Armored vehicle transmission system fault diagnosis method and system based on sound signals

Abstract

The invention discloses an armored vehicle transmission system fault diagnosis method and system based on sound signals. The method comprises the steps that the sound signals in the operation process of a transmission system and the sound signals in the normal operation state are obtained through a sound signal sensor; segmenting the collected sound signal; respectively calculating tail-cut average square envelope negentropies by using the acquired sound signals and making difference values; selecting a frequency band with a relatively large difference value as a center frequency, and determining a search bandwidth according to a center frequency range; filtering the sound signal according to a frequency band determined by the center frequency and the bandwidth, and calculating a composite weighting index of the filtered sound signal; and selecting the frequency band with the maximum composite weighted index value as a final frequency band, drawing an envelope spectrum image of the frequency band, and marking the fault characteristic frequency in the envelope spectrum. Compared with a traditional transmission system fault diagnosis method, the method has the advantages that various indexes representing fault characteristics are fused, the sensitivity, the reliability and the accuracy of fault diagnosis are effectively improved, and the missing report rate and the false report rate of the fault diagnosis are reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of operation state monitoring of mechanical transmission systems, and particularly relates to a fault diagnosis method and system for an armored vehicle transmission system based on acoustic signals. Background Technique

[0002] The transmission system is a key component in mechanical equipment and is widely used in various fields such as aerospace, automotive transmission, and wind power equipment. The working environment of the armored vehicle transmission system is harsh and complex. Due to reasons such as overload and fatigue, key components such as gears, rotors, and bearings are prone to damage, which in turn affects the normal operation of the entire transmission system, causing serious safety problems and economic losses. Therefore, in order to ensure the normal and stable operation of the transmission system, it is very necessary to carry out state monitoring and fault diagnosis on it.

[0003] Fault diagnosis of transmission systems still faces significant challenges in practice. The fault diagnosis method needs to ensure both its robustness and accuracy at the same time. During the operation of the equipment, a large amount of state information will be generated. Acoustic signal sensors have the advantages of non-contact measurement, easy acquisition, and no need to paste sensors in advance, and are widely used in the fault diagnosis of transmission systems. The acoustic signal itself has the characteristics of a wide frequency band and a low signal-to-noise ratio. Therefore, the key to fault diagnosis of transmission systems based on acoustic signals is to find the sensitive frequency band rich in fault information in the acoustic signal to achieve fault diagnosis.

[0004] The Chinese patent with the publication number: CN117554062A provides a gearbox fault identification method based on the time-frequency spectrogram of acoustic signals to solve the problem of diagnosing broken tooth faults of gears. However, this method has poor noise resistance. Kurtosis is a fourth-order dimensionless parameter that can quickly and accurately locate the position of non-stationary signals such as impacts in the frequency domain, and is particularly suitable for diagnosing early faults of transmission systems. Finding the sensitive frequency band rich in fault information based on the maximum kurtosis value is a current research hotspot. Antoni first proposed a spectral kurtosis calculation method based on the short-time Fourier transform and a fast spectral kurtosis calculation method based on a binary tree and a filter bank; the Chinese patent with the publication number: CN113743338A provides a gearbox fault diagnosis method and diagnosis system based on the SVN-Kurtosis diagram to solve the problems of fault feature enhancement and fault diagnosis of gearboxes. However, this method does not consider the influence of single-pulse interference on kurtosis calculation; to avoid the influence of single-pulse interference on the fast spectral kurtosis algorithm, Dai Shichao et al. proposed an improvement to the fast spectral kurtosis diagram algorithm based on the average of sub-band spectral kurtosis; Wang L et al. introduced the dual-tree complex wavelet packet transform (DTCWPT) to replace the filter bank in the fast spectral kurtosis diagram algorithm, improving the calculation efficiency and accuracy of the algorithm.

[0005] It is found from the existing retrieval literature that there are the following two problems in the currently commonly used fault diagnosis method for the transmission system based on the selection of the sensitive frequency band of the acoustic signal: 1) The influence of sporadic impacts in the signal on the selection of the sensitive frequency band cannot be well eliminated; 2) Most of them only consider extracting the impact in the signal to select the sensitive frequency band, without considering the cyclostationarity in the signal and the influence of the healthy reference signal on the selection of the sensitive frequency band. Summary of the Invention

[0006] In order to solve the problems existing in the prior art, the present invention provides a fault diagnosis method for the transmission system of an armored vehicle based on the acoustic signal, which is used to solve the technical problems that it is difficult to eliminate the influence of sporadic impacts and the influence of cyclostationarity and healthy reference signal on the selection of the sensitive frequency band in the currently commonly used fault diagnosis method for the transmission system based on the selection of the sensitive frequency band of the acoustic signal. The method introduces the index calculation method of sub-band trimmed mean, comprehensively considers the influence of impact, cyclostationarity and healthy reference signal in the signal on the selection of the sensitive frequency band, and establishes a composite weighted index.

[0007] In order to achieve the above object, the technical solution adopted by the present invention is: a fault diagnosis method for the transmission system of an armored vehicle based on the acoustic signal, including the following steps:

[0008] S1, perform average segmentation processing on the acoustic signal during the operation of the transmission system and the acoustic signal in the normal operation state, and obtain n segments of acoustic signals during the operation process and n segments of acoustic signals in the normal operation state.

[0009] S2, perform band-pass filtering on each segment of the acoustic signal during the operation process and the acoustic signal in the normal operation state respectively with a set frequency band, and obtain filtered signals corresponding to the number of frequency bands;

[0010] S3, calculate the difference between the corresponding values of the trimmed mean square envelope negative entropy of the obtained acoustic signal during the operation process and the acoustic signal in the normal operation state, and select the frequency band with a larger difference as the center frequency band;

[0011] S4, determine the center frequency and the search bandwidth;

[0012] S5, use the frequency band composed of the center frequency and the search bandwidth as the frequency band in S2, and execute S2 and S3 again to obtain the trimmed mean square envelope negative entropy and the trimmed mean square envelope spectrum negative entropy of the acoustic signal during the operation process and the acoustic signal in the normal operation state;

[0013] S6, calculate the composite weighted index by using the trimmed mean square envelope negative entropy and the trimmed mean square envelope spectrum negative entropy of the acoustic signal during the operation process and the acoustic signal in the normal operation state obtained in S5, and select the filtered frequency band corresponding to the maximum value in the composite weighted index as the final frequency band;

[0014] S7. Filter the acoustic signal during the operation process according to the final frequency band to obtain the filtered signal, calculate the envelope signal of the filtered signal, and obtain the fault characteristic frequency of the transmission system from the envelope signal.

[0015] Further, in S1, the acoustic signal during the operation process of the transmission system and the acoustic signal in the normal operation state include four time-domain signals, namely periodic impact signal, sine signal, single impact, and Gaussian noise.

[0016] Further, in S1, the acoustic signal during the operation process of the transmission system and the acoustic signal in the normal operation state are obtained through simulation or from the monitoring of the actually operating transmission system.

[0017] Further, in S3, calculate the difference between the corresponding values of the trimmed mean square envelope negative entropy of the obtained acoustic signal during the operation process and the acoustic signal in the normal operation state, and select the frequency band with a larger difference as the central frequency band, including:

[0018] The square envelope negative entropy index is:

[0019]

[0020] where N C represents the length of the calculated signal, and ε x is the square envelope of the calculated signal;

[0021] The calculation result of the first segment of the acoustic signal during the operation process is expressed as The calculation result of the nth segment is expressed as m represents the number of filtering frequency bands in S2, and the trimmed mean square envelope negative entropy of the acoustic signal during the operation process is expressed as:

[0022]

[0023] The calculation result of the first segment of the acoustic signal in the normal operation state is expressed as The calculation result of the nth segment is expressed as The trimmed mean square envelope negative entropy of the acoustic signal in the normal operation state is expressed as:

[0024]

[0025] Take the difference S TA_NESE -H TA_NESE between the corresponding values of the trimmed mean square envelope negative entropy of the obtained acoustic signal during the operation process and the acoustic signal in the normal operation state, and select the frequency band with a larger difference as the central frequency band.

[0026] Further, in S5, the calculation methods of the frequency band B i,j , square envelope spectrum negative entropy, trimmed mean square envelope negative entropy, and trimmed mean square envelope spectrum negative entropy are as follows:

[0027]

[0028] Among them, C i is the i-th center frequency; W j is the j-th search bandwidth; B i,j is the bandpass filter frequency band determined by C i and W j ; N C represents the length of the signal for calculating the negative entropy of the squared envelope spectrum, ε x is the squared envelope of the calculated signal, and E x is the energy spectrum of ε x ; x is the collected acoustic signal; Trimmed Average NESE(x|B i,j ) is to perform bandpass filtering on the signal x using B i,j and calculate the trimmed average negative entropy of the squared envelope; Trimmed Average NESES(x|B i,j ) is to perform bandpass filtering on the signal x using B i,j and calculate the trimmed average negative entropy of the squared envelope spectrum.

[0029] Furthermore, in S6, the composite weighted index is calculated using the trimmed average negative entropy of the squared envelope and the trimmed average negative entropy of the squared envelope spectrum of the running process acoustic signal and the normal running state acoustic signal, which is expressed as:

[0030]

[0031] Select the filter frequency band corresponding to the maximum value max(WI i×j ) in the WI matrix as the final frequency band B O .

[0032] The present invention also provides an armored vehicle transmission system fault diagnosis system based on acoustic signals, including a segmentation processing module, a filtering module, a center frequency band acquisition module, a center frequency determination module, a calculation module, and a fault characteristic frequency acquisition module;

[0033] The segmentation processing module is used to perform average segmentation processing on the running process acoustic signal and the normal running state acoustic signal of the transmission system to obtain n running process acoustic signals and n normal running state acoustic signals.

[0034] The filtering module is used to perform bandpass filtering on each running process acoustic signal and the normal running state acoustic signal of the running process respectively with a set frequency band to obtain filtering signals corresponding to the number of frequency bands;

[0035] The center frequency band acquisition module is used to calculate the difference between the corresponding values of the trimmed average negative entropy of the squared envelope of the obtained running process acoustic signal and the normal running state acoustic signal, and select the frequency band with a larger difference as the center frequency band;

[0036] The center frequency determination module is used to determine the center frequency and the search bandwidth;

[0037] The calculation module uses the frequency band composed of the center frequency and the search bandwidth as the filtering frequency band, and performs filtering and obtains the center frequency band again to obtain the trimmed mean square envelope negative entropy and the trimmed mean square envelope spectrum negative entropy of the sound signal during operation and the sound signal in the normal operation state; calculates the composite weighted index by using the obtained trimmed mean square envelope negative entropy and the trimmed mean square envelope spectrum negative entropy of the sound signal during operation and the sound signal in the normal operation state, and selects the filtering frequency band corresponding to the maximum value in the composite weighted index as the final frequency band;

[0038] The fault characteristic frequency acquisition module filters the sound signal during operation according to the final frequency band to obtain the filtered signal, calculates the envelope signal of the filtered signal, and obtains the fault characteristic frequency of the transmission system from the envelope signal.

[0039] Further, the fault characteristic frequency acquisition module includes an envelope spectrogram drawing unit, filters the sound signal during operation according to the final frequency band to obtain the filtered signal, calculates the envelope signal of the filtered signal, draws the envelope spectrogram and marks the fault characteristic frequency of the transmission system, and identifies the fault frequency and harmonic frequency based on the envelope spectrogram.

[0040] The present invention can also provide a computer device, including a processor and a memory. The memory is used to store computer-executable programs. The processor reads the computer-executable programs from the memory and executes them. When the processor executes the computer-executable programs, it can implement the fault diagnosis method for the armored vehicle transmission system based on sound signals described in the present invention.

[0041] At the same time, a computer-readable storage medium is provided. A computer program is stored in the computer-readable storage medium. When the computer program is executed by a processor, it can implement the fault diagnosis method for the armored vehicle transmission system based on sound signals described in the present invention.

[0042] Compared with the prior art, the present invention has at least the following beneficial effects: In order to reduce the influence of sporadic impulses existing in the signal on the frequency band selection result, the present invention introduces an index calculation method of sub-band trimmed mean, which reduces the influence of sporadic impulses in the signal on this method; comprehensively considers the influence of impulse, cyclic stationarity and healthy reference signal in the signal on the selection of sensitive frequency bands and establishes a composite weighted index, which improves the robustness and accuracy of the selection of sensitive frequency bands; compared with the traditional fault diagnosis method for the transmission system, the method of the present invention integrates a variety of indexes characterizing fault characteristics, effectively improves the sensitivity, reliability and accuracy of the fault diagnosis of the transmission system, and reduces the false negative rate and false positive rate of the fault diagnosis. Description of the Drawings

[0043] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the attached drawings required for the description of the embodiments. Obviously, the attached drawings described below are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other attached drawings can also be obtained based on these attached drawings.

[0044] Figure 1 The sound signal for the operation process of the simulation is S = [s(1), s(2), … s(N)].

[0045] Figure 2 The sound signal for the normal operation state of the simulation is H = [h(1), h(2), … h(N)].

[0046] Figure 3 Schematic diagram of evenly dividing the sound signal during the operation process into 5 segments S1, S2, S3, S4, S5.

[0047] Figure 4 Schematic diagram of evenly dividing the sound signal in the normal operation state into 5 segments H1, H2, H3, H4, H5.

[0048] Figure 5 Schematic diagram of the result of taking the difference between the trimmed mean square envelope negative entropy of the sound signal during the operation process and the sound signal in the normal operation state S TA_NESE -H TA_NESE .

[0049] Figure 6 3D schematic diagram of the calculated composite weighted index matrix.

[0050] Figure 7 2D schematic diagram of the calculated composite weighted index matrix.

[0051] Figure 8 Based on the final frequency band B O Filter the sound signal S during the operation process to obtain the filtered signal s, and calculate and draw the envelope spectrum of the signal s and mark the fault characteristic frequencies. Specific implementation manner

[0052] To make the purpose, technical solutions, and advantages of the implementation of this application clearer, the following will describe the technical solutions in the embodiments of this application in more detail with reference to the accompanying drawings in the embodiments of this application. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions from beginning to end. The described embodiments are part of the embodiments of this application, not all of them. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain this application and should not be construed as a limitation of this application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this application without creative efforts belong to the scope of protection of this application. The following will explain the embodiments of this application in detail with reference to the accompanying drawings.

[0053] The following will further describe in detail a method for diagnosing faults in the transmission system of an armored vehicle based on acoustic signals in combination with the accompanying drawings.

[0054] A method for diagnosing faults in the transmission system of an armored vehicle based on acoustic signals includes the following steps:

[0055] (1) Obtain the acoustic signal of the transmission system

[0056] Obtain the acoustic signal during the operation of the transmission system and the acoustic signal in the normal operation state through simulation or real-time monitoring of the actually operating transmission system. Represent the obtained acoustic signal during the operation process as S = [s(1), s(2),..., s(N)], refer to Figure 1 , and represent the acoustic signal in the normal operation state as H = [h(1), h(2),..., h(N)], refer to Figure 2 . The sampling frequency of the simulation signal Fs = 10000Hz, the sampling length N = 30000, and a total of four time-domain signals are set, namely the periodic impulse signal x(t), the sine signal c(t), the single impulses i(t)1, i(t)2, and the Gaussian noise n(t). The specific mathematical formulas are as follows:

[0057]

[0058] where t is time, k is the number of periodic impulses, randn is to generate random numbers with a standard normal distribution, and N is the number of sampling points.

[0059] (2) Segment the signal

[0060] To avoid the influence of sporadic impulses in the collected acoustic signal on the calculated indicators, perform average segmentation processing on the acoustic signal during the operation process and the acoustic signal in the normal operation state. The obtained 5 segments of acoustic signals during the operation process are represented as S1, S2, S3, S4, S5, refer to Figure 3 , and the obtained 5 segments of acoustic signals in the normal operation state are represented as H1, H2, H3, H4, H5, refer to Figure 4 .

[0061] (3) Perform band-pass filtering on the signal

[0062] Filter at intervals of 50 Hz, and perform band-pass filtering on each segment of the acoustic signal during operation in the frequency bands of [0, 50], [50, 100], …, [4950, 5000]. After the first segment of filtering, it is denoted as S 1,1 , S 1,2 , … S 1,100 , where 100 represents the number of filtering frequency bands, and after the 5th segment of filtering, it is denoted as S 5,1 , S 5,2 , …, S 5,100 . Perform the same operation on the acoustic signal in the normal operation state after segmentation. After the first segment of filtering, it is denoted as H 1,1 , H 1,2 , … H 1,100 , and after the 5th segment of filtering, it is denoted as H 5,1 , H 5,2 , …, H 5,100 .

[0063] (4) Calculate the trimmed mean square envelope negative entropy of the signal and take the difference

[0064] Calculate the square envelope negative entropy of the filtered signal. The square envelope negative entropy index is:

[0065]

[0066] where N C represents the length of the calculated signal, and ε x is the square envelope of the calculated signal. The calculation result of the first segment of the acoustic signal during operation is denoted as The calculation result of the 5th segment is denoted as The trimmed mean square envelope negative entropy of the acoustic signal during operation is denoted as:

[0067]

[0068] The calculation result of the first segment of the acoustic signal in the normal operation state is denoted as The calculation result of the 5th segment is denoted as The trimmed mean square envelope negative entropy of the acoustic signal in the normal operation state is denoted as:

[0069]

[0070] Take the difference S TA_NESE -H TA_NESE of the corresponding values of the trimmed mean square envelope negative entropy of the calculated acoustic signal during operation and the acoustic signal in the normal operation state, and select the frequency band with a larger difference as the center frequency band. The center frequency band range is denoted as [2300, 3550] for reference Figure 5 .

[0071] (5) Determine the center frequency and search bandwidth

[0072] The center frequency band range is [2300, 3550], and the center frequency sequence C i is expressed as 2300, 2350, …, 3500, 3550, where i is the number of center frequencies. The search bandwidth sequence W j is expressed as 50×[1, 2, 3, 4, 6, 8, 12, 16, 24], where j is the number of search bandwidths, and the array [1, 2, 3, 4, 6, 8, 12, 16, 24] is composed of the array 2 n (0 ≤ n ≤ 4) and 3·2 n (0 ≤ n ≤ 3). To ensure that the maximum search bandwidth basically covers the center frequency range, it is necessary to ensure that |1200 - 1250| ≤ 200.

[0073] (6) Filter the signal according to different frequency band combinations and calculate the indicators

[0074] Using the center frequency C i and the search bandwidth W j to form the frequency band B i,j to replace the filtering frequency band in step (3) and repeat steps (3) and (4), and calculate the trimmed average squared envelope spectrum negative entropy of the signal according to step (4). The calculation methods of the frequency band B i,j , squared envelope spectrum negative entropy, trimmed average squared envelope negative entropy, and trimmed average squared envelope spectrum negative entropy are as follows:

[0075]

[0076] Among them, C i is the i-th center frequency; W j is the j-th search bandwidth; B i,j is the band-pass filter frequency band determined by C i and W j ; N C represents the length of the signal for calculating the squared envelope spectrum negative entropy, ε x is the squared envelope of the calculated signal, E x is the energy spectrum of ε x ; x is the collected acoustic signal; Trimmed Average NESE(x|B i,j ) is to perform band-pass filtering on the signal x using B i,j and calculate the trimmed average squared envelope negative entropy; Trimmed Average NESES(x|B i,j ) is to perform band-pass filtering on the signal x using B i,j and calculate the trimmed average squared envelope spectrum negative entropy.

[0077] The trimmed mean square envelope negative entropy and the trimmed mean square envelope spectrum negative entropy of the calculated operating process acoustic signal and the normal operating state acoustic signal are denoted as S TA_NESE , H TA_NESE , S TA_NESES , H TA_NESES , both are 9×26 specification arrays, the search bandwidth sequence length is 9, and the center frequency band sequence length is 26.

[0078] (7) Calculate the composite weighted index and select the final frequency band

[0079] Using the trimmed mean square envelope negative entropy and the trimmed mean square envelope spectrum negative entropy of the operating process acoustic signal and the normal operating state acoustic signal obtained in step (6), calculate the composite weighted index, which is denoted as:

[0080]

[0081] Select the filtering frequency band [3025, 3075] corresponding to the maximum value max(WI 9×26 ) in the WI matrix as the final frequency band B O , referring to Figure 6 and Figure 7 .

[0082] (8) Plot the envelope spectrum diagram and mark the fault characteristic frequencies

[0083] Filter the operating process acoustic signal S according to the final frequency band [3025, 3075] to obtain the filtered signal s, calculate the envelope signal e of the filtered signal s, referring to Figure 8 , plot the envelope spectrum diagram and find and mark the fault characteristic frequencies of the transmission system. From Figure 8 , the fault frequency 25Hz and its harmonic frequencies can be clearly identified.

[0084] Based on the concept of the above method, the present invention also provides a fault diagnosis system for the transmission system of an armored vehicle based on acoustic signals, including a segmentation processing module, a filtering module, a center frequency band acquisition module, a center frequency determination module, a calculation module, and a fault characteristic frequency acquisition module;

[0085] The segmentation processing module is used to perform average segmentation processing on the acoustic signals during the operation process of the transmission system and the acoustic signals in the normal operation state, and obtain n acoustic signals during the operation process and n acoustic signals in the normal operation state.

[0086] The filtering module is used to perform band-pass filtering on each acoustic signal during the operation process and the acoustic signal in the normal operation state with a set frequency band, and obtain filtered signals corresponding to the number of frequency bands;

[0087] The center frequency band acquisition module is used to calculate the difference between the corresponding values of the trimmed mean square envelope negative entropy of the obtained running process sound signal and the normal running state sound signal, and select the frequency band with a larger difference as the center frequency band;

[0088] The center frequency determination module is used to determine the center frequency and the search bandwidth;

[0089] The calculation module uses the frequency band composed of the center frequency and the search bandwidth as the filtering frequency band, and filters and acquires the center frequency band again to obtain the trimmed mean square envelope negative entropy and the trimmed mean square envelope spectrum negative entropy of the running process sound signal and the normal running state sound signal; calculates the composite weighted index using the obtained trimmed mean square envelope negative entropy and the trimmed mean square envelope spectrum negative entropy of the running process sound signal and the normal running state sound signal, and selects the filtering frequency band corresponding to the maximum value in the composite weighted index as the final frequency band

[0090] The fault characteristic frequency acquisition module filters the running process sound signal according to the final frequency band to obtain a filtered signal, calculates the envelope signal of the filtered signal, and acquires the transmission system fault characteristic frequency from the envelope signal.

[0091] The fault characteristic frequency acquisition module includes an envelope spectrum diagram drawing unit, filters the running process sound signal according to the final frequency band to obtain a filtered signal, calculates the envelope signal of the filtered signal, draws an envelope spectrum diagram and marks the transmission system fault characteristic frequency, and identifies the fault frequency and harmonic frequency based on the envelope spectrum diagram.

[0092] On the other hand, the present invention provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, it can implement the method for diagnosing faults in the transmission system of an armored vehicle based on sound signals according to the present invention.

[0093] The present invention can also provide a computer device, including a processor and a memory. The memory is used to store computer-executable programs. The processor reads the computer-executable programs from the memory and executes them. When the processor executes the computer-executable programs, it can implement the method for diagnosing faults in the transmission system of an armored vehicle based on sound signals according to the present invention.

[0094] The computer device can be a laptop computer, a desktop computer or a workstation.

[0095] The processor can be a central processing unit (CPU), a digital signal processor (DSP), an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA) ready for use.

[0096] For the memory described in the present invention, it can be an internal storage unit of a laptop, a desktop computer or a workstation, such as a memory or a hard disk; or an external storage unit can be adopted, such as a mobile hard disk or a flash card.

[0097] A computer-readable storage medium may include a computer storage medium and a communication medium. The computer storage medium includes volatile and non-volatile, removable and non-removable media implemented by any method or technology for storing information such as computer-readable instructions, data structures, program modules or other data. The computer-readable storage medium may include: read-only memory (ROM, Read Only Memory), random access memory (RAM, Random Access Memory), solid state drives (SSD, Solid State Drives) or optical discs, etc. Among them, the random access memory may include resistive random access memory (ReRAM, Resistance Random Access Memory) and dynamic random access memory (DRAM, Dynamic Random Access Memory).

[0098] The present invention is described by way of examples. For those skilled in the art, various changes or equivalent replacements can be made to these features and examples without departing from the spirit and scope of the present invention. Additionally, under the teachings of the present invention, these features and examples can be modified to adapt to specific situations and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application belong to the protection scope of the present invention.

Claims

1. A fault diagnosis method for armored vehicle transmission system based on acoustic signals, characterized in that: The following steps are involved: S1, performing average segmentation processing on the transmission system operation process sound signal and the normal operation state sound signal to obtain n segments of operation process sound signal and n segments of normal operation state sound signal; S2, performing bandpass filtering on the sound signal of each operation process and the sound signal of the normal operation state in the set frequency band respectively, to obtain a filtering signal corresponding to the number of frequency bands; S3, calculating the difference between the corresponding values ​​of the trimmed mean square envelope negative entropy of the sound signal of the operation process and the sound signal of the normal operation state, and selecting the frequency band with the larger difference as the central frequency band; S4, determine the center frequency and search bandwidth; S5, using the frequency band composed of the center frequency and the search bandwidth as the frequency band of S2, and executing S2 and S3 again to obtain the trimmed mean square envelope negative entropy and trimmed mean square envelope spectrum negative entropy of the running process sound signal and the normal running state sound signal; S6, using the trimmed mean square envelope negentropy and the trimmed mean square envelope spectrum negentropy of the operation process sound signal and the normal operation state sound signal obtained in S5 to calculate the composite weighted index, and selecting the filter frequency band corresponding to the maximum value in the composite weighted index as the final frequency band; S7, filtering the running process sound signal according to the final frequency band to obtain a filtered signal, calculating an envelope signal of the filtered signal, and obtaining a transmission system fault characteristic frequency from the envelope signal.

2. The method for fault diagnosis of armored vehicle transmission system based on acoustic signals according to claim 1 is characterized in that: In S1, the acoustic signals of the transmission system during operation and the acoustic signals of the normal operating state include four time domain signals, namely, periodic impulse signals, sinusoidal signals, single impulses and Gaussian noise.

3. The method for fault diagnosis of armored vehicle transmission system based on acoustic signals according to claim 1 is characterized in that: In S1, the sound signal of the transmission system during operation and the sound signal of the normal operating state are obtained through simulation or from monitoring of the transmission system in actual operation.

4. The method for fault diagnosis of armored vehicle transmission system based on acoustic signals according to claim 1 is characterized in that: In S3, the corresponding values ​​of the trimmed mean square envelope negative entropy of the calculated operation process sound signal and the normal operation state sound signal are subtracted, and the frequency band with the larger difference is selected as the central frequency band, including: The squared envelope negative entropy index is: Where N C Indicates the length of the calculated signal, ε x To calculate the square envelope of the signal; The calculation result of the first section of the running process sound signal is expressed as The calculation result of the nth segment is expressed as m represents the number of filter bands in S2. The mean square envelope negative entropy of the sound signal during operation is expressed as: The calculation result of the first section of the acoustic signal in normal operation is expressed as The calculation result of the nth segment is expressed as The negative entropy of the tail-cut average square envelope of the acoustic signal in normal operation is expressed as: The difference S is calculated between the corresponding values ​​of the trimmed mean square envelope negative entropy of the running process sound signal and the normal running state sound signal. TA_NESE -H TA_NESE , select the frequency band with the larger difference as the center frequency band.

5. The method for fault diagnosis of armored vehicle transmission system based on acoustic signals according to claim 1 is characterized in that: S5, Band B i,j The calculation methods of , square envelope spectrum negative entropy, trimmed average square envelope negative entropy and trimmed average square envelope spectrum negative entropy are as follows: Among them, C i is the i-th center frequency; W j is the j-th search bandwidth; B i,j Because C i and W j Determine the bandpass filter frequency band; N C Indicates the length of calculating the square envelope spectrum negative entropy signal, ε x To calculate the square envelope of the signal, E x is x Energy spectrum; x is the collected sound signal; Trimmed Average NESE (x|B i,j ) is to use B i,j Bandpass filter the signal x and calculate the trimmed average square envelope negative entropy; Trimmed Average NESES(x|B i,j ) is to use B i,j Bandpass filter the signal x and calculate the trimmed mean square envelope spectrum negentropy.

6. The method for fault diagnosis of armored vehicle transmission system based on acoustic signals according to claim 1 is characterized in that: In S6, the composite weighted index is calculated using the trimmed mean square envelope negentropy and the trimmed mean square envelope spectrum negentropy of the running process sound signal and the normal running state sound signal, and is expressed as: Select the maximum value max(WI i×j ) is used as the final frequency band B O .

7. The armored vehicle transmission system fault diagnosis system based on acoustic signals is characterized by: It includes a segment processing module, a filtering module, a center frequency band acquisition module, a center frequency determination module, a calculation module and a fault characteristic frequency acquisition module; The segment processing module is used to perform average segment processing on the transmission system operation process sound signal and the normal operation state sound signal, and obtain n segments of operation process sound signal and n segments of normal operation state sound signal; The filtering module is used to perform bandpass filtering on the sound signal of each operation process and the sound signal of the normal operation state in the set frequency band respectively, to obtain a filtering signal corresponding to the number of frequency bands; The central frequency band acquisition module is used to calculate the difference between the corresponding values ​​of the trimmed mean square envelope negative entropy of the running process sound signal and the normal running state sound signal, and select the frequency band with the larger difference as the central frequency band; The center frequency determination module is used to determine the center frequency and search bandwidth; The calculation module is used to use the frequency band composed of the center frequency and the search bandwidth as the filtering frequency band, and perform filtering again and obtain the center frequency band to obtain the trimmed mean square envelope negative entropy and the trimmed mean square envelope spectrum negative entropy of the running process sound signal and the normal running state sound signal; The composite weighted index is calculated by using the trimmed mean square envelope negentropy and trimmed mean square envelope spectrum negentropy of the obtained operation process sound signal and normal operation state sound signal, and the filter frequency band corresponding to the maximum value in the composite weighted index is selected as the final frequency band; The fault characteristic frequency acquisition module filters the running process sound signal according to the final frequency band to obtain a filtered signal, calculates the envelope signal of the filtered signal, and obtains the transmission system fault characteristic frequency from the envelope signal.

8. The armored vehicle transmission system fault diagnosis system based on acoustic signals according to claim 7 is characterized in that: The fault characteristic frequency acquisition module includes an envelope spectrum drawing unit, which filters the operating process sound signal according to the final frequency band to obtain a filtered signal, calculates the envelope signal of the filtered signal, draws an envelope spectrum and marks the transmission system fault characteristic frequency, and identifies the fault frequency and harmonic frequency based on the envelope spectrum.

9. A computer device, characterized in that: It includes a processor and a memory, the memory is used to store a computer executable program, the processor reads part or all of the computer executable program from the memory and executes it, and when the processor executes part or all of the computer executable program, it can implement the armored vehicle transmission system fault diagnosis method based on acoustic signals as described in any one of claims 1-6.

10. A computer-readable storage medium, characterized in that: A computer program is stored in the computer-readable storage medium. When the computer program is executed by the processor, it can implement the armored vehicle transmission system fault diagnosis method based on acoustic signals as described in any one of claims 1-6.

Images