Automobile fastener torque evaluation method and system based on intelligent sensor

Abstract

The present application relates to the technical field of fastener torque measurement, in particular to a method and system for evaluating the torque of automobile fasteners based on intelligent sensors, comprising the following steps: collecting time series of multiple source sensors, extracting key features of the tightening process, identifying slope difference and rhythm deviation, analyzing direction reversal and synchronization mismatch, combining staggered node structure, removing rhythm repeat segments, linearly integrating propulsion segments, and outputting the evaluation results of the torque of fasteners. In the present application, the tightening process is segmented by extracting the torque slope, angular displacement amplitude and wave crest variation trend, combining the propulsion rhythm combination mode, constructing the tightening sequence with time index and analyzing the rhythm consistency and direction connection trend, which can identify abnormal fluctuations and synchronization deviation, combine the staggered node calibration mismatch area, remove the repeated propulsion segments and realize linear integration, so that the torque evaluation results reflect the coordination and variation characteristics of the propulsion state, and the identification ability of fastening abnormalities is improved.

Description

Technical Field

[0001] This invention relates to the field of fastener torque measurement technology, and in particular to a method and system for evaluating the torque of automotive fasteners based on intelligent sensors. Background Technology

[0002] The field of fastener torque measurement technology involves the detection, evaluation, and control of torque applied to various mechanical connection components during assembly. Core aspects include monitoring the stress state of fasteners such as bolts and nuts in the connected state, controlling assembly torque, detecting preload, and analyzing torque-angle relationships. This technology is widely used in industries such as automotive, aerospace, and rail transportation, aiming to ensure the reliability and safety of mechanical connections. It primarily relies on torque sensors, strain gauges, and bridge circuits for real-time torque data acquisition and analysis. With increasing demands for assembly quality, torque measurement technology is trending towards intelligence and systematization, integrating multiple sensing methods and improving measurement accuracy and efficiency through data analysis. Traditional automotive fastener torque evaluation methods involve applying tightening force to bolts, nuts, and other connecting components during vehicle assembly, and then measuring the torque during the tightening process using tools such as torque wrenches and mechanical torque testers to determine whether the assembly meets standard requirements. These methods typically involve manual measurement, with the determination of whether the set target value has been reached based on instrument readings. They often employ mechanical friction torque measuring tools or strain gauge structures built into electric wrenches for testing, and the evaluation process is mostly based on the torque value after the final tightening is completed.

[0003] Existing technologies rely on torque readings at fixed points in time to determine the tightness of fasteners. The detection method focuses on the torque result after tightening is completed, which cannot dynamically track the entire assembly process. It lacks the ability to identify changes in torque slope, angle advancement rhythm, and frequency fluctuations. It cannot effectively extract and respond to potential rhythm disorders, torque reversals, and local disturbances during the assembly process. Especially in multi-axis synchronous or confined space application environments, small fluctuations in torque response are difficult to reflect hidden abnormal behavior, resulting in some fastening defects not being accurately identified, affecting the stability and safety of the connection structure during operation. Summary of the Invention

[0004] To achieve the above objectives, the present invention adopts the following technical solution: a method for evaluating the torque of automotive fasteners based on intelligent sensors, comprising the following steps: S1: Collect time series data from torque sensor, angular displacement sensor, and vibration channel; divide the torque slope change into segments; cut the angle advance according to the tightening displacement interval; extract the peak position of the vibration frequency peak; and generate a set of tightening process segments. S2: Using the segmented set of the tightening process, analyze the slope change direction and difference trend, determine the stability of the angular displacement amplitude of the angular displacement segment, judge the consistency of the peak value change trend of the disturbance segment, and output the abnormal propulsion structure segment. S3: Based on the abnormal propulsion structure segment, determine whether the directions of adjacent segments of the torque curve are reversed, classify the angle interval change in the angle change rate feature, mark the combination structure with inconsistent wave order, and output the abnormal rhythm mark segment. S4: From the abnormal rhythm marked sections, analyze the pairing relationship between angle and torque slope, filter out sections with asynchronous angles but shifted torque direction, combine staggered nodes, and output the tightening rhythm staggered area; S5: Using the aforementioned tightening rhythm staggered zone, determine the trend of the direction connection of the advancement section, clear the rhythm repetition segment, perform time sequence connection according to direction, and output the torque evaluation result of the automotive fastener.

[0005] As a further aspect of the present invention, the tightening process segment set includes torque growth slope segments, angle propulsion segments divided according to tightening displacement, and frequency peak indexes within the time window; the abnormal propulsion structure segments include slope change direction sequences, angular displacement amplitude stability indicators, and adjacent peak position trends; the abnormal rhythm marking segments include propulsion direction reversal patterns, pitch change type labels, and fluctuation sequence numbering information; the tightening rhythm interlacing zone includes angle and torque slope pairing combinations, asynchronous change segment combinations, and interlaced propulsion node sequences; and the automotive fastener torque evaluation results include propulsion direction connection characteristics, rhythm repetition interval patterns, and linear combination propulsion structures.

[0006] As a further aspect of the present invention, the trend of connection of the propulsion section direction refers to the continuity and consistency trend of adjacent tightening propulsion sections in the direction of torque change.

[0007] As a further aspect of the present invention, the angular displacement amplitude stability refers to the fact that within an angular displacement segment, the fluctuation range of the angular change amplitude in the time series is less than a preset amplitude fluctuation variance threshold.

[0008] As a further aspect of the present invention, the specific steps of S1 are as follows: S101: Obtain the time series of the intelligent torque sensor, calculate the torque difference between adjacent sampling points, compare the change rate sequence obtained by dividing the difference by the time interval with the set torque growth change rate threshold, divide the slope change segment, and generate the torque growth slope segment division structure. S102: Call the time index of the torque growth slope segment division structure, extract the corresponding angular displacement sequence, calculate the displacement increment, and cut continuous segments according to the set tightening displacement interval threshold to generate a continuous angular segment sequence of displacement interval. S103: Call the time index of the continuous angle segment sequence of the displacement interval, extract the corresponding vibration signal, perform spectrum analysis, locate the frequency point where the amplitude exceeds the set vibration peak threshold, and establish a set of tightening process segments.

[0009] As a further aspect of the present invention, the specific steps of S2 are as follows: S201: Call the set of segments of the tightening process, calculate the direction of change and the slope difference of each slope sequence, determine the trend of change, and obtain the slope transformation trend matrix; S202: Based on the slope transformation trend matrix, extract the positioning angle advancement segment, calculate the adjacent angle displacement amplitude and compare the difference with the set angle offset stability threshold to obtain the angle displacement judgment identifier set; S203: Call the frequency peak index corresponding to the segment in the segment set of the tightening process, analyze the consistency of the direction of change of adjacent peak positions, filter synchronous fluctuation segments, and establish abnormal propulsion structure segments.

[0010] As a further aspect of the present invention, the specific steps of S3 are as follows: S301: Call the abnormal propulsion structure segment, extract the torque propulsion curve, perform pairwise combination judgment on the slope change direction of adjacent segments, filter out the combination sequence with direction reversal, and generate a direction reversal combination identifier set; S302: Based on the direction reversal combination identifier set, extract the angle advancement rhythm sequence, calculate the adjacent angle shift pitch and the pitch change trend to obtain the pitch change classification label set; S303: Call the set of classification labels for pitch changes, extract the combination structure of inconsistent rhythms, number and identify it according to the order of advancement, aggregate the index intervals corresponding to the identified segments, and output the abnormal rhythm marked segments.

[0011] As a further aspect of the present invention, the specific steps of S4 are as follows: S401: Call the numbered segments in the abnormal rhythm marker section, extract the angle sequence and torque slope sequence of each segment during the tightening process, calculate the slope direction change of adjacent segments and match and compare it with the angle rhythm change, filter the combination segments where the angle rhythm is not synchronized but the torque direction has shifted, and generate a rhythm slope offset matching set. S402: Based on the rhythm slope offset matching set, extract the combined segment number and corresponding index, sort them according to the order of advancement, divide them into consecutive number groups, arrange the node positions in each group of segments, and obtain the rhythm offset combined node sequence. S403: Call the rhythm offset combination node sequence, aggregate the position indices of the interlaced occurrences according to the index sequence, identify the structure type and interlacing attributes within the position markers, and output the tightening rhythm interlacing area.

[0012] As a further aspect of the present invention, the specific steps of S5 are as follows: S501: Call the tightening rhythm interlacing area, extract the directional connection trend of each advancement segment, analyze the change of rhythm before and after based on the trend, filter and process the advancement segments with repeated rhythms, and generate a set of rhythm repeating advancement segment identifiers. S502: Based on the rhythm repeating advance segment identifier set, the advance segments in the identifiers are extracted and cleared at intervals, and the continuity of the remaining segments is calculated to obtain the advance segments that meet the set direction connection threshold, and a set of advance segments after clearing is generated. S503: Call the set of cleared and advanced segments, perform sequential splicing of the segments according to the direction information of each segment, arrange and combine the interlaced nodes in sequence, and output the torque evaluation result of the automotive fastener.

[0013] A smart sensor-based automotive fastener torque evaluation system includes: The data acquisition and segmentation module is used to implement S1: acquiring the time series of torque sensor, angular displacement sensor and vibration channel, performing segmentation on the change of torque slope, intercepting the angle advance according to the tightening displacement interval, extracting the peak position of vibration frequency peak, and generating a set of tightening process segments; The stability and anomaly detection module is used to implement S2: using the segment set of the tightening process, analyze the slope change direction and difference trend, determine the stability of the angular displacement amplitude of the angular displacement segment, determine the consistency of the peak value change trend of the disturbance segment, and output the abnormal propulsion structure segment; The rhythm anomaly marking module is used to implement S3: based on the abnormal propulsion structure segment, determine whether the direction of adjacent segments of the torque curve is reversed, classify the angle interval change in the angle change rate feature, mark the combination structure with inconsistent wave order, and output the abnormal rhythm marking segment. The rhythm interlacing identification module is used to implement S4: from the abnormal rhythm marking section, analyze the pairing relationship between angle and torque slope, filter the sections with asynchronous angles but shifted torque direction, combine interlacing nodes, and output the tightening rhythm interlacing area; The torque evaluation output module is used to implement S5: using the tightening rhythm staggered zone, determine the connection trend of the propulsion section direction, clear the rhythm repetition segment, perform timing connection according to direction, and output the torque evaluation result of the automotive fastener.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, the assembly process is segmented by extracting the torque slope, angular displacement amplitude, and peak change trend, combined with the combination of propulsion rhythm, to construct a tightening sequence with time index and analyze the rhythm consistency and direction connection trend. Abnormal fluctuations and synchronization deviations can be identified. By combining staggered nodes to calibrate mismatch areas, duplicate propulsion segments are eliminated and linear integration is achieved, so that the torque evaluation results reflect the coordination and change characteristics of the propulsion state, and the ability to identify tightening abnormalities is improved. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a schematic diagram of the steps of the present invention; Figure 2 This is a detailed schematic diagram of S1 of the present invention; Figure 3 This is a detailed schematic diagram of S2 of the present invention; Figure 4 This is a detailed schematic diagram of S3 of the present invention; Figure 5 This is a detailed schematic diagram of S4 of the present invention; Figure 6 This is a detailed schematic diagram of S5 of the present invention; Figure 7 This is a system module diagram of the present invention. Detailed Implementation

[0017] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0018] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0019] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent.

[0020] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0021] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0022] Please see Figure 1 This invention provides a method for evaluating the torque of automotive fasteners based on smart sensors, comprising the following steps: S1: Acquire the time series outputs of the intelligent torque sensor, angular displacement sensor, and vibration channel sensing, divide the torque growth slope into segments, continuously extract the angle advance record according to the tightening displacement interval, extract the peak position of the frequency peak within the time window, construct a tightening sequence structure with time index, and generate a set of tightening process segments. S2: Call the torque segment in the tightening process segment set, analyze the slope change direction and slope difference trend in the continuous propulsion segment, perform offset judgment on the stability of the forward and backward angular displacement amplitude in the corresponding angle propulsion segment, and then perform trend consistency judgment on the change trend of adjacent peak positions in the disturbance segment, extract the segment that meets the synchronous fluctuation characteristics, and obtain the abnormal propulsion structure segment. S3: Call the continuous segments in the abnormal propulsion structure segment, determine whether there is a reversal in the combination of change direction between adjacent segments in the torque propulsion curve, classify and mark the pitch change trend in the corresponding angle propulsion rhythm, and number and mark the combination structure with inconsistent fluctuation order according to the propulsion order, and output the abnormal rhythm marked segment. S4: Call the numbered segments in the abnormal rhythm marker section, analyze the pairing relationship between the angle and torque slope of the preceding and following rhythms during the tightening process, filter the segments where the angle of advancement does not change synchronously but the torque growth shows a direction shift, arrange and combine the interlaced nodes according to the consecutive segments, and output the tightening rhythm interlaced area. S5: Call the torque propulsion structure before the node in the tightening rhythm intersection zone, locate the directional connection trend of the propulsion segment in the force application rhythm, extract and clear the propulsion segment with rhythm repetition, and then splice the propulsion segment in the current direction to output the torque evaluation result of the automotive fastener.

[0023] The tightening process segment set includes torque growth slope segments, angle propulsion segments divided by tightening displacement, and frequency peak indexes within the time window. Abnormal propulsion structure segments include slope change direction sequences, angular displacement amplitude stability indicators, and adjacent peak position trends. Abnormal rhythm marking segments include propulsion direction reversal patterns, pitch change type labels, and fluctuation sequence numbering information. Tightening rhythm interlacing zones include angle and torque slope pairing combinations, asynchronous change segment combinations, and interlaced propulsion node sequences. The automotive fastener torque evaluation results include propulsion direction connection characteristics, rhythm repetition interval patterns, and linear combination propulsion structures.

[0024] Please see Figure 2 The specific steps of S1 are as follows: S101: Obtain the time series of the intelligent torque sensor, calculate the torque difference between adjacent sampling points, compare the change rate sequence obtained by dividing the difference by the time interval with the set torque growth change rate threshold, divide the slope change segment, and generate the torque growth slope segment division structure. Before processing begins, it is necessary to confirm that the collected data is continuous, without missed samples, and free from jitter interference, at a high frequency. For example, if the sampling frequency is 1000Hz, the total recording time is 10 seconds, and a total of 10,000 data points are acquired, first extract each set of torque data in time series order, and then arrange them sequentially according to their index positions as T0, T1, T2…T n Then, the difference is calculated for each pair of adjacent data points to obtain the difference sequence ΔT0, ΔT1…ΔT n -1, for example, when T0 is 1.5 Nm and T1 is 1.9 Nm, then ΔT0 is 0.4 Nm. After performing the difference calculation on all data points in sequence, a torque difference sequence with a length differing from the original time series by 1 is formed. Then, each difference value in this difference sequence is classified and judged. The set torque growth rate threshold Rt is set according to the workpiece material, bolt specifications, and feedback characteristics of the tightening tool in actual operation. For example, when tightening M8 bolts using an automatic tightening device, the torque growth trend usually varies between 0.2 Nm / ms and 0.5 Nm / ms. Therefore, Rt can be set as follows: t The threshold is set to 0.3 Nm / ms. For each ΔT value, its rate of change is calculated, and it is determined whether it exceeds this threshold. When the rate is greater than 0.3 Nm / ms, it is identified as a slope abrupt change point, and the corresponding time index is added to the candidate set. Then, consecutive time index points that meet this condition are merged. For example, if consecutive index points are 15, 16, 17, and 18, and their corresponding time differences are all within 5 ms, they are merged into one slope segment, and its start time is recorded as t. 15 The termination time is t 18Continue traversing the entire sequence in this way to complete the segmentation of all slope change points. The final torque growth slope segmentation structure includes the start and end times of each segment, the average slope value, the maximum slope change point, and the segment duration.

[0025] S102: Call the time index of the torque growth slope segment to divide the structure, extract the corresponding angular displacement sequence, calculate the displacement increment, and cut continuous segments according to the set tightening displacement interval threshold to generate a continuous angular segment sequence of displacement interval. First, the start and end time indices of each segment are retrieved, and the corresponding angle data points are located in the original angular displacement sensor time series to form continuous angular displacement data segments. For example, a segment time index is [t]. 400 , t 480 If θ is obtained within that interval, then... 400 to θ 480 All the angular displacement values ​​in between are arranged in sequence to form an angular displacement sequence. Then, the incremental values ​​of adjacent angles in this sequence are calculated to obtain the incremental sequence Δθ. n Further evaluation is conducted to determine whether each increment exceeds the set tightening displacement interval threshold θ_thresh. For example, in a high-precision tightening process in an electronic torque-controlled tightening device, the minimum significant angle change can be set to 0.25°, so the threshold θ_thresh is set to 0.3°. The judgment process then filters out Δθ. n For a range >0.3°, if multiple consecutive data points meet the condition, it is considered a valid tightening angle range, and its time range is recorded. For example, in θ... 420 to θ 440 If the incremental threshold condition is met continuously, the time index interval is extracted as a valid segment. The above process is repeated in all segments, and the resulting angle segment structure is standardized. The structure includes the start and end angle values, start and end time index, duration and incremental mean of the angle segment. The final generated displacement interval continuous angle segment sequence records multiple stages of continuous effective angular displacement changes during tightening and can reflect the rhythmic characteristics of angle changes.

[0026] S103: Call the time index of the continuous angle segment sequence of displacement interval, extract the corresponding vibration signal, perform spectrum analysis, locate the frequency point where the amplitude exceeds the set vibration peak threshold, and establish a set of tightening process segments. Extract the corresponding time index interval as a time window, locate it in the original vibration signal sequence, and centrally process the vibration acceleration signal within this time window. For example, the time range of a certain continuous angle segment is t. 700 To t 760 Then extract the vibration signal sequence A. 700 To A 760The vibration data sequence is segmented for frequency distribution detection. During this process, a Fast Fourier Analysis (FFA) is performed using a sliding window of 64 points per segment to obtain a frequency-amplitude comparison table. The amplitude of each frequency point is then individually assessed. A vibration amplitude threshold A_thresh is set based on the maximum allowable acceleration value of the standard operating equipment. For example, if the maximum allowable vibration of the equipment is 0.4g, A_thresh can be set to 0.3g. This is used to determine whether the amplitude of all frequency points exceeds the threshold. If the amplitude of a frequency point is 0.35g, that frequency point is marked as an anomaly, and the process continues to check whether it exceeds the current threshold. If there are multiple abnormal frequency points in a segment, and multiple frequency points exceeding the threshold exist, the time window corresponding to that segment is marked as an abnormal segment. Then, all segments marked as abnormal segments are integrated. If the time interval between two adjacent abnormal segments is less than the set merging interval ΔT_merge, for example, ΔT_merge=50ms, the two segments are merged into one whole segment, ultimately forming a set of tightening process segments. Each segment in this set contains the corresponding time index range, abnormal frequency values ​​in the spectrum, corresponding amplitude values, and distribution of abnormal points, thus completing the accurate division of vibration abnormal segments in the tightening process.

[0027] Please see Figure 3 The specific steps of S2 are as follows: S201: Call the set of segments in the tightening process, calculate the direction of change and the difference in slope for each segment, determine the trend of change, and obtain the slope transformation trend matrix; First, the corresponding torque slope sequence is read segment by segment from the set. Each segment's slope sequence is then expanded into a set of scalar data that can be directly compared, according to the time index. For example, if a segment has 80 sampling points, its slope sequence is recorded as K1 to K... 80The sequence is then processed by subtracting adjacent slope values ​​from each other to obtain the difference needed for direction determination. For each difference, the current slope value is subtracted from the slope value at the next moment, resulting in a continuous difference. For example, if K1 is 0.18 and K2 is 0.22, the difference is 0.04. This difference is then compared to zero. If the difference is greater than zero, the direction is marked as "ascending"; if less than zero, it's marked as "descending"; and if zero, it's marked as "straight". After obtaining the direction mark sequence, the sequence is repeatedly traversed to extract direction changes. Starting from the beginning of the sequence, each direction mark is read and compared with the direction mark at the next moment. If they match, it's recorded as "direction unchanged"; otherwise, it's recorded as "direction changed". The algorithm iterates through segments, recording the corresponding index position when a turning point occurs for subsequent trend statistics. It also calculates the maximum consecutive direction length within each segment. The process involves accumulating the number of consecutive identical direction markers starting from the beginning of the sequence. If a direction change occurs, the accumulated count is written into the consecutive direction length table for that segment, and a new count begins. For example, if a direction marker sequence is [up, up, up, down, down, up], the consecutive lengths are recorded as 3, 2, and 1, and the number of direction changes is counted as 2. This process is repeated across all segments. Finally, the average slope difference, the number of direction changes, and the length of direction duration for each segment are summed and written into a unified matrix structure. Matrix rows correspond to segment numbers, and matrix columns correspond to the direction record sequence, the number of direction changes, the average difference, and the maximum consecutive direction length. The slope change trend matrix is ​​obtained through these steps.

[0028] S202: Based on the slope transformation trend matrix, extract the positioning angle advancement segment, calculate the adjacent angular displacement amplitude and compare the difference with the set angular offset stability threshold to obtain the angular displacement judgment identifier set; Each row of data is screened line by line. The process involves reading the number of direction changes, the maximum continuous direction length, and the average difference of a segment, and comparing them with preset angle advancement stability requirements. In this example, segments with no more than two direction changes, a maximum continuous direction length of no less than five sampling points, and an average difference between 0.02 and 0.10 are considered relatively stable in terms of advancement direction. Therefore, during screening, if a segment meets these three conditions, it is recorded as an angle advancement segment. Subsequently, the time index of each advancement segment is retrieved again to locate the corresponding angle point in the original angular displacement sequence, and each angular displacement value is extracted in chronological order to form the angle sequence θ1 to θ2. mAfter obtaining the sequence, an amplitude difference operation is performed on adjacent angle values. This is done by directly subtracting the current angle data from the angle data at the next moment to obtain the angular displacement amplitude value. For example, if θ5 is 36.80° and θ6 is 37.15° in a certain data segment, the difference is -0.35°. This difference value is written into the angular displacement amplitude sequence. Subsequently, the sequence is processed for amplitude fluctuation. To make a judgment, an angular displacement stability threshold needs to be set. The threshold setting reference object is the acceptable angular displacement fluctuation in the operation. For example, in precision assembly, the angular displacement fluctuation is usually no more than 0.05°, so an angular displacement stability threshold can be set. The qualitative threshold is 0.06°. The judgment action is to traverse the angular displacement amplitude sequence and count whether the fluctuation of all differences remains within 0.06°. Specifically, the angular displacement amplitude value is read one by one and compared with 0.06°. If all differences in a certain segment are less than or equal to the threshold, the offset label of the segment is set as "stable". If any difference is greater than the threshold, it is marked as "offset". The judgment results of all segments are recorded in sequence as elements of the label set, and the segment number, angular displacement amplitude range, maximum angular displacement amplitude value and judgment label are written into it, thus obtaining the angular displacement offset judgment label set.

[0029] S203: Call the frequency peak index corresponding to the segment in the tightening process segment set, analyze the consistency of the direction of change of adjacent peak positions, filter synchronous fluctuation segments, and establish abnormal propulsion structure segments; The segments marked "offset" are processed one by one, and the disturbance segment data of the corresponding time period is retrieved. The action is to locate the peak point of the vibration signal within the disturbance segment. The method is to traverse the vibration signal sequence of the segment and compare the amplitude of each sampling point with the amplitude of the two sampling points before and after it. When the amplitude of a point is greater than the amplitude of its two adjacent amplitudes, the point is recorded as a peak point, and its time position and amplitude value are obtained. All peak points are constructed into a peak position sequence in chronological order. For example, the peak sequence [(1.20s, 0.42g), (1.36s, 0.45g), (1.53s, 0.47g), (1.70s, 0.44g)] is obtained. Then, the direction consistency of the sequence is checked. The action is to read the time interval of adjacent peak points and perform a direct subtraction operation. For example, 1.36s-1.20s results in 0.16s, which is then compared with the next group of 1.53s- A time interval of 1.36s is compared with a time interval of 0.17s. If the time intervals differ by no more than 0.03s, they are considered to be consistent. At the same time, the direction of the peak amplitude change is read. The action is to compare the magnitudes of two adjacent amplitudes. If the amplitude of the later peak is greater than that of the earlier peak, it is marked as "increasing"; if it is less, it is marked as "decreasing". Consistency judgment is performed on adjacent items. If three consecutive sets of time intervals are consistent and the direction of amplitude change is consistent in the sequence, the sequence segment is recorded as a synchronous fluctuation segment. When performing the filtering action, the overlap between the synchronous fluctuation segment and the offset segment time range needs to be checked segment by segment. The overlap judgment action is to compare the magnitudes of the start times of the two time ranges and determine whether there is an intersection interval. If there is an intersection, the fluctuation segment and the angular shift offset segment are integrated, and its peak change direction, peak amplitude sequence, time range and corresponding offset segment number are recorded to establish an abnormal advancement structure segment.

[0030] Please see Figure 4 The specific steps of S3 are as follows: S301: Call the abnormal propulsion structure segment, extract the torque propulsion curve, perform pairwise combination judgment on the slope change direction of adjacent segments, filter out the combination sequence with direction reversal, and generate a set of direction reversal combination identifiers; First, extract the torque variation data within the time index range of the segment, and construct a complete torque sequence in chronological order. For example, if a segment corresponds to the time interval t1=1.2s to t2=2.0s, then obtain all torque values ​​within this interval from the sensor data and label them sequentially as T1, T2, ..., T nThe slope value is calculated for every two adjacent time periods. This involves dividing the torque difference by the time difference for each pair of consecutive sampling points to obtain the propulsion slope for each segment. After the slope value sequence is constructed, adjacent slope segments are paired, and a direction judgment operation is performed on their values. When the subsequent slope value is greater than the previous slope value, it is considered that the direction is consistent; when the subsequent slope value is less than the previous slope value, it is considered that the direction is reversed. During the judgment, the above judgment logic is executed group by group for each pair of slope segments, and the judgment result is recorded. The judgment record is a Boolean value, with "0" indicating consistent direction and "1" indicating reversed direction. For example, if the slope sequence is [0.15, 0.12, 0.18, 0.14], then the slope pairs are respectively ( 0.15, 0.12), (0.12, 0.18), (0.18, 0.14), the first pair is reversed, the second pair is consistent, and the third pair is reversed, the direction mark sequence is [1, 0, 1]. Further filter out the sequence segments with two or more direction reversal marks, record the start segment number and end segment number of the sequence segment, and use them as candidates for direction reversal combinations. Set the filtering threshold to include the segment if the number of direction reversals is not less than 2. If there are 5 sets of slope pairs in a certain structural segment, 3 of which are direction reversals, then the segment meets the filtering conditions. Finally, the indexes of all structural segments that meet the conditions and the direction mark sequences are formed into a unified structure to generate a direction reversal combination identifier set.

[0031] S302: Based on the direction reversal combination identifier set, extract the angle advancement rhythm sequence, calculate the adjacent angle shift pitch and analyze the pitch change trend to obtain the pitch change classification label set; The angle advance data for each combined structure is read one by one. An angle advance rhythm sequence is extracted within the time range of each combined segment. First, angular displacement sampling points are extracted from the rhythm sequence in chronological order, and then an angle advance rhythm sequence is constructed. This sequence consists of the time label and angle value corresponding to each instantaneous angular displacement. For example, if the combined segment time is t... a If t = 3.0s to t_b = 4.0s, then extract all θ values ​​within this interval and arrange them in order as θ1 to θ2. mThe propulsion pitch between adjacent angle values ​​is then calculated. This involves reading adjacent angular displacement points at sampling time intervals and using the difference as the propulsion pitch. For example, if θ1 = 22.0° and θ2 = 22.8°, the pitch is 0.8°. All pitches are sequentially arranged to form a propulsion pitch sequence. After obtaining the complete pitch sequence, pitch change trend processing is performed. This involves comparing the differences between two adjacent pitch values. If the later pitch is greater than the earlier pitch, the group is marked as "increasing"; if they are equal, it is marked as "stable"; and if the later pitch is less than the earlier pitch, it is marked as "decreasing". This judgment logic is continued across the entire sequence, and the corresponding values ​​for each group are recorded. The change labels are used to classify the rhythm trends of each combination segment. When multiple directions of change occur, such as the pattern of "increasing → decreasing → increasing", the trend is recorded as "fluctuating". When the change labels are continuously "increasing" or "decreasing", it is recorded as "unidirectional". If the number of "stable" labels exceeds 80%, it is marked as "stable". To determine the stability, a threshold needs to be set. The threshold is set to a pitch change of no more than 0.05°. That is, when the difference between adjacent pitches is less than or equal to 0.05°, it is considered that there is no significant change. After classifying the rhythm trends of each combination segment, the classification labels are recorded and matched with the combination number to form a set of pitch change classification labels.

[0032] S303: Call the pitch change classification label set, extract the rhythm inconsistency combination structure, number and label it according to the advancement order, aggregate the index interval corresponding to the labeled segment, and output the abnormal rhythm marked segment. All rhythm segments identified as "fluctuating" are selected, and their structure numbers are extracted from the corresponding combination structures. The elements marked "fluctuating" within these segments are then assigned a progression order number based on their original time index order. This involves reading their starting index position on the time axis and sorting them in the entire set from smallest to largest time, assigning numbers 1, 2, 3…n according to the sorting order. Subsequently, the start and end positions of the time index corresponding to each numbered fluctuation segment are summarized to construct an identifier segment index table. This table records the segment number, starting index, ending index, and category label for each segment. After the construction is completed, interval aggregation processing is performed on the consecutively numbered fluctuation segments. That is, when two or more fluctuation segments are continuous in time index, they are merged into an abnormal rhythm combination segment. The criterion for judging continuity is that the time index interval between two segments does not exceed 10 sampling points. If the end time of two adjacent segments is 1000 and the start time of the next segment is 1007, they are judged to be continuous, and the aggregation operation is performed to merge their start and end times into 1000 to the end time of the next segment, for example, 1025. Finally, all segments that meet the aggregation conditions are written into the structure list, and the list is output as the abnormal rhythm marked segment.

[0033] Please see Figure 5 The specific steps of S4 are as follows: S401: Call the numbered segments in the abnormal rhythm marker section, extract the angle sequence and torque slope sequence of each segment during the tightening process, calculate the slope direction change of adjacent segments and match and compare it with the angle rhythm change, filter the combination segments where the angle rhythm is not synchronized but the torque direction has shifted, and generate a rhythm slope offset matching set. For each numbered segment, the data interval corresponding to its time index is extracted one by one. Within this interval, the angle progression sequence is extracted first, and then the torque sequence is extracted and the corresponding slope sequence is calculated. The angle sequence is constructed by arranging the angular displacement values ​​at each sampling time in the order of the time index. For example, if the time range of a segment is from t=2.00s to t=2.80s and the sampling interval is 5ms, then 161 sets of angular displacement values ​​are obtained and arranged in ascending order of time to form θ1 to θ2. 160 The angle sequence is then used to extract torque data within the same time interval. The difference between adjacent sampling points in the torque value sequence is calculated and divided by the time interval to form a slope sequence. The action is T. n +1–T n Subtract the numerical values ​​and then use Δt as the divisor to obtain the instantaneous slope value, thus obtaining the slope sequence K1 to K2. 159 After obtaining the angular progression rhythm sequence, rhythm difference processing is performed again, that is, for θ n With θ n The difference of +1 constitutes the rhythm sequence Δθ1 to Δθ 159 The rhythm sequence and slope sequence are paired and compared. First, the difference between two adjacent rhythm values ​​in each segment and the change in slope direction are used to make a pairwise judgment. The judgment condition is set as follows: if the rhythm change trend remains unchanged but the slope change direction is reversed, the combination is judged as "rhythm is not synchronized but slope is reversed". In the judgment, the rhythm change label of each pair of adjacent segments is extracted and compared with the slope change label. The rhythm change label is "rising", "falling" or "stable", and the slope change label is also defined as "rising", "falling" or "stable". When the two labels are inconsistent, the combination segment is recorded as a candidate segment. The above pairing action is continued until all numbered segments are processed. Finally, all combination segments that meet the judgment condition are recorded according to their original number and index position information and a unified data structure is generated to form a rhythm slope offset matching set.

[0034] S402: Based on the rhythm slope offset matching set, extract the combined segment number and corresponding index, sort them according to the order of advancement, divide them into consecutive number groups, arrange the node positions in each group of segments, and obtain the rhythm offset combined node sequence. Read the unique number and corresponding time index range of each combined segment, and sort all numbers in chronological order. The sorting process involves extracting the start time represented by each number and arranging them in ascending order. After sorting, new advance number identifiers are assigned sequentially, for example, the earliest segment is number 1, the second segment is number 2, and so on. Then, the continuity of these renumbered combined segments is determined. The determination criteria are that the difference between the numbers of two combined segments is 1 and the interval between the time indices does not exceed the specified threshold Δt_group. The threshold value can be set to 50ms according to the normal continuous duration of the tightening rhythm. If the start and end times of the two combined segments with numbers 5 and 6 are 1, 2, 3, 4, 5, 6, 7, 8, 9, 1 ... The time intervals from 0.50s to 1.70s and from 1.71s to 1.95s are 10ms, which is less than 50ms, satisfying the continuous combination condition and forming a numbered group. All combination segments that meet this continuous condition are clustered to form multiple combination groups. For each group of segments, a node position extraction operation is performed, that is, the start index and end index of each segment in the group are extracted and recorded as combination nodes. During the execution, all nodes are summarized according to the time index in the original segment and deduplicated to obtain a set of combination nodes. Then, each node is sorted to form a rhythm offset combination node sequence. Each element in this sequence is a time index position, along with its source segment number and corresponding sequence number in the group.

[0035] S403: Call the rhythm offset combination node sequence, aggregate the position indices of the interlaced occurrences according to the index sequence, identify the structure type and interlacing attributes within the position markers, and output the tightening rhythm interlacing area; The system performs continuity checks on the time index values ​​in the node sequence and executes interleaved aggregation. The aggregation logic starts from the first node of the sequence, sequentially reading the time index values ​​of adjacent nodes, calculating their intervals, and determining if any discontinuities occur. If the difference between a node's index and the previous node is greater than a set threshold Δt_cross, it is recorded as the interleaving start point. The system then searches for the next node that meets the breakpoint condition as the interleaving end point and marks this index range as an interleaved segment. After completing the interleaving of the entire sequence, the structure type in each interleaved segment is extracted. The structure type originates from the combination to which the segment belongs. The torque change pattern and angle advancement rhythm identifier in the process are executed by mapping the combination segment number of the node to the structural attribute field recorded in the rhythm slope offset matching set. For example, if the source segment of a node is segment number 7 and its structural type is "stable rhythm + decreasing slope", then this type is written into the interlaced segment attribute. Further, all structural types in each interlaced segment are integrated and statistically analyzed, and the interlaced attribute information of the interlaced segment is output, including the types of structural types that appear, the quantity distribution of each type, the length of the interlaced segment, the start and end index, and other parameters. Finally, all interlaced segments are sorted in chronological order and output as the tightening rhythm interlaced zone.

[0036] Please see Figure 6The specific steps of S5 are as follows: S501: Call the tightening rhythm interlacing area, extract the directional connection trend of each advancement segment, analyze the changes in the rhythm before and after based on the trend, filter and process the advancement segments with repeated rhythms, and generate a set of rhythm repeating advancement segment identifiers. After locating each numbered segment node, the torque propulsion structure information preceding that node is extracted by backtracking. Specifically, based on the start time index of the rhythm-interlacing section, the termination index of the adjacent propulsion segment is queried forward. Two or more consecutive segments of torque slope value sequences are extracted from the corresponding propulsion structure. Adjacent direction changes are judged on the extracted slope sequences. This judgment involves comparing the start and termination slopes of each propulsion segment. If the termination slope is higher than the start slope, the direction is marked as "ascending"; if lower, it is marked as "descending"; if equal, it is marked as "level". By performing the above judgment operation on multiple propulsion segments, a set of directional connection trend sequences can be obtained. For example, if the direction of three consecutive propulsion segments is "ascending → ascending → descending", the trend shows a stable initial state followed by a change. Further rhythm comparison processing is performed on this trend, which involves comparing the trend with the subsequent... The rhythm progression sequence is paired segment by segment to determine whether the rhythm changes with the slope direction. That is, if the slope changes direction but the rhythm remains the same, it is considered that the rhythm is repeated. In the judgment, a rhythm difference threshold θ_repeat_thresh needs to be set to determine whether the rhythm is repeated. This threshold can be set to 0.1°, that is, if the difference in the average rhythm value between two adjacent progression segments is less than or equal to 0.1°, the rhythm is considered repeated. During execution, the average progression angle rhythm value of each pair of progression segments is extracted and the difference is compared with the threshold. If the difference is within ±0.1° and the corresponding slope direction changes, the progression segment number is recorded in the identifier set. Finally, all progression segment numbers that meet the condition, along with their start and end time indices, slope direction change markers, rhythm repetition ratios, and other parameters, are integrated to generate a rhythm repetition progression segment identifier set.

[0037] S502: Based on the rhythm repeating advance segment identifier set, the advance segments in the identifiers are extracted and cleared at intervals, and the continuity of the remaining segments is calculated to obtain the advance segments that meet the set direction connection threshold, and a set of advance segments after clearing is generated. The recorded propulsion segments undergo an interval extraction and clearing operation. Specifically, each propulsion segment number and its corresponding time index interval are read sequentially. This interval is marked as a "removable segment" in the original propulsion structure. The time intervals between these segments are extracted and judged. If a propulsion segment exists between two adjacent "removable segments" and the start-end time interval of that segment exceeds the set minimum propulsion segment threshold Δt_min, then that segment is retained for subsequent analysis. Δt_min can be set to 100ms to ensure the remaining propulsion segments have a continuous structure. Subsequently, the propulsion sequence is reconstructed in chronological order for all remaining propulsion segments after clearing, and its continuity is evaluated. The operation involves traversing the start and end time indices of the remaining segments, and for any two phases... The time difference between adjacent propulsion segments is used for judgment. If the time difference is less than the set directional connection time threshold Δt_dir_thresh, it is considered continuous. Δt_dir_thresh is set to 50ms based on the mechanical system response characteristics. If two propulsion segments terminate at t=2.20s and start at t=2.25s, the time interval is 50ms, which meets the continuity condition, and the connection is retained. If it exceeds the threshold, the connection is broken, and the remaining segments are traversed until all propulsion segments are processed. Finally, the sequence of consecutive propulsion segments and its structural information are recorded as propulsion segments, and the cleared propulsion segment set is output. Each propulsion segment in this set contains segment number, directional trend, time index range, and source segment index information.

[0038] S503: Call the set of cleared and advanced segments, perform sequential splicing of segments according to the direction information of each segment, arrange and combine interlaced nodes in sequence, and output the torque evaluation result of automotive fasteners; Each segment in the set is sorted in ascending order by time index. The direction marker information of each segment is read sequentially, and the segments are spliced ​​together according to the progression trend. Specifically, the ending direction of the previous progressive segment is compared with the starting direction of the next segment. If the two directions are the same, the two segments are merged into a linear progression structure. If they are not the same, they are not merged, but their time order is preserved. The same judgment is performed again until the entire set is traversed. In the progress structure after the splicing is completed, the corresponding crossover node positions are extracted for each linear structure. This operation is performed by cross-referencing the time index list of the progress structure with the rhythm offset combination node sequence. The method involves filtering out all interlaced node index positions within the propulsion structure and arranging these interlaced positions according to their relative positions within the propulsion structure to construct an interlaced index sequence. After arranging the interlaced nodes, the torque value corresponding to each interlaced node is read and compared with the slope direction of its propulsion structure to determine whether the torque at that node shows a turning trend. If the turning point coincides with the node, it is recorded as a critical torque node. Finally, the position, time index, torque value, and direction information of all critical torque interlaced nodes in the propulsion structure are organized into structured result data, and the automotive fastener torque evaluation result is output.

[0039] Please see Figure 7 A smart sensor-based automotive fastener torque evaluation system includes: The data acquisition and segmentation module is used to achieve S1: acquiring the time series of the intelligent torque sensor, angular displacement sensor, and vibration channel, dividing the torque slope change into segments, truncating the angle advance according to the tightening displacement interval, extracting the peak position of the vibration frequency peak, and generating a set of tightening process segments; The stability and anomaly detection module is used to implement S2: by using the segment set of the tightening process, analyze the slope change direction and difference trend, determine the stability of the angular displacement amplitude of the angular displacement segment, judge the consistency of the peak value change trend of the disturbance segment, and output the abnormal propulsion structure segment; The rhythm anomaly marking module is used to implement S3: based on the abnormal propulsion structure fragments, determine whether the directions of adjacent segments of the torque curve are reversed, classify the angle interval changes in the angle change rate features, mark the combination structure with inconsistent wave order, and output the abnormal rhythm marking segment. The rhythm interlacing recognition module is used to implement S4: based on the abnormal rhythm marker section, analyze the pairing relationship between angle and torque slope, filter the sections with asynchronous angles but shifted torque direction, combine interlacing nodes, and output the tightening rhythm interlacing area; The torque assessment output module is used to implement S5: using the front torque structure of the tightening rhythm staggered zone, it judges the connection trend of the propulsion section direction, clears the rhythm repetition segment, performs the timing connection according to the direction, and outputs the torque assessment result of the automotive fastener.

[0040] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for evaluating the torque of automotive fasteners based on intelligent sensors, characterized in that, Includes the following steps: S1: Collect time series data from torque sensor, angular displacement sensor, and vibration channel; divide the torque slope change into segments; cut the angle advance according to the tightening displacement interval; extract the peak position of the vibration frequency peak; and generate a set of tightening process segments. S2: Using the segmented set of the tightening process, analyze the slope change direction and difference trend, determine the stability of the angular displacement amplitude of the angular displacement segment, judge the consistency of the peak value change trend of the disturbance segment, and output the abnormal propulsion structure segment. S3: Based on the abnormal propulsion structure segment, determine whether the directions of adjacent segments of the torque curve are reversed, classify the angle interval change in the angle change rate feature, mark the combination structure with inconsistent wave order, and output the abnormal rhythm mark segment. S4: From the abnormal rhythm marked sections, analyze the pairing relationship between angle and torque slope, filter out sections with asynchronous angles but shifted torque direction, combine staggered nodes, and output the tightening rhythm staggered area; S5: Using the aforementioned tightening rhythm staggered zone, determine the trend of the direction connection of the advancement section, clear the rhythm repetition segment, connect the timing according to the direction, and output the torque evaluation result of the automotive fastener.

2. The method for evaluating the torque of automotive fasteners based on intelligent sensors according to claim 1, characterized in that, The tightening process segment set includes torque growth slope segments, angle propulsion segments divided by tightening displacement, and frequency peak indices within the time window. The abnormal propulsion structure segments include slope change direction sequences, angular displacement amplitude stability indicators, and adjacent peak position trends. The abnormal rhythm marking segments include propulsion direction reversal patterns, pitch change type labels, and fluctuation sequence numbering information. The tightening rhythm interlacing zone includes angle and torque slope pairing combinations, asynchronous change segment combinations, and interlaced propulsion node sequences. The automotive fastener torque evaluation results include propulsion direction connection characteristics, rhythm repetition interval patterns, and linear combination propulsion structures.

3. The method for evaluating the torque of automotive fasteners based on intelligent sensors according to claim 1, characterized in that, The trend of connection between the propulsion sections refers to the continuity and consistency of adjacent tightening propulsion sections in the direction of torque change.

4. The method for evaluating the torque of automotive fasteners based on intelligent sensors according to claim 1, characterized in that, The stability of angular displacement amplitude refers to the fact that within an angular displacement range, the fluctuation range of the angular change amplitude in the time series is less than a preset amplitude fluctuation variance threshold.

5. The method for evaluating the torque of automotive fasteners based on intelligent sensors according to claim 1, characterized in that, The specific steps of S1 are as follows: S101: Obtain the time series of the intelligent torque sensor, calculate the torque difference between adjacent sampling points, compare the change rate sequence obtained by dividing the difference by the time interval with the set torque growth change rate threshold, divide the slope change segment, and generate the torque growth slope segment division structure. S102: Call the time index of the torque growth slope segment division structure, extract the corresponding angular displacement sequence, calculate the displacement increment, and cut continuous segments according to the set tightening displacement interval threshold to generate a continuous angular segment sequence of displacement interval. S103: Call the time index of the continuous angle segment sequence of the displacement interval, extract the corresponding vibration signal, perform spectrum analysis, locate the frequency point where the amplitude exceeds the set vibration peak threshold, and establish a set of tightening process segments.

6. The method for evaluating the torque of automotive fasteners based on intelligent sensors according to claim 1, characterized in that, The specific steps of S2 are as follows: S201: Call the set of segments of the tightening process, calculate the direction of change and the slope difference of each slope sequence, determine the trend of change, and obtain the slope transformation trend matrix; S202: Based on the slope transformation trend matrix, extract the positioning angle advancement segment, calculate the adjacent angle displacement amplitude and compare the difference with the set angle offset stability threshold to obtain the angle displacement judgment identifier set; S203: Call the frequency peak index corresponding to the segment in the tightening process segment set, analyze the consistency of the direction of change of adjacent peak positions, filter synchronous fluctuation segments, and establish abnormal propulsion structure segments.

7. The method for evaluating the torque of automotive fasteners based on intelligent sensors according to claim 1, characterized in that, The specific steps for S3 are as follows: S301: Call the abnormal propulsion structure segment, extract the torque propulsion curve, perform pairwise combination judgment on the slope change direction of adjacent segments, filter out the combination sequence with direction reversal, and generate a direction reversal combination identifier set; S302: Based on the direction reversal combination identifier set, extract the angle advancement rhythm sequence, calculate the adjacent angle shift pitch and the pitch change trend to obtain the pitch change classification label set; S303: Call the set of classification labels for pitch changes, extract the combination structure of inconsistent rhythms, number and identify it according to the order of advancement, aggregate the index intervals corresponding to the identified segments, and output the abnormal rhythm marked segments.

8. The method for evaluating the torque of automotive fasteners based on intelligent sensors according to claim 1, characterized in that, The specific steps of S4 are as follows: S401: Call the numbered segments in the abnormal rhythm marker section, extract the angle sequence and torque slope sequence of each segment during the tightening process, calculate the slope direction change of adjacent segments and match and compare it with the angle rhythm change, filter the combination segments where the angle rhythm is not synchronized but the torque direction has shifted, and generate a rhythm slope offset matching set. S402: Based on the rhythm slope offset matching set, extract the combined segment number and corresponding index, sort them according to the order of advancement, divide them into consecutive number groups, arrange the node positions in each group of segments, and obtain the rhythm offset combined node sequence. S403: Call the rhythm offset combination node sequence, aggregate the position indices of the interlaced occurrences according to the index sequence, identify the structure type and interlacing attributes within the position markers, and output the tightening rhythm interlacing area.

9. The method for evaluating the torque of automotive fasteners based on intelligent sensors according to claim 1, characterized in that, The specific steps of S5 are as follows: S501: Call the tightening rhythm interlacing area, extract the directional connection trend of each advancement segment, analyze the change of rhythm before and after based on the trend, filter and process the advancement segments with repeated rhythms, and generate a set of rhythm repeating advancement segment identifiers. S502: Based on the rhythm repeating advance segment identifier set, the advance segments in the identifiers are extracted and cleared at intervals, and the continuity of the remaining segments is calculated to obtain the advance segments that meet the set direction connection threshold, and a set of advance segments after clearing is generated. S503: Call the set of cleared and advanced segments, perform sequential splicing of the segments according to the direction information of each segment, arrange and combine the interlaced nodes in sequence, and output the torque evaluation result of the automotive fastener.

10. A torque evaluation system for automotive fasteners based on intelligent sensors, characterized in that, The system is used to implement the automotive fastener torque evaluation method based on intelligent sensors as described in any one of claims 1-9, and the system comprises: The data acquisition and segmentation module is used to implement S1: acquiring the time series of torque sensor, angular displacement sensor and vibration channel, performing segmentation on the change of torque slope, intercepting the angle advance according to the tightening displacement interval, extracting the peak position of vibration frequency peak, and generating a set of tightening process segments; The stability and anomaly detection module is used to implement S2: using the segment set of the tightening process, analyze the slope change direction and difference trend, determine the stability of the angular displacement amplitude of the angular displacement segment, determine the consistency of the peak value change trend of the disturbance segment, and output the abnormal propulsion structure segment; The rhythm anomaly marking module is used to implement S3: based on the abnormal propulsion structure segment, determine whether the direction of adjacent segments of the torque curve is reversed, classify the angle interval change in the angle change rate feature, mark the combination structure with inconsistent wave order, and output the abnormal rhythm marking segment. The rhythm interlacing identification module is used to implement S4: from the abnormal rhythm marking section, analyze the pairing relationship between angle and torque slope, filter the sections with asynchronous angles but shifted torque direction, combine interlacing nodes, and output the tightening rhythm interlacing area; The torque evaluation output module is used to implement S5: using the tightening rhythm staggered zone, determine the connection trend of the propulsion section direction, clear the rhythm repetition segment, perform timing connection according to direction, and output the torque evaluation result of the automotive fastener.

Images