Vehicle Abnormal Noise Source Identification Ability Training Method, Device, Equipment and Storage Medium

Abstract

The present invention discloses a method, device, equipment and storage medium for training the ability to identify vehicle abnormal noise sources. The method obtains the analysis result of factors affecting the ability to identify the azimuth of vehicle abnormal noise sources, determines the target dimension according to the analysis result; determines the corresponding target training mode according to the target dimension; conducts vehicle abnormal noise source identification training on the abnormal noise source identification equipment of the test vehicle according to the target training mode, and obtains the training result; improves the accuracy of abnormal noise identification, shortens the time for troubleshooting abnormal noise sources, avoids a large amount of consumption of human and material resources, and improves the speed and efficiency of abnormal noise source identification.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle abnormal noise identification, and particularly to a method, device, equipment and storage medium for training the ability to identify the source of vehicle abnormal noise. Background Art

[0002] With the development of the times, people's awareness of quality is getting stronger and stronger; in the field of automobile consumption, the abnormal noise performance of the whole vehicle is an important performance index of the automobile; now, people are paying more and more attention to the abnormal noise performance of the whole vehicle, and major automobile manufacturers are sparing no effort to improve the abnormal noise performance of vehicles; when encountering vehicle abnormal noise problems, it is necessary to quickly and efficiently lock the source of the abnormal noise, and then quickly eliminate the abnormal noise.

[0003] The existing vehicle abnormal noise identification methods generally identify abnormal noises through manual inspections by engineers. However, because the abnormal noise propagates very fast and attenuates little in the vehicle body structure, it brings great difficulties and blindness to the inspections by engineers. Moreover, due to the characteristics of the sheet metal structure, some simple measures are difficult to implement and verify. Therefore, to solve the abnormal noise problem, it mainly relies on the experience of engineers, and gradually checks by welding and adjusting gaps in the areas where abnormal noises may occur, resulting in a long time required to check the source of the abnormal noise and a high requirement for the experience of engineers; the accuracy of abnormal noise identification is relatively low, the time for checking the source of the abnormal noise is long, a large amount of manpower and material resources are consumed, and the efficiency of identifying the source of the abnormal noise is low. Summary of the Invention

[0004] The main purpose of the present invention is to provide a method, device, equipment and storage medium for training the ability to identify the source of vehicle abnormal noise, aiming to solve the technical problems in the prior art that the accuracy of abnormal noise identification is relatively low, the time for checking the source of the abnormal noise is long, a large amount of manpower and material resources are consumed, and the efficiency of identifying the source of the abnormal noise is relatively high.

[0005] In a first aspect, the present invention provides a method for training the ability to identify the source of vehicle abnormal noise, and the method for training the ability to identify the source of vehicle abnormal noise includes the following steps:

[0006] Obtain the analysis result of the factors affecting the ability to identify the orientation of the vehicle abnormal noise source, and determine the target dimension according to the analysis result;

[0007] Determine the corresponding target training mode according to the target dimension;

[0008] Perform vehicle abnormal noise source identification training on the abnormal noise source identification device of the test vehicle according to the target training mode, and obtain the training result.

[0009] Optionally, the obtaining the analysis result of the factors affecting the ability to identify the orientation of the vehicle abnormal noise source and determining the target dimension according to the analysis result includes:

[0010] Obtain the historical monitoring data of the orientation identification of the vehicle abnormal noise source;

[0011] Analyze the historical monitoring data to obtain the analysis results of the factors affecting the ability to identify the azimuth of the vehicle abnormal sound source;

[0012] Determine the target dimension that directly affects the ability to identify the azimuth of the abnormal sound source according to the analysis results.

[0013] Optionally, the determining the target dimension that directly affects the ability to identify the azimuth of the abnormal sound source according to the analysis results includes:

[0014] Determine the target influencing factors in the analysis results whose influence degree on the accuracy of the abnormal sound source identification device in identifying the azimuth of the abnormal sound source is higher than the preset degree threshold;

[0015] Take the target influencing factors as the target dimensions directly affecting the measurement of the ability to identify the azimuth of the abnormal sound source.

[0016] Optionally, the taking the target influencing factors as the target dimensions directly affecting the measurement of the ability to identify the azimuth of the abnormal sound source includes:

[0017] Take the sound field environment where the abnormal sound source identification device is located, the relative distance between the speaker and the abnormal sound source identification device, the relative position between the speaker and the abnormal sound source identification device, and the distance between the speakers in the target influencing factors as the target dimensions directly affecting the measurement of the ability to identify the azimuth of the abnormal sound source.

[0018] Optionally, the determining the corresponding target training mode according to the target dimensions includes:

[0019] Combine each current dimension in the target dimensions to obtain the corresponding target training mode.

[0020] Optionally, the training the abnormal sound source identification device of the test vehicle for vehicle abnormal sound source identification according to the target training mode to obtain the training results includes:

[0021] Determine the corresponding sound test data according to the target training mode, and play the sound test data through the speaker group;

[0022] Obtain the sound identification answer data of the abnormal sound source identification device of the test vehicle for identifying the sound;

[0023] Calculate the score of the correct answers in the sound identification answer data, and take the score as the training result under the current training mode.

[0024] Optionally, after training the abnormal sound source identification device of the test vehicle for vehicle abnormal sound source identification according to the target training mode to obtain the training results, the vehicle abnormal sound source identification ability training method further includes:

[0025] Determine the identification parameter fine-tuning data according to the training result, and fine-tune the relevant parameters in the abnormal sound source identification device of the test vehicle according to the identification parameter fine-tuning data.

[0026] In a second aspect, to achieve the above object, the present invention further provides a training device for vehicle abnormal sound source identification ability, and the training device for vehicle abnormal sound source identification ability includes:

[0027] A dimension determination module, configured to obtain an analysis result of factors affecting the vehicle abnormal sound source orientation identification ability, and determine a target dimension according to the analysis result;

[0028] A training mode determination module, configured to determine a corresponding target training mode according to the target dimension;

[0029] A training module, configured to perform vehicle abnormal sound source identification training on the abnormal sound source identification device of the test vehicle according to the target training mode, and obtain a training result.

[0030] In a third aspect, to achieve the above object, the present invention further provides a training device for vehicle abnormal sound source identification ability, and the training device for vehicle abnormal sound source identification ability includes: a memory, a processor, and a vehicle abnormal sound source identification ability training program stored on the memory and executable on the processor, and the vehicle abnormal sound source identification ability training program is configured to implement the steps of the vehicle abnormal sound source identification ability training method as described above.

[0031] In a fourth aspect, to achieve the above object, the present invention further provides a storage medium, on which a vehicle abnormal sound source identification ability training program is stored, and when the vehicle abnormal sound source identification ability training program is executed by a processor, the steps of the vehicle abnormal sound source identification ability training method as described above are implemented.

[0032] The vehicle abnormal sound source identification ability training method proposed by the present invention determines a target dimension according to the analysis result by obtaining the analysis result of factors affecting the vehicle abnormal sound source orientation identification ability; determines a corresponding target training mode according to the target dimension; performs vehicle abnormal sound source identification training on the abnormal sound source identification device of the test vehicle according to the target training mode, and obtains a training result; improves the accuracy of abnormal sound identification, shortens the time for troubleshooting the abnormal sound source, avoids a large amount of consumption of manpower and material resources, and improves the speed and efficiency of abnormal sound source identification. Description of the Drawings

[0033] Figure 1 It is a schematic diagram of the device structure of the hardware operating environment related to the embodiment solution of the present invention;

[0034] Figure 2 It is a schematic flowchart of the first embodiment of the vehicle abnormal sound source identification ability training method of the present invention;

[0035] Figure 3 It is a schematic flowchart of the second embodiment of the method for training the vehicle abnormal noise source identification ability of the present invention;

[0036] Figure 4 It is a schematic flowchart of the third embodiment of the method for training the vehicle abnormal noise source identification ability of the present invention;

[0037] Figure 5 It is a schematic flowchart of the fourth embodiment of the method for training the vehicle abnormal noise source identification ability of the present invention;

[0038] Figure 6 It is a schematic diagram of the 1-6 training modes in the method for training the vehicle abnormal noise source identification ability of the present invention;

[0039] Figure 7 It is a schematic diagram of the 7-12 training modes in the method for training the vehicle abnormal noise source identification ability of the present invention;

[0040] Figure 8 It is a schematic diagram of the 13-20 training modes in the method for training the vehicle abnormal noise source identification ability of the present invention;

[0041] Figure 9 It is a schematic flowchart of the fifth embodiment of the method for training the vehicle abnormal noise source identification ability of the present invention;

[0042] Figure 10 It is a schematic flowchart of the sixth embodiment of the method for training the vehicle abnormal noise source identification ability of the present invention;

[0043] Figure 11 It is a functional module diagram of the first embodiment of the device for training the vehicle abnormal noise source identification ability of the present invention.

[0044] The realization, functional characteristics and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. Specific embodiments

[0045] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0046] The solution of the embodiment of the present invention is mainly: by obtaining the analysis result of the factors affecting the vehicle abnormal noise source orientation identification ability, determining the target dimension according to the analysis result; determining the corresponding target training mode according to the target dimension; performing vehicle abnormal noise source identification training on the abnormal noise source identification device of the test vehicle according to the target training mode to obtain the training result; improving the abnormal noise recognition accuracy, shortening the time for troubleshooting the abnormal noise source, avoiding a large amount of consumption of manpower and material resources, enhancing the speed and efficiency of abnormal noise source identification, and solving the technical problems of low abnormal noise recognition accuracy, long troubleshooting time for the abnormal noise source, large consumption of manpower and material resources, and low efficiency of abnormal noise source identification in the prior art.

[0047] Reference Figure 1 , Figure 1 is a schematic diagram of the device structure of the hardware operating environment involved in the embodiment solution of the present invention.

[0048] As Figure 1 shown, the device may include: a processor 1001, such as a CPU, a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display) and an input unit such as a keyboard (Keyboard). Optionally, the user interface 1003 may further include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a Wi-Fi interface). The memory 1005 may be a high-speed RAM memory or a stable memory (Non-Volatile Memory), such as a disk memory. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0049] Those skilled in the art can understand that Figure 1 the device structure shown in

[0050] does not constitute a limitation on the device, and may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements. Figure 1 As

[0051] shown, the memory 1005, as a storage medium, may include an operating device, a network communication module, a user interface module, and a vehicle abnormal sound source identification ability training program.

[0052] The device of the present invention calls the vehicle abnormal sound source identification ability training program stored in the memory 1005 through the processor 1001 and performs the following operations:

[0053] Obtain the analysis result of the factors affecting the vehicle abnormal sound source orientation identification ability, and determine the target dimension according to the analysis result;

[0054] Determine the corresponding target training mode according to the target dimension;

[0055] Train the vehicle abnormal sound source identification device of the test vehicle according to the target training mode to obtain a training result.

[0056] The device of the present invention calls the vehicle abnormal sound source identification ability training program stored in the memory 1005 through the processor 1001 and also performs the following operations:

[0057] Obtain the historical monitoring data of the vehicle abnormal sound source orientation identification;Analyze the historical monitoring data to obtain the analysis result of the factors affecting the vehicle abnormal sound source orientation identification ability;

[0058] Determine the target dimension that directly affects the abnormal sound source orientation identification ability according to the analysis result.

[0059] The device of the present invention calls the vehicle abnormal sound source identification ability training program stored in the memory 1005 through the processor 1001, and also performs the following operations:

[0060] Determine from the analysis result the target influencing factors whose influence degree on the accuracy rate of the abnormal sound source identification device in identifying the abnormal sound source orientation is higher than the preset degree threshold;

[0061] Use the target influencing factor as the target dimension directly affecting the measurement of the abnormal sound source orientation identification ability.

[0062] The device of the present invention calls the vehicle abnormal sound source identification ability training program stored in the memory 1005 through the processor 1001, and also performs the following operations:

[0063] Use the sound field environment where the abnormal sound source identification device is located, the relative distance between the speaker and the abnormal sound source identification device, the relative position between the speaker and the abnormal sound source identification device, and the distance between the speakers in the target influencing factor as the target dimension directly affecting the measurement of the abnormal sound source orientation identification ability.

[0064] The device of the present invention calls the vehicle abnormal sound source identification ability training program stored in the memory 1005 through the processor 1001, and also performs the following operations:

[0065] Combine each current dimension in the target dimension to obtain the corresponding target training mode.

[0066] The device of the present invention calls the vehicle abnormal sound source identification ability training program stored in the memory 1005 through the processor 1001, and also performs the following operations:

[0067] Determine the corresponding sound test data according to the target training mode, and play the sound test data through the speaker group;

[0068] Obtain the sound identification answering data of the abnormal sound source identification device of the test vehicle for identifying the sound;

[0069] Calculate the score of the correct answers in the sound identification answering data, and use the score as the training result under the current training mode.

[0070] The device of the present invention calls the vehicle abnormal sound source identification ability training program stored in the memory 1005 through the processor 1001, and also performs the following operations:

[0071] Determine the identification parameter fine-tuning data according to the training result, and fine-tune the relevant parameters in the abnormal sound source identification device of the test vehicle according to the identification parameter fine-tuning data.

[0072] In this embodiment, through the above solution, by obtaining the analysis result of the factors affecting the vehicle abnormal sound source orientation identification ability, determine the target dimension according to the analysis result; determine the corresponding target training mode according to the target dimension; perform vehicle abnormal sound source identification training on the abnormal sound source identification device of the test vehicle according to the target training mode, and obtain the training result; improve the accuracy of abnormal sound recognition, shorten the time for troubleshooting the abnormal sound source, avoid a large amount of consumption of manpower and material resources, and improve the speed and efficiency of abnormal sound source identification.

[0073] Based on the above hardware structure, an embodiment of the vehicle abnormal sound source identification ability training method of the present invention is proposed.

[0074] Refer to Figure 2 , Figure 2 which is a schematic flowchart of the first embodiment of the vehicle abnormal sound source identification ability training method of the present invention.

[0075] In the first embodiment, the vehicle abnormal sound source identification ability training method includes the following steps:

[0076] Step S10: Obtain the analysis result of the factors affecting the vehicle abnormal sound source orientation identification ability, and determine the target dimension according to the analysis result.

[0077] It should be noted that through the analysis of the factors affecting the vehicle abnormal sound source orientation identification ability, multiple dimensions of the vehicle abnormal sound source orientation identification ability system can be determined.

[0078] Step S20: Determine the corresponding target training mode according to the target dimension.

[0079] It can be understood that by further analyzing the target dimension, the corresponding training mode can be determined, and different training dimensions correspond to different training modes.

[0080] Step S30: Perform vehicle abnormal sound source identification training on the abnormal sound source identification device of the test vehicle according to the target training mode, and obtain the training result.

[0081] It should be understood that through the target training mode, vehicle abnormal sound source identification training can be performed on the abnormal sound source identification device of the test vehicle, so as to obtain the corresponding training result, and these training modes can comprehensively and deeply train the orientation identification ability of the abnormal sound source identification device.

[0082] In this embodiment, through the above solution, by obtaining the analysis result of the factors affecting the vehicle abnormal noise source orientation identification ability, determining the target dimension according to the analysis result; determining the corresponding target training mode according to the target dimension; training the abnormal noise source identification device of the test vehicle according to the target training mode to obtain the training result; improving the accuracy of abnormal noise identification, shortening the time for troubleshooting the abnormal noise source, avoiding a large amount of consumption of manpower and material resources, and enhancing the speed and efficiency of abnormal noise source identification.

[0083] Further, Figure 3 It is a schematic flowchart of the second embodiment of the vehicle abnormal noise source identification ability training method of the present invention. As Figure 3 shown, based on the first embodiment, the second embodiment of the vehicle abnormal noise source identification ability training method of the present invention is proposed. In this embodiment, the step S10 specifically includes the following steps:

[0084] Step S11, obtain the historical monitoring data of vehicle abnormal noise source orientation identification.

[0085] It should be noted that the historical monitoring data is the data generated during the identification process of vehicle abnormal noise source orientation identification for a period of time.

[0086] Step S12, analyze the historical monitoring data to obtain the analysis result of the factors affecting the vehicle abnormal noise source orientation identification ability.

[0087] It can be understood that by analyzing the historical monitoring data, the analysis result corresponding to the relevant factors affecting the vehicle abnormal noise source orientation identification ability can be obtained.

[0088] Step S13, determine the target dimension that directly affects the abnormal noise source orientation identification ability according to the analysis result.

[0089] It should be understood that through the analysis result, the relevant factors that have a greater impact on the abnormal noise source orientation identification ability can be determined, and then the relevant factors can be determined as the target dimension.

[0090] In this embodiment, through the above solution, by obtaining the historical monitoring data of vehicle abnormal noise source orientation identification; analyzing the historical monitoring data to obtain the analysis result of the factors affecting the vehicle abnormal noise source orientation identification ability; determining the target dimension that directly affects the abnormal noise source orientation identification ability according to the analysis result; it can accurately determine the abnormal noise source identification dimension and enhance the speed and efficiency of abnormal noise source identification.

[0091] Further, Figure 4 It is a schematic flowchart of the third embodiment of the vehicle abnormal noise source identification ability training method of the present invention. As Figure 4As shown, based on the second embodiment, the third embodiment of the method for training the vehicle abnormal noise source identification ability of the present invention is proposed. In this embodiment, the step S13 specifically includes the following steps:

[0092] Step S131: Determine, from the analysis results, a target influencing factor whose influence degree on the accuracy of the abnormal noise source identification device in identifying the azimuth of the abnormal noise source is higher than a preset degree threshold.

[0093] It should be noted that relevant factors for the abnormal noise source identification device to identify the azimuth of the abnormal noise source can be determined from the analysis results, and the factors with a relatively high influence degree on the accuracy of identifying the azimuth of the abnormal noise source among the relevant factors are used as target influencing factors. The preset degree threshold is a threshold for different factors to affect the accuracy at different levels, which can be set to other values such as 80%, 90%, and 95%, and this embodiment does not limit this.

[0094] Step S132: Use the target influencing factor as the target dimension directly affecting the measurement of the abnormal noise source azimuth identification ability.

[0095] It can be understood that after determining the target influencing factor, the target influencing factor can be used as the target dimension directly affecting the measurement of the abnormal noise source azimuth identification ability.

[0096] Further, the step S132 specifically includes the following steps:

[0097] Use the sound field environment where the abnormal noise source identification device is located, the relative distance between the speaker and the abnormal noise source identification device, the relative position between the speaker and the abnormal noise source identification device, and the distance between the speakers among the target influencing factors as the target dimensions directly affecting the measurement of the abnormal noise source azimuth identification ability.

[0098] It can be understood that the factors that have a greater impact on the measurement of the abnormal noise source azimuth identification ability include the sound field environment where the abnormal noise source identification device is located, the relative distance between the speaker and the abnormal noise source identification device, the relative position between the speaker and the abnormal noise source identification device, and the distance between the speakers. Therefore, the above factors can be used as the target dimensions directly affecting the measurement of the abnormal noise source azimuth identification ability.

[0099] In a specific implementation, the first dimension is the sound field environment in which the abnormal sound source identification device to be trained is located. Generally, there are two types: free sound field and reverberant sound field, which can be named S1 and S2 respectively; the second dimension is the distance between the speaker and the abnormal sound source identification device to be trained; in this embodiment, 2 distances can be set, named R1 and R2 respectively; the distance corresponding to R1 is less than the distance corresponding to R2; the third dimension is the relative position type between the speaker and the abnormal sound source identification device to be trained; in this embodiment, 8 orientations can be set, namely front, rear, left, right, front left, front right, rear left, and rear right; correspondingly, 3 relative position types can be set: the first type includes 4 orientations: front, rear, left, and right, and can be named P1; the second type includes 4 orientations: front left, front right, rear left, and rear right, and can be named P2; the third type includes all the above 8 orientations and can be named P3; the fourth dimension is the distance between the speakers. In this embodiment, 2 distances can be set, named D1 and D2 respectively; the distance corresponding to D1 is less than the distance corresponding to D2.

[0100] Through the above solution in this embodiment, by determining from the analysis results the target influencing factors whose influence degree on the accuracy of identifying the abnormal sound source orientation by the abnormal sound source identification device is higher than the preset degree threshold; and taking the target influencing factors as the target dimensions directly affecting the measurement of the abnormal sound source orientation identification ability, the abnormal sound source identification dimensions can be accurately determined, and the speed and efficiency of abnormal sound source identification are improved.

[0101] Furthermore, Figure 5 is a flowchart of the fourth embodiment of the method for training the abnormal sound source identification ability of the vehicle of the present invention. As Figure 5 shown, based on the first embodiment, the fourth embodiment of the method for training the abnormal sound source identification ability of the vehicle of the present invention is proposed. In this embodiment, the step S20 specifically includes the following steps:

[0102] Step S21, combine each current dimension in the target dimension to obtain the corresponding target training mode.

[0103] It should be noted that by further analyzing and combining each current dimension in the target dimension, multiple training modes can be combined, and these training modes can comprehensively and deeply train the orientation identification ability of the abnormal sound source identification device.

[0104] In a specific implementation, it can be set that the first dimension is in the S1 (free sound field) state, and the second and third dimensions are combined to obtain 6 training modes. As Figure 6 shown, Figure 6 is a schematic diagram of the 1-6 training modes in the method for training the abnormal sound source identification ability of the vehicle of the present invention; Figure 6It contains schematic diagrams of 6 training modes. All the schematic diagrams are top views of the training system; the abnormal sound source identification device to be trained is located at the center of the circle; the black squares are schematic diagrams of the speaker groups, and each speaker is located on the same circumference; the sound emission directions of each speaker all point to the center of the circle.

[0105] In mode 1, the second dimension is the R1 state (the distance between the abnormal sound source identification device to be trained and the speaker is relatively close), and the third dimension is the P1 state. Considering that the first dimension is the S1 state, mode 1 can be named: S1-R1-P1.

[0106] In mode 1, the 4 speakers are respectively located at the front, rear, left, and right of the abnormal sound source identification device to be trained.

[0107] Similarly, mode 2 is named: S1-R1-P2; mode 3 is named: S1-R1-P3.

[0108] In mode 2, the 4 speakers are respectively located at the front left, front right, rear left, and rear right of the abnormal sound source identification device to be trained; in mode 3, the 8 speakers are respectively located at the front, rear, left, right, front left, front right, rear left, and rear right of the abnormal sound source identification device to be trained.

[0109] In mode 4, the second dimension is the R2 state (the distance between the abnormal sound source identification device to be trained and the sound source is relatively far), and the third dimension is the P1 state. Mode 1 can be named: S1-R2-P1.

[0110] Similarly, mode 5 is named: S1-R2-P2; mode 6 is named: S1-R2-P3.

[0111] Suppose the first dimension is in the S2 (reverberant sound field) state. By combining the second and third dimensions, 6 more training modes can be obtained, as Figure 7 shown, Figure 7 This is the schematic diagram of the 7-12 training modes in the method for training the abnormal sound source identification ability of the vehicle of the present invention.

[0112] Figure 7 It contains schematic diagrams of 6 training modes; all the schematic diagrams are top views of the training system; the abnormal sound source identification device to be trained is located at the center of the circle; the black squares are schematic diagrams of the speaker groups, and each speaker is located on the same circumference; the sound emission directions of each speaker all point to the center of the circle.

[0113] Compared with Figure 6 , Figure 7 each mode in it has a black rectangle around it, which represents the reverberant sound field to distinguish it from the Figure 6 free sound field in

[0114] Following the naming rules of Pattern 1 to Pattern 6, Pattern 7 to Pattern 12 are respectively named as: S2-R1-P1, S2-R1-P2, S2-R1-P3, S2-R2-P1, S2-R2-P2, S2-R2-P3.

[0115] Furthermore, by combining the first, second, and fourth dimensions, 8 training patterns are obtained, as Figure 8 shown; Figure 8 This is the schematic diagram of Training Patterns 13 - 20 in the vehicle abnormal noise source identification ability training method of the present invention.

[0116] Figure 8 It contains schematic diagrams of 8 training patterns. In each pattern, the black square is the schematic diagram of the speaker, and each speaker is arranged in a planar array form, and the plane is perpendicular to the horizontal plane; the abnormal noise source identification device to be trained faces (is directly opposite to) the speaker planar array; the sound emission direction of each speaker points to the abnormal noise source identification device to be trained; Figure 8 All the schematic diagrams in

[0117] Figure 8 In Pattern 13, the first dimension is the S1 state, the second dimension is the R1 state (the distance between the abnormal noise source identification device to be trained and the speaker is relatively close), and the fourth dimension is the D1 state; it can be considered that Pattern 13 is named as: S1-R1-D1.

[0118] Similarly, Pattern 14 to Pattern 20 are respectively named as: S1-R1-D2, S1-R2-D1, S1-R2-D2, S2-R1-D1, S2-R1-D2, S2-R2-D1, S2-R2-D2.

[0119] Compared with Pattern 13 to Pattern 16, there is a black rectangle around Pattern 17 to Pattern 20, which represents the reverberant sound field to distinguish it from the free sound field in Pattern 13 to Pattern 16.

[0120] Figure 8 In each pattern, the relative position of the speaker array and the abnormal noise source identification device to be trained is that the abnormal noise source identification device to be trained faces (is directly opposite to) the speaker array; if necessary, the relative position of the speaker array and the abnormal noise source identification device to be trained can be changed; for example, make the speaker array be located at the left front of the abnormal noise source identification device to be trained. At this time, the abnormal noise source identification device to be trained can be rotated 45° to the right around the original position.

[0121] Furthermore, a fifth dimension - azimuth dimension - can be added to the training pattern; the 8 azimuths of front, back, left, right, left front, right front, left rear, and right rear are respectively named as O1, O2, O3, O4, O5, O6, O7, O8; thus, Figure 8The naming of Mode 13 to Mode 20 in the medium mode are S1-R1-D1-O1, S1-R1-D2-O1, S1-R2-D1-O1, S1-R2-D2-O1, S2-R1-D1-O1, S2-R1-D2-O1, S2-R2-D1-O1, and S2-R2-D2-O1 respectively.

[0122] It should be noted that the third dimension and the fifth dimension are both related and different. The third dimension is a further classification based on the fifth dimension.

[0123] So far, without adding the fifth dimension, the total number of training modes is Figures 6 to 8 a total of 20. If the fifth dimension is added, it is Figure 8 the 8 in Figures 6 to 7 multiplied by 8 (8 directions), plus

[0124] Through the above solution, in this embodiment, by combining each current dimension in the target dimension, the corresponding target training mode is obtained, and different training modes can be determined for subsequent abnormal sound source identification ability training, improving the accuracy of abnormal sound identification and shortening the time for abnormal sound source investigation.

[0125] Furthermore, Figure 9 is a schematic flowchart of the fifth embodiment of the method for training the abnormal sound source identification ability of the vehicle according to the present invention. As Figure 9 shown, based on the first embodiment, the fifth embodiment of the method for training the abnormal sound source identification ability of the vehicle according to the present invention is proposed. In this embodiment, the step S30 specifically includes the following steps:

[0126] Step S31: Determine the corresponding sound test data according to the target training mode, and play the sound test data through the speaker group.

[0127] It should be noted that the corresponding sound data for training can be determined through the target training mode, and the sound test data can be played through the speaker group.

[0128] In specific implementation, for each training mode, the microprocessor randomly selects a certain loudspeaker from the corresponding loudspeaker group, that is: assigns a specific value to each loudspeaker, and each value corresponds to each loudspeaker one by one; randomly selects a value from these values, and sends sound data to it for playback; this embodiment involves two sound field environments, namely free sound field and reverberant sound field; the free sound field can be realized by installing sound-absorbing cotton on the surrounding and top walls inside a closed room, and the reverberant sound field is realized by the smooth and hard walls around and up and down inside the closed room; the role of the microprocessor (including the power module) is to receive the answering data sent by the abnormal sound source identification device (including the display device), and return the judgment result of the answering data to the abnormal sound source identification device (including the display device); another role of the microprocessor (including the power module) is to send sound data to a certain loudspeaker in the loudspeaker group, so that a certain loudspeaker plays a sound signal; at the same time, the microprocessor (including the power module) can only send sound data to one loudspeaker in the loudspeaker group; another role of the microprocessor (including the power module) is to supply power to the loudspeaker group and the abnormal sound source identification device (including the display device).

[0129] Step S32: Obtain the sound identification answering data of the abnormal sound source identification device of the test vehicle for identifying sounds.

[0130] It can be understood that the abnormal sound source identification device can identify the sound generated by the loudspeaker group playing the sound test data, and thus can judge the abnormal sound source and generate the answering data for the abnormal sound source judgment.

[0131] In specific implementation, the role of the loudspeaker group is to receive the sound data sent by the microprocessor (including the power module) and play it; the role of the abnormal sound source identification device (including the display device) is to send the answering data to the microprocessor (including the power module); after the microprocessor (including the power module) returns the answering judgment result, the result is displayed on the display device of the abnormal sound source identification device.

[0132] Step S33: Calculate the score of the correct answers in the sound identification answering data, and use the score as the training result in the current training mode.

[0133] It should be understood that after obtaining the sound identification answering data, the corresponding scores of each sound identification answering data can be calculated, and thus the corresponding scores can be used as the training result in the current training mode.

[0134] In a specific implementation, the abnormal sound source identification device sends the answering data to the microprocessor. The microprocessor processes the answering data, determines whether the answer of the trained abnormal sound source identification device is correct, and returns the determination result (whether it is correct and what the correct answer is) to the abnormal sound source identification device to strengthen or correct the judgment of the trained abnormal sound source identification device on the sound source direction recognition. After each playback of the sound data, answering questions and immediately feedback of the answering results are carried out, and then the training of the next mode is carried out.

[0135] After all the training modes are completed, the training score can be calculated. Assuming that the full score of the training score is 100 points and the total number of training modes is model_num, then the score of each mode model_score is:

[0136] model_score = 100 / model_num

[0137] Among all the training modes, the number of modes in the free sound field and the number of modes in the reverberant sound field each account for half. Assuming that the number of modes answered correctly in the free sound field is free_num and the number of modes answered correctly in the reverberant sound field is reverb_num.

[0138] In actual work, when analyzing the abnormal sound source of a vehicle, the sound field environment where the abnormal sound performance development abnormal sound source identification device is located is closer to the reverberant sound field. Considering this situation, it is necessary to set weight coefficients for the scores in the two cases of the free sound field and the reverberant sound field. The weight coefficient of the reverberant sound field is higher than that of the free sound field. Assuming that the weight coefficient of the free sound field is free_ratio, its value is between 0 and 1; the weight coefficient of the reverberant sound field is reverb_ratio, its value is between 0 and 1; the sum of free_ratio and reverb_ratio is 1; the typical value of free_ratio is 0.4, and the typical value of reverb_ratio is 0.6.

[0139] The training score total_score is calculated according to the following formula:

[0140] total_score = free_num * model_score * free_ratio + reverb_num * model_score * reverb_ratio

[0141] The calculation of the training score is completed by the microprocessor; after the calculation is completed, it is displayed on the display device of the abnormal sound source identification device; among the five dimensions of the training mode, the number of types in each dimension can be added; for example, in the second dimension, 3 distance categories can be set.

[0142] In this embodiment, through the above solution, the corresponding sound test data is determined according to the target training mode, and the sound test data is played through the speaker group; the sound identification answer data of the sound identified by the abnormal sound source identification device of the test vehicle is obtained; the score of the correct answer in the sound identification answer data is calculated, and the score is used as the training result in the current training mode; the accuracy of abnormal sound identification is improved, the time for troubleshooting the abnormal sound source is shortened, a large amount of human and material resources are avoided, and the speed and efficiency of abnormal sound source identification are improved.

[0143] Further, Figure 10 is a schematic flowchart of the sixth embodiment of the method for training the abnormal sound source identification ability of a vehicle according to the present invention. As Figure 10 shown, based on the first embodiment, the sixth embodiment of the method for training the abnormal sound source identification ability of a vehicle according to the present invention is proposed. In this embodiment, after the step S30, the method for training the abnormal sound source identification ability of the vehicle further includes the following steps:

[0144] Step S40: Determine the identification parameter fine-tuning data according to the training result, and fine-tune the relevant parameters in the abnormal sound source identification device of the test vehicle according to the identification parameter fine-tuning data.

[0145] It should be understood that the identification parameter fine-tuning data can be determined through the training result, that is, different training results correspond to different parameter fine-tuning data. Different parameter fine-tuning data can be used to fine-tune the relevant parameters in the abnormal sound source identification device of the test vehicle to different degrees, and timely correct and adjust the identification and judgment ability of the abnormal sound source identification device of the test vehicle, further improving the identification efficiency of the abnormal sound source.

[0146] In this embodiment, through the above solution, the identification parameter fine-tuning data is determined through the training result, and the relevant parameters in the abnormal sound source identification device of the test vehicle are fine-tuned according to the identification parameter fine-tuning data, improving the accuracy of abnormal sound identification, shortening the time for troubleshooting the abnormal sound source, avoiding a large amount of human and material resources, and enhancing the speed and efficiency of abnormal sound source identification.

[0147] Correspondingly, the present invention further provides a device for training the abnormal sound source identification ability of a vehicle.

[0148] Referring to Figure 11 , Figure 11 is a functional module diagram of the first embodiment of the device for training the abnormal sound source identification ability of a vehicle according to the present invention.

[0149] In the first embodiment of the device for training the abnormal sound source identification ability of a vehicle according to the present invention, the device for training the abnormal sound source identification ability of a vehicle includes:

[0150] A dimension determination module 10, configured to obtain an analysis result of factors affecting the vehicle abnormal noise source orientation identification ability, and determine a target dimension according to the analysis result.

[0151] A training mode determination module 20, configured to determine a corresponding target training mode according to the target dimension.

[0152] A training module 30, configured to perform vehicle abnormal noise source identification training on the abnormal noise source identification device of a test vehicle according to the target training mode, and obtain a training result.

[0153] The dimension determination module 10 is further configured to obtain historical monitoring data of vehicle abnormal noise source orientation identification; analyze the historical monitoring data to obtain an analysis result of factors affecting the vehicle abnormal noise source orientation identification ability; determine a target dimension directly affecting the abnormal noise source orientation identification ability according to the analysis result.

[0154] The dimension determination module 10 is further configured to determine a target influencing factor in the analysis result whose influence degree on the accuracy rate of the abnormal noise source identification device for identifying the abnormal noise source orientation is higher than a preset degree threshold; use the target influencing factor as the target dimension directly affecting the measurement of the abnormal noise source orientation identification ability.

[0155] The dimension determination module 10 is further configured to use the sound field environment where the abnormal noise source identification device is located, the relative distance between the speaker and the abnormal noise source identification device, the relative position between the speaker and the abnormal noise source identification device, and the distance between the speakers among the target influencing factors as the target dimension directly affecting the measurement of the abnormal noise source orientation identification ability.

[0156] The training mode determination module 20 is further configured to combine each current dimension in the target dimension to obtain a corresponding target training mode.

[0157] The training module 30 is further configured to determine corresponding sound test data according to the target training mode, play the sound test data through a speaker group; obtain sound identification answering data of the abnormal noise source identification device of the test vehicle for identifying sounds; calculate the score of the correct answers in the sound identification answering data, and use the score as the training result under the current training mode.

[0158] The training module 30 is further configured to determine identification parameter fine-tuning data according to the training result, and fine-tune relevant parameters in the abnormal noise source identification device of the test vehicle according to the identification parameter fine-tuning data.

[0159] Wherein, the steps implemented by each functional module of the vehicle abnormal noise source identification ability training device can refer to each embodiment of the vehicle abnormal noise source identification ability training method of the present invention, which will not be elaborated here.

[0160] In addition, an embodiment of the present invention further provides a storage medium, on which a vehicle abnormal sound source identification ability training program is stored. When the vehicle abnormal sound source identification ability training program is executed by a processor, the following operations are implemented:

[0161] Obtain the analysis result of the factors affecting the vehicle abnormal sound source orientation identification ability, and determine the target dimension according to the analysis result;

[0162] Determine the corresponding target training mode according to the target dimension;

[0163] Perform vehicle abnormal sound source identification training on the abnormal sound source identification device of the test vehicle according to the target training mode, and obtain the training result.

[0164] Further, when the vehicle abnormal sound source identification ability training program is executed by a processor, the following operations are also implemented:

[0165] Obtain the historical monitoring data of vehicle abnormal sound source orientation identification;

[0166] Analyze the historical monitoring data to obtain the analysis result of the factors affecting the vehicle abnormal sound source orientation identification ability;

[0167] Determine the target dimension that directly affects the abnormal sound source orientation identification ability according to the analysis result.

[0168] Further, when the vehicle abnormal sound source identification ability training program is executed by a processor, the following operations are also implemented:

[0169] Determine the target influencing factor in the analysis result whose influence degree on the accuracy of the abnormal sound source identification device in identifying the abnormal sound source orientation is higher than the preset degree threshold;

[0170] Use the target influencing factor as the target dimension directly affecting the measurement of the abnormal sound source orientation identification ability.

[0171] Further, when the vehicle abnormal sound source identification ability training program is executed by a processor, the following operations are also implemented:

[0172] Use the sound field environment where the abnormal sound source identification device is located, the relative distance between the speaker and the abnormal sound source identification device, the relative position between the speaker and the abnormal sound source identification device, and the distance between the speakers among the target influencing factors as the target dimension directly affecting the measurement of the abnormal sound source orientation identification ability.

[0173] Further, when the vehicle abnormal sound source identification ability training program is executed by a processor, the following operations are also implemented:

[0174] Combine each current dimension in the target dimension to obtain the corresponding target training mode.

[0175] Further, when the vehicle abnormal noise source identification ability training program is executed by a processor, the following operations are also implemented:

[0176] Determine corresponding sound test data according to the target training mode, and play the sound test data through the speaker group;

[0177] Obtain the sound identification answer data of the abnormal noise source identification device of the test vehicle for identifying the sound;

[0178] Calculate the score of the correct answers in the sound identification answer data, and use the score as the training result under the current training mode.

[0179] Further, when the vehicle abnormal noise source identification ability training program is executed by a processor, the following operations are also implemented:

[0180] Determine the identification parameter fine-tuning data according to the training result, and fine-tune the relevant parameters in the abnormal noise source identification device of the test vehicle according to the identification parameter fine-tuning data.

[0181] In this embodiment, through the above solution, by obtaining the analysis result of the factors affecting the vehicle abnormal noise source orientation identification ability, determining the target dimension according to the analysis result; determining the corresponding target training mode according to the target dimension; performing vehicle abnormal noise source identification training on the abnormal noise source identification device of the test vehicle according to the target training mode, and obtaining the training result; improving the abnormal noise recognition accuracy, shortening the time for troubleshooting the abnormal noise source, avoiding a large amount of consumption of manpower and material resources, and enhancing the speed and efficiency of abnormal noise source identification.

[0182] It should be noted that in this article, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the phrase "including a..." does not exclude the existence of additional identical elements in the process, method, article or device including the element.

[0183] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments.

[0184] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. A method for training the ability to identify the source of abnormal vehicle noises, characterized in that, the method for training the ability to identify the source of abnormal vehicle noises includes: obtaining an analysis result of factors affecting the ability to identify the azimuth of the source of abnormal vehicle noises, and determining a target dimension according to the analysis result; determining a corresponding target training mode according to the target dimension; performing vehicle abnormal noise source identification training on the abnormal noise source identification device of the test vehicle according to the target training mode to obtain a training result; wherein, the obtaining an analysis result of factors affecting the ability to identify the azimuth of the source of abnormal vehicle noises and determining a target dimension according to the analysis result includes: obtaining historical monitoring data of the azimuth identification of the source of abnormal vehicle noises; analyzing the historical monitoring data to obtain an analysis result of factors affecting the ability to identify the azimuth of the source of abnormal vehicle noises; determining a target dimension that directly affects the ability to identify the azimuth of the source of abnormal noises according to the analysis result; wherein, the determining a target dimension that directly affects the ability to identify the azimuth of the source of abnormal noises according to the analysis result includes: determining a target influencing factor in the analysis result whose influence degree on the accuracy rate of the abnormal noise source identification device in identifying the azimuth of the abnormal noise source is higher than a preset degree threshold; using the target influencing factor as the target dimension directly affecting the measurement of the ability to identify the azimuth of the abnormal noise source; wherein, the using the target influencing factor as the target dimension directly affecting the measurement of the ability to identify the azimuth of the abnormal noise source includes: using the sound field environment where the abnormal noise source identification device is located, the relative distance between the speaker and the abnormal noise source identification device, the relative position between the speaker and the abnormal noise source identification device, and the distance between the speakers in the target influencing factor as the target dimension directly affecting the measurement of the ability to identify the azimuth of the abnormal noise source; wherein, the determining a corresponding target training mode according to the target dimension includes: combining each current dimension in the target dimension to obtain a corresponding target training mode; wherein, after performing vehicle abnormal noise source identification training on the abnormal noise source identification device of the test vehicle according to the target training mode to obtain a training result, the method for training the ability to identify the source of abnormal vehicle noises further includes: determining fine-tuning data for identification parameters according to the training result, and fine-tuning relevant parameters in the abnormal noise source identification device of the test vehicle according to the fine-tuning data for identification parameters.

2. The method for training the ability to identify the source of abnormal vehicle noises according to claim 1, characterized in that, the performing vehicle abnormal noise source identification training on the abnormal noise source identification device of the test vehicle according to the target training mode to obtain a training result includes: determining corresponding sound test data according to the target training mode, and playing the sound test data through a speaker group; obtaining sound identification answer data of the abnormal noise source identification device of the test vehicle for identifying sounds; calculating the score of the correct answers in the sound identification answer data, and using the score as the training result under the current training mode.

3. A device for training the ability to identify the source of abnormal vehicle noises, characterized in that, the device for training the ability to identify the source of abnormal vehicle noises includes: a dimension determination module, configured to obtain an analysis result of factors affecting the ability to identify the azimuth of the source of abnormal vehicle noises, and determine a target dimension according to the analysis result; A training mode determination module, configured to determine a corresponding target training mode according to the target dimension; A training module, configured to perform vehicle abnormal sound source identification training on the abnormal sound source identification device of the test vehicle according to the target training mode, and obtain a training result; The dimension determination module is further configured to obtain historical monitoring data of vehicle abnormal sound source orientation identification; analyze the historical monitoring data to obtain an analysis result of factors affecting the vehicle abnormal sound source orientation identification ability; determine a target dimension directly affecting the abnormal sound source orientation identification ability according to the analysis result; The dimension determination module is further configured to determine a target influencing factor in the analysis result whose influence degree on the accuracy of the abnormal sound source identification device in identifying the abnormal sound source orientation is higher than a preset degree threshold; use the target influencing factor as a target dimension directly affecting the measurement of the abnormal sound source orientation identification ability; The dimension determination module is further configured to use the sound field environment in which the abnormal sound source identification device is located, the relative distance between the speaker and the abnormal sound source identification device, the relative position between the speaker and the abnormal sound source identification device, and the distance between the speakers among the target influencing factors as target dimensions directly affecting the measurement of the abnormal sound source orientation identification ability; The training mode determination module is further configured to combine each current dimension in the target dimension to obtain a corresponding target training mode; The training module is further configured to determine identification parameter fine-tuning data according to the training result, and fine-tune relevant parameters in the abnormal sound source identification device of the test vehicle according to the identification parameter fine-tuning data.

4. A vehicle abnormal sound source identification ability training device Characterized in that The vehicle abnormal sound source identification ability training device includes: a memory, a processor, and a vehicle abnormal sound source identification ability training program stored on the memory and executable on the processor, and the vehicle abnormal sound source identification ability training program is configured to implement the steps of the vehicle abnormal sound source identification ability training method according to any one of claims 1 to 2.

5. A storage medium Characterized in that The storage medium stores a vehicle abnormal sound source identification ability training program, and when the vehicle abnormal sound source identification ability training program is executed by a processor, it implements the steps of the vehicle abnormal sound source identification ability training method according to any one of claims 1 to 2.

Images