An automobile part noise testing system and testing method

Abstract

This invention discloses a noise testing system and method for automotive components, relating to the field of noise testing technology. The method includes the following steps: acquiring the sound pressure amplitude, frequency position, time markers, and excitation rhythm intervals of automotive components as their speed increases during noise testing; calculating the amplitude change rate and frequency offset corresponding to each time marker; and forming a continuous recording segment containing traces of multi-source changes. This invention, by constructing continuous recording segments and introducing amplitude change rate and frequency offset, achieves continuous characterization of the dynamic changes of multi-source noise, avoiding the misjudgment of superimposed signals as stable states. Simultaneously, by reverse expansion and segmented rearrangement to form a time evolution path and guiding subsequent changes forward, it separates and presents multi-source changes temporally, thereby improving anomaly identification capabilities and enhancing the accuracy of fault trend analysis.

Description

Technical Field

[0001] This invention relates to the field of noise testing technology, and specifically to a noise testing system and method for automotive parts. Background Technology

[0002] Noise testing of automotive components refers to the process of quantitatively collecting and analyzing the sound radiation and structural vibration generated by individual components such as engine accessories, transmission mechanisms, braking components, and interior components, or their assembly states, under controlled operating conditions and specified environmental conditions. This is achieved through sound pressure sensors, vibration acquisition devices, and supporting analysis systems. The testing typically incorporates boundary conditions such as rotational speed, load, temperature, or assembly constraints to decompose the noise signal in the time and frequency domains, extracting key indicators such as sound pressure level, spectral distribution, and characteristic peak values. This allows for the identification of noise sources, assessment of structural transmission paths, and determination of abnormal vibration coupling relationships, ultimately providing a basis for component structural optimization, material selection, and improvement of overall vehicle sound quality.

[0003] The existing technology has the following shortcomings:

[0004] During noise testing of automotive components, when the test object is subjected to continuous periodic excitation, the abnormal noise generated in different parts usually exhibits a dispersed and varied state. However, within a specific time period, the changing rhythms of each abnormal source will briefly align, causing multiple sets of abnormal signals that were originally independent to superimpose synchronously in time. In the acquisition results, this appears as a single signal with stable amplitude and consistent fluctuation patterns. When the test system analyzes the noise data, it will identify this superimposed signal as a stable operating state, thereby ignoring the multi-source abnormal information contained within it. This leads to the masking of multi-source coupling characteristics, ultimately affecting the accuracy of abnormal identification and weakening the ability to judge potential fault risks.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a noise testing system and method for automotive parts to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for testing the noise of automotive components, comprising the following steps:

[0008] The system acquires the sound pressure amplitude, frequency position, time markers, and excitation rhythm intervals of automotive parts as the speed increases during noise testing, and calculates the amplitude change rate and frequency offset corresponding to each time marker to form a continuous recording segment containing multi-source change traces.

[0009] Based on the amplitude change rate and frequency offset in continuous recording segments, difference processing is performed on adjacent time markers to extract segments with similar changes. By comparing the frequency offset, segments with synchronously increasing amplitude and continuously changing frequency offset differences are selected to determine suspected synchronous superposition segments and their corresponding duration range.

[0010] For the amplitude change rate in the continuous recording segment corresponding to the suspected synchronous superposition section, reverse expansion processing is performed according to the frequency offset in the continuous recording segment. The overlapping change trajectory is staggered according to the time sequence and pulled apart point by point according to the frequency difference to restore the multi-segment progressive change pattern.

[0011] By utilizing a multi-segment progressive change pattern, the sound pressure amplitude within the corresponding time range is segmented and rearranged, and the concentrated change amplitude is distributed to adjacent time markers in an increasing order, while maintaining the continuous change relationship of frequency position, thus forming a time evolution path of multi-source change.

[0012] Based on the time evolution path, the amplitude changes that subsequently enter the testing process are guided to move forward, embedding changes that conform to the progressive characteristics into the unfolded segment, and continuously advancing along the time marker, thereby suppressing the synchronous superposition phenomenon under the periodic excitation.

[0013] Preferably, in the process of continuously characterizing and extracting the changing trend of sound pressure information, by synchronously acquiring the sound pressure amplitude and frequency position and expanding the recording of the changing characteristics, the multi-source change traces are continuously described, and continuous recording segments are constructed. The steps are as follows:

[0014] Noise is collected at different speed change nodes to obtain sound pressure amplitude and frequency position and assign time stamps. At the same time, the excitation rhythm interval between adjacent time stamps is recorded to establish a correspondence.

[0015] The sound pressure amplitude is processed point by point along the time marker sequence. The rate of change of amplitude is determined by the difference between adjacent time markers and the excitation rhythm interval. The frequency offset is obtained by comparing the frequency position and then the two are correlated.

[0016] The sound pressure amplitude, frequency position, amplitude change rate and frequency offset are uniformly organized and bound according to the time marker, and the excitation rhythm interval is inserted to obtain a continuously arranged data set.

[0017] The dataset is spliced ​​along the time marker direction to obtain a continuous recording segment covering the speed increase process, so that each data item maintains a corresponding relationship for multi-source change trace identification.

[0018] Preferably, the sound pressure amplitude, frequency position, amplitude change rate, and frequency offset in the continuous recording segment are bound point by point through the time stamp sequence, and the excitation rhythm interval is inserted between adjacent time stamps. The data corresponding to each time stamp are arranged continuously, thereby maintaining the one-to-one correspondence between the sound pressure amplitude and the frequency position and change rate and maintaining the continuous distribution of the time span relationship.

[0019] Preferably, in the process of identifying convergent change segments and determining synchronous overlay segments, the synchronous change segments are extracted and their duration range is determined by comparing the difference in amplitude change rates with the frequency offset. The steps are as follows:

[0020] The amplitude change rate is read point by point along the time markers in the continuous recording segment. The difference between the amplitude change rates corresponding to adjacent time markers is calculated, and the change difference is bound to the time marker and recorded.

[0021] Based on the change difference, the time markers are continuously scanned, and time markers with consistent difference characteristics are connected and divided into segments with similar change. At the same time, the rate of change of amplitude and the change difference within the segment are recorded.

[0022] For segments with similar changes, the frequency offset is read point by point. Alignment comparison is performed on the frequency offsets corresponding to adjacent time markers, and time markers with continuously increasing amplitude change rates and consistent frequency offset change directions are selected as synchronous change segments.

[0023] By combining synchronous change segments with time markers, a suspected synchronous overlay segment is obtained. At the same time, the time interval corresponding to the start time marker and the end time marker is recorded as the duration range.

[0024] Preferably, during the frequency offset alignment and comparison process, the frequency offsets corresponding to adjacent time markers are compared point by point and the direction and magnitude of change are recorded. Time markers with the same direction of frequency offset change are filtered out, and time markers with a continuously increasing amplitude change rate are combined. Synchronous change segments are obtained by continuously connecting time markers, and the amplitude change rate and frequency offset corresponding to each time marker are retained.

[0025] Preferably, for the decomposition and progressive change pattern recovery of overlapping trajectories, the overlapping trajectories are misaligned and sequentially unfolded by leveraging the correspondence between the amplitude change rate and the frequency offset. The steps are as follows:

[0026] The amplitude change rate is read point by point along the time marker sequence of the continuous recording segment corresponding to the suspected synchronous superposition section, and the frequency offset is read synchronously. The time markers with the same change direction and overlapping values ​​of the amplitude change rate are marked and connected to obtain the overlapping change trajectory interval.

[0027] By comparing the frequency offset point by point within the overlapping trajectory interval, time markers with the same direction of change and consecutive adjacent differences are grouped into the same group, and the amplitude change rate is rearranged according to the order of frequency offset change to unfold the overlapping trajectory.

[0028] Based on the unfolding results, the amplitude change rate is moved to a different position. The amplitude change rate with a continuous change relationship between the frequency offsets is placed at adjacent time marker positions, and the amplitude change rate with a progressive change relationship between the frequency offsets is arranged in chronological order to separate the change trajectories.

[0029] The position movement results are used to reconnect each change trajectory in time sequence and record the start time and end time, so that the amplitude change rate and frequency offset maintain the corresponding relationship to obtain a multi-segment progressive change pattern.

[0030] Preferably, during the positional movement of each change trajectory, the amplitude change rate that maintains a continuous relationship between the differences between frequency offsets is arranged in adjacent time markers, while the amplitude change rate that shows a progressive relationship between frequency offsets is moved sequentially along the time markers, and the one-to-one correspondence between the amplitude change rate and the frequency offset is maintained during the movement, so as to ensure the continuous distribution of each change trajectory on the time markers.

[0031] Preferably, during the process of organizing the multi-segment progressive changes, the temporal evolution relationship of the multi-source changes is reconstructed by rearranging the segments around the sound pressure amplitude and continuously adjusting the frequency position. The steps are as follows:

[0032] The sound pressure amplitude is read point by point along the time marker sequence in the multi-segment progressive change pattern, and the frequency position is read simultaneously. The sound pressure amplitudes within the same time range are collected and arranged in the time marker sequence to maintain the correspondence between the sound pressure amplitude and the frequency position.

[0033] By combining the centrally collected sound pressure amplitude values, point-by-point comparisons are performed, the sound pressure amplitude values ​​are sorted according to their numerical values, and the sorting results are redistributed to the time marker positions to achieve segmented rearrangement;

[0034] The adjacent time ranges are connected according to the segmented rearrangement results. The end time marker of the previous time range is connected with the start time marker of the next time range, and the sound pressure amplitude is adjusted point by point to maintain the progressive change relationship.

[0035] By connecting the processing results, the sound pressure amplitude and frequency position are uniformly arranged so that the time markers are continuously distributed and maintain the corresponding relationship, so as to obtain the time evolution path of multi-source changes.

[0036] Preferably, the processing of the extension and amplitude change oriented towards the time evolution path, by guiding the embedding of the new changes and advancing along the time marker, suppresses the synchronous superposition phenomenon. The steps are as follows:

[0037] The sound pressure amplitude is read point by point along the time markers in the time evolution path, and the frequency position is read simultaneously. The sound pressure amplitudes corresponding to the continuous time markers at the end are arranged and the direction and magnitude of change are recorded to obtain the reference segment.

[0038] By comparing the sound pressure amplitude corresponding to the new time marker with the reference section, the new time marker is positioned between adjacent sound pressure amplitudes, and the corresponding frequency position is recorded to maintain continuity.

[0039] Based on the positioning results, the time stamps after the insertion position are shifted sequentially to embed the new time stamps into the time evolution path while maintaining the correspondence between sound pressure amplitude and frequency position.

[0040] By repeatedly reading, locating, and shifting the sound pressure amplitude of subsequent incoming sound pressure values ​​through the embedded processing results, the sound pressure amplitude is continuously advanced along the time marker to suppress the phenomenon of synchronous superposition.

[0041] A noise testing system for automotive components includes a data modeling module, a superimposed section identification module, a trajectory unfolding module, a multi-source path reconstruction module, and an evolutionary traction control module.

[0042] The data modeling module acquires the sound pressure amplitude, frequency position, time markers, and excitation rhythm intervals of automotive parts as the speed increases during noise testing, and calculates the amplitude change rate and frequency offset corresponding to each time marker to form a continuous recording segment containing multi-source change traces.

[0043] The superimposed segment identification module performs difference processing on adjacent time markers based on the amplitude change rate and frequency offset in continuous recording segments, extracts segments with similar changes, and filters segments with synchronously rising amplitude and continuously changing frequency offset differences through frequency offset alignment comparison, thereby determining suspected synchronous superimposed segments and their corresponding duration range.

[0044] The trajectory unfolding module targets the amplitude change rate in the continuous recording segments corresponding to the suspected synchronous superposition section, performs reverse unfolding processing according to the frequency offset in the continuous recording segments, misaligns the overlapping change trajectories according to the time sequence, and gradually pulls them apart according to the frequency difference to restore the multi-segment progressive change pattern.

[0045] The multi-source path reconstruction module uses a multi-segment progressive change pattern to segment and rearrange the sound pressure amplitude within the corresponding time range, distributing the concentrated change amplitude to adjacent time markers in an increasing order, while maintaining the continuous change relationship of frequency position, thus forming a time evolution path of multi-source changes.

[0046] The evolutionary traction control module, based on the time evolution path, guides the amplitude changes that subsequently enter the test process to move forward, embeds changes that conform to the progressive characteristics into the unfolded segment, and continues to advance along the time marker, thereby suppressing the synchronous superposition phenomenon under the periodic excitation.

[0047] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0048] This invention constructs continuous recording segments containing sound pressure amplitude, frequency position, time markers, and excitation rhythm intervals, and further introduces amplitude change rate and frequency offset to characterize the change process. This transforms the original analysis method, which was mainly based on single time-domain or frequency-domain features, into an analysis method based on dynamic change trajectories. Under periodic excitation, it can continuously express the change process of multiple sources, thereby avoiding the misjudgment of multi-source superimposed signals as stable signals, improving the ability to express complex noise change processes, and enhancing the identifiability of abnormal information in the time dimension.

[0049] This invention reverses the trajectory of changes within suspected synchronous superposition sections, rearranges the sound pressure amplitude in segments by combining multiple progressive change patterns, and guides subsequent amplitude changes forward using time evolution paths. This allows the changes from multiple sources to form a continuous distribution relationship in terms of time markers, thereby suppressing the recurrence of synchronous superposition phenomena during subsequent testing. It also enables the change paths of each abnormal source to be presented independently, improving the ability to identify anomalies under multi-source coupling conditions and enhancing the accuracy of analyzing potential fault change trends. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0051] Figure 1 This is a flowchart of a method for testing the noise of automotive parts according to the present invention.

[0052] Figure 2 This is a schematic diagram of a noise testing system for automotive parts according to the present invention. Detailed Implementation

[0053] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0054] This invention provides, for example Figure 1 The method for testing the noise of automotive components, as shown, includes the following steps:

[0055] The system acquires the sound pressure amplitude, frequency position, time markers, and excitation rhythm intervals of automotive parts as the speed increases during noise testing, and calculates the amplitude change rate and frequency offset corresponding to each time marker to form a continuous recording segment containing multi-source change traces.

[0056] In the process of noise testing of automotive components, in order to continuously characterize the sound pressure information that changes with speed, a complete continuous recording segment is formed by synchronously acquiring and processing the sound pressure amplitude, frequency position, time marker, and excitation rhythm interval. The specific implementation method is as follows:

[0057] When automotive components are operating under progressively varying speeds, their noise is continuously collected. At each speed change point, the sound pressure level (SPL) signal is acquired using acoustic acquisition methods, and the corresponding SPL amplitude is extracted from the signal. Simultaneously, frequency analysis is performed to determine the frequency position at that moment, and a unique time stamp is assigned to each acquisition action. This time stamp is recorded sequentially according to the actual acquisition order. While recording the time stamp, the time interval between two adjacent speed changes is also recorded synchronously and stored as the excitation rhythm interval. This ensures that each time stamp corresponds to the SPL amplitude, frequency position, and excitation rhythm interval between adjacent time stamps, thereby establishing a correspondence between SPL amplitude, frequency position, time stamp, and excitation rhythm interval on the same time dimension. This ensures that the acquired data remains continuous in time and contains complete information about the progressive speed changes.

[0058] Based on the established time correspondence data, the sound pressure amplitude corresponding to adjacent time markers is processed one by one according to the chronological order of the time markers. By calculating the difference between the sound pressure amplitude of the current time marker and the previous time marker, and combining the excitation rhythm interval between the two, the amplitude change rate corresponding to the time marker is determined, so that each time marker corresponds to a change rate reflecting the amplitude change trend. At the same time, the frequency position in the same time series is continuously tracked. The frequency position corresponding to the current time marker is compared with the frequency position corresponding to the previous time marker, and the difference between the two is calculated to obtain the frequency offset corresponding to the time marker. The amplitude change rate and the frequency offset are synchronously associated according to the time marker, so that each time marker not only retains the original sound pressure amplitude and frequency position information, but also further includes the amplitude change rate and frequency offset describing the change trend, thereby constructing an extended data set that can reflect the dynamic change process.

[0059] After obtaining the amplitude change rate and frequency offset, the sound pressure amplitude, frequency position, amplitude change rate, and frequency offset corresponding to each time marker are uniformly organized according to the arrangement order of the time markers. The above data are bound to the corresponding time markers as indexes and arranged according to the chronological order of the time markers, so that the data under different time markers are arranged continuously according to the actual acquisition order. At the same time, the excitation rhythm interval is inserted between adjacent time markers to mark the time span relationship between each time point. In this way, the sound pressure amplitude, frequency position, amplitude change rate, frequency offset, and excitation rhythm interval are presented in the same data structure, so that the entire data set can reflect the synchronous change relationship and mutual influence between the parameters during the speed progression.

[0060] After data organization and unified arrangement, all data are continuously spliced ​​along the time marker direction, connecting all data arranged in chronological order one by one to form a continuous recording segment covering the complete speed change process. In this continuous recording segment, each time marker corresponds to a complete set of data items, including sound pressure amplitude, frequency position, amplitude change rate, frequency offset, and excitation rhythm interval, while maintaining the one-to-one correspondence between each data item. Through the construction of this continuous recording segment, the change trajectory generated by multi-source signals under periodic excitation can be completely presented in the same time dimension, and the change relationship between each time marker is continuous and traceable, providing a continuous and complete data foundation for subsequent identification and processing of multi-source change traces.

[0061] Based on the amplitude change rate and frequency offset in continuous recording segments, difference processing is performed on adjacent time markers to extract segments with similar changes. By comparing the frequency offset, segments with synchronously increasing amplitude and continuously changing frequency offset differences are selected to determine suspected synchronous superposition segments and their corresponding duration range.

[0062] Given that the continuous recording segments already contain the amplitude change rate and frequency offset, the method involves interpolation and alignment comparison of adjacent time markers to extract segments with similar changes and further determine suspected synchronous overlay segments and their corresponding duration ranges. The specific implementation method is as follows:

[0063] Following the time markers in the continuous recording segment, the data is read sequentially from the first time marker. The amplitude change rate of each current time marker is compared point by point with the amplitude change rate of the previous time marker. The amplitude change rate of the current time marker is subtracted from the amplitude change rate of the previous time marker to obtain the difference between the two. This difference is then bound to the current time marker and recorded. During the recording process, the same process is performed on each group of adjacent time markers, so that each time marker in the entire continuous recording segment is accompanied by an amplitude change rate difference information with the previous time marker. At the same time, these differences are arranged continuously in the order of the time markers to form a difference data set reflecting the continuous fluctuation of the amplitude change rate in the time dimension.

[0064] The aforementioned difference data set is scanned point by point along the time marker sequence. Multiple time markers whose amplitude change rate differences between adjacent time markers remain within a close range are continuously connected. These time markers are then merged and divided into candidate segments with similar changes. During the merging process, the start and end time markers of each candidate segment are recorded, and the amplitude change rate and its difference for each time marker within the segment are also retained. This ensures that the candidate segments contain both temporal continuity and amplitude change rate variations. By continuously expanding backward, time markers that meet the characteristic of consistent difference changes are continuously included in the same candidate segment until a deviation in the difference change occurs, at which point the merging of the candidate segment ends. This process extracts multiple segments with similar changes from the continuous recording segments.

[0065] Within each convergent change segment, the corresponding frequency offset is read sequentially according to the time markers. The frequency offset corresponding to the current time marker is compared point by point with the frequency offset corresponding to the previous time marker. The direction and magnitude of the frequency offset change between adjacent time markers are recorded. Time markers with continuous and consistent frequency offset changes are further filtered. During the filtering process, time markers that simultaneously satisfy the condition of a continuously increasing amplitude change rate and a consistent frequency offset change direction are grouped together. These time markers are then centrally marked within the same convergent change segment, so that these marked time markers have both synchronous amplitude change characteristics and a frequency position convergence relationship. This allows for the selection of segments that meet the synchronous change characteristics within the convergent change segment.

[0066] For the selected synchronous change segments, adjacent time markers are connected according to the continuity of time markers. Temporally adjacent and continuous synchronous change segments are merged to form complete suspected synchronous overlay segments. During the formation process, the start and end time markers of each suspected synchronous overlay segment are recorded, and the duration range of the segment is determined based on the time interval between the two. At the same time, the amplitude change rate and frequency offset corresponding to each time marker within the segment are also retained, so that each suspected synchronous overlay segment has complete time range information and corresponding change characteristic data, thereby completing the determination of the suspected synchronous overlay segments and their duration range.

[0067] For the amplitude change rate in the continuous recording segment corresponding to the suspected synchronous superposition section, reverse expansion processing is performed according to the frequency offset in the continuous recording segment. The overlapping change trajectory is staggered according to the time sequence and pulled apart point by point according to the frequency difference to restore the multi-segment progressive change pattern.

[0068] Given that the suspected synchronous overlay section has been determined, the overlapping change trajectory is split and restored by processing the amplitude change rate and frequency offset in the corresponding continuous recording segments point by point, so as to form a multi-segment progressive change pattern. The specific implementation method is as follows:

[0069] Following the time marker sequence in the continuous recording segments corresponding to the suspected synchronous overlay section, data is read sequentially from the starting time marker of the segment. The amplitude change rate corresponding to each time marker is extracted, and the frequency offset corresponding to the same time marker is read simultaneously. Each set of amplitude change rate and frequency offset is bound and recorded, and arranged according to the time marker sequence. During the reading process, the parts where the amplitude change rate changes in the same direction and the change values ​​are consistent between adjacent time markers are marked point by point. These time markers are continuously connected to form overlapping change trajectory intervals, and the corresponding amplitude change rate and frequency offset are retained at each time marker position, so that each data point in the interval has complete change information, providing continuous data support for subsequent processing.

[0070] Around the marked overlapping change trajectory interval, the frequency offset corresponding to each time marker is compared point by point in chronological order. Time markers with consistent frequency offset change direction and continuous adjacent differences are grouped together. Within the same group, the amplitude change rate is rearranged according to the change order of the frequency offset. This allows the amplitude change rates that originally overlapped at the same time marker position to be unfolded sequentially according to the change relationship of the frequency offset. During the unfolding process, amplitude change rates from different sources are separated according to the magnitude of the frequency offset and arranged to different time marker positions. This makes the change information that was originally superimposed at the same time point form a staggered arrangement on the time axis, thus completing the initial reverse unfolding process.

[0071] After the initial unfolding is completed, each separated change trajectory is adjusted point by point. According to the specific difference between the frequency offsets, the amplitude change rate at adjacent time markers is moved. Amplitude change rates with similar frequency offsets are arranged at adjacent time markers, and amplitude change rates with gradually changing frequency offsets are arranged in chronological order. By moving point by point, the change trajectories that were originally concentrated in the same time interval are gradually separated, so that different change trajectories form an intermittent distribution on the time axis. At the same time, the correspondence between the amplitude change rate and the frequency offset corresponding to each time marker is kept unchanged, so that each change trajectory forms a continuously distributed independent trajectory in the time dimension.

[0072] After positional adjustments, all change trajectories were systematically organized and reconnected according to their time signatures. The start and end times of each trajectory were recorded and arranged chronologically to ensure a continuous connection between them on the timeline. During this process, the one-to-one correspondence between the amplitude change rate and frequency offset corresponding to each time signature was maintained. Simultaneously, the time range corresponding to each trajectory was fully recorded, restoring the overlapping trajectories within the previously suspected synchronous superposition zone into multiple progressively unfolding segments, thus forming a multi-segment progressive change pattern. This provides a clear temporal distribution structure for further processing of multi-source change paths.

[0073] By utilizing a multi-segment progressive change pattern, the sound pressure amplitude within the corresponding time range is segmented and rearranged, and the concentrated change amplitude is distributed to adjacent time markers in an increasing order, while maintaining the continuous change relationship of frequency position, thus forming a time evolution path of multi-source change.

[0074] Given that a multi-segment progressive change pattern has already been formed, the sound pressure amplitude within the corresponding time range is segmented and rearranged, while maintaining the continuous change relationship of frequency position during the rearrangement process, to form a complete multi-source change time evolution path. The specific implementation method is as follows:

[0075] Following the time marker sequence within a multi-segment progressive change pattern, the time range corresponding to each progressive change pattern is located one by one. Within each time range, the corresponding sound pressure amplitude is read point by point according to the arrangement of the time markers, and the corresponding frequency position is recorded simultaneously with each sound pressure amplitude reading, so that each time marker corresponds to a set of sound pressure amplitude and frequency position data pairs. During the reading process, all sound pressure amplitudes within the same time range are collected and aggregated according to the original time marker sequence, and arranged in ascending order of the time markers, so that the sound pressure amplitudes within this time range form a continuous arrangement. At the same time, the corresponding frequency positions are arranged synchronously according to the same time marker sequence, so that the sound pressure amplitude and frequency position maintain a one-to-one correspondence in the time dimension, thereby constructing a complete data set corresponding to each progressive change pattern.

[0076] For each time range, the collected sound pressure amplitude data is compared point by point. All sound pressure amplitudes within the current time range are sorted according to their numerical values, and a new arrangement order is generated based on the sorting results. During the sorting process, the data is arranged upwards from the minimum sound pressure amplitude to the maximum sound pressure amplitude. After the sorting results are formed, the sorted sound pressure amplitudes are redistributed to the various time marker positions within the time range. In the redistribution process, starting from the beginning time marker of the time range, the first sorted sound pressure amplitude is assigned to the beginning time marker position, then the second sorted sound pressure amplitude is assigned to the next time marker position, and so on until all time markers within the entire time range are assigned. At the same time, during the redistribution of sound pressure amplitudes, the frequency positions corresponding to each time marker are kept from being cross-adjusted, so that the arrangement of frequency positions on the time markers still maintains the original continuous change relationship.

[0077] After redistributing the sound pressure amplitude within each time range, the connection between adjacent time ranges is processed. The end time marker of the previous time range is sequentially connected with the start time marker of the next time range. During the connection process, the sound pressure amplitude corresponding to the end time marker of the previous time range is compared with the sound pressure amplitude corresponding to the start time marker of the next time range, so that the sound pressure amplitude at the start position of the next time range continues the trend of change at the end position of the previous time range. The sound pressure amplitude at the boundary of adjacent time ranges is adjusted point by point, so that the sound pressure amplitude forms a continuous progressive relationship at the connection position of the cross-time range. At the same time, the corresponding frequency positions are continuously arranged according to the time marker order, so that the frequency positions maintain a continuous change state at the connection point of the cross-time range, thereby ensuring that a complete and continuous change structure is formed between multiple time ranges after connection.

[0078] Around the entire time marker range, all sound pressure amplitudes and frequency positions after segmentation, rearrangement, and connection are uniformly organized. The sound pressure amplitudes corresponding to each time marker are arranged in chronological order, maintaining the one-to-one correspondence between the frequency position and sound pressure amplitude for each time marker. This ensures that all time markers form a continuous distribution on the same time axis. During this continuous distribution, the sound pressure amplitude gradually changes from low to high along the time direction, while the frequency position changes continuously without interruption on the time markers. This forms a complete multi-source change time evolution path, allowing the multi-segment progressive change pattern to unfold continuously in the time dimension, providing a clear time evolution structure for further processing of the multi-source change process.

[0079] Based on the time evolution path, the amplitude changes that subsequently enter the test process are guided to move forward, embedding changes that conform to the progressive characteristics into the unfolded segment, and continuously advancing along the time marker, thereby suppressing the synchronous superposition phenomenon under the periodic excitation.

[0080] Given that the time evolution path has been formed and is continuously updated, the amplitude changes entering the test process are guided forward to embed them into the existing time evolution path and continue to advance along the time marker, thereby reducing the possibility of synchronous superposition under periodic excitation. The specific implementation method is as follows:

[0081] Following the time marker sequence along the time evolution path, several consecutive time markers at the end of the current time range are selected to form a reference segment. Within this reference segment, the sound pressure amplitude corresponding to each time marker is read point by point, and the corresponding frequency position is read simultaneously. The read sound pressure amplitude values ​​are arranged according to the time marker sequence, and the direction and magnitude of sound pressure amplitude changes between adjacent time markers are recorded. This ensures that the sound pressure amplitudes within the reference segment form a continuous and progressive change relationship, and maintains the continuous change of the corresponding frequency position on the time markers without interruption. By organizing the data within this reference segment, the amplitude changes entering the subsequent testing process can have a clear reference sequence when entering the time evolution path.

[0082] For each new time marker entering the testing process, its corresponding sound pressure level (SPL) amplitude is read point by point. This SPL amplitude is then compared with the SPL amplitudes corresponding to each time marker within the reference segment. During the comparison, the search proceeds sequentially from the beginning of the reference segment. When a newly read SPL amplitude falls between the SPL amplitudes of two adjacent time markers, the new time marker is positioned between these two time markers. Simultaneously, the frequency positions of the time markers before and after this position are recorded. This ensures that the new time marker maintains a continuous frequency position with the preceding and following time markers during the embedding process. Through this point-by-point comparison and positioning method, amplitude changes that conform to progressive characteristics are accurately embedded into the corresponding positions within the reference segment.

[0083] After the new time marker is located, all time markers after that position are sequentially shifted backward. The time markers that were originally after the insertion position are shifted one position backward so that the sound pressure amplitude corresponding to the new time marker can be inserted into the time evolution path. During the shifting process, the position of each time marker is adjusted point by point, while keeping the correspondence between the sound pressure amplitude and frequency position of each time marker unchanged. At the same time, the shifted time markers are renumbered so that the entire time evolution path remains in a continuous arrangement after expansion, and the arrangement of sound pressure amplitude on the time markers still conforms to the progressive change relationship.

[0084] Around the updated time evolution path, the above reading, comparison, positioning, and follow-up processing are repeatedly performed on subsequent continuous sound pressure amplitude changes. This ensures that each newly entering sound pressure amplitude is embedded into the time evolution path according to the progressive relationship in the reference segment and continues to advance along the time marker direction. During the advancement, the progressive arrangement of sound pressure amplitude on the time marker and the continuous change relationship of frequency position are maintained. This makes the amplitude changes from different sources dispersed on the time axis, thereby avoiding the superposition of multiple amplitude changes at the same time marker position when the periodic excitation effect continues to exist, and thus suppressing the phenomenon of synchronous superposition.

[0085] This invention constructs continuous recording segments containing sound pressure amplitude, frequency position, time markers, and excitation rhythm intervals, and further introduces amplitude change rate and frequency offset to characterize the change process. This transforms the original analysis method, which was mainly based on single time-domain or frequency-domain features, into an analysis method based on dynamic change trajectories. Under periodic excitation, it can continuously express the change process of multiple sources, thereby avoiding the misjudgment of multi-source superimposed signals as stable signals, improving the ability to express complex noise change processes, and enhancing the identifiability of abnormal information in the time dimension.

[0086] This invention reverses the trajectory of changes within suspected synchronous superposition sections, rearranges the sound pressure amplitude in segments by combining multiple progressive change patterns, and guides subsequent amplitude changes forward using time evolution paths. This allows the changes from multiple sources to form a continuous distribution relationship in terms of time markers, thereby suppressing the recurrence of synchronous superposition phenomena during subsequent testing. It also enables the change paths of each abnormal source to be presented independently, improving the ability to identify anomalies under multi-source coupling conditions and enhancing the accuracy of analyzing potential fault change trends.

[0087] This invention provides, for example Figure 2 The illustrated automotive component noise testing system includes a data modeling module, a superimposed section identification module, a trajectory unfolding module, a multi-source path reconstruction module, and an evolutionary traction control module.

[0088] The data modeling module acquires the sound pressure amplitude, frequency position, time markers, and excitation rhythm intervals of automotive parts as the speed increases during noise testing, and calculates the amplitude change rate and frequency offset corresponding to each time marker to form a continuous recording segment containing multi-source change traces.

[0089] The superimposed segment identification module performs difference processing on adjacent time markers based on the amplitude change rate and frequency offset in continuous recording segments, extracts segments with similar changes, and filters segments with synchronously rising amplitude and continuously changing frequency offset differences through frequency offset alignment comparison, thereby determining suspected synchronous superimposed segments and their corresponding duration range.

[0090] The trajectory unfolding module targets the amplitude change rate in the continuous recording segments corresponding to the suspected synchronous superposition section, performs reverse unfolding processing according to the frequency offset in the continuous recording segments, misaligns the overlapping change trajectories according to the time sequence, and gradually pulls them apart according to the frequency difference to restore the multi-segment progressive change pattern.

[0091] The multi-source path reconstruction module uses a multi-segment progressive change pattern to segment and rearrange the sound pressure amplitude within the corresponding time range, distributing the concentrated change amplitude to adjacent time markers in an increasing order, while maintaining the continuous change relationship of frequency position, thus forming a time evolution path of multi-source changes.

[0092] The evolutionary traction control module, based on the time evolution path, guides the amplitude changes that subsequently enter the test process to move forward, embeds changes that conform to the progressive characteristics into the unfolded segment, and continues to advance along the time marker, thereby suppressing the synchronous superposition phenomenon under the periodic excitation.

[0093] The present invention provides a method for testing the noise of automotive parts, which is implemented by the above-mentioned automotive parts noise testing system. For details of the specific method and process of the automotive parts noise testing system, please refer to the above-mentioned embodiment of the method for testing the noise of automotive parts, which will not be repeated here.

[0094] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method of testing for noise in an automotive component, the method comprising: Includes the following steps: The system acquires the sound pressure amplitude, frequency position, time markers, and excitation rhythm intervals of automotive parts as the speed increases during noise testing, and calculates the amplitude change rate and frequency offset corresponding to each time marker to form a continuous recording segment containing multi-source change traces. Based on the amplitude change rate and frequency offset in continuous recording segments, difference processing is performed on adjacent time markers to extract segments with similar changes. By comparing the frequency offset, segments with synchronously increasing amplitude and continuously changing frequency offset differences are selected to determine suspected synchronous superposition segments and their corresponding duration range. For the amplitude change rate in the continuous recording segment corresponding to the suspected synchronous superposition section, reverse expansion processing is performed according to the frequency offset in the continuous recording segment. The overlapping change trajectory is staggered according to the time sequence and pulled apart point by point according to the frequency difference to restore the multi-segment progressive change pattern. By utilizing a multi-segment progressive change pattern, the sound pressure amplitude within the corresponding time range is segmented and rearranged, and the concentrated change amplitude is distributed to adjacent time markers in an increasing order, while maintaining the continuous change relationship of frequency position, thus forming a time evolution path of multi-source change. Based on the time evolution path, the amplitude changes that subsequently enter the testing process are guided to move forward, embedding changes that conform to the progressive characteristics into the unfolded segment, and continuing to push along the time marker.

2. The method of claim 1, wherein, In the process of continuously characterizing and extracting the changing trends of sound pressure information, the multi-source change traces are continuously described by synchronously acquiring the sound pressure amplitude and frequency position and expanding the recording of the change characteristics, thus constructing a continuous recording segment. The steps are as follows: Noise is collected at different speed change nodes to obtain sound pressure amplitude and frequency position and assign time stamps. At the same time, the excitation rhythm interval between adjacent time stamps is recorded to establish a correspondence. The sound pressure amplitude is processed point by point along the time marker sequence. The rate of change of amplitude is determined by the difference between adjacent time markers and the excitation rhythm interval. The frequency offset is obtained by comparing the frequency position and then the two are correlated. The sound pressure amplitude, frequency position, amplitude change rate and frequency offset are uniformly organized and bound according to the time marker, and the excitation rhythm interval is inserted to obtain a continuously arranged data set. The data set is spliced ​​along the time marker direction to obtain a continuous recording segment covering the speed progression process.

3. The method of claim 2, wherein, By sequentially binding the sound pressure amplitude, frequency position, amplitude change rate, and frequency offset in the continuous recording segments according to the time stamp sequence, and inserting the excitation rhythm interval between adjacent time stamps, the data corresponding to each time stamp are continuously arranged, thereby maintaining the one-to-one correspondence between the sound pressure amplitude and the frequency position and change rate, and maintaining the continuous distribution of the time span relationship.

4. The method for testing the noise of automotive parts according to claim 2, characterized in that, In the process of identifying convergent change segments and determining synchronous overlay segments, the extraction of synchronous change segments and the determination of their duration range are completed by processing the difference in amplitude change rates and aligning and comparing them with frequency offsets. The steps are as follows: The amplitude change rate is read point by point along the time markers in the continuous recording segment. The difference between the amplitude change rates corresponding to adjacent time markers is calculated, and the change difference is bound to the time marker and recorded. Based on the change difference, the time markers are continuously scanned, and time markers with consistent difference characteristics are connected and divided into segments with similar change. At the same time, the rate of change of amplitude and the change difference within the segment are recorded. For segments with similar changes, the frequency offset is read point by point. Alignment comparison is performed on the frequency offsets corresponding to adjacent time markers, and time markers with continuously increasing amplitude change rates and consistent frequency offset change directions are selected as synchronous change segments. By combining synchronous change segments with time markers, a suspected synchronous overlay segment is obtained. At the same time, the time interval corresponding to the start time marker and the end time marker is recorded as the duration range.

5. The method of claim 4, wherein, During the frequency offset alignment and comparison process, the frequency offsets corresponding to adjacent time markers are compared point by point, and the direction and magnitude of change are recorded. Time markers with the same direction of frequency offset change are filtered out, and time markers with a continuously increasing amplitude change rate are combined. Synchronous change segments are obtained by continuously connecting time markers, and the amplitude change rate and frequency offset corresponding to each time marker are retained.

6. The method of claim 4, wherein, To decompose and restore the progressive change pattern of overlapping trajectories, the overlapping trajectories are misaligned and sequentially unfolded by leveraging the correspondence between the amplitude change rate and the frequency offset. The steps are as follows: The amplitude change rate is read point by point along the time marker sequence of the continuous recording segment corresponding to the suspected synchronous superposition section, and the frequency offset is read synchronously. The time markers with the same change direction and overlapping values ​​of the amplitude change rate are marked and connected to obtain the overlapping change trajectory interval. By comparing the frequency offset point by point within the overlapping trajectory interval, time markers with the same direction of change and consecutive adjacent differences are grouped into the same group, and the amplitude change rate is rearranged according to the order of frequency offset change to unfold the overlapping trajectory. Based on the unfolding results, the amplitude change rate is moved to a different position. The amplitude change rate with a continuous change relationship between the frequency offsets is placed at adjacent time marker positions, and the amplitude change rate with a progressive change relationship between the frequency offsets is arranged in chronological order to separate the change trajectories. The position movement results are used to reconnect each change trajectory in time sequence and record the start time and end time, so that the amplitude change rate and frequency offset maintain the corresponding relationship to obtain a multi-segment progressive change pattern.

7. The method of claim 6, wherein, During the positional shift of each trajectory, the amplitude change rate that maintains a continuous relationship between the differences between frequency offsets is arranged in adjacent time markers. At the same time, the amplitude change rate that shows a progressive relationship between frequency offsets is moved sequentially along the time markers, and the one-to-one correspondence between the amplitude change rate and the frequency offset is maintained during the shift.

8. The method of claim 6, wherein, During the process of organizing the multi-segment progressive changes, the temporal evolution relationship of the multi-source changes is reconstructed by rearranging the segments around the sound pressure amplitude and continuously adjusting the frequency position. The steps are as follows: The sound pressure amplitude is read point by point along the time marker sequence in the multi-segment progressive change pattern, and the frequency position is read simultaneously. The sound pressure amplitudes within the same time range are collected and arranged in the time marker sequence to maintain the correspondence between the sound pressure amplitude and the frequency position. By combining the centrally collected sound pressure amplitude values, point-by-point comparisons are performed, the sound pressure amplitude values ​​are sorted according to their numerical values, and the sorting results are redistributed to the time marker positions to achieve segmented rearrangement; The adjacent time ranges are connected according to the segmented rearrangement results. The end time marker of the previous time range is connected with the start time marker of the next time range, and the sound pressure amplitude is adjusted point by point to maintain the progressive change relationship. By connecting the processing results, the sound pressure amplitude and frequency position are uniformly arranged so that the time markers are continuously distributed and maintain the corresponding relationship, so as to obtain the time evolution path of multi-source changes.

9. The method of claim 1, wherein, The process of extending the time evolution path and introducing amplitude changes involves guiding and embedding the new changes and advancing along the time marker to suppress the synchronous superposition phenomenon. The steps are as follows: The sound pressure amplitude is read point by point along the time markers in the time evolution path, and the frequency position is read simultaneously. The sound pressure amplitudes corresponding to the continuous time markers at the end are arranged and the direction and magnitude of change are recorded to obtain the reference segment. By comparing the sound pressure amplitude corresponding to the new time marker with the reference section, the new time marker is positioned between adjacent sound pressure amplitudes, and the corresponding frequency position is recorded to maintain continuity. Based on the positioning results, the time markers after the insertion position are shifted sequentially to embed the new time markers into the time evolution path while maintaining the correspondence between sound pressure amplitude and frequency position. By repeatedly reading, locating, and shifting the sound pressure amplitude of subsequent incoming sound pressure values ​​through the embedded processing results, the sound pressure amplitude is continuously advanced along the time marker to suppress the phenomenon of synchronous superposition.

10. A noise testing system for automotive components, used to implement the noise testing method for automotive components according to any one of claims 1-9, characterized in that, It includes a data modeling module, an overlay segment identification module, a trajectory unfolding module, a multi-source path reconstruction module, and an evolutionary traction control module. The data modeling module acquires the sound pressure amplitude, frequency position, time markers, and excitation rhythm intervals of automotive parts as the speed increases during noise testing, and calculates the amplitude change rate and frequency offset corresponding to each time marker to form a continuous recording segment containing multi-source change traces. The superimposed segment identification module performs difference processing on adjacent time markers based on the amplitude change rate and frequency offset in continuous recording segments, extracts segments with similar changes, and filters segments with synchronously rising amplitude and continuously changing frequency offset differences through frequency offset alignment comparison, thereby determining suspected synchronous superimposed segments and their corresponding duration range. The trajectory unfolding module targets the amplitude change rate in the continuous recording segments corresponding to the suspected synchronous superposition section, performs reverse unfolding processing according to the frequency offset in the continuous recording segments, misaligns the overlapping change trajectories according to the time sequence, and gradually pulls them apart according to the frequency difference to restore the multi-segment progressive change pattern. The multi-source path reconstruction module uses a multi-segment progressive change pattern to segment and rearrange the sound pressure amplitude within the corresponding time range, distributing the concentrated change amplitude to adjacent time markers in an increasing order, while maintaining the continuous change relationship of frequency position, thus forming a time evolution path of multi-source changes. The evolutionary traction control module, based on the time evolution path, guides the amplitude changes that subsequently enter the test process to move forward, embeds changes that conform to the progressive characteristics into the unfolded segment, and continues to advance along the time marker.

Images