variable speeds.

[0003] Traditional Fourier analysis is suitable for stationary vibration signals because it can analyze the global frequency and energy distribution of the signal, but it cannot analyze the time-varying patterns and local characteristics of non-stationary vibration signals. If Fourier analysis is directly performed on non-stationary vibration signals, the resulting spectrum will be inaccurate due to the overlapping spectral bandwidths of the components, making it difficult to identify fault frequencies. With the increasing demand for non-stationary vibration signal analysis in industrial applications, order analysis techniques have become widely used due to the limitations of traditional Fourier analysis methods.

[0004] Theoretically, this problem can be solved by employing order analysis and converting a non-stationary signal into a stationary signal in the angular domain through time-to-angle domain transformation. In order analysis, we convert a non-stationary signal in the time domain into a stationary signal in the angular domain through equal-angle resampling. At this point, all order components in the signal become sine waves of constant frequency in the angular domain.

[0005] However, in practical implementation, due to the non-ideal nature of hardware angular sampling, the number of physically collected data points is not an integer multiple of the rotational speed, or the frequencies of each order are not integer multiples of the order resolution when performing Fourier transforms in the angular domain. This directly leads to synchronization errors, causing order energy to diffuse to adjacent frequencies. Visually, this manifests in the order spectrum as prominent order peaks becoming broad and wide with side lobes, and peak amplitudes lower than the true values. Therefore, researching a method for measuring the vibration order of rotating machinery that improves accuracy and reduces the impact of synchronization errors on analysis results is of great significance in ensuring the safe and stable operation of rotating machinery. Summary of the Invention

[0006] The purpose of this invention is to provide a method for analyzing the vibration order of rotating machinery based on quasi-synchronous iteration in the angle domain, so as to solve the problems in the prior art.

[0007] To achieve the above objectives, this invention provides a method for measuring the order of vibration of rotating machinery based on quasi-synchronous iteration in the angle domain. The method utilizes an angle-domain quasi-synchronous iterative algorithm to reconstruct the acquired signal in the angle domain, which has a certain synchronization error, obtaining a reconstructed sequence after quasi-synchronous processing. Then, a Fourier transform is performed on this reconstructed sequence to obtain the order spectrum of the rotating machinery vibration signal, resulting in an improved order measurement result. The specific technical solution is as follows: Step 1: Acquire angular domain signals at equal angles to form the initial A / D angular domain acquisition sequence; Step 2: Calculate the iterative coefficients of the initial A / D angle domain acquisition sequence to obtain the quasi-synchronous reconstruction sequence of the vibration signal after quasi-synchronous iteration; Step 3: Based on the quasi-synchronous reconstruction sequence of the vibration signal after quasi-synchronous iteration, perform Fourier transform to obtain the order spectrum of the rotating machinery vibration signal; Step 4: Further analyze the order spectrum of the vibration signal of the rotating machinery to obtain the amplitude of each order of the vibration signal.

[0008] According to the above technical solution, step 1 further includes: The real-time speed signal is obtained by the tachometer. The speed signal is then passed through a frequency multiplier phase-locked loop circuit to generate an external trigger pulse after frequency multiplication. This pulse serves as the acquisition control signal for the A / D converter. The vibration signal is acquired with a constant angular increment. At the same time, a filter is added to the sampled signal to filter out high-frequency harmonics and noise interference, and the angular domain signal acquired at equal angles is obtained as the original A / D angular domain acquisition sequence.

[0009] According to the above technical solution, it also includes: Determine the filter cutoff frequency and the angle domain sampling rate of the external trigger pulse after the rotational speed signal is multiplied by the frequency multiplier phase-locked loop circuit.

[0010] According to the above technical solution, determining the filter cutoff frequency includes: Once the range of reference shaft speeds for the rotating machinery and the highest order of analysis required are determined, then the highest order of analysis required can be determined. The frequency range of the order components within is: ; The filter's cutoff frequency is: ; in, This is the maximum rotational speed of the reference shaft of the rotating machinery.

[0011] According to the above technical solution, the determination of the angle domain sampling rate of the external trigger pulse after the frequency doubling phase-locked loop generates the rotational speed signal includes: After the time-domain vibration signal of rotating machinery is filtered, it is greater than or equal to The frequency components are filtered out, k-th order component signal The corresponding frequency The range is:

[0012] Therefore, the highest order contained in the filtered signal for: ; Further determine the angular frequency sampling rate for: ; in, This is the minimum reference shaft speed for rotating machinery.

[0013] According to the above technical solution, the iteration parameters include a system-preset selection rule, which includes the number of sampling points N per revolution and the number of iterations p determined by the system.

[0014] Compared with the prior art, the beneficial effects of the present invention are: it has higher accuracy in the corresponding order components in the measurement of vibration of rotating machinery, and can reduce the impact of synchronization error caused by sampling at non-integer multiples of rotational speed on accuracy when the number of sampling points in the angle domain is not an integer multiple of the rotational speed. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating the steps of a method for analyzing the order of vibrations in rotating machinery based on quasi-synchronous iteration in the angle domain, as described in this invention. Figure 2 The figure shows the iteration coefficient curve of the vibration order analysis method of rotating machinery based on quasi-synchronous iteration in the angle domain according to the present invention. Figure 3 This is a schematic diagram illustrating the generation of the original angle domain acquisition sequence from the rotational speed and vibration signal acquisition of a rotating machinery vibration order analysis method based on quasi-synchronous iteration in the angle domain according to the present invention. Figure 4 The figure shows the results of the vibration order analysis method for rotating machinery based on quasi-synchronous iteration in the angle domain under a synchronization error of 0.5% according to the present invention. Figure 5 The graph shows the results under a synchronization error of 0.5% for the conventional order analysis method. Detailed Implementation

[0016] 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.

[0017] Example: Figures 1-5 As shown, the present invention provides a technical solution: An initial A / D angle domain acquisition sequence is formed by acquiring angular domain signals at equal angles. The initial A / D angle domain acquisition sequence is then iteratively coefficient-calculated to obtain a quasi-synchronous reconstruction sequence of the vibration signal after quasi-synchronous iteration. Based on the quasi-synchronous reconstruction sequence of the vibration signal after quasi-synchronous iteration, a Fourier transform is performed to obtain the order spectrum of the rotating machinery vibration signal. The amplitude of each order of the vibration signal is further analyzed based on the order spectrum of the rotating machinery vibration signal.

[0018] If a periodic deviation occurs during data acquisition, a synchronization error will be introduced into the measurement results. If the actual signal is... Calculated value for:

[0019] Let the starting point of integration be... ,make , Therefore For a periodic function, it is possible to The mean is obtained by integration over the period. ,and .

[0020] Let the starting point of integration be... ,but ; Similarly, since it is not possible to strictly achieve integrals over an entire period, let: ; Following this pattern, we can define the following recursive formula:

[0021] in, In this context, "n" represents the nth recursive operation. Let be the integral variable for the nth recursive operation.

[0022] After multiple iterations, we can obtain:

[0023] Discretize the algorithm: In width The integration interval [ On the sample, N+1 data points are collected. ( And perform the following operations according to the quadrature formula for a composite rectangle:

[0024] in, The "1" in the text indicates the first product operation; ; indicates the starting point of each product operation; N represents the number of sampling points in each period; The weight coefficients are determined by the formula for quadrature of the corresponding complex rectangle.

[0025] If the width is Equal interval sampling on the interval Data ( ).

[0026] The first product operation is performed using... The calculations performed starting from this point show that... Sampling start point The function.

[0027] The subsequent recursive operations are defined as follows:

[0028] In the formula, In this context, "n" indicates the nth product operation; ; It is a periodic signal The result of the nth recursion is a set of vectors; if the recursion is performed p times in total, the final result obtained through iterative calculation is a set of weight coefficients. The sum of all numbers can be written in the following form:

[0029] in, For data The weighting coefficients. For a given numerical quadrature formula and pre-given N and n, the weighting coefficients... It can be calculated in advance and is independent of the data itself.

[0030] Vibration signals sampled at equal angles can be represented as:

[0031] in, Let m be the angle and m be the order. , represents the amplitude coefficients of the m-th order cosine and sine components.

[0032]

[0033]

[0034] make , It is also based on A periodic function with periodicity, therefore

[0035] By comparison, we can obtain

[0036] Therefore, as long as the exact result is obtained Then you can get accurate results. Value. Determine the number of iterations. Then we can use the following formula to obtain Value. That is:

[0037] As for the weighting coefficients, as long as we obtain value, The value can be obtained. That is, any angle The value is equal to that angle. Value and that angle The product of:

[0038] Seek Value, further obtained The value of can be obtained similarly. value.

[0039]

[0040]

[0041] make ;

[0042]

[0043] The discrete Fourier transform is as follows:

[0044] in, , These are the nth-order cosine and sine components.

[0045] Therefore, it can be seen that the equiangular sampling values ​​can be obtained. Multiply by the normalized iteration coefficients to reconstruct a new sequence, and then perform a discrete Fourier transform on the new sequence to obtain the quasi-synchronous iterative sequence. , .

[0046] After obtaining the result, we can then use:

[0047]

[0048] The magnitude and phase of the m-th order are obtained, where Let m be the amplitude of the m-th order component. Let be the phase of the m-th order component.

[0049] In this embodiment, the above method is specifically demonstrated as follows: The first step, after determining the number of sampling points, the number of iterations, and the numerical quadrature method based on specific measurement requirements, is to determine the weight coefficients for quasi-synchronous iteration. Establish a quasi-synchronous weight coefficient array. Based on the previously derived iterative expansion, expand the recursive operation, and according to the determined numerical quadrature formula and given N and p, determine the weight coefficients. It can be calculated in advance and is independent of the data itself.

[0050] The specific iterative process can be performed using computer-aided tools to obtain the weighting coefficients.

[0051] ,in =

[0052] The number of sampling points N per revolution is 128, and the number of iterations p is 3. The weight coefficients are determined iteratively using a composite rectangular integral method as follows: Let =

[0053]

[0054] The second step, as Figure 2 As shown, a speed sensor and a vibration sensor are used to acquire the speed and vibration signals of the rotating machinery, respectively. The speed signal is multiplied by a phase-locked loop to generate a multiplied acquisition pulse, which serves as the acquisition clock for the A / D converter, achieving signal acquisition at equal angles. The vibration signal is filtered and then sent to the analog input of the A / D converter. After conversion, the original angle domain acquisition sequence is generated. ,in = .

[0055] The selection principles for the filter cutoff frequency and the angle domain sampling rate of the frequency multiplier circuit trigger pulse are as follows: The reference shaft speed range of rotating machinery is The highest order of analysis required is The following are the specific steps for signal anti-aliasing operations: Determining the filter cutoff frequency: Required order of analysis The frequency range of the order components within is: ; The filter's cutoff frequency is: .

[0056] Determining the corner sampling rate: After the time-domain vibration signal of rotating machinery is filtered by a filter, The frequency components are filtered out. The frequency range corresponding to the k-th order component signal is... The higher-order components are filtered out, so the highest-order component in the filtered signal can be calculated. for: The angular frequency sampling rate is determined to be: .

[0057] The third step is to process the angle domain acquisition sequence obtained in the second step. Compared with the weight coefficient array obtained in the first iteration Perform a multiplication operation to obtain the reconstructed quasi-synchronous sequence. .

[0058] Step 4: Reconstruct the quasi-synchronous sequence obtained in Step 3. Perform a DFT to obtain the order sequence. = .right To conduct analysis, in The peak values ​​corresponding to each order are found, and the amplitude and phase of each order component are obtained by analyzing the peak values.

[0059] Thus, the vibration order analysis of rotating machinery based on quasi-synchronous iteration in the angle domain was completed.

[0060] The following simulation experiments using MATLAB are used to verify the method provided by this invention.

[0061] The relevant simulation parameter values ​​are as follows: The number of sampling points per week is N=128, the number of iterations is p=3, and the complex rectangular integral iteration calculation is used, that is, the iteration coefficients use those provided in this invention. ; The signal model is as follows: x_theta = A1 * cos(1*theta + phi1)+A2 * cos(2*theta + phi2)+ A3 * cos(4*theta + phi3) +A4 * cos(15*theta + phi4); In the formula, theta represents sampling at equal angles, with a step size of 2*pi / N*(1 + err_percent), and err_percent represents the synchronization error. The magnitude of the first-order component A1 is 1, and the phase phi1 is 0; The magnitude of the second-order component A2 is 0.5, and the phase phi2 is pi / 4; The magnitude A3 of the 4th-order component is 0.3, and the phase phi3 = pi / 2; The amplitude A4 of the 15th-order component is 0.2, and the phase phi4 = pi / 3; like Figure 4 As shown, the final order analysis results are obtained through simulation tools, and the order amplitude and corresponding error of each simulation are exported. Specific results are shown in Tables 1, 2, and 3. By modifying the magnitude of the synchronization error parameter, the influence of synchronization error on the order analysis results under different synchronization error conditions is verified, as well as the effectiveness of the method of this invention.

[0062] The simulation results comparing the conventional method and the method provided by this invention under the condition of a synchronization error of 0.1% are shown in Table 1 below.

[0063] Table 1. Comparison of simulation experiments between conventional methods and the method provided by this invention under the condition of a synchronization error of 0.1%.

[0064] The simulation results comparing the conventional method and the method provided by this invention under the condition of a synchronization error of 0.2% are shown in Table 2 below.

[0065] Table 2 compares the simulation experiments of the conventional method and the method provided by this invention under the condition of a synchronization error of 0.2%.

[0066] The simulation results comparing the conventional method and the method provided by this invention under a synchronization error of 0.7% are shown in Table 3 below.

[0067] Table 3. Comparison of simulation experiments between conventional methods and the method provided by this invention under the condition of a synchronization error of 0.7%.

[0068] Simulation results show that the method provided by this invention can reduce the impact of synchronization error on the vibration order analysis of rotating machinery and improve the accuracy of the order analysis.

[0069] like Figure 5 As shown, this application also provides a result graph of the conventional order analysis method with a synchronization error of 0.5%, through... Figure 4 , Figure 5 As can be seen from the comparison, the present invention reduces the impact of synchronization errors caused by sampling at non-integer multiples of rotational speed on accuracy.

[0070] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for analyzing the vibration order of rotating machinery based on quasi-synchronous iteration in the angle domain, characterized in that: The method includes: Step 1: Acquire angular domain signals at equal angles to form the initial A / D angular domain acquisition sequence; Step 2: Calculate the iterative coefficients of the initial A / D angle domain acquisition sequence to obtain the quasi-synchronous reconstruction sequence of the vibration signal after quasi-synchronous iteration; Step 3: Based on the quasi-synchronous reconstruction sequence of the vibration signal after quasi-synchronous iteration, perform Fourier transform to obtain the order spectrum of the rotating machinery vibration signal; Step 4: Further analyze the order spectrum of the vibration signal of the rotating machinery to obtain the amplitude of each order of the vibration signal.

2. The method for analyzing the vibration order of rotating machinery based on quasi-synchronous iteration in the angle domain according to claim 1, characterized in that: Step 1 also includes: The real-time speed signal is obtained by the tachometer. The speed signal is then passed through a frequency multiplier phase-locked loop circuit to generate an external trigger pulse after frequency multiplication. This pulse serves as the acquisition control signal for the A / D converter. The vibration signal is acquired with a constant angular increment. At the same time, a filter is added to the sampled signal to filter out high-frequency harmonics and noise interference, and the angular domain signal acquired at equal angles is obtained as the original A / D angular domain acquisition sequence.

3. The method for analyzing the vibration order of rotating machinery based on quasi-synchronous iteration in the angle domain according to claim 2, characterized in that: Also includes: Determine the filter cutoff frequency and the angle domain sampling rate of the external trigger pulse after the rotational speed signal is multiplied by the frequency multiplier phase-locked loop circuit.

4. The method for analyzing the vibration order of rotating machinery based on quasi-synchronous iteration in the angle domain according to claim 3, characterized in that: Determining the filter cutoff frequency includes: Once the range of reference shaft speeds for the rotating machinery and the highest order of analysis required are determined, then the highest order of analysis required can be determined. The frequency range of the order components within is: ; The filter's cutoff frequency is: ; in, This is the maximum rotational speed of the reference shaft of the rotating machinery.

5. The method for analyzing the vibration order of rotating machinery based on quasi-synchronous iteration in the angle domain according to claim 4, characterized in that: The determination of the angle domain sampling rate of the external trigger pulse after the frequency doubling phase-locked loop generates the rotational speed signal includes: After the time-domain vibration signal of rotating machinery is filtered, it is greater than or equal to The frequency components are filtered out, k-th order component signal The corresponding frequency The range is: ； Therefore, the highest order contained in the filtered signal for: ; Further determine the angular frequency sampling rate for: ; in, This is the minimum reference shaft speed for rotating machinery.

6. The method for analyzing the order of vibration of rotating machinery based on quasi-synchronous iteration in the angle domain according to claim 1, characterized in that: The iteration parameters include system-preset selection rules, which include the number of sampling points N per revolution and the number of iterations p determined by the system.

Images