Method and device for diagnosing abnormal sound fault of valve mechanism of engine

Abstract

The invention provides a diagnosis method and a diagnosis device for abnormal sound faults of a valve mechanism of an engine. The method comprises the steps that acoustic signals of a target area in the engine test run process and classification results corresponding to the acoustic signals are obtained, the acoustic signals are signals transmitted by a valve mechanism through sound waves in the engine test run process, and the classification results include that the valve mechanism has an abnormal sound fault and does not have the abnormal sound fault; performing feature extraction processing on the acoustic signal to obtain a feature signal, and obtaining a signal-to-noise ratio of the feature signal and kurtosis of the feature signal; obtaining a signal-to-noise ratio and kurtosis of a to-be-diagnosed acoustic signal to obtain a to-be-diagnosed signal-to-noise ratio and a to-be-diagnosed kurtosis, and inputting the to-be-diagnosed signal-to-noise ratio and the to-be-diagnosed kurtosis into the classification model to obtain a classification result corresponding to the to-be-diagnosed acoustic signal; and according to a classification result, determining whether the valve mechanism has an abnormal sound fault or not. According to the method, the problems of low accuracy and low efficiency of the abnormal sound fault diagnosis mode of the valve mechanism of the engine in the prior art are solved.

Description

Technical Field

[0001] The present application relates to the field of engines, and in particular, to a method for diagnosing abnormal noise faults of a valve train of an engine, a device for diagnosing abnormal noise faults of a valve train of an engine, a computer-readable storage medium, and an electronic device. Background Art

[0002] After the engine is assembled, the test technician usually uses the human ear to judge whether there is a valve train abnormal noise fault caused by excessive valve clearance through bench testing in the test workshop. However, due to the noisy background noise in the test workshop working environment and the direct influence of the test technician's subjective experience, this method has a huge risk of missing faults, which in turn leads to the flow of engines with abnormal valve train noise faults into the market, affecting the use of engines.

[0003] Therefore, there is an urgent need for a method to solve the problem of low accuracy and efficiency in the prior art of diagnosing abnormal noise faults in the valve train of an engine. Summary of the invention

[0004] The main purpose of the present application is to provide a method for diagnosing abnormal noise faults of the engine's valve mechanism, a device for diagnosing abnormal noise faults of the engine's valve mechanism, a computer-readable storage medium and an electronic device, so as to at least solve the problem of low accuracy and efficiency of the method for diagnosing abnormal noise faults of the engine's valve mechanism in the prior art.

[0005] According to one aspect of the present application, a method for diagnosing abnormal noise faults of a valve train of an engine is provided, comprising: obtaining an acoustic signal of a target area during an engine test run and a classification result corresponding to the acoustic signal, wherein the target area is the area above the cylinder head of the engine, the acoustic signal is a signal transmitted by the valve train through sound waves during the engine test run, and the classification result includes abnormal noise faults occurring in the valve train and no abnormal noise faults occurring in the valve train; performing feature extraction processing on the acoustic signal to obtain a feature signal, and obtaining a signal-to-noise ratio of the feature signal and a kurtosis of the feature signal, wherein the signal-to-noise ratio of the feature signal is used to characterize the periodic characteristics of the acoustic signal. The kurtosis of the characteristic signal is used to characterize the impact characteristics of the acoustic signal; the signal-to-noise ratio and kurtosis of the acoustic signal to be diagnosed are obtained, the signal-to-noise ratio to be diagnosed and the kurtosis to be diagnosed are obtained, the signal-to-noise ratio to be diagnosed and the kurtosis to be diagnosed are input into the classification model, and the classification result corresponding to the acoustic signal to be diagnosed is obtained, wherein the classification model is trained using the signal-to-noise ratio of the characteristic signal, the kurtosis of the characteristic signal and the classification result corresponding to the acoustic signal, the signal-to-noise ratio to be diagnosed is the signal-to-noise ratio corresponding to the characteristic signal of the acoustic signal to be diagnosed, and the kurtosis to be diagnosed is the kurtosis corresponding to the characteristic signal of the acoustic signal to be diagnosed; according to the classification result, it is determined whether the valve mechanism has an abnormal noise fault.

[0006] Optionally, feature extraction processing is performed on the acoustic signal to obtain a feature signal, including: filtering the acoustic signal using a finite impulse response filter to obtain a first processed signal; and down-sampling the first processed signal to obtain the feature signal.

[0007] Optionally, the acoustic signal is subjected to feature extraction processing to obtain a feature signal, including: performing wavelet decomposition processing on the acoustic signal using a wavelet transform algorithm to obtain a second processed signal; and processing the second processed signal using a particle swarm optimization algorithm to obtain the feature signal.

[0008] Optionally, a particle swarm optimization algorithm is used to process the second processing signal to obtain the characteristic signal, including: a generation step of randomly generating a plurality of initial particles in a predetermined parameter space, wherein the initial particles have an initial velocity and an initial position, and the initial position of the initial particles is used to characterize the frequency center of the second processing signal and the bandwidth of the second processing signal; a calculation step of calculating the objective function of the second processing signal, wherein the objective function is used to characterize the periodic characteristics of the demodulated signal of the second processing signal; an updating step of updating the initial velocity of the initial particles to obtain an updated velocity, and updating the initial position of the initial particles to obtain an updated position; a repetition step of sequentially repeating the calculation step and the updating step at least once, and in the repetition process, updating the initial velocity in the updating step to the updated velocity in the updating step obtained in the last repetition process, and updating the initial position in the updating step to the updated position in the updating step obtained in the last repetition process, until a preset condition is met, wherein the preset condition is that the number of repetitions is greater than a first threshold or the function value of the objective function is less than a second threshold.

[0009] Optionally, obtaining the signal-to-noise ratio of the characteristic signal includes: obtaining an envelope spectrum signal of the characteristic signal; and obtaining the signal-to-noise ratio of the characteristic signal according to the formula Calculate the signal-to-noise ratio Snr of the characteristic signal x (f c ,B w ), where f p is the characteristic frequency corresponding to the abnormal noise failure of the valve train, X w is the envelope spectrum signal of the characteristic signal, f n is the characteristic frequency corresponding to the supercharger of the engine, floor(·) is a floor function, and i and j are both positive integers greater than or equal to 1 and less than or equal to 10.

[0010] Optionally, obtaining the envelope spectrum signal of the characteristic signal includes: performing Hilbert transform processing on the acoustic signal to obtain the envelope signal; and performing Fourier transform on the envelope signal to obtain the envelope spectrum signal.

[0011] Optionally, obtaining the kurtosis of the characteristic signal includes: according to the formula Calculate the kurtosis K of the characteristic signal x ,in, is the nth acoustic signal x(n) at the center frequency f c and bandwidth B w The corresponding characteristic signal is The characteristic signal The fourth moment of The characteristic signal The second moment of .

[0012] According to another aspect of the present application, a diagnostic device for abnormal noise fault of a valve mechanism of an engine is provided, comprising: an acquisition unit, used to acquire an acoustic signal of a target area during an engine test and a classification result corresponding to the acoustic signal, wherein the target area is the area above the cylinder head of the engine, the acoustic signal is a signal transmitted by the valve mechanism through sound waves during the engine test, and the classification result includes abnormal noise faults occurring in the valve mechanism and no abnormal noise faults occurring in the valve mechanism; a processing unit, used to perform feature extraction processing on the acoustic signal to obtain a feature signal, and acquire a signal-to-noise ratio of the feature signal and a kurtosis of the feature signal, wherein the signal-to-noise ratio of the feature signal is used to characterize the periodic characteristics of the acoustic signal. , the kurtosis of the characteristic signal is used to characterize the impact characteristics of the acoustic signal; an input unit is used to obtain the signal-to-noise ratio and kurtosis of the acoustic signal to be diagnosed, obtain the signal-to-noise ratio to be diagnosed and the kurtosis to be diagnosed, input the signal-to-noise ratio to be diagnosed and the kurtosis to be diagnosed into the classification model, and obtain the classification result corresponding to the acoustic signal to be diagnosed, wherein the classification model is trained using the signal-to-noise ratio of the characteristic signal, the kurtosis of the characteristic signal and the classification result corresponding to the acoustic signal, the signal-to-noise ratio to be diagnosed is the signal-to-noise ratio corresponding to the characteristic signal of the acoustic signal to be diagnosed, and the kurtosis to be diagnosed is the kurtosis corresponding to the characteristic signal of the acoustic signal to be diagnosed; a determination unit is used to determine whether the valve mechanism has an abnormal noise fault according to the classification result.

[0013] According to another aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute any one of the methods described.

[0014] According to another aspect of the present application, an electronic device is provided, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include methods for executing any one of the described methods.

[0015] The technical solution of the present application is applied to provide a method for diagnosing abnormal noise faults of the valve mechanism of an engine. First, the acoustic signal of the target area during the engine test and the classification result corresponding to the acoustic signal are obtained. The acoustic signal is the signal transmitted by the valve mechanism through sound waves during the engine test. The classification result includes abnormal noise faults of the valve mechanism and no abnormal noise faults of the valve mechanism; then, feature extraction processing is performed on the acoustic signal to obtain a feature signal, and the signal-to-noise ratio of the feature signal and the kurtosis of the feature signal are obtained; then the signal-to-noise ratio and kurtosis of the acoustic signal to be diagnosed are obtained to obtain the signal-to-noise ratio and kurtosis to be diagnosed, and the signal-to-noise ratio and kurtosis to be diagnosed are input into the classification model to obtain the classification result corresponding to the acoustic signal to be diagnosed; finally, according to the classification result, it is determined whether the valve mechanism has an abnormal noise fault. The acoustic signal is obtained by collecting the signal above the cylinder head during the operation of the engine; and the signal analysis method is used to extract the signal-to-noise ratio and kurtosis of the feature signal of the fault characteristics representing the periodic characteristics and the impact characteristics. Then, the classification model for abnormal noise fault diagnosis of the engine valve train is trained based on the signal-to-noise ratio and kurtosis of the characteristic signal to realize automatic detection of abnormal noise cylinder head of the engine. Compared with the human ear judgment method in the prior art, this method can effectively improve the accuracy and efficiency of detection, and solve the problem of low accuracy and efficiency of manual fault diagnosis in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings constituting part of the present application are used to provide a further understanding of the present application. The exemplary embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0017] Figure 1 A hardware structure block diagram of a mobile terminal for executing a method for diagnosing abnormal noise fault of a valve train of an engine provided in an embodiment of the present application is shown;

[0018] Figure 2 A schematic flow chart of a method for diagnosing abnormal noise fault of a valve train of an engine provided according to an embodiment of the present application is shown;

[0019] Figure 3 A hardware structure block diagram of a structure for acquiring an acoustic signal provided in an embodiment of the present application is shown;

[0020] Figure 4 A schematic diagram of a process of using a particle swarm optimization algorithm to process a second processing signal to obtain a characteristic signal according to an embodiment of the present application is shown;

[0021] Figure 5 A specific flow chart of a method for diagnosing abnormal noise fault of a valve train of an engine provided according to an embodiment of the present application is shown;

[0022] Figure 6 A structural block diagram of a device for diagnosing abnormal noise faults of a valve train of an engine provided according to an embodiment of the present application is shown.

[0023] The above drawings include the following reference numerals:

[0024] 102. Processor; 104. Memory; 106. Transmission device; 108. Input / output device; 110. Acoustic sensor; 112. First-level guide rail; 114. Second-level guide rail; 116. Column. DETAILED DESCRIPTION

[0025] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0026] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0027] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0028] As introduced in the background technology, the method of diagnosing abnormal noise fault of the engine's valve mechanism in the prior art has low accuracy and efficiency. To solve the above problems, the embodiments of the present application provide a method for diagnosing abnormal noise fault of the engine's valve mechanism, a device for diagnosing abnormal noise fault of the engine's valve mechanism, a computer-readable storage medium and an electronic device.

[0029] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0030] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 1 is a hardware structure block diagram of a mobile terminal for diagnosing abnormal noise faults of a valve train of an engine according to an embodiment of the present invention. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is for illustration only and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0031] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the diagnostic method of abnormal sound fault of the valve mechanism of the engine in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, the above method is realized. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories can be connected to the mobile terminal through a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network and a combination thereof. The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0032] In this embodiment, a method for diagnosing abnormal noise failure of the valve mechanism of an engine running on a mobile terminal, a computer terminal or a similar computing device is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0033] Figure 2 FIG. 1 is a flow chart of a method for diagnosing abnormal noise faults of a valve train of an engine according to an embodiment of the present application. Figure 2 As shown, the method comprises the following steps:

[0034] Step S201, obtaining an acoustic signal of a target area during an engine test and a classification result corresponding to the acoustic signal, wherein the target area is an area above a cylinder head of the engine, the acoustic signal is a signal transmitted by a valve mechanism through sound waves during the engine test, and the classification result includes an abnormal sound fault occurring in the valve mechanism and a non-abnormal sound fault occurring in the valve mechanism;

[0035] Specifically, the engine test process refers to the process of starting, running and testing the engine after it is assembled to ensure the performance, stability and reliability of the engine. This process is very important for newly manufactured engines, overhauled engines and regularly maintained engines. Through the test run, it can be ensured that the engine meets the design requirements and safety standards before use. The cylinder head of the engine, also known as the cylinder head, is an important component of the engine. It is mainly responsible for sealing the cylinder and forming a combustion chamber together with the top of the piston and the cylinder block. According to the mechanical structure and operating mechanism of the engine, when the valve clearance of the engine is too large, it will produce large valve seating vibrations and valve tappets hitting the valve rocker arms and other vibrations. According to the meshing relationship of the engine tooth clearance, the frequency of this type of vibration should be half of the engine rotation frequency, so it also has periodic characteristics. According to the acoustic-vibration coupling mechanism of the engine, the sound waves generated by this vibration coupling are more significant near the cylinder head. In the process of obtaining the acoustic signal of the target area during the engine test run, the following methods can be used: Figure 3 In the structure shown, two sliding rails, the first rail 112 and the second rail 114, are designed on the column 116 to realize non-contact collection of acoustic data. A windproof sponge ball can be installed on the head of the acoustic sensor 110 to reduce the influence of airflow disturbance in the test workshop.

[0036] Step S202, performing feature extraction processing on the acoustic signal to obtain a feature signal, and obtaining a signal-to-noise ratio of the feature signal and a kurtosis of the feature signal, wherein the signal-to-noise ratio of the feature signal is used to characterize the periodic feature of the acoustic signal, and the kurtosis of the feature signal is used to characterize the impact feature of the acoustic signal;

[0037] Specifically, the acoustic signal has both impact characteristics and periodic characteristics. When the valve clearance is too large and the valve mechanism makes abnormal noise, the impact characteristics and periodic characteristics are significantly enhanced. In the feature extraction process, Fourier transform, Mel frequency cepstral coefficients, cepstral coefficients, zero crossing rate, chromaticity characteristics, energy characteristics, entropy and complexity can be used; then principal component analysis, linear discriminant analysis and other methods can be used to reduce the dimension of the features to improve processing efficiency and performance.

[0038] Step S203, obtaining the signal-to-noise ratio and kurtosis of the acoustic signal to be diagnosed, obtaining the signal-to-noise ratio to be diagnosed and the kurtosis to be diagnosed, inputting the above-mentioned signal-to-noise ratio to be diagnosed and the above-mentioned kurtosis to be diagnosed into the classification model, and obtaining the classification result corresponding to the acoustic signal to be diagnosed, wherein the above-mentioned classification model is trained using the signal-to-noise ratio of the above-mentioned characteristic signal and the kurtosis of the above-mentioned characteristic signal and the classification result corresponding to the above-mentioned acoustic signal, the above-mentioned signal-to-noise ratio to be diagnosed is the signal-to-noise ratio corresponding to the characteristic signal of the above-mentioned acoustic signal to be diagnosed, and the above-mentioned kurtosis to be diagnosed is the kurtosis corresponding to the characteristic signal of the above-mentioned acoustic signal to be diagnosed;

[0039] Specifically, feature engineering is established based on the signal-to-noise ratio of the above-mentioned characteristic signal, the kurtosis of the above-mentioned characteristic signal, and the classification results corresponding to the above-mentioned acoustic signal, which can characterize the difference between normal acoustic signals and abnormal acoustic signals of the valve train. The above-mentioned classification model can adopt machine learning classification algorithms such as support vector machines, neural networks, Bayesian classifiers, decision trees, and random forests. By using the kurtosis and signal-to-noise ratio indicators to establish feature engineering, and further training to obtain the above-mentioned classification model, the impact characteristics and periodic characteristics in the acoustic signal can be directly reflected, making the feature engineering have significant interpretability.

[0040] Step S204: determining whether an abnormal noise fault occurs in the valve train according to the classification result.

[0041] In practical applications, the accuracy of the above classification results can be judged by manual re-examination. If the results are wrong, the above steps are repeated to update the classification model. In addition, the above method is not only applicable to the abnormal noise fault diagnosis of the valve mechanism of diesel engines, but also to the abnormal noise fault diagnosis of the valve mechanism of gasoline, natural gas and other internal combustion engines.

[0042] Through this embodiment, a method for diagnosing abnormal noise faults of a valve mechanism of an engine is provided. First, an acoustic signal of a target area during an engine test and a classification result corresponding to the acoustic signal are obtained. The acoustic signal is a signal transmitted by the valve mechanism through sound waves during the engine test. The classification result includes abnormal noise faults of the valve mechanism and no abnormal noise faults of the valve mechanism. Then, feature extraction processing is performed on the acoustic signal to obtain a feature signal, and the signal-to-noise ratio and kurtosis of the feature signal are obtained. Then, the signal-to-noise ratio and kurtosis of the acoustic signal to be diagnosed are obtained to obtain the signal-to-noise ratio and kurtosis to be diagnosed. The signal-to-noise ratio and kurtosis to be diagnosed are input into a classification model to obtain the classification result corresponding to the acoustic signal to be diagnosed. Finally, according to the classification result, it is determined whether the valve mechanism has an abnormal noise fault. The acoustic signal is obtained by collecting the signal above the cylinder head during the operation of the engine. The signal-to-noise ratio and kurtosis of the feature signal of the fault characteristics representing the periodic characteristics and the impact characteristics are extracted using a signal analysis method. Then, the classification model for abnormal noise fault diagnosis of the engine valve train is trained based on the signal-to-noise ratio and kurtosis of the characteristic signal to realize automatic detection of abnormal noise cylinder head of the engine. Compared with the human ear judgment method in the prior art, this method can effectively improve the accuracy and efficiency of detection, and solve the problem of low accuracy and efficiency of manual fault diagnosis in the prior art.

[0043] In the specific implementation process, the above step S202 can be implemented by the following steps: step S2021, using a finite impulse response filter to filter the above acoustic signal to obtain a first processed signal; step S2022, down-sampling the above first processed signal to obtain the above characteristic signal. This method can further quickly obtain the above characteristic signal.

[0044] Specifically, the steps of combining the finite impulse response (FIR) filter with downsampling to process the original acoustic signal are as follows: First, design a finite impulse response filter to meet specific filtering characteristics. Select the appropriate order according to the required filtering effect. The filter coefficients can be determined by methods such as the window function method or the least squares method. Apply the designed FIR filter to the original acoustic signal. The filtered signal obtained is a representation of the original signal within a specific frequency range. The sampling rate of the signal is reduced by downsampling to further reduce the bandwidth of the signal. The downsampling rate is determined according to the bandwidth of the required narrowband signal. Take a sample every N sampling points, where N is the downsampling rate.

[0045] In order to further quickly obtain the above-mentioned characteristic signal, the above-mentioned step S202 of the present application can also be implemented through the following steps: step S2023, using a wavelet transform algorithm to perform wavelet decomposition processing on the above-mentioned acoustic signal to obtain a second processed signal; step S2024, using a particle swarm optimization algorithm to process the above-mentioned second processed signal to obtain the above-mentioned characteristic signal.

[0046] Specifically, in addition to using a finite impulse response filter to filter the above acoustic signal or using a wavelet transform algorithm to perform wavelet decomposition on the above acoustic signal, algorithms such as empirical mode decomposition, variational mode decomposition, and infinite impulse response filter (Infinite Impulse Response Filter, abbreviated as IIR) can also be used.

[0047] like Figure 4 As shown, the above step S2024 can also be implemented in other ways, for example: step S20241, a generation step, randomly generating a plurality of initial particles in a predetermined parameter space, wherein the above initial particles have an initial velocity and an initial position, and the initial position of the above initial particles is used to characterize the frequency center of the above second processed signal and the bandwidth of the above second processed signal; step S20242, a calculation step, calculating the objective function of the above second processed signal, wherein the above objective function is used to characterize the periodic characteristics of the signal after the demodulation of the above second processed signal; step S20243, an update step, updating the initial velocity of the above initial particles Update to obtain the update speed, and update the initial position of the initial particle to obtain the update position; step S20244, repeat the step, repeat the calculation step and the update step at least once in sequence, and in the repetition process, update the initial speed in the update step to the update speed in the update step obtained in the last repetition process, and update the initial position in the update step to the update position in the update step obtained in the last repetition process until the preset condition is met, and the preset condition is that the number of repetitions is greater than the first threshold or the function value of the objective function is less than the second threshold. Constructing a wavelet filter group in the wavelet transform framework and combining it with a particle swarm optimization algorithm can further improve the accuracy of extracting the above-mentioned feature signal.

[0048] Specifically, the above steps of combining the wavelet filter bank with particle swarm optimization to process the original acoustic signal are as follows: First, define the objective function as "Signal-to-noise ratio index based on envelope spectrum". This index is used to measure the periodic characteristics of the envelope signal obtained after demodulation of the narrowband signal after bandpass filtering. Set the optimization variables. Select the frequency center fc and bandwidth Bw as optimization variables, and encode these coefficients as the position of the particle. Each particle represents a set of possible filter coefficients. Initialize the particle swarm. Randomly generate a set of initial particles in a given parameter space. The position of each particle corresponds to a set of parameters. Set the initial speed for each particle. The speed determines the direction and speed of the particle's search in the parameter space. Iterate optimization. For each particle, use the filter coefficients it generates to perform wavelet decomposition on the signal, and calculate the objective function of the decomposed signal. Update particles: According to the evaluation result of the objective function, update the speed and position of each particle. This process follows the rules of the particle swarm optimization algorithm, and the particles will be adjusted according to the individual best position and the global best position. Repeat the evaluation and update process until the termination condition is met (such as reaching the maximum number of iterations or the objective function improvement is less than a certain threshold).

[0049] In some embodiments, the above step S202 can also be implemented by the following steps: step S2025, obtaining the envelope spectrum signal of the above characteristic signal; step S2026, according to the formula Calculate the signal-to-noise ratio Snr of the above characteristic signal x (f c ,B w ), where f p is the characteristic frequency corresponding to the abnormal noise failure of the above valve train, X w is the envelope spectrum signal of the above characteristic signal, f n is the characteristic frequency corresponding to the supercharger of the above engine, floor(·) is a floor rounding function, and i and j are both positive integers greater than or equal to 1 and less than or equal to 10. This method can further obtain an accurate signal-to-noise ratio of the characteristic signal.

[0050] Specifically, the signal-to-noise ratio is a measure of the ratio of useful signals to noise signals in a signal. In acoustic signal processing, a high signal-to-noise ratio means better signal quality and less noise. The specific steps for calculating the signal-to-noise ratio of the characteristic signal are as follows: first, sum the squares of the signal sequence and then divide it by the signal length to calculate the signal power; then sum the squares of the noise signal and then divide it by the length of the noise signal to calculate the noise power, and finally, calculate the signal-to-noise ratio. It should be noted that the calculation of the signal-to-noise ratio depends on the power of the signal and the noise, not the amplitude of the signal and the noise. In practical applications, the noise power can be estimated by filters, autocorrelation functions, and other methods.

[0051] The above step S2025 can be implemented by the following steps: step S20251, performing Hilbert transform processing on the above acoustic signal to obtain an envelope signal; step S20252, performing Fourier transform on the above envelope signal to obtain the above envelope spectrum signal. This method can further quickly obtain the above envelope spectrum signal, thereby revealing the time-frequency characteristics of the signal.

[0052] In practical applications, before performing Hilbert transform (also known as Hilbert transform), the DC component in the signal can be removed, and then filtering is performed to remove noise or highlight the frequency band of interest. The preprocessed signal is subjected to Hilbert transform, where Hilbert transform is a linear operation that can convert a real-valued signal into an analytical signal. The analytical signal is a complex signal, the real part of which is the original signal, and the imaginary part is the Hilbert transform of the signal. The real part and the imaginary part are extracted from the analytical signal, and then their absolute values ​​are calculated to obtain the amplitude envelope of the signal. The real part and the imaginary part of the analytical signal are used to calculate the instantaneous frequency. The instantaneous frequency is the frequency of the signal at a certain moment. The amplitude envelope is Fourier transformed to obtain the above-mentioned envelope spectrum signal. The above-mentioned envelope spectrum signal can identify the change of the frequency component in the signal over time, and can also be used to identify the periodic components in the signal, the change of the instantaneous frequency and other characteristics. The above-mentioned Hilbert envelope spectrum is a frequency domain method for analyzing nonlinear and non-stationary signals. It obtains the instantaneous amplitude and instantaneous frequency of the signal through Hilbert transform, thereby revealing the time-frequency characteristics of the signal.

[0053] In some other embodiments, the above step S202 can also be implemented by the following steps: Step S2027, according to the formula Calculate the kurtosis K of the above characteristic signal x ,in, is the nth acoustic signal x(n) at the center frequency f c and bandwidth B w The corresponding characteristic signal is The characteristic signal The fourth moment of The characteristic signal This method can further obtain the kurtosis of the accurate characteristic signal.

[0054] Specifically, kurtosis is a statistic that measures the sharpness of a signal's peak value, which describes the "peak" and "heavy tail" characteristics of the signal distribution. In acoustic signal processing, kurtosis can be used to distinguish different types of sounds. The kurtosis of the above characteristic signal can also be calculated using the following steps: first, calculate the mean and standard deviation of the signal sequence; then calculate the difference between each sample and the mean; calculate the fourth power of the difference, calculate the expected value of the fourth power, and finally, calculate the kurtosis value.

[0055] In order to enable those skilled in the art to more clearly understand the technical solution of the present application, the implementation process of the method for diagnosing abnormal noise fault of the valve mechanism of the engine of the present application will be described in detail below in combination with specific embodiments.

[0056] This embodiment relates to a specific method for diagnosing abnormal noise faults of a valve train of an engine, such as Figure 5 As shown, the following steps are included:

[0057] Training process:

[0058] Step S11: Acquiring acoustic signals of a target area during engine test;

[0059] Step S12: Processing the acoustic signal by combining wavelet decomposition and particle swarm optimization algorithm to obtain a characteristic signal;

[0060] Step S13: obtaining the kurtosis of the characteristic signal, and obtaining the signal-to-noise ratio of the characteristic signal based on the envelope spectrum;

[0061] Step S14: using a classification algorithm to process the kurtosis of the characteristic signal and the signal-to-noise ratio of the characteristic signal to establish a classification model.

[0062] Model iterative update process:

[0063] Step S21: judging whether the diagnosis result of the classification model is accurate through manual re-examination. If the result is wrong, iteratively updating the model by using the feature engineering of the signal;

[0064] Step S22: acquiring new acoustic signals of the target area during the engine test;

[0065] Step S23: Processing the acoustic signal by combining wavelet decomposition and particle swarm optimization algorithm to obtain a new characteristic signal;

[0066] Step S24: obtaining the kurtosis of the new characteristic signal, and obtaining the signal-to-noise ratio of the new characteristic signal based on the envelope spectrum;

[0067] Step S25: using a classification algorithm to process the kurtosis of the new feature signal and the signal-to-noise ratio of the new feature signal, and iteratively updating the classification model.

[0068] Reasoning process:

[0069] Step S31: Acquiring an acoustic signal to be diagnosed in a target area during an engine test;

[0070] Step S32: Processing the acoustic signal by combining wavelet decomposition and particle swarm optimization algorithm to obtain a characteristic signal to be diagnosed;

[0071] Step S33: obtaining the kurtosis of the characteristic signal to be diagnosed, and obtaining the signal-to-noise ratio of the characteristic signal to be diagnosed based on the envelope spectrum;

[0072] Step S34: input the kurtosis of the characteristic signal to be diagnosed and the signal-to-noise ratio of the characteristic signal to be diagnosed into the above classification model to perform fault diagnosis, and output the diagnosis result if it is the same as the result of manual re-inspection.

[0073] The embodiment of the present application also provides a diagnostic device for abnormal noise failure of the valve mechanism of an engine. It should be noted that the diagnostic device for abnormal noise failure of the valve mechanism of an engine in the embodiment of the present application can be used to execute the diagnostic method for abnormal noise failure of the valve mechanism of an engine provided in the embodiment of the present application. The device is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation in hardware, or a combination of software and hardware, is also possible and conceivable.

[0074] The following is an introduction to the diagnostic device for abnormal noise fault of the engine's valve train provided in an embodiment of the present application.

[0075] Figure 6 Schematic diagram of a diagnostic device for abnormal noise fault of a valve train of an engine according to an embodiment of the present application. Figure 6 As shown, the device comprises:

[0076] An acquisition unit 10 is used to acquire an acoustic signal of a target area during an engine test and a classification result corresponding to the acoustic signal, wherein the target area is an area above a cylinder head of the engine, the acoustic signal is a signal transmitted by a valve mechanism through sound waves during the engine test, and the classification result includes an abnormal sound fault occurring in the valve mechanism and a non-abnormal sound fault occurring in the valve mechanism;

[0077] Specifically, the engine test process refers to the process of starting, running and testing the engine after it is assembled to ensure the performance, stability and reliability of the engine. This process is very important for newly manufactured engines, overhauled engines and regularly maintained engines. Through the test run, it can be ensured that the engine meets the design requirements and safety standards before use. The cylinder head of the engine, also known as the cylinder head, is an important component of the engine. It is mainly responsible for sealing the cylinder and forming a combustion chamber together with the top of the piston and the cylinder block. According to the mechanical structure and operating mechanism of the engine, when the valve clearance of the engine is too large, it will produce large valve seating vibrations and valve tappets hitting the valve rocker arms and other vibrations. According to the meshing relationship of the engine tooth clearance, the frequency of this type of vibration should be half of the engine rotation frequency, so it also has periodic characteristics. According to the acoustic-vibration coupling mechanism of the engine, the sound waves generated by this vibration coupling are more significant near the cylinder head. In the process of obtaining the acoustic signal of the target area during the engine test run, the following methods can be used: Figure 3 In the structure shown, two sliding rails, the first rail 112 and the second rail 114, are designed on the column 116 to realize non-contact collection of acoustic data. A windproof sponge ball can be installed on the head of the acoustic sensor 110 to reduce the influence of airflow disturbance in the test workshop.

[0078] The processing unit 20 is used to perform feature extraction processing on the acoustic signal to obtain a feature signal, and obtain a signal-to-noise ratio of the feature signal and a kurtosis of the feature signal, wherein the signal-to-noise ratio of the feature signal is used to characterize the periodic feature of the acoustic signal, and the kurtosis of the feature signal is used to characterize the impact feature of the acoustic signal;

[0079] Specifically, the acoustic signal has both impact characteristics and periodic characteristics. When the valve clearance is too large and the valve mechanism makes abnormal noise, the impact characteristics and periodic characteristics are significantly enhanced. In the feature extraction process, Fourier transform, Mel frequency cepstral coefficients, cepstral coefficients, zero crossing rate, chromaticity characteristics, energy characteristics, entropy and complexity can be used; then principal component analysis, linear discriminant analysis and other methods can be used to reduce the dimension of the features to improve processing efficiency and performance.

[0080] An input unit 30 is used to obtain the signal-to-noise ratio and kurtosis of the acoustic signal to be diagnosed, obtain the signal-to-noise ratio to be diagnosed and the kurtosis to be diagnosed, input the above-mentioned signal-to-noise ratio to be diagnosed and the above-mentioned kurtosis to be diagnosed into a classification model, and obtain a classification result corresponding to the acoustic signal to be diagnosed, wherein the above-mentioned classification model is trained using the signal-to-noise ratio of the above-mentioned characteristic signal and the kurtosis of the above-mentioned characteristic signal and the classification result corresponding to the above-mentioned acoustic signal, the above-mentioned signal-to-noise ratio to be diagnosed is the signal-to-noise ratio corresponding to the characteristic signal of the above-mentioned acoustic signal to be diagnosed, and the above-mentioned kurtosis to be diagnosed is the kurtosis corresponding to the characteristic signal of the above-mentioned acoustic signal to be diagnosed;

[0081] Specifically, feature engineering is established based on the signal-to-noise ratio of the above-mentioned characteristic signal, the kurtosis of the above-mentioned characteristic signal, and the classification results corresponding to the above-mentioned acoustic signal, which can characterize the difference between normal acoustic signals and abnormal acoustic signals of the valve train. The above-mentioned classification model can adopt machine learning classification algorithms such as support vector machines, neural networks, Bayesian classifiers, decision trees, and random forests. By using the kurtosis and signal-to-noise ratio indicators to establish feature engineering, and further training to obtain the above-mentioned classification model, the impact characteristics and periodic characteristics in the acoustic signal can be directly reflected, making the feature engineering have significant interpretability.

[0082] The determination unit 40 is used to determine whether the valve train has an abnormal noise fault according to the classification result.

[0083] In practical applications, the accuracy of the above classification results can be judged by manual re-examination. If the results are wrong, the above steps are repeated to update the classification model. In addition, the above method is not only applicable to the abnormal noise fault diagnosis of the valve mechanism of diesel engines, but also to the abnormal noise fault diagnosis of the valve mechanism of gasoline, natural gas and other internal combustion engines.

[0084] Through this embodiment, a diagnostic device for abnormal noise fault of the valve mechanism of an engine is provided, wherein the acquisition unit acquires the acoustic signal of the target area during the engine test and the classification result corresponding to the acoustic signal, the acoustic signal is the signal transmitted by the valve mechanism through sound waves during the engine test, and the classification result includes abnormal noise fault of the valve mechanism and no abnormal noise fault of the valve mechanism; the processing unit performs feature extraction processing on the acoustic signal to obtain a feature signal, and obtains the signal-to-noise ratio and kurtosis of the feature signal; the input unit acquires the signal-to-noise ratio and kurtosis of the acoustic signal to be diagnosed, obtains the signal-to-noise ratio and kurtosis to be diagnosed, inputs the signal-to-noise ratio and kurtosis to be diagnosed into the classification model, and obtains the classification result corresponding to the acoustic signal to be diagnosed; the determination unit determines whether the valve mechanism has an abnormal noise fault according to the classification result. The acoustic signal is obtained by collecting the signal above the cylinder head during the operation of the engine; and the signal analysis method is used to extract the signal-to-noise ratio and kurtosis of the feature signal of the fault characteristics representing the periodic characteristics and the impact characteristics. Then, the classification model for abnormal noise fault diagnosis of the engine valve train is trained based on the signal-to-noise ratio and kurtosis of the characteristic signal to realize automatic detection of abnormal noise cylinder head of the engine. Compared with the human ear judgment method in the prior art, this method can effectively improve the accuracy and efficiency of detection, and solve the problem of low accuracy and efficiency of manual fault diagnosis in the prior art.

[0085] In a specific implementation process, the processing unit includes a first processing module and a second processing module, wherein the first processing module is used to filter the acoustic signal using a finite impulse response filter to obtain a first processed signal; the second processing module is used to downsample the first processed signal to obtain the characteristic signal. The device can further quickly obtain the characteristic signal.

[0086] Specifically, the steps of combining the finite impulse response (FIR) filter with downsampling to process the original acoustic signal are as follows: First, design a finite impulse response filter to meet specific filtering characteristics. Select the appropriate order according to the required filtering effect. The filter coefficients can be determined by methods such as the window function method or the least squares method. Apply the designed FIR filter to the original acoustic signal. The filtered signal obtained is a representation of the original signal within a specific frequency range. The sampling rate of the signal is reduced by downsampling to further reduce the bandwidth of the signal. The downsampling rate is determined according to the bandwidth of the required narrowband signal. Take a sample every N sampling points, where N is the downsampling rate.

[0087] In order to further quickly obtain the above-mentioned characteristic signal, the above-mentioned processing unit of the present application also includes a third processing module and a fourth processing module, wherein the third processing module is used to use a wavelet transform algorithm to perform wavelet decomposition processing on the above-mentioned acoustic signal to obtain a second processed signal; the fourth processing module is used to use a particle swarm optimization algorithm to process the above-mentioned second processed signal to obtain the above-mentioned characteristic signal.

[0088] Specifically, in addition to using a finite impulse response filter to filter the above acoustic signal or using a wavelet transform algorithm to perform wavelet decomposition on the above acoustic signal, algorithms such as empirical mode decomposition, variational mode decomposition, and infinite impulse response filter (Infinite Impulse Response Filter, abbreviated as IIR) can also be used.

[0089] The fourth processing module includes a generating submodule, a calculating submodule, an updating submodule and a repeating submodule, wherein the generating submodule is used for a generating step, randomly generating a plurality of initial particles in a predetermined parameter space, wherein the initial particles have an initial velocity and an initial position, and the initial position of the initial particles is used to characterize the frequency center of the second processed signal and the bandwidth of the second processed signal; the calculating submodule is used for a calculating step, calculating the objective function of the second processed signal, wherein the objective function is used to characterize the periodic characteristics of the demodulated signal of the second processed signal; the updating submodule is used for an updating step, updating the initial velocity of the initial particles to obtain an updated velocity, and updating the initial position of the initial particles to obtain an updated position; the repeating submodule is used for a repeating step, repeating the calculating step and the updating step at least once in sequence, and in the repeating process, updating the initial velocity in the updating step to the updated velocity in the updating step obtained in the last repeating process, and updating the initial position in the updating step to the updated position in the updating step obtained in the last repeating process, until a preset condition is met, wherein the preset condition is that the number of repetitions is greater than a first threshold or the function value of the objective function is less than a second threshold. Constructing a wavelet filter bank under the wavelet transform framework and combining it with the particle swarm optimization algorithm can further improve the accuracy of extracting the above-mentioned characteristic signals.

[0090] Specifically, the above steps of combining the wavelet filter bank with particle swarm optimization to process the original acoustic signal are as follows: First, define the objective function as "Signal-to-noise ratio index based on envelope spectrum". This index is used to measure the periodic characteristics of the envelope signal obtained after demodulation of the narrowband signal after bandpass filtering. Set the optimization variables. Select the frequency center fc and bandwidth Bw as optimization variables, and encode these coefficients as the position of the particle. Each particle represents a set of possible filter coefficients. Initialize the particle swarm. Randomly generate a set of initial particles in a given parameter space. The position of each particle corresponds to a set of parameters. Set the initial speed for each particle. The speed determines the direction and speed of the particle's search in the parameter space. Iterate optimization. For each particle, use the filter coefficients it generates to perform wavelet decomposition on the signal, and calculate the objective function of the decomposed signal. Update particles: According to the evaluation result of the objective function, update the speed and position of each particle. This process follows the rules of the particle swarm optimization algorithm, and the particles will be adjusted according to the individual best position and the global best position. Repeat the evaluation and update process until the termination condition is met (such as reaching the maximum number of iterations or the objective function improvement is less than a certain threshold).

[0091] In some embodiments, the processing unit includes an acquisition module and a first calculation module, wherein the acquisition module is used to acquire the envelope spectrum signal of the characteristic signal; the first calculation module is used to calculate the envelope spectrum signal according to the formula Calculate the signal-to-noise ratio Snr of the above characteristic signal x (f c ,B w ), where f p is the characteristic frequency corresponding to the abnormal noise failure of the above valve train, X w is the envelope spectrum signal of the above characteristic signal, f n is the characteristic frequency corresponding to the supercharger of the above engine, floor(·) is a floor rounding function, and i and j are both positive integers greater than or equal to 1 and less than or equal to 10. The device can further obtain a precise signal-to-noise ratio of the characteristic signal.

[0092] Specifically, the signal-to-noise ratio is a measure of the ratio of useful signals to noise signals in a signal. In acoustic signal processing, a high signal-to-noise ratio means better signal quality and less noise. The specific steps for calculating the signal-to-noise ratio of the characteristic signal are as follows: first, sum the squares of the signal sequence and then divide it by the signal length to calculate the signal power; then sum the squares of the noise signal and then divide it by the length of the noise signal to calculate the noise power, and finally, calculate the signal-to-noise ratio. It should be noted that the calculation of the signal-to-noise ratio depends on the power of the signal and the noise, not the amplitude of the signal and the noise. In practical applications, the noise power can be estimated by filters, autocorrelation functions, and other methods.

[0093] The acquisition module includes a first processing submodule and a second processing submodule, wherein the first processing submodule is used to perform Hilbert transform processing on the acoustic signal to obtain an envelope signal; the second processing submodule is used to perform Fourier transform on the envelope signal to obtain the envelope spectrum signal. The device can further quickly obtain the envelope spectrum signal, thereby revealing the time-frequency characteristics of the signal.

[0094] In practical applications, before performing Hilbert transform (also known as Hilbert transform), the DC component in the signal can be removed, and then filtering is performed to remove noise or highlight the frequency band of interest. The preprocessed signal is subjected to Hilbert transform, where Hilbert transform is a linear operation that can convert a real-valued signal into an analytical signal. The analytical signal is a complex signal, the real part of which is the original signal, and the imaginary part is the Hilbert transform of the signal. The real part and the imaginary part are extracted from the analytical signal, and then their absolute values ​​are calculated to obtain the amplitude envelope of the signal. The real part and the imaginary part of the analytical signal are used to calculate the instantaneous frequency. The instantaneous frequency is the frequency of the signal at a certain moment. The amplitude envelope is Fourier transformed to obtain the above-mentioned envelope spectrum signal. The above-mentioned envelope spectrum signal can identify the change of the frequency component in the signal over time, and can also be used to identify the periodic components in the signal, the change of the instantaneous frequency and other characteristics. The above-mentioned Hilbert envelope spectrum is a frequency domain method for analyzing nonlinear and non-stationary signals. It obtains the instantaneous amplitude and instantaneous frequency of the signal through Hilbert transform, thereby revealing the time-frequency characteristics of the signal.

[0095] In some other embodiments, the processing unit includes a calculation unit for calculating the Calculate the kurtosis K of the above characteristic signal x ,in, is the nth acoustic signal x(n) at the center frequency f c and bandwidth B w The corresponding characteristic signal is The characteristic signal The fourth moment of The characteristic signal The device can further obtain the kurtosis of the accurate characteristic signal.

[0096] Specifically, kurtosis is a statistic that measures the sharpness of a signal's peak value, which describes the "peak" and "heavy tail" characteristics of the signal distribution. In acoustic signal processing, kurtosis can be used to distinguish different types of sounds. The kurtosis of the above characteristic signal can also be calculated using the following steps: first, calculate the mean and standard deviation of the signal sequence; then calculate the difference between each sample and the mean; calculate the fourth power of the difference, calculate the expected value of the fourth power, and finally, calculate the kurtosis value.

[0097] The above-mentioned diagnostic device for abnormal noise fault of the valve train of the engine includes a processor and a memory. The above-mentioned acquisition unit, processing unit, input unit and determination unit are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize the corresponding functions. The above-mentioned modules are all located in the same processor; or, the above-mentioned modules are located in different processors in the form of any combination.

[0098] The processor includes a kernel, which retrieves the corresponding program unit from the memory. One or more kernels can be set, and the abnormal noise fault of the valve train of the engine can be diagnosed by adjusting the kernel parameters.

[0099] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0100] An embodiment of the present invention provides a computer-readable storage medium, which includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute a method for diagnosing abnormal noise faults of the valve mechanism of the engine.

[0101] Specifically, the method for diagnosing abnormal noise fault of the valve train of the engine includes:

[0102] Step S201, obtaining an acoustic signal of a target area during an engine test and a classification result corresponding to the acoustic signal, wherein the target area is an area above a cylinder head of the engine, the acoustic signal is a signal transmitted by a valve mechanism through sound waves during the engine test, and the classification result includes an abnormal sound fault occurring in the valve mechanism and a non-abnormal sound fault occurring in the valve mechanism;

[0103] Step S202, performing feature extraction processing on the acoustic signal to obtain a feature signal, and obtaining a signal-to-noise ratio of the feature signal and a kurtosis of the feature signal, wherein the signal-to-noise ratio of the feature signal is used to characterize the periodic feature of the acoustic signal, and the kurtosis of the feature signal is used to characterize the impact feature of the acoustic signal;

[0104] Step S203, obtaining the signal-to-noise ratio and kurtosis of the acoustic signal to be diagnosed, obtaining the signal-to-noise ratio to be diagnosed and the kurtosis to be diagnosed, inputting the above-mentioned signal-to-noise ratio to be diagnosed and the above-mentioned kurtosis to be diagnosed into the classification model, and obtaining the classification result corresponding to the acoustic signal to be diagnosed, wherein the above-mentioned classification model is trained using the signal-to-noise ratio of the above-mentioned characteristic signal and the kurtosis of the above-mentioned characteristic signal and the classification result corresponding to the above-mentioned acoustic signal, the above-mentioned signal-to-noise ratio to be diagnosed is the signal-to-noise ratio corresponding to the characteristic signal of the above-mentioned acoustic signal to be diagnosed, and the above-mentioned kurtosis to be diagnosed is the kurtosis corresponding to the characteristic signal of the above-mentioned acoustic signal to be diagnosed;

[0105] Step S204: determining whether an abnormal noise fault occurs in the valve train according to the classification result.

[0106] An embodiment of the present invention provides a processor, which is used to run a program, wherein the program executes a method for diagnosing abnormal noise faults of the valve train of the engine when the program is running.

[0107] Specifically, the method for diagnosing abnormal noise fault of the valve train of the engine includes:

[0108] Step S201, obtaining an acoustic signal of a target area during an engine test and a classification result corresponding to the acoustic signal, wherein the target area is an area above a cylinder head of the engine, the acoustic signal is a signal transmitted by a valve mechanism through sound waves during the engine test, and the classification result includes an abnormal sound fault occurring in the valve mechanism and a non-abnormal sound fault occurring in the valve mechanism;

[0109] Step S202, performing feature extraction processing on the acoustic signal to obtain a feature signal, and obtaining a signal-to-noise ratio of the feature signal and a kurtosis of the feature signal, wherein the signal-to-noise ratio of the feature signal is used to characterize the periodic feature of the acoustic signal, and the kurtosis of the feature signal is used to characterize the impact feature of the acoustic signal;

[0110] Step S203, obtaining the signal-to-noise ratio and kurtosis of the acoustic signal to be diagnosed, obtaining the signal-to-noise ratio to be diagnosed and the kurtosis to be diagnosed, inputting the above-mentioned signal-to-noise ratio to be diagnosed and the above-mentioned kurtosis to be diagnosed into the classification model, and obtaining the classification result corresponding to the acoustic signal to be diagnosed, wherein the above-mentioned classification model is trained using the signal-to-noise ratio of the above-mentioned characteristic signal and the kurtosis of the above-mentioned characteristic signal and the classification result corresponding to the above-mentioned acoustic signal, the above-mentioned signal-to-noise ratio to be diagnosed is the signal-to-noise ratio corresponding to the characteristic signal of the above-mentioned acoustic signal to be diagnosed, and the above-mentioned kurtosis to be diagnosed is the kurtosis corresponding to the characteristic signal of the above-mentioned acoustic signal to be diagnosed;

[0111] Step S204: determining whether an abnormal noise fault occurs in the valve train according to the classification result.

[0112] An embodiment of the present invention provides a device, the device comprising a processor, a memory, and a program stored in the memory and executable on the processor, and when the processor executes the program, at least the following steps are implemented:

[0113] Step S201, obtaining an acoustic signal of a target area during an engine test and a classification result corresponding to the acoustic signal, wherein the target area is an area above a cylinder head of the engine, the acoustic signal is a signal transmitted by a valve mechanism through sound waves during the engine test, and the classification result includes an abnormal sound fault occurring in the valve mechanism and a non-abnormal sound fault occurring in the valve mechanism;

[0114] Step S202, performing feature extraction processing on the acoustic signal to obtain a feature signal, and obtaining a signal-to-noise ratio of the feature signal and a kurtosis of the feature signal, wherein the signal-to-noise ratio of the feature signal is used to characterize the periodic feature of the acoustic signal, and the kurtosis of the feature signal is used to characterize the impact feature of the acoustic signal;

[0115] Step S203, obtaining the signal-to-noise ratio and kurtosis of the acoustic signal to be diagnosed, obtaining the signal-to-noise ratio to be diagnosed and the kurtosis to be diagnosed, inputting the above-mentioned signal-to-noise ratio to be diagnosed and the above-mentioned kurtosis to be diagnosed into the classification model, and obtaining the classification result corresponding to the acoustic signal to be diagnosed, wherein the above-mentioned classification model is trained using the signal-to-noise ratio of the above-mentioned characteristic signal and the kurtosis of the above-mentioned characteristic signal and the classification result corresponding to the above-mentioned acoustic signal, the above-mentioned signal-to-noise ratio to be diagnosed is the signal-to-noise ratio corresponding to the characteristic signal of the above-mentioned acoustic signal to be diagnosed, and the above-mentioned kurtosis to be diagnosed is the kurtosis corresponding to the characteristic signal of the above-mentioned acoustic signal to be diagnosed;

[0116] Step S204: determining whether an abnormal noise fault occurs in the valve train according to the classification result.

[0117] The devices in this article can be servers, PCs, PADs, mobile phones, etc.

[0118] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program for initializing at least the following method steps:

[0119] Step S201, obtaining an acoustic signal of a target area during an engine test and a classification result corresponding to the acoustic signal, wherein the target area is an area above a cylinder head of the engine, the acoustic signal is a signal transmitted by a valve mechanism through sound waves during the engine test, and the classification result includes an abnormal sound fault occurring in the valve mechanism and a non-abnormal sound fault occurring in the valve mechanism;

[0120] Step S202, performing feature extraction processing on the acoustic signal to obtain a feature signal, and obtaining a signal-to-noise ratio of the feature signal and a kurtosis of the feature signal, wherein the signal-to-noise ratio of the feature signal is used to characterize the periodic feature of the acoustic signal, and the kurtosis of the feature signal is used to characterize the impact feature of the acoustic signal;

[0121] Step S203, obtaining the signal-to-noise ratio and kurtosis of the acoustic signal to be diagnosed, obtaining the signal-to-noise ratio to be diagnosed and the kurtosis to be diagnosed, inputting the above-mentioned signal-to-noise ratio to be diagnosed and the above-mentioned kurtosis to be diagnosed into the classification model, and obtaining the classification result corresponding to the acoustic signal to be diagnosed, wherein the above-mentioned classification model is trained using the signal-to-noise ratio of the above-mentioned characteristic signal and the kurtosis of the above-mentioned characteristic signal and the classification result corresponding to the above-mentioned acoustic signal, the above-mentioned signal-to-noise ratio to be diagnosed is the signal-to-noise ratio corresponding to the characteristic signal of the above-mentioned acoustic signal to be diagnosed, and the above-mentioned kurtosis to be diagnosed is the kurtosis corresponding to the characteristic signal of the above-mentioned acoustic signal to be diagnosed;

[0122] Step S204: determining whether an abnormal noise fault occurs in the valve train according to the classification result.

[0123] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0124] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0125] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0126] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0127] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0128] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0129] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0130] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0131] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0132] From the above description, it can be seen that the above embodiments of the present application achieve the following technical effects:

[0133] 1) The diagnostic method of abnormal noise fault of the valve mechanism of the engine of the present application, first, obtain the acoustic signal of the target area during the engine test and the classification result corresponding to the acoustic signal, the acoustic signal is the signal transmitted by the valve mechanism through sound waves during the engine test, and the classification result includes the abnormal noise fault of the valve mechanism and the abnormal noise fault of the valve mechanism; then, perform feature extraction processing on the acoustic signal to obtain the feature signal, and obtain the signal-to-noise ratio and kurtosis of the feature signal; then obtain the signal-to-noise ratio and kurtosis of the acoustic signal to be diagnosed, obtain the signal-to-noise ratio and kurtosis to be diagnosed, input the signal-to-noise ratio and kurtosis to be diagnosed into the classification model, and obtain the classification result corresponding to the acoustic signal to be diagnosed; finally, determine whether the valve mechanism has an abnormal noise fault according to the classification result. The acoustic signal is obtained by collecting the signal above the cylinder head during the operation of the engine; and the signal analysis method is used to extract the signal-to-noise ratio and kurtosis of the feature signal of the fault characteristics representing the periodic characteristics and the impact characteristics. Then, the classification model for abnormal noise fault diagnosis of the engine valve train is trained based on the signal-to-noise ratio and kurtosis of the characteristic signal to realize automatic detection of abnormal noise cylinder head of the engine. Compared with the human ear judgment method in the prior art, this method can effectively improve the accuracy and efficiency of detection, and solve the problem of low accuracy and efficiency of manual fault diagnosis in the prior art.

[0134] 2) The diagnostic device for abnormal noise fault of the valve mechanism of the engine of the present application, the acquisition unit acquires the acoustic signal of the target area during the engine test and the classification result corresponding to the acoustic signal, the acoustic signal is the signal transmitted by the valve mechanism through sound waves during the engine test, and the classification result includes abnormal noise fault of the valve structure and no abnormal noise fault of the valve mechanism; the processing unit performs feature extraction processing on the acoustic signal to obtain a feature signal, and obtains the signal-to-noise ratio of the feature signal and the kurtosis of the feature signal; the input unit acquires the signal-to-noise ratio and kurtosis of the acoustic signal to be diagnosed, obtains the signal-to-noise ratio and kurtosis to be diagnosed, inputs the signal-to-noise ratio and kurtosis to be diagnosed into the classification model, and obtains the classification result corresponding to the acoustic signal to be diagnosed; the determination unit determines whether the valve mechanism has an abnormal noise fault according to the classification result. The acoustic signal is obtained by collecting the signal above the cylinder head during the operation of the engine; and the signal analysis method is used to extract the signal-to-noise ratio and kurtosis of the feature signal of the fault characteristics representing the periodic characteristics and the impact characteristics. Then, the classification model for abnormal noise fault diagnosis of the engine valve train is trained based on the signal-to-noise ratio and kurtosis of the characteristic signal to realize automatic detection of abnormal noise cylinder head of the engine. Compared with the human ear judgment method in the prior art, this method can effectively improve the accuracy and efficiency of detection, and solve the problem of low accuracy and efficiency of manual fault diagnosis in the prior art.

[0135] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for diagnosing abnormal noise fault of a valve train of an engine, characterized in that: include: Acquiring an acoustic signal of a target area during an engine test and a classification result corresponding to the acoustic signal, wherein the target area is an area above a cylinder head of the engine, the acoustic signal is a signal transmitted by a valve mechanism through sound waves during the engine test, and the classification result includes an abnormal noise fault occurring in the valve mechanism and a non-abnormal noise fault occurring in the valve mechanism; Performing feature extraction processing on the acoustic signal to obtain a feature signal, and acquiring a signal-to-noise ratio of the feature signal and a kurtosis of the feature signal, wherein the signal-to-noise ratio of the feature signal is used to characterize a periodic feature of the acoustic signal, and the kurtosis of the feature signal is used to characterize an impact feature of the acoustic signal; Acquiring a signal-to-noise ratio and a kurtosis of an acoustic signal to be diagnosed, obtaining the signal-to-noise ratio to be diagnosed and the kurtosis to be diagnosed, inputting the signal-to-noise ratio to be diagnosed and the kurtosis to be diagnosed into a classification model, and obtaining a classification result corresponding to the acoustic signal to be diagnosed, wherein the classification model is trained using the signal-to-noise ratio of the characteristic signal and the kurtosis of the characteristic signal and the classification result corresponding to the acoustic signal, the signal-to-noise ratio to be diagnosed is the signal-to-noise ratio corresponding to the characteristic signal of the acoustic signal to be diagnosed, and the kurtosis to be diagnosed is the kurtosis corresponding to the characteristic signal of the acoustic signal to be diagnosed; According to the classification result, it is determined whether the valve train has an abnormal noise fault.

2. The method according to claim 1, characterized in that Performing feature extraction processing on the acoustic signal to obtain a feature signal includes: Using a finite impulse response filter to filter the acoustic signal to obtain a first processed signal; Down-sampling is performed on the first processed signal to obtain the characteristic signal.

3. The method according to claim 1, characterized in that Performing feature extraction processing on the acoustic signal to obtain a feature signal includes: The acoustic signal is subjected to wavelet decomposition processing by adopting a wavelet transform algorithm to obtain a second processed signal; and the second processed signal is processed by adopting a particle swarm optimization algorithm to obtain the characteristic signal.

4. The method according to claim 3, characterized in that The second processed signal is processed by a particle swarm optimization algorithm to obtain the characteristic signal, including: A generating step of randomly generating a plurality of initial particles in a predetermined parameter space, wherein the initial particles have an initial velocity and an initial position, and the initial position of the initial particles is used to characterize the frequency center of the second processed signal and the bandwidth of the second processed signal; A calculation step, calculating an objective function of the second processed signal, wherein the objective function is used to characterize a periodic feature of a signal after demodulation of the second processed signal; An updating step of updating the initial velocity of the initial particle to obtain an updated velocity, and updating the initial position of the initial particle to obtain an updated position; The repeating step sequentially repeats the calculation step and the updating step at least once, and in the repeating process, updates the initial speed in the updating step to the updated speed in the updating step obtained in the last repeating process, and updates the initial position in the updating step to the updated position in the updating step obtained in the last repeating process, until a preset condition is met, and the preset condition is that the number of repetitions is greater than a first threshold or the function value of the objective function is less than a second threshold.

5. The method according to claim 1, characterized in that Acquiring the signal-to-noise ratio of the characteristic signal includes: Acquiring an envelope spectrum signal of the characteristic signal; According to the formula Calculate the signal-to-noise ratio Snr of the characteristic signal x (f c ,B w ), where f p is the characteristic frequency corresponding to the abnormal noise failure of the valve train, X w is the envelope spectrum signal of the characteristic signal, f n is the characteristic frequency corresponding to the supercharger of the engine, floor(·) is a floor function, and i and j are both positive integers greater than or equal to 1 and less than or equal to 10.

6. The method according to claim 5, characterized in that Acquiring an envelope spectrum signal of the characteristic signal, comprising: Performing Hilbert transform processing on the acoustic signal to obtain an envelope signal; Performing Fourier transform on the envelope signal to obtain the envelope spectrum signal.

7. The method according to claim 1, characterized in that Obtaining the kurtosis of the characteristic signal includes: According to the formula Calculate the kurtosis K of the characteristic signal x ,in, is the nth acoustic signal x(n) at the center frequency f c and bandwidth B w The corresponding characteristic signal is The characteristic signal The fourth moment of The characteristic signal The second moment of .

8. A diagnostic device for abnormal noise fault of a valve train of an engine, characterized in that: include: an acquisition unit, used for acquiring an acoustic signal of a target area during an engine test and a classification result corresponding to the acoustic signal, wherein the target area is an area above a cylinder head of the engine, the acoustic signal is a signal transmitted by a valve mechanism through sound waves during the engine test, and the classification result includes an abnormal noise fault occurring in the valve mechanism and a non-abnormal noise fault occurring in the valve mechanism; a processing unit, configured to perform feature extraction processing on the acoustic signal to obtain a feature signal, and acquire a signal-to-noise ratio of the feature signal and a kurtosis of the feature signal, wherein the signal-to-noise ratio of the feature signal is used to characterize a periodic feature of the acoustic signal, and the kurtosis of the feature signal is used to characterize an impact feature of the acoustic signal; An input unit, used to obtain the signal-to-noise ratio and kurtosis of the acoustic signal to be diagnosed, obtain the signal-to-noise ratio to be diagnosed and the kurtosis to be diagnosed, input the signal-to-noise ratio to be diagnosed and the kurtosis to be diagnosed into a classification model, and obtain a classification result corresponding to the acoustic signal to be diagnosed, wherein the classification model is trained using the signal-to-noise ratio of the characteristic signal and the kurtosis of the characteristic signal and the classification result corresponding to the acoustic signal, the signal-to-noise ratio to be diagnosed is the signal-to-noise ratio corresponding to the characteristic signal of the acoustic signal to be diagnosed, and the kurtosis to be diagnosed is the kurtosis corresponding to the characteristic signal of the acoustic signal to be diagnosed; A determination unit is used to determine whether the valve mechanism has an abnormal noise fault according to the classification result.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include methods for executing any one of claims 1 to 7.

Images