Engine abnormal sound detection method, device, equipment and medium

Abstract

The invention discloses an engine abnormal sound detection method, device and equipment and a medium. The engine abnormal sound detection method comprises the following steps: acquiring actual noise audio data of a to-be-detected engine; processing the actual noise audio data to generate actual noise audio feature data; establishing a standard audio database; wherein the standard audio database comprises standard noise audio feature data under various combination conditions of various influence parameters; determining matched noise audio feature data according to the feature information of the to-be-detected engine and the standard noise audio feature data; according to similarity parameters of the matched noise audio feature data and the actual noise audio feature data, abnormal sound faults of the engine are recognized; and when the abnormal sound fault of the engine is identified, the abnormal sound fault type of the engine is diagnosed according to the actual noise audio feature data.

Description

Technical Field

[0001] Embodiments of the present invention relate to the field of engine technology, and in particular to a method, device, equipment, and medium for detecting abnormal engine noise. Background Art

[0002] With technological advancements and increasing customer demands for quality, abnormal engine noise is receiving increasing attention. While it remains a difficult problem among existing engine failures, it has numerous causes, making troubleshooting, diagnosis, and resolution difficult and costly.

[0003] The most common method of detecting abnormal noises is still manual detection through the human ear. This method is limited by people's subjective feelings and relies on personnel experience. The reliability and traceability of such manual detection methods are poor. Because people are close to moving mechanical parts and high noise sources during the detection process, there are also potential safety risks. At the same time, in the face of the various types of abnormal noise faults currently existing in engines, traditional manual detection through the human ear is even more unable to identify the type of abnormal noise fault, and the reliability of identifying the fault type is low. Summary of the Invention

[0004] The present invention provides an engine abnormal noise detection method, device, equipment and medium, which realize automatic detection of engine abnormal noise, adapt to abnormal noise detection under various influencing parameters, and diagnose the type of engine abnormal noise fault.

[0005] To achieve the above objectives, in a first aspect, an embodiment of the present invention provides a method for detecting abnormal engine noise, the method comprising:

[0006] Collect actual noise audio data of the engine to be tested;

[0007] Processing the actual noise audio data to generate actual noise audio feature data;

[0008] Establishing a standard audio database; wherein the standard audio database includes audio feature data of various standard noises under various combination conditions of multiple influencing parameters;

[0009] Determining matching noise audio feature data based on the feature information of the engine to be tested and each of the standard noise audio feature data;

[0010] Identifying an abnormal noise fault in the engine according to a similarity parameter between the matched noise audio feature data and the actual noise audio feature data;

[0011] When it is identified that an abnormal noise fault occurs in the engine, the abnormal noise fault type of the engine is diagnosed according to the actual noise audio feature data.

[0012] Optionally, a standard audio database is established; wherein the standard audio database includes audio feature data of various standard noises under various combinations of various influencing parameters; including:

[0013] determining a plurality of influencing parameters of the noise audio data;

[0014] Under any combination of the plurality of influencing parameter conditions, standard noise audio data of a plurality of preset standard engines are collected;

[0015] Performing standardization processing on each of the standard noise audio data;

[0016] Performing feature extraction on each of the standard noise audio data after standard processing to determine parameters for generating similarity comparison;

[0017] Processing each of the generated similarity comparison parameters to determine each noise similarity sequence corresponding to each of the standard noise audio data;

[0018] Each of the noise similarity sequences is processed to output standard noise audio feature data under any combination of the plurality of influencing parameters.

[0019] Optionally, processing each of the generated similarity comparison parameters to determine each noise similarity sequence corresponding to each of the standard noise audio data comprises:

[0020] Performing DTW distance processing on the parameters used for generating similarity comparison;

[0021] A similarity conversion is performed using parameters based on the generated similarities after the DTW distance processing to determine the corresponding noise similarity sequences of the standard noise audio data.

[0022] Optionally, processing each of the noise similarity sequences to output standard noise audio feature data under any combination of multiple influencing parameters includes:

[0023] Determining each similarity standard deviation according to the noise similarities in each noise similarity sequence;

[0024] The standard noise audio feature data under any combination of the plurality of influencing parameters is determined and outputted according to the similarity standard deviations.

[0025] Optionally, processing the actual noise audio data to generate actual noise audio feature data includes:

[0026] Performing standardization processing on the actual noise audio data;

[0027] Feature extraction is performed on the actual noise audio data after standard processing to generate actual noise audio feature data.

[0028] Optionally, when an abnormal noise fault is identified in the engine, diagnosing the abnormal noise fault type of the engine according to the actual noise audio feature data includes:

[0029] Determining actual abnormal sound audio conversion data according to the matched noise audio feature data and the actual noise audio feature data;

[0030] Performing feature extraction on the actual abnormal sound audio conversion data to generate actual abnormal sound audio feature data;

[0031] Determine an abnormal sound similarity sequence by processing the actual abnormal sound audio feature data and reference abnormal sound audio feature data of different fault categories in the abnormal sound fault audio database;

[0032] The abnormal noise fault category of the engine is diagnosed by sorting the abnormal noise similarity sequence into high and low order.

[0033] Optionally, the method further includes: establishing an abnormal sound fault audio database; wherein the abnormal sound fault audio database includes reference abnormal sound audio feature data of different fault categories; including:

[0034] Determine the detection parameters that affect the abnormal noise fault engine; wherein the detection parameters include detection scene parameters and detection operating condition parameters;

[0035] Under different detection parameters, collecting reference abnormal sound audio data of engines with abnormal noise of different fault categories;

[0036] Determine matching abnormal sound audio data based on the detection parameters of the abnormal sound fault engine and each of the standard noise audio feature data;

[0037] generating reference abnormal sound audio conversion data for the reference abnormal sound audio data and the matched abnormal sound audio data;

[0038] Feature extraction is performed on the reference abnormal sound audio conversion data to generate reference abnormal sound audio diagnostic data.

[0039] In a second aspect, an embodiment of the present invention further provides an engine abnormal noise detection device, the detection device comprising:

[0040] An acquisition module, used to collect actual noise audio data of the engine to be tested;

[0041] A first processing module, configured to process the actual noise audio data to generate actual noise audio feature data;

[0042] The first database establishment module is used to establish a standard audio database; wherein the standard audio database includes various standard noise audio feature data under various combination conditions of multiple influencing parameters;

[0043] a second processing module, configured to determine matching noise audio feature data based on the feature information of the engine to be tested and each of the standard noise audio feature data;

[0044] an identification module, configured to identify an abnormal noise fault occurring in the engine based on a similarity parameter between the matched noise audio feature data and the actual noise audio feature data;

[0045] The diagnostic module is used to diagnose the type of abnormal noise fault of the engine according to the actual noise audio feature data when it is identified that the engine has an abnormal noise fault.

[0046] In a third aspect, an embodiment of the present invention further provides a computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the engine abnormal noise detection method as described in the first aspect when executing the program.

[0047] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the engine abnormal noise detection method described in the first aspect.

[0048] According to an embodiment of the present invention, actual noise audio data of an engine to be tested is collected; the actual noise audio data is processed to generate actual noise audio feature data; a standard audio database is established; wherein the standard audio database includes various standard noise audio feature data under various combinations of influencing parameters; matching noise audio feature data is determined based on feature information of the engine to be tested and each of the standard noise audio feature data; an abnormal noise fault of the engine is identified based on a similarity parameter between the matching noise audio feature data and the actual noise audio feature data; when an abnormal noise fault of the engine is identified, the type of the abnormal noise fault of the engine is diagnosed based on the actual noise audio feature data, thereby realizing automatic detection of abnormal noise of the engine and adapting to abnormal noise detection under various influencing parameters; and diagnosing the type of abnormal noise fault of the engine.

[0049] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0051] Figure 1 This is a flow chart of a method for detecting abnormal engine noise provided by an embodiment of the present invention;

[0052] Figure 2 This is a flow chart of another engine abnormal noise detection method provided by an embodiment of the present invention;

[0053] Figure 3 This is a flow chart of another engine abnormal noise detection method provided by an embodiment of the present invention;

[0054] Figure 4 This is a flow chart of another engine abnormal noise detection method provided by an embodiment of the present invention;

[0055] Figure 5 This is a schematic structural diagram of an engine abnormal noise detection device provided by an embodiment of the present invention;

[0056] Figure 6 It is a structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative work should fall within the scope of protection of the present invention.

[0058] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes 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.

[0059] Figure 1 This is a flow chart of an engine abnormal noise detection method provided by an embodiment of the present invention. This embodiment is applicable to detecting abnormal engine noise and diagnosing the type of abnormal noise. The method can be executed by an engine abnormal noise detection device, such as Figure 1 As shown, the detection method specifically includes the following steps:

[0060] S110 : Collecting actual noise audio data of the engine to be tested.

[0061] In this embodiment, one or more noise sensors or microphones can be deployed to collect actual noise audio data of the engine under test in specific scenarios (such as test benches, production lines, roads, open spaces, etc.), specific operating conditions (such as acceleration conditions, idling conditions, etc.), and specific states (such as boundary conditions such as engine oil temperature, fuel temperature, inlet and outlet water temperature, and intake intercooler temperature);

[0062] Take the factory bench test of the engine to be tested as an example: the engine to be tested flows normally to the bench test station in the production line. The operator controls the start of the engine to be tested through the bench measurement and control system, and performs the test according to the established test procedure in the engine program-controlled test system. The engine to be tested operates to a specific operating condition under the control of the test program. When the engine runs to a specific operating condition steady state during the test process (the engine is in a specific state when reaching steady state), the actual noise audio data of the engine to be tested is collected through one or more noise sensors pre-arranged in the bench.

[0063] S120: Process the actual noise audio data to generate actual noise audio feature data.

[0064] Among them, the actual noise audio feature data is a key indicator data that can eliminate the differences caused by different sampling frequencies and sampling amplitudes, and can also describe the noise energy distribution properties; the actual noise audio feature data can be used as key data to determine abnormal noise failures in the engine.

[0065] S130 , establishing a standard audio database; wherein the standard audio database includes audio feature data of various standard noises under various combination conditions of multiple influencing parameters.

[0066] The various influencing parameters are factors that affect the actual noise audio collected, such as the engine model, operating conditions, status, and environment; or may include sampling equipment, sampling points, parameter differences, etc.; of course, some factors can be controlled in actual testing to eliminate differences; this embodiment does not limit the types of the various influencing parameters;

[0067] The multiple influencing parameter conditions may include multiple combination conditions, and each combination condition may correspond to a corresponding set of standard noise audio feature data. For example, taking two influencing parameter conditions as an example, the idle condition under the operating condition influencing parameters and the bench test under the scenario influencing parameters correspond to standard noise audio feature data 1; the idle condition under the operating condition influencing parameters and the road test under the scenario influencing parameters correspond to standard noise audio feature data 2. It is understood that the multiple combination conditions under the multiple influencing parameter conditions may include multiple sets, and this embodiment does not limit this. Each set of standard noise audio feature data is audio data of the engine operating normally under each combination condition. Such multiple standard noise audio feature data facilitates accurate subsequent determination of abnormal engine noise faults, avoiding the inaccurate determination of abnormal engine noise faults caused by the fixed standard noise audio feature data in the prior art.

[0068] S140 : Determine matching noise audio feature data based on feature information of the engine to be tested and each standard noise audio feature data.

[0069] Among them, the characteristic information of the engine to be tested may include: the model of the engine to be tested, the scenario in which the engine to be tested is located, the working conditions performed by the transmitter to be tested and the status information of the engine to be tested; specifically, the matching noise audio feature data can be determined based on the characteristic information of the engine to be tested and the standard noise audio feature data containing the characteristic information (i.e., each combination condition) in each standard noise audio feature data; the matching noise audio feature data can be used as the standard noise audio feature data corresponding to the engine to be tested; since the matching noise audio feature data is audio feature data under multiple influencing parameters, the fault judgment of the engine to be tested under multiple influencing parameters can be performed based on the matching noise audio feature data.

[0070] S150: Identify an abnormal noise fault in the engine based on a similarity parameter between the matched noise audio feature data and the actual noise audio feature data.

[0071] The similarity parameter between the matched noise audio feature data and the actual noise audio feature data can be a difference parameter, a ratio parameter, or other similarity-representing parameter; this embodiment is not limited thereto. Specifically, when the similarity parameter is higher than a preset threshold, the engine is determined to be qualified and free of abnormal noise faults; if the similarity parameter is lower than the preset threshold, the engine is determined to be unqualified and with an abnormal noise fault. The preset threshold is a fixed value that can be determined by calibration.

[0072] S160: When an abnormal noise fault is identified in the engine, diagnose the type of the abnormal noise fault in the engine based on actual noise audio characteristic data.

[0073] Among them, the abnormal noise data category of the engine can be diagnosed based on the actual noise audio feature data, so that the abnormal noise fault category of the engine can also be diagnosed after the abnormal noise fault is identified; the method of diagnosing the abnormal noise fault category of the engine based on the actual noise audio feature data is not limited in this embodiment.

[0074] According to an embodiment of the present invention, actual noise audio data of an engine to be tested is collected; the actual noise audio data is processed to generate actual noise audio feature data; a standard audio database is established; wherein the standard audio database includes various standard noise audio feature data under various combinations of influencing parameters; matching noise audio feature data is determined based on feature information of the engine to be tested and various standard noise audio feature data; an abnormal noise fault of the engine is identified based on similarity parameters between the matching noise audio feature data and the actual noise audio feature data; when an abnormal noise fault of the engine is identified, the type of the abnormal noise fault of the engine is diagnosed based on the actual noise audio feature data, thereby realizing automatic detection of abnormal noise of the engine, adapting to abnormal noise detection under various influencing parameters, improving detection accuracy, and reducing the difficulty of abnormal noise detection; and the type of abnormal noise fault of the engine is diagnosed.

[0075] Optionally, based on the above embodiment, the steps S120 and S130 of establishing a standard audio database are further refined. Figure 2 This is a flow chart of another engine abnormal noise detection method provided by an embodiment of the present invention. Figure 2 As shown, the detection method specifically includes the following steps:

[0076] S210: Collect actual noise audio data of the engine to be tested.

[0077] S220: Process the actual noise audio data to generate actual noise audio feature data.

[0078] The processing of the actual noise audio data to generate the actual noise audio feature data includes: performing standardization processing on the actual noise audio data; and performing feature extraction on the actual noise audio data after standardization processing to generate the actual noise audio feature data.

[0079] Considering that the collected audio files were not sampled using the same detection hardware or according to preset standard sampling parameters, the actual noise audio data needs to be normalized. Specifically, this process can include resampling, amplitude normalization, and silent segment cropping. Resampling involves unifying the sampling rate to a preset target value (such as 22.05kHz, 44.1kHz, etc.) to minimize differences between different sampling rates. Amplitude normalization involves scaling the amplitude of the actual noise audio data to the range [-1, 1] to eliminate the impact of different collected volume levels. Silence segment cropping involves removing the leading and trailing silent segments through energy threshold detection to retain the valid audio.

[0080] The feature extraction of the actual noise audio data after standard processing to generate the actual noise audio feature data includes: performing frame windowing, Fourier transform, Mel filtering, logarithmic operation and discrete cosine transform processing on the actual noise audio data after standard processing in sequence, so as to generate the actual noise audio feature data; specifically, frame windowing is to divide the audio into short time frames (frame length 25ms, frame shift 10ms); Fourier transform is to perform Fourier transform on each short time frame signal to obtain the spectrum energy distribution; Mel filtering is to map the spectrum energy distribution to the Mel scale (simulating the auditory characteristics of the human ear) and extract 20-40 Mel band energies; logarithmic operation is to take the logarithm of the Mel energy to enhance the low-frequency discrimination; discrete cosine transform processing is to perform discrete cosine transform on the logarithmic Mel spectrum, retaining the first 12-13 coefficients as MFCC features, which is the actual noise audio feature data.

[0081] S230. Establish a standard audio database based on the standard noise audio data under various combinations of the multiple influencing parameter conditions; wherein the standard audio database includes the standard noise audio feature data under various combinations of the multiple influencing parameter conditions.

[0082] The process of establishing a standard audio database based on various standard noise audio data under various combination conditions of multiple influencing parameter conditions includes the following steps:

[0083] S1. Determine multiple influencing parameters of noisy audio data.

[0084] Among them, the various influencing parameters are factors that affect the actual noise audio collected, such as the model, operating conditions, status and environment of the engine itself; or they may also include sampling equipment, points, parameter differences, etc.; of course, some factors can be controlled in actual detection to eliminate differences; this embodiment does not limit the types of the various influencing parameters.

[0085] S2. Under any combination of multiple influencing parameter conditions, collect standard noise audio data of multiple preset standard engines;

[0086] The standard engine is an engine that does not produce abnormal noise faults. This embodiment may include multiple preset engines, so that standard noise audio data corresponding to multiple standard engines can be collected. The multiple influencing parameter conditions may include multiple combination conditions. The standard noise audio data is: the noise audio data of the standard engine under each combination condition can be collected by one or more noise sensors or microphones.

[0087] For example, noise audio data of a standard engine can be collected in different scenarios (such as test benches, production lines, roads, open spaces, etc.), different operating conditions (such as acceleration conditions, idling conditions, etc.), and different states (such as engine oil temperature, fuel temperature, inlet and outlet water temperatures, intake intercooler temperature, and other boundary conditions). Specifically, take the factory bench test of a standard engine as an example: the standard engine flows normally to the bench test station in the production line. The operator controls the start of the standard engine through the bench measurement and control system, and performs the test according to the established test procedure in the engine program-controlled test system. The standard engine operates to different working conditions under the control of the test program. When the engine runs to different steady-state working conditions during the test process (the engine is in a specific state when reaching the steady state), the actual noise audio data of the standard engine is collected through one or more noise sensors pre-arranged in the bench; and so on, the noise audio data of the standard engine is collected in different scenarios (such as bench test rooms, production lines, roads, open spaces, etc.), different working conditions (such as acceleration conditions, idling conditions, etc.) and different states (such as engine oil temperature, fuel temperature, inlet and outlet water temperature, intake intercooler temperature and other boundary conditions).

[0088] S3. Perform standardization processing on each standard noise audio data.

[0089] The process of standardizing each standard noise audio data may also include: standardizing each standard noise audio data may include: resampling processing, amplitude normalization processing) and silent segment cropping processing; the steps of each processing process are the same as the standardization process of the actual noise audio feature data, which will not be repeated here.

[0090] S4. Feature extraction is performed on each standard noise audio data after standard processing to determine parameters for generating similarity comparison. The standard noise audio data after standard processing is sequentially subjected to frame windowing, Fourier transform, Mel filtering, logarithmic operation, and discrete cosine transform processing to generate parameters for generating similarity comparison. That is, the mature MFCC (Mel-Frequency Cepstral Coefficient) feature extraction technology is used on each standard noise audio data after standard processing to extract a multidimensional parameter vector representing the acoustic characteristics of the engine from the standardized audio. The process of frame windowing, Fourier transform, Mel filtering, logarithmic operation, and discrete cosine transform processing is the same as the process of sequentially performing frame windowing, Fourier transform, Mel filtering, logarithmic operation, and discrete cosine transform processing on actual noise audio data, and is not explained here.

[0091] S5. Processing each generated similarity comparison parameter to determine each noise similarity sequence corresponding to each standard noise audio data;

[0092] Among them, each noise similarity sequence is a set of similarity sizes of each standard noise audio data relative to other standard noise audio data; in some embodiments, each generated similarity comparison parameter is processed to determine each corresponding noise similarity sequence of each standard noise audio data; including: processing each generated similarity comparison parameter by DTW distance processing; performing similarity conversion based on each generated similarity comparison parameter after DTW distance processing to determine each corresponding noise similarity sequence of the standard noise audio data.

[0093] DTW distance processing is performed on each generated similarity comparison parameter. This involves using the Dynamic Time Warping (DTW) algorithm to align the MFCC sequences of different lengths for each standard noise audio data relative to other standard noise audio data. The Euclidean distance is then calculated to quantify the differences and obtain the global optimal matching distance. Specifically, the following steps are performed: 1. Sequence alignment: Dynamic programming is used to align two MFCC sequences of unequal lengths to eliminate time axis scaling differences. 2. Local distance calculation: The Euclidean distance (or other distance metric) is calculated between each pair of frames. 3. Cumulative path optimization: The path with the minimum cumulative distance is selected to obtain the global optimal matching distance.

[0094] Based on the parameters used for the generated similarity comparisons after DTW distance processing, similarity conversion is performed to determine the corresponding noise similarity sequences for each standard noise audio data set. This involves converting the DTW distances into similarity values ​​ranging from 0% to 100%. Specifically, the DTW distances are converted into similarity percentages using the formula: Similarity = (1-Distance) / (Distance+Decay)×100% (the default Decay attenuation coefficient is 600). Distance is the parameter used for generating similarity comparisons, and Decay is the attenuation coefficient that adjusts the sensitivity of similarity to changes in distance. When Distance is much smaller than Decay, the similarity approaches 100%. When Distance is much larger than Decay, the similarity approaches 0%, preventing drastic changes in similarity caused by small distance fluctuations.

[0095] Exemplarily, under combination condition 1, each standard noise audio data includes A standard noise audio data, B standard noise audio data and C standard noise audio data; each noise similarity sequence is: (A / B similarity, A / C similarity), (B / A similarity, B / C similarity), (C / A similarity, C / B similarity).

[0096] S6. Process each noise similarity sequence and output standard noise audio feature data under any combination condition.

[0097] In some embodiments, processing each noise similarity sequence and outputting standard noise audio feature data under any combination of conditions includes: determining each similarity standard deviation based on the noise similarities within each noise similarity sequence; and determining and outputting the standard noise audio feature data under any combination of conditions based on each similarity standard deviation. Specifically, the standard noise audio data corresponding to the smallest of the similarity standard deviations is used as the standard noise audio feature data under any combination of conditions.

[0098] For example, under combination condition 1, each standard noise audio data includes standard noise audio data A, standard noise audio data B, and standard noise audio data C; each noise similarity sequence is: (A / B similarity, A / C similarity), (B / A similarity, B / C similarity), (C / A similarity, C / B similarity). The standard deviations of the A / B similarity and the A / C similarity, the standard deviations of the B / A similarity and the B / C similarity, and the standard deviations of the C / A similarity and the C / B similarity are determined, and the corresponding standard noise audio data with the smallest similarity standard deviation is used as the standard noise audio feature data under any combination condition.

[0099] According to the above steps S1-S6, the audio characteristic data of each standard noise under various combination conditions under multiple influencing parameters can be determined.

[0100] S240 : Determine matching noise audio feature data based on feature information of the engine to be tested and each standard noise audio feature data.

[0101] S250: Identify an abnormal noise fault in the engine based on a similarity parameter between the matched noise audio feature data and the actual noise audio feature data.

[0102] S260: When an abnormal noise fault is detected in the engine, diagnose the type of the abnormal noise fault in the engine based on actual noise audio characteristic data.

[0103] The embodiment of the present invention, based on the above embodiment, further processes the actual noise audio data to generate actual noise audio feature data; and refines the process of establishing a standard audio database, thereby realizing automatic detection of abnormal engine noise, while adapting to abnormal noise detection under various influencing parameters, improving detection accuracy, and reducing the difficulty of abnormal noise detection; and diagnosing the type of abnormal engine noise fault.

[0104] Optionally, based on the above embodiment, S260 is further refined. Figure 3 This is a flow chart of another engine abnormal noise detection method provided by an embodiment of the present invention. Figure 3 As shown, the detection method specifically includes the following steps:

[0105] S310: Collect actual noise audio data of the engine to be tested.

[0106] In some embodiments, before collecting the actual noise audio data of the engine under test, the process also includes: collecting the test bench noise audio when there is no engine on the test bench and all test bench equipment is not working, and comparing it with the standard empty test bench audio collected under the same conditions. If the obtained similarity value is greater than 80%, the software self-test is considered successful, and then the actual noise audio data of the engine under test is collected again to ensure the accuracy of subsequent engine abnormal noise detection;

[0107] Of course, in other embodiments, before collecting the actual noise audio data of the engine to be tested, another engine to be tested can be used to reversely drag through an electric dynamometer, such as setting the speed to 500 rpm, and testing without starting the engine; the noise audio of the other engine to be tested under this working condition is collected at this time and compared with the standard audio collected under the same conditions. If the obtained similarity value is higher than 80%, it is considered that the software self-test is successful; at this time, the actual noise audio data of the engine to be tested is collected again to ensure the accuracy of subsequent engine abnormal noise detection.

[0108] S320: Process the actual noise audio data to generate actual noise audio feature data.

[0109] S330. Establish a standard audio database based on the standard noise audio data under various combinations of multiple influencing parameter conditions; wherein the standard audio database includes the standard noise audio feature data under various combinations of multiple influencing parameter conditions.

[0110] S340 : Determine matching noise audio feature data based on feature information of the engine to be tested and each standard noise audio feature data.

[0111] S350: Identify an abnormal noise fault in the engine based on a similarity parameter between the matched noise audio feature data and the actual noise audio feature data.

[0112] S360: When an abnormal noise fault is identified in the engine, diagnose the type of the abnormal noise fault in the engine according to actual noise audio feature data based on the abnormal noise fault audio database.

[0113] When an abnormal noise fault is detected in the engine, the abnormal noise fault type of the engine is diagnosed based on the actual noise audio feature data in the abnormal noise fault audio database, specifically including the following steps:

[0114] S10. Determine actual abnormal sound audio conversion data according to the matched noise audio feature data and the actual noise audio feature data.

[0115] Among them, when an abnormal noise fault is identified in the engine, the actual noise audio feature data is the actual abnormal noise audio feature data, and the actual abnormal noise audio conversion data is determined according to the difference between the matching noise audio feature data and the actual noise audio feature data. This reduces the background noise of the matching noise audio data and avoids the background noise contained in the actual abnormal noise audio conversion data affecting the subsequent diagnosis of the abnormal noise fault type.

[0116] S20, performing feature extraction on the actual abnormal sound audio conversion data to generate actual abnormal sound audio feature data;

[0117] Among them, feature extraction is performed on the actual abnormal sound audio conversion data to generate actual abnormal sound audio feature data, specifically including sequentially performing frame windowing, Fourier transform, Mel filtering, logarithmic operation and discrete cosine transform on the actual abnormal sound audio conversion data, thereby generating the actual abnormal sound audio feature data; the process of frame windowing, Fourier transform, Mel filtering, logarithmic operation and discrete cosine transform is the same as the process of sequentially performing frame windowing, Fourier transform, Mel filtering, logarithmic operation and discrete cosine transform on the actual noise audio data, and will not be explained here.

[0118] S30, determining an abnormal sound similarity sequence based on the actual abnormal sound audio feature data and reference abnormal sound audio feature data of different fault categories in the abnormal sound fault audio database;

[0119] The abnormal sound similarity sequence is determined by processing the actual abnormal sound audio feature data and reference abnormal sound audio feature data of different fault categories in the abnormal sound fault audio database. Specifically, the actual abnormal sound audio feature data and the reference abnormal sound audio feature data of different fault categories are subjected to DTW distance processing and similarity conversion to determine the abnormal sound similarity sequence. The specific method of performing DTW distance processing and similarity conversion on the actual abnormal sound audio feature data and the reference abnormal sound audio feature data of different fault categories is the same as the method of performing DTW distance processing on each generated similarity comparison parameter in S5; and performing similarity conversion based on each generated similarity comparison parameter after DTW distance processing, and will not be repeated here.

[0120] Exemplarily, the reference abnormal sound audio feature data for different fault categories include reference abnormal sound audio feature data 1, reference abnormal sound audio feature data 2, and reference abnormal sound audio feature data 3; by performing DTW distance processing and similarity conversion on the actual abnormal sound audio feature data and the reference abnormal sound audio feature data 1, reference abnormal sound audio feature data 2, and reference abnormal sound audio feature data 3, an abnormal sound similarity sequence can be determined as: similarity 1 between the actual abnormal sound audio feature data and the reference abnormal sound audio feature data 1, similarity 1 between the actual abnormal sound audio feature data and the reference abnormal sound audio feature data 2, and similarity 3 between the actual abnormal sound audio feature data and the reference abnormal sound audio feature data 3.

[0121] S40: diagnose the abnormal noise fault category of the engine by sorting the abnormal noise similarity sequence.

[0122] Specifically, the abnormal sound similarity sequence may be sorted in order of high and low order, and the fault type with the highest similarity in the abnormal sound similarity sequence may be determined as the abnormal sound fault category of the engine.

[0123] In the embodiment of the present invention, based on the above embodiment, when an abnormal noise fault of the engine is identified, the abnormal noise fault type of the engine is diagnosed based on the abnormal noise fault audio database according to the actual noise audio feature data. In this way, the abnormal noise of the engine can be automatically detected, and the abnormal noise detection under various influencing parameters can be adapted, thereby improving the detection accuracy and reducing the difficulty of abnormal noise detection; and the type of abnormal noise fault of the engine can be diagnosed.

[0124] Optionally, based on the above embodiment, the establishment of an abnormal sound fault audio database in step S360 is further described. Figure 4 The present invention provides a flowchart of another method for detecting abnormal engine noise. Figure 4 As shown, the engine abnormal noise detection method includes the following steps:

[0125] S410: Collect actual noise audio data of the engine to be tested.

[0126] S420: Process the actual noise audio data to generate actual noise audio feature data.

[0127] S430. Establish a standard audio database based on various standard noise audio data under various combinations of multiple influencing parameter conditions; wherein the standard audio database includes various standard noise audio feature data under multiple influencing parameters.

[0128] S440 : Determine matching noise audio feature data based on feature information of the engine to be tested and each standard noise audio feature data.

[0129] S450: Identify an abnormal noise fault in the engine based on a similarity parameter between the matched noise audio feature data and the actual noise audio feature data.

[0130] S460: Establish an abnormal sound fault audio database; wherein the abnormal sound fault audio database includes reference abnormal sound audio feature data of different fault categories;

[0131] The establishment of an abnormal sound fault audio database includes the following steps:

[0132] S01. Determine detection parameters that affect the engine with abnormal noise; wherein the detection parameters include detection scene parameters and detection operating condition parameters;

[0133] When an engine has an abnormal noise fault, the scene in which it is located often varies, resulting in differences in the actual noise. When an abnormal engine noise fault is actually discovered, it is necessary to determine the detection scene in which the engine is located. That is, it is necessary to determine the detection scene parameters that affect the abnormal noise engine, such as the laboratory, production line, road, open space, etc., and ensure stable audio collection in the same detection environment. In particular, if the engine is installed on a vehicle and is driving on a road with poor road conditions or noisy vehicles, the audio collection will be unstable, and this detection scene can be excluded.

[0134] In addition, the noise performance of the engine abnormal noise fault is different under different working conditions; therefore, it is necessary to determine the detection working condition parameters that affect the engine abnormal noise fault, such as: idling condition and acceleration condition, etc.; considering that the human ear cannot perceive some working conditions, in practice one or more characteristic working condition points with obvious abnormal noise fault performance should be selected to collect the engine noise audio.

[0135] However, the engine state during noise collection can also affect engine noise performance. For example, engine noise differs when the engine is cold and hot. In practice, the engine can be controlled using an ECU or test bench to maintain boundary condition parameters such as engine oil temperature, fuel temperature, inlet and outlet water temperature, and intake intercooler temperature within the same range. For some abnormal noises that occur during cold engine startup, the ambient temperature must also be kept similar. Therefore, the detection scenario parameters do not need to be included in determining the detection parameters that affect abnormal noise in engines.

[0136] It should be noted that due to practical limitations, it's difficult to maintain consistent noise collection point locations and hardware equipment across different testing scenarios and operating conditions. Maintaining consistent noise collection point locations, hardware equipment, and collection parameters across all testing scenarios and operating conditions minimizes the impact of these locations and hardware on engine noise performance. Optional noise collection hardware includes various noise sensors, microphones, and smartphones.

[0137] S02. Under different detection parameters, collect reference abnormal noise audio data of engines with abnormal noise of different fault categories.

[0138] The "noise-producing engine" refers to an engine that has experienced a noise fault. Reference noise audio data can be collected from noise-producing engines with different fault categories under different detection parameters. Specifically, the reference noise audio data is collected using one or more noise sensors or microphones under different combinations of detection parameters. For example, different fault categories may include aerodynamic noise, combustion noise, mechanical friction, and mechanical knocking. The noise can be categorized by the location of the noise, including the cylinder head, intake and exhaust pipes, supercharger, flywheel, and pulley system.

[0139] For example, under different fault categories, the abnormal noise audio data of the abnormal noise engine can be collected in different scenarios (such as test benches, production lines, roads, open spaces, etc.) and under different operating conditions (such as acceleration conditions, idling conditions, etc.). Specifically, taking the factory test bench test of abnormal noise engines as an example: abnormal noise engines of the same fault category are normally transferred to the test bench station in the production line. The operator controls the start of the abnormal noise engine through the test bench measurement and control system and tests it according to the established test procedures in the engine program control test system. The abnormal noise engine operates under the control of the test procedure to different operating conditions. When the engine reaches the steady state of different operating conditions during the test process (the engine is in a specific state when reaching the steady state), the actual noise audio data of the abnormal noise engine is collected through one or more noise sensors pre-deployed in the test bench. By analogy, the abnormal noise audio data of the abnormal noise engine in different scenarios (such as test benches, production lines, roads, open spaces, etc.), different operating conditions (such as acceleration conditions, idling conditions, etc.) and different fault categories can be collected.

[0140] S03. Determine matching abnormal sound audio data based on the detection parameters of the abnormal sound fault engine and various standard noise audio feature data.

[0141] Specifically, by matching the detection parameters of the abnormal noise fault engine with the standard noise audio feature data containing the detection parameters in each standard noise audio feature data, the matched abnormal noise audio data can be determined; the matched abnormal noise audio data can be used as the standard noise audio feature data of the corresponding abnormal noise engine; since the matched abnormal noise audio data is the audio feature data under multiple detection parameters, the fault category of the abnormal noise engine under multiple scenarios and multiple working conditions can be accurately judged based on the matched abnormal noise audio data.

[0142] S04. Generate reference abnormal sound audio conversion data based on the reference abnormal sound audio data and the matching abnormal sound audio data;

[0143] The reference abnormal sound audio conversion data is determined based on the difference between the reference abnormal sound audio data and the matching abnormal sound audio data, thereby eliminating the background noise of the matching abnormal sound audio data and avoiding the background noise contained in the reference abnormal sound audio data affecting the determination of the subsequent reference abnormal sound audio feature data.

[0144] S05. Perform feature extraction on the reference abnormal sound audio conversion data to generate reference abnormal sound audio feature data.

[0145] Among them, feature extraction is performed on the reference abnormal sound audio conversion data to generate reference abnormal sound audio feature data, specifically including sequentially performing frame windowing, Fourier transform, Mel filtering, logarithmic operation and discrete cosine transform on the reference abnormal sound audio conversion data, thereby generating the reference abnormal sound audio feature data; the process of frame windowing, Fourier transform, Mel filtering, logarithmic operation and discrete cosine transform is the same as the process of sequentially performing frame windowing, Fourier transform, Mel filtering, logarithmic operation and discrete cosine transform on the actual abnormal sound audio conversion data, and will not be explained here.

[0146] S470: When an abnormal noise fault is identified in the engine, diagnose the type of the abnormal noise fault in the engine according to actual noise audio feature data based on the abnormal noise fault audio database.

[0147] The embodiment of the present invention, based on the above embodiment, specifically describes the process of establishing an abnormal noise fault audio database, and further based on the abnormal noise fault audio database, when an abnormal noise fault is identified in the engine, the abnormal noise fault type of the engine is diagnosed according to the actual noise audio feature data, thereby realizing automatic detection of abnormal noise of the engine, and adapting to abnormal noise detection under various influencing parameters, thereby improving detection accuracy and reducing the difficulty of abnormal noise detection; and realizing diagnosis of the type of abnormal noise fault of the engine.

[0148] An embodiment of the present invention further provides an engine abnormal noise detection device; the engine abnormal noise detection device provided by the embodiment of the present invention can execute an engine abnormal noise detection method provided by any embodiment of the present invention, and has corresponding functional modules and beneficial effects of the execution method. Figure 5 : is a structural diagram of an engine abnormal noise detection device provided by an embodiment of the present invention; Figure 5 As shown, the engine abnormal noise detection device includes:

[0149] An acquisition module, used to collect actual noise audio data of the engine to be tested;

[0150] A first processing module is used to process actual noise audio data to generate actual noise audio feature data;

[0151] The first database establishment module is used to establish a standard audio database; wherein the standard audio database includes various standard noise audio feature data under various combination conditions of multiple influencing parameters;

[0152] The second processing module is used to determine matching noise audio feature data based on the feature information of the engine to be tested and each standard noise audio feature data;

[0153] An identification module is used to identify abnormal noise faults of the engine based on similarity parameters between the matched noise audio feature data and the actual noise audio feature data;

[0154] The diagnostic module is used to diagnose the type of abnormal noise fault of the engine based on the actual noise audio feature data when an abnormal noise fault of the engine is identified.

[0155] Optionally, the first database establishment module includes:

[0156] a determination unit, configured to determine a plurality of influencing parameters of the noise audio data;

[0157] A collection unit, configured to collect standard noise audio data of a plurality of preset standard engines under any combination of a plurality of influencing parameter conditions;

[0158] A standardization unit, used for performing standardization processing on each standard noise audio data;

[0159] A parameter unit is used to extract features from each standard noise audio data after standard processing to determine parameters for generating similarity comparison;

[0160] A similarity unit is used to process each generated similarity comparison parameter to determine each noise similarity sequence corresponding to each standard noise audio data;

[0161] The output unit is used to process each noise similarity sequence and output standard noise audio feature data under any combination of multiple influencing parameters.

[0162] Optional similarity unit, specifically:

[0163] The parameters for each generated similarity comparison are processed using DTW distance processing;

[0164] The generated similarities after the DTW distance processing are compared with parameters for similarity conversion to determine the corresponding noise similarity sequences of the standard noise audio data.

[0165] Optional output unit, specifically:

[0166] Determine each similarity standard deviation according to the noise similarity in each noise similarity sequence;

[0167] The standard noise audio feature data under any combination of multiple influencing parameters is output according to the standard deviation of each similarity.

[0168] Optionally, the first processing module is specifically:

[0169] Normalize the actual noise audio data;

[0170] Feature extraction is performed on the actual noise audio data after standard processing to generate actual noise audio feature data.

[0171] Optionally, the identification module includes:

[0172] A conversion unit, configured to determine actual abnormal sound audio conversion data based on the matched noise audio feature data and the actual noise audio feature data;

[0173] An extraction unit, configured to extract features from the actual abnormal sound audio conversion data to generate actual abnormal sound audio feature data;

[0174] A matching unit is used to process the actual abnormal sound audio feature data and the reference abnormal sound audio feature data of different fault categories in the abnormal sound fault audio database to determine an abnormal sound similarity sequence;

[0175] The diagnosis unit is used to diagnose the abnormal noise fault category of the engine by sorting the abnormal noise similarity sequence.

[0176] Optionally, the identification module further includes: a database establishment unit, wherein the abnormal sound fault audio database includes reference abnormal sound audio feature data of different fault categories; the database establishment unit specifically includes:

[0177] Determine the detection parameters that affect the abnormal noise fault engine; wherein the detection parameters include detection scene parameters and detection operating condition parameters;

[0178] Under different detection parameters, collect reference abnormal noise audio data of engines with different fault categories;

[0179] Determine matching abnormal sound audio data based on detection parameters of the abnormal sound fault engine and various standard noise audio feature data;

[0180] generating reference abnormal sound audio conversion data for the reference abnormal sound audio data and the matched abnormal sound audio data;

[0181] Feature extraction is performed on the reference abnormal sound audio conversion data to generate reference abnormal sound audio diagnostic data.

[0182] An embodiment of the present invention further provides an electronic device, Figure 6 1 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. The electronic device is intended to represent various forms of digital computers, such as embedded computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are only examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0183] like Figure 6 As shown, the electronic device 01 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., which is communicatively connected to the at least one processor 11. The memory stores a computer program that can be executed by the at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. Various programs and data required for the operation of the electronic device 100 can also be stored in the RAM 13. The processor 11, ROM 12, and RAM 13 are connected to each other via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0184] Multiple components in the electronic device 100 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 100 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0185] The processor 11 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any other suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a method for detecting abnormal engine noise.

[0186] In some embodiments, a method for detecting abnormal engine noise can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 01 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the method for detecting abnormal engine noise described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to execute the method for detecting abnormal engine noise via any other suitable means (e.g., via firmware).

[0187] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0188] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0189] In the context of the present invention, computer-readable storage media can be tangible media that can contain or store a computer program for use with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Computer-readable storage media can include but are not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage media can be machine-readable signal media. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0190] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0191] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0192] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a hosting product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

[0193] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0194] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

[0195] Note that the above are only preferred embodiments of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments, and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments and may include many other equivalent embodiments without departing from the concept of the present invention. The scope of the present invention is determined by the scope of the appended claims.

Claims

1. A method for detecting abnormal engine noise, characterized in that: include: Collect actual noise audio data of the engine to be tested; Processing the actual noise audio data to generate actual noise audio feature data; Establishing a standard audio database; wherein the standard audio database includes audio feature data of various standard noises under various combination conditions of multiple influencing parameters; Determining matching noise audio feature data based on the feature information of the engine to be tested and each of the standard noise audio feature data; Identifying an abnormal noise fault in the engine according to a similarity parameter between the matched noise audio feature data and the actual noise audio feature data; When it is identified that an abnormal noise fault occurs in the engine, the abnormal noise fault type of the engine is diagnosed according to the actual noise audio feature data.

2. The engine abnormal noise detection method according to claim 1, characterized in that: Establish a standard audio database; wherein the standard audio database includes various standard noise audio feature data under various combinations of influencing parameters; including: determining a plurality of influencing parameters of the noise audio data; Under any combination of the plurality of influencing parameter conditions, standard noise audio data of a plurality of preset standard engines are collected; Performing standardization processing on each of the standard noise audio data; Performing feature extraction on each of the standard noise audio data after standard processing to determine parameters for generating similarity comparison; Processing each of the generated similarity comparison parameters to determine each noise similarity sequence corresponding to each of the standard noise audio data; Each of the noise similarity sequences is processed to output standard noise audio feature data under any combination of the plurality of influencing parameters.

3. The engine abnormal noise detection method according to claim 2, characterized in that: Processing each of the generated similarity comparison parameters to determine each noise similarity sequence corresponding to each of the standard noise audio data; comprising: Performing DTW distance processing on the parameters used for generating similarity comparison; A similarity conversion is performed using parameters based on the generated similarities after the DTW distance processing to determine the corresponding noise similarity sequences of the standard noise audio data.

4. The engine abnormal noise detection method according to claim 2, characterized in that: Processing each of the noise similarity sequences to output standard noise audio feature data under any combination of multiple influencing parameters includes: Determining each similarity standard deviation according to the noise similarities in each noise similarity sequence; The standard noise audio feature data under any combination of the plurality of influencing parameters is determined and outputted according to the similarity standard deviations.

5. The engine abnormal noise detection method according to claim 1, characterized in that: Processing the actual noise audio data to generate actual noise audio feature data includes: Performing standardization processing on the actual noise audio data; Feature extraction is performed on the actual noise audio data after standard processing to generate actual noise audio feature data.

6. The engine abnormal noise detection method according to claim 1, characterized in that: When an abnormal noise fault is identified in the engine, diagnosing the abnormal noise fault type of the engine according to the actual noise audio feature data includes: Determining actual abnormal sound audio conversion data according to the matched noise audio feature data and the actual noise audio feature data; Performing feature extraction on the actual abnormal sound audio conversion data to generate actual abnormal sound audio feature data; Determine an abnormal sound similarity sequence by processing the actual abnormal sound audio feature data and reference abnormal sound audio feature data of different fault categories in the abnormal sound fault audio database; The abnormal noise fault category of the engine is diagnosed by sorting the abnormal noise similarity sequence into high and low order.

7. The engine abnormal noise detection method according to claim 6, characterized in that: Also includes: Establish an abnormal sound fault audio database; wherein the abnormal sound fault audio database includes reference abnormal sound audio feature data of different fault categories; including: Determine the detection parameters that affect the abnormal noise fault engine; wherein the detection parameters include detection scene parameters and detection operating condition parameters; Under different detection parameters, collecting reference abnormal sound audio data of engines with abnormal noise of different fault categories; Determine matching abnormal sound audio data based on the detection parameters of the abnormal sound fault engine and each of the standard noise audio feature data; generating reference abnormal sound audio conversion data for the reference abnormal sound audio data and the matched abnormal sound audio data; Feature extraction is performed on the reference abnormal sound audio conversion data to generate reference abnormal sound audio diagnostic data.

8. An engine abnormal noise detection device, characterized in that: include: An acquisition module, used to collect actual noise audio data of the engine to be tested; A first processing module, configured to process the actual noise audio data to generate actual noise audio feature data; The first database establishment module is used to establish a standard audio database; wherein the standard audio database includes various standard noise audio feature data under various combination conditions of multiple influencing parameters; a second processing module, configured to determine matching noise audio feature data based on the feature information of the engine to be tested and each of the standard noise audio feature data; an identification module, configured to identify an abnormal noise fault occurring in the engine based on a similarity parameter between the matched noise audio feature data and the actual noise audio feature data; The diagnostic module is used to diagnose the type of abnormal noise fault of the engine according to the actual noise audio feature data when it is identified that the engine has an abnormal noise fault.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the engine abnormal noise detection method as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the engine abnormal noise detection method as described in any one of claims 1 to 7 is implemented.

Images