Automobile wire harness quality detection method and system based on multi-source heterogeneous data

Abstract

The invention discloses an automobile wire harness quality detection method and system based on multi-source heterogeneous data, and relates to the technical field of automobile wire harness quality detection.The voltage waveform of an automobile communication wire harness is collected, edge mutation recognition is carried out, waveform response offset characteristics are accurately extracted, and rapid early warning of early risks is achieved in combination with a preset threshold value; on the basis, an impedance drift factor and a phase symmetry factor are extracted and fused to form a communication waveform dissimilation index, the influence of a wiring structure on communication performance is accurately quantified, dissimilation grade evaluation is completed, when structural degradation is identified, wire harness segmentation analysis is automatically executed, a structure-induced degradation risk segment is identified, and an abnormal result analysis table is generated; maintenance suggestions are pushed and repair data are fed back, so that structural closed-loop feedback is formed, and the diagnosis depth and the maintenance guidance capability of the automobile communication wire harness under the operation condition are effectively improved.

Description

Technical Field

[0001] This invention relates to the field of automotive wiring harness quality inspection technology, specifically to an automotive wiring harness quality inspection method and system based on multi-source heterogeneous data. Background Technology

[0002] With the continuous improvement of the integration and intelligence level of automotive electronic systems, communication harnesses, as a key signal transmission medium in the vehicle's electrical architecture, have a decisive impact on the functional safety of the entire vehicle due to their operational stability and anti-interference capabilities. In particular, high-speed differential communication, such as CAN-FD and 100Base-T1 Ethernet, is widely used in advanced driver assistance systems (ADAS) and intelligent connected systems. As a result, the quality inspection of automotive communication harnesses has been extended from traditional continuity testing to communication integrity assessment and structural reliability identification.

[0003] Chinese invention patent application CN117368223A discloses a machine vision-based method and system for inspecting the quality of automotive wiring harnesses, relating to the automotive industry. The inspection method includes appearance inspection, tensile testing, compression testing, and communication testing. The inspection system includes a tensile testing module, a compression testing module, a communication testing module, and a machine vision inspection module, and the system is matched to the inspection method. Using this machine vision-based method and system, the performance of automotive wiring harnesses, such as appearance, tensile strength, compressive strength, and communication quality, can be inspected based on machine vision technology. The machine vision inspection module provides corresponding evaluation reports, achieving versatility of the equipment. Furthermore, it can obtain reliable inspection results based on the data processing and analysis capabilities of machine vision, thus reducing costs and improving inspection quality.

[0004] The above methods can perform quality inspection of automotive wiring harnesses when they leave the factory. However, apart from this, the existing quality inspection methods for automotive wiring harnesses mainly focus on static testing before leaving the factory.

[0005] However, existing automotive wiring harness testing methods generally employ continuity testing, resistance testing, or static insulation assessment in the unfolded state of the wiring harness. While these methods can identify physical open circuits, short circuits, or poor contact issues, they struggle to detect communication degradation issues caused by wiring structure features such as small-radius bends, damaged shielding layers, and stress concentration at clamping points. In actual vehicle assembly, due to space constraints, automotive communication wiring harnesses are often arranged in an "S-shaped," "U-shaped," or irregular three-dimensional bend configuration, resulting in inconsistent conductor paths, impedance discontinuities, and electromagnetic coupling imbalances. This, in turn, leads to anomalies such as differential signal phase drift, edge distortion, and impedance jumps. These phenomena are often difficult to detect during functional continuity testing, posing significant engineering risks.

[0006] Therefore, the present invention provides a method and system for detecting the quality of automotive wiring harnesses based on multi-source heterogeneous data. Summary of the Invention

[0007] To address the shortcomings of existing technologies, this invention provides a method and system for detecting automotive wiring harness quality based on multi-source heterogeneous data, thus solving the problems mentioned in the background.

[0008] To achieve the above objectives, the present invention provides the following technical solution: a method for inspecting the quality of automotive wiring harnesses based on multi-source heterogeneous data, comprising the following steps:

[0009] S1. Use the high-speed ADC interface inside the vehicle ECU to collect relevant data of the vehicle communication harness in real time during vehicle operation, perform edge calibration, and extract the waveform response offset coefficient Bx in the calibration edge segment.

[0010] S2. Preset waveform response offset threshold BXyz, compare and analyze the waveform response offset coefficient Bx and the waveform response offset threshold BXyz, evaluate the safety status of the automotive communication harness, and trigger the risk alarm mechanism.

[0011] S3. Real-time collection of communication waveform electrical signal data and feature extraction to obtain the impedance drift factor ΔZ. diff and phase symmetry factor Φ CM Then, a weighted fusion calculation is performed to obtain the communication waveform heterogeneity index F. DST Classify automotive communication harnesses according to their degree of differentiation;

[0012] S4. Divide the automotive communication harness of the third differentiation level into N harness segments, and combine them with the communication waveform differentiation index F. DST It identifies structurally induced degradation risk segments and generates an anomaly analysis table;

[0013] S5. Based on the abnormal result analysis table, generate repair suggestions for the automotive communication wiring harness and collect relevant data on the automotive communication wiring harness during the repair process for feedback in real time.

[0014] Preferably, step S1 specifically includes:

[0015] S11. During vehicle operation, an automotive-grade ADC data acquisition unit is connected to the vehicle controller to continuously sample data, obtain the voltage waveform of the automotive communication harness over a period of time in real time, and construct a data set H related to the automotive communication harness.

[0016] S12. Based on the relevant data set H of the automotive communication wiring harness, perform edge calibration. The specific process of edge calibration is as follows:

[0017] Set a fixed time sampling window T wAnd extract the sampling window T at a fixed time. w The voltage waveform of the automotive communication harness inside the vehicle is obtained, and the voltage difference ΔV(t) at each data sampling time point is calculated. i ), generating a differential sequence, where the voltage difference ΔV(t) i The specific method of obtaining it is as follows:

[0018] ΔV(t i )=V(t i+1 )-V(t i );

[0019] In the formula, ΔV(t) i ) represents the data sampling time point t i The voltage difference of the automotive communication harness, V(t) i+1 ) and V(t i ) represent the data sampling time points t and t, respectively. i+1 and t i The voltage of the automotive communication harness at that time;

[0020] Set the judgment threshold ∈ s And in the difference sequence, the judgment threshold ∈ s and voltage difference ΔV(t) i A comparative evaluation was conducted to analyze the voltage abrupt change state at the data sampling time point, where the data sampling time point t was used as the basis for the evaluation. i For example, if ΔV(t) i )≥∈ s This indicates the data sampling time point t. i This is the mutation point, at which the data sampling time point t is marked. i If ΔV(t) is "1", then... i )<∈ s This indicates the data sampling time point t. i This is the normal point; at this time, the data sampling time point t is marked. i The value is "0", and all mutation points are collected to form a mutation marker sequence;

[0021] In the mutation marker sequence, M consecutive data sampling time points marked as "1" are extracted to construct a candidate edge segment, thereby obtaining several candidate edge segments. For each candidate edge segment, the mutation score PF is calculated, and the candidate edge segment with the largest mutation score PF is selected as the edge labeling segment. Taking candidate edge segment j as an example, the mutation score PF of candidate edge segment j is... (j) The specific method of obtaining it is as follows:

[0022]

[0023] In the formula, This represents the mean of the voltage differences at M data sampling time points within the candidate edge segment j. Let represent the variance of the voltage difference at M data sampling time points within the candidate edge segment j, ∈ represent the disturbance adjustment constant, and ln represent the natural logarithm function.

[0024] Preferably, step S1 further includes,

[0025] S13. Based on the relevant data set H of the automotive communication wiring harness, extract the voltage waveform of the automotive communication wiring harness within the edge calibration section, and compare and analyze the voltage waveform of the automotive communication wiring harness within the edge calibration section with the calibration voltage waveform to obtain the waveform response offset coefficient Bx. The specific method for obtaining the waveform response offset coefficient Bx is as follows:

[0026]

[0027] In the formula, V sig (t m ) represents the data sampling time point t m The voltage of the automotive communication harness at that time, V ref (t m ) represents the data sampling time point t m The standard voltage of the automotive communication harness at that time, M represents the total number of data sampling time points within the edge calibration section, and m represents the index of the data sampling time point within the edge calibration section.

[0028] Preferably, step S2 specifically includes:

[0029] S21. Preset the waveform response offset threshold BXyz for the automotive communication wiring harness, and compare and analyze the waveform response offset threshold BXyz and the waveform response offset coefficient Bx to evaluate the safety status of the automotive communication wiring harness. The specific evaluation process is as follows:

[0030] If the waveform response offset coefficient Bx of the automotive communication harness is less than or equal to the waveform response offset threshold BXyz, i.e., Bx≤BXyz, then the automotive communication harness is judged to be in a safe state. At this time, the relevant data of the automotive communication harness during the operation of the vehicle are continuously collected and the collected relevant data of the automotive communication harness are stored in the automotive quality inspection database.

[0031] If the waveform response offset coefficient Bx of the automotive communication harness is greater than the waveform response offset threshold BXyz, i.e., Bx > BXyz, then the automotive communication harness is determined to be in a risky state. At this time, the risk alarm mechanism is automatically triggered, and the communication waveform abnormal response identification process is carried out.

[0032] Preferably, step S3 specifically includes:

[0033] S31. Call the vehicle electronic wiring topology database, read the wiring path model and communication type of the vehicle communication harness, install differential sampling probes on the vehicle communication harness, collect the communication waveform electrical signal data of the vehicle communication harness in real time, and construct a differential waveform data set. The communication waveform electrical signal data includes the current difference I between conductor A and conductor B in the vehicle communication harness. AB (t), the voltage difference V between conductor A and conductor B AB (t) and voltage waveforms of conductors A and B in the automotive communication harness.

[0034] Preferably, step S3 further includes,

[0035] S32. Based on the differential waveform data set, calculate the deviation between the actual differential transmission path of the automotive communication harness and the theoretical design impedance within the time window [tW, t], to obtain the impedance drift factor ΔZ. diff Among them, the impedance drift factor ΔZ diff (t) is obtained in the following way:

[0036]

[0037] In the formula, Z ref V represents the design reference impedance of the automotive communication harness. AB (τ) and I AB (τ) represents the voltage difference and current difference between conductor A and conductor B in the automotive communication harness at the data sampling time point τ, respectively. τ represents the index of the data sampling time point, τ∈{tW, t}, W represents the length of the time window [tW, t], and k represents the adjustment coefficient.

[0038] S33. Perform FFT analysis on the voltage waveforms of conductors A and B in the automotive communication wiring harness to extract the dominant frequency phase Φ of conductors A and B in the automotive communication wiring harness. A (t) and Φ B (t), and calculate the average absolute value of the phase difference between the main frequency components of conductors A and B in the automotive communication harness within the time window [tW, t], in order to obtain the phase symmetry factor Φ. CM Wherein, the phase symmetry factor Φ CM The specific method of obtaining it is as follows:

[0039]

[0040] In the formula, Φ A (t) represents the dominant frequency phase of conductor A in the automotive communication harness, Φ B (t) represents the dominant frequency phase of conductor B in the automotive communication harness.

[0041] Preferably, step S3 further includes,

[0042] S34. Based on the obtained impedance drift factor ΔZ diff and phase symmetry factor Φ CM A weighted summation calculation is performed to obtain the communication waveform heterogeneity index F of the automotive communication harness. DST Among them, the communication waveform distortion index F DST The specific method of obtaining it is as follows:

[0043]

[0044] In the formula, α and β represent the impedance drift factor ΔZ, respectively. diff and phase symmetry factor Φ CM The weighting coefficients, where C represents the first correction constant;

[0045] S35. Preset the first communication waveform distortion threshold δ1 and the second communication waveform distortion threshold δ2, and set the communication waveform distortion index F DST The distortion level of the automotive communication harness is evaluated by comparing it with the first communication waveform distortion threshold δ1 and the second communication waveform distortion threshold δ2. The specific evaluation content is as follows:

[0046] If the communication waveform heterogeneity index F DST If the value is less than or equal to the first communication waveform distortion threshold δ1, the automotive communication harness is determined to be at the first distortion level. At this time, the automotive communication harness has no significant structural interference and is marked as “C1”.

[0047] If the communication waveform heterogeneity index F DST If the value is greater than the first communication waveform distortion threshold δ1 and less than the second communication waveform distortion threshold δ2, then the automotive communication harness is determined to be at the second distortion level. At this time, structural interference of the automotive communication harness initially appears, and the automotive communication harness is marked as "C2".

[0048] If the communication waveform heterogeneity index F DST If the value is greater than or equal to the second communication waveform distortion threshold δ2, the automotive communication harness is determined to be at the third distortion level. At this time, the automotive communication harness wiring structure is in a distortion state, automatically triggering the automotive wiring structure risk inspection process and marking the automotive communication harness as "C3".

[0049] Preferably, step S4 specifically includes:

[0050] S41. After determining that the automotive communication wiring harness is of the third level of alienation, read the CAD drawing of the wiring harness of the vehicle, divide the automotive communication wiring harness into N wiring harness segments according to a fixed physical length, and establish a virtual mapping model of the wiring harness segments. The virtual mapping model of the wiring harness segments specifically includes the spatial position of each wiring harness segment in the vehicle body, the minimum bending radius of the wiring harness segment, and the maximum torsion angle of the wiring harness segment.

[0051] S42. Inject a short-period synchronous communication pulse into one end of the automotive communication harness, and use a time-domain reflectometry (TDR) system to collect the communication waveform electrical signal data of the automotive communication harness in real time at the end of each harness segment, and obtain the communication waveform distortion index F of each harness segment. DST If the communication waveform distortion index F of the harness segment DST Greater than μ FDST +2σ FDST If the wire harness segment is marked as a "structure-induced degradation risk segment", recommended measures are generated, including adjusting the wire harness bending radius to 120% of the actual bending radius and adjusting the clamping point density to 80% of the actual clamping point density. If the communication waveform alienation index F of the wire harness segment is... DST Less than or equal to μ FDST +2σ FDST If so, no processing is required;

[0052] S43. For each “structure-induced degradation risk segment”, generate an anomaly analysis table and push the generated anomaly analysis table to the automotive MES system. The anomaly analysis table includes the complete radius, bending shape and communication alienation level of each “structure-induced degradation risk segment”.

[0053] Preferably, step S5 specifically includes:

[0054] S51. Receive and generate equipment maintenance suggestions based on the abnormal result analysis table, including increasing the wiring radius, removing clamping points, replacing the shielding structure, replacing the entire wiring harness, and encapsulating local stress buffers. Bind the equipment maintenance instructions to the "structural induced degradation risk segment" and send the equipment maintenance suggestions to the vehicle owner terminal and the maintenance personnel terminal simultaneously. During the execution of the maintenance task, collect relevant data of the "structural induced degradation risk segment" and provide data feedback.

[0055] Preferably, an automotive wiring harness quality inspection system based on multi-source heterogeneous data includes a data acquisition and edge calibration module, a preliminary early warning module, a heterogeneity analysis module, a fault source identification module, and a feedback module;

[0056] The data acquisition and edge calibration module is used to acquire relevant data of the vehicle communication harness in real time during vehicle operation using the high-speed ADC interface inside the vehicle ECU, and to perform edge calibration, and extract the waveform response offset coefficient Bx in the calibrated edge segment;

[0057] The preliminary warning module is used to preset the waveform response offset threshold BXyz, compare and analyze the waveform response offset coefficient Bx and the waveform response offset threshold BXyz, assess the safety status of the automotive communication harness, and trigger the risk alarm mechanism.

[0058] The heterogeneity analysis module is used to collect communication waveform electrical signal data in real time and perform feature extraction to obtain the impedance drift factor ΔZ. diff and phase symmetry factor Φ CM Then, a weighted fusion calculation is performed to obtain the communication waveform heterogeneity index F. DST Classify automotive communication harnesses according to their degree of differentiation;

[0059] The fault source identification module is used to divide the automotive communication harness of the third level of alienation into N harness segments, and combine this with the communication waveform alienation index F. DST It identifies structurally induced degradation risk segments and generates an anomaly analysis table;

[0060] The feedback module is used to generate repair suggestions for automotive communication wiring harnesses based on the abnormal result analysis table, and to collect relevant data on automotive communication wiring harnesses during the repair process in real time for feedback.

[0061] This invention provides a method and system for inspecting the quality of automotive wiring harnesses based on multi-source heterogeneous data, which has the following advantages:

[0062] (1) By integrating high-speed ADC real-time sampling data, voltage edge mutation feature identification and structural segmentation communication waveform evaluation mechanism, it can accurately identify communication waveform distortion problems caused by physical factors such as wiring bends, shield offsets or structural stress concentration. Compared with existing methods based on functional on / off judgment or static impedance testing, this method does not depend on the wiring harness unfolding state and performs dynamic data acquisition and distortion identification during vehicle operation. This solves the technical defect of existing detection systems that are difficult to identify structurally induced communication degradation hazards and improves the coverage of communication reliability diagnosis during vehicle assembly.

[0063] (2) By constructing the communication waveform heterogeneity index F DST Incorporating impedance drift factor ΔZ diff and phase symmetry factor Φ CM Two complementary feature indices in the frequency and time domains are used, and a unified time window [tW, t] is applied for feature processing. This allows for the simultaneous measurement of differential signal asymmetry caused by structure and dynamic waveform degradation characteristics. Furthermore, the communication waveform distortion index F is used to measure these characteristics. DST Compared with the preset classification threshold, the system realizes the classification of the degree of alienation of automotive communication wiring harnesses under the influence of wiring structure, and automatically triggers the wiring structure risk analysis process according to the degree of alienation. This enhances the detection method's ability to quantify and respond to the degree of wiring anomalies, and improves the accuracy of differentiated fault management and control.

[0064] (3) By integrating multi-source heterogeneous data, a closed-loop control mechanism was constructed for the entire process, from voltage edge identification, waveform distortion analysis, structural segment location, maintenance suggestion generation to maintenance data collection and feedback. The diagnostic results can not only automatically generate an anomaly analysis table for structural degradation risk segments, but also link with the automotive MES system to synchronize maintenance suggestions, supporting intelligent prompts and task confirmation for car owners and maintenance terminals. At the same time, during the maintenance process, the system can also collect waveform data after structural repair, archive it to form a case library, and continuously optimize the diagnostic model and process suggestions, which has high engineering adaptability and quality management value. Attached Figure Description

[0065] Figure 1 This is a schematic diagram of the process for a method for detecting the quality of automotive wiring harnesses based on multi-source heterogeneous data according to the present invention.

[0066] Figure 2 This is a block diagram of an automotive wiring harness quality inspection system based on multi-source heterogeneous data according to the present invention.

[0067] Figure 3 This is a schematic diagram of the classification process for the different levels of automotive communication wiring harnesses according to the present invention;

[0068] Figure 4 This is a schematic diagram of the communication waveform electrical signal data of the present invention. Detailed Implementation

[0069] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0070] Example 1

[0071] Please see Figure 1 This invention provides a method for quality inspection of automotive wiring harnesses based on multi-source heterogeneous data, comprising the following steps:

[0072] S1. Use the high-speed ADC interface inside the vehicle ECU to collect relevant data of the vehicle communication harness in real time during vehicle operation, perform edge calibration, and extract the waveform response offset coefficient Bx in the calibration edge segment.

[0073] S2. Preset waveform response offset threshold BXyz, compare and analyze the waveform response offset coefficient Bx and the waveform response offset threshold BXyz, evaluate the safety status of the automotive communication harness, and trigger the risk alarm mechanism.

[0074] S3. Real-time collection of communication waveform electrical signal data and feature extraction to obtain the impedance drift factor ΔZ. diff and phase symmetry factor Φ CM Then, a weighted fusion calculation is performed to obtain the communication waveform heterogeneity index F. DST Classify automotive communication harnesses according to their degree of differentiation;

[0075] S4. Divide the automotive communication harness of the third differentiation level into N harness segments, and combine them with the communication waveform differentiation index F. DST It identifies structurally induced degradation risk segments and generates an anomaly analysis table;

[0076] S5. Based on the abnormal result analysis table, generate repair suggestions for the automotive communication wiring harness and collect relevant data on the automotive communication wiring harness during the repair process for feedback in real time.

[0077] In this embodiment, by continuously acquiring and identifying edge mutations in the voltage waveform of the automotive communication harness using the high-speed ADC interface inside the vehicle ECU, waveform response offset characteristics can be accurately extracted. Combined with preset thresholds, this enables rapid early warning of risks. Based on this, a differential waveform data set is further constructed to extract impedance drift factors and phase symmetry factors, which are then fused to form a communication waveform distortion index. This accurately quantifies the impact of wiring structure on communication performance and completes distortion level assessment. When structural degradation is identified, harness segmentation analysis is automatically performed to identify structurally induced degradation risk segments and generate an abnormal result analysis table. Finally, maintenance suggestions are pushed and repair data is collected, forming a structural closed-loop feedback. This effectively improves the diagnostic depth and maintenance guidance capability of automotive communication harnesses under operating conditions, overcoming the technical limitations of existing methods that rely on single signal anomaly judgment and have difficulty locating physical structural faults.

[0078] Example 2

[0079] Please refer to Figure 1 Specifically: S1 includes the following steps:

[0080] S11. During vehicle operation, an automotive-grade ADC data acquisition unit is connected to the vehicle controller to continuously sample data, obtain the voltage waveform of the automotive communication harness over a period of time in real time, and construct a data set H related to the automotive communication harness.

[0081] S12. Based on the relevant data set H of the automotive communication wiring harness, perform edge calibration. The specific process of edge calibration is as follows:

[0082] Set a fixed time sampling window T w And extract the sampling window T at a fixed time. w The voltage waveform of the automotive communication harness inside the vehicle is obtained, and the voltage difference ΔV(t) at each data sampling time point is calculated.i ), generating a differential sequence, where the voltage difference ΔV(t) i The specific method of obtaining it is as follows:

[0083] ΔV(t i )=V(t i+1 )-V(t i );

[0084] In the formula, ΔV(t) i ) represents the data sampling time point t i The voltage difference of the automotive communication harness, V(t) i+1 ) and V(t i ) represent the data sampling time points t and t, respectively. i+1 and t i The voltage of the automotive communication harness at that time;

[0085] Set the judgment threshold ∈ s And in the difference sequence, the judgment threshold ∈ s and voltage difference ΔV(t) i A comparative evaluation was conducted to analyze the voltage abrupt change state at the data sampling time point, where the data sampling time point t was used as the basis for the evaluation. i For example, if ΔV(t) i )≥∈ s This indicates the data sampling time point t. i This is the mutation point, at which the data sampling time point t is marked. i If ΔV(t) is "1", then... i )<∈ s This indicates the data sampling time point t. i This is the normal point; at this time, the data sampling time point t is marked. i The value is "0", and all mutation points are collected to form a mutation marker sequence;

[0086] In the mutation marker sequence, M consecutive data sampling time points marked as "1" are extracted to construct a candidate edge segment, thereby obtaining several candidate edge segments. For each candidate edge segment, the mutation score PF is calculated, and the candidate edge segment with the largest mutation score PF is selected as the edge labeling segment. Taking candidate edge segment j as an example, the mutation score PF of candidate edge segment j is... (j) The specific method of obtaining it is as follows:

[0087]

[0088] In the formula, This represents the mean of the voltage differences at M data sampling time points within the candidate edge segment j. Let represent the variance of the voltage difference at M data sampling time points within the candidate edge segment j, ∈ represent the disturbance adjustment constant, and ln represent the natural logarithm function. The disturbance adjustment constant is used to prevent the denominator from being 0.

[0089] In wiring harness signal analysis or vehicle diagnostic systems, voltage waveforms are often analyzed using methods such as fixed-window statistics, threshold triggering, and periodic averaging. However, without calibration, abnormal transitions can be confused with normal pulse transitions, affecting the accuracy of subsequent signal identification and diagnostic results. For example, in a real-world scenario, such as starting a vehicle in the morning at low temperatures, the battery's internal resistance is high, current surges, and wiring harness voltage drops significantly. Without edge markings, the system might misjudge a short circuit. Current technologies generally use fixed-time-window voltage waveform processing, which makes it difficult to accurately identify abrupt changes and edge regions in wiring harness signals. This results in insufficient ability to identify short-term interference, poor contact, or environmental stress. Edge segment markings, on the other hand, can anchor and classify voltage abrupt changes, providing highly reliable basic data for subsequent system health assessments, anomaly warnings, and behavior modeling. This is especially suitable for edge-sensitive scenarios such as cold starts, loose plugs, and high-frequency interference, and has significant engineering application value and practical necessity.

[0090] S1 specific steps also include,

[0091] S13. Based on the relevant data set H of the automotive communication wiring harness, extract the voltage waveform of the automotive communication wiring harness within the edge calibration section, and compare and analyze the voltage waveform of the automotive communication wiring harness within the edge calibration section with the calibration voltage waveform to obtain the waveform response offset coefficient Bx. The specific method for obtaining the waveform response offset coefficient Bx is as follows:

[0092]

[0093] In the formula, V sig (t m ) represents the data sampling time point t m The voltage of the automotive communication harness at that time, V ref (t m ) represents the data sampling time point t m The standard voltage of the automotive communication harness at that time, M represents the total number of data sampling time points within the edge calibration section, and m represents the index of the data sampling time point within the edge calibration section.

[0094] The waveform response offset coefficient Bx represents the cosine similarity of the angle between voltage signal waveform vectors. It is used to measure the "overall response offset" between the current voltage signal waveform and the reference standard waveform. During the transmission of voltage signals in automotive communication harnesses, the overall waveform of the signal may "drift" due to contact aging, resistance mutation, or interference, but this will not cause functional interruption. The waveform response offset coefficient Bx can directly quantify the offset between the current waveform and the standard waveform.

[0095] In this embodiment, by continuously sampling the voltage waveform data of the automotive communication harness in real time and constructing a differential sequence based on a fixed-time sampling window, combined with abrupt change point identification and abrupt change score calculation, the edge segment with the most structural response characteristics in the voltage waveform can be accurately identified. Furthermore, a waveform response offset coefficient Bx is introduced. By comparing and analyzing with the standard voltage waveform, the degree of waveform response offset within the edge segment can be quantified, effectively identifying early weak anomalies caused by harness wiring status, environmental interference, or loose connections. Compared with traditional methods based on overall voltage trend assessment, this method has higher positioning accuracy and anti-interference capability, can detect communication quality degradation trends in advance, improves the sensitivity of harness structural anomaly response identification and early warning capability, and provides a high-quality feature basis for subsequent anomaly analysis and fault level classification.

[0096] Example 3

[0097] Please refer to Figure 1 Specifically: The specific steps of S2 include,

[0098] S21. Preset the waveform response offset threshold BXyz for the automotive communication wiring harness, and compare and analyze the waveform response offset threshold BXyz and the waveform response offset coefficient Bx to evaluate the safety status of the automotive communication wiring harness. The specific evaluation process is as follows:

[0099] If the waveform response offset coefficient Bx of the automotive communication harness is less than or equal to the waveform response offset threshold BXyz, i.e., Bx≤BXyz, then the automotive communication harness is judged to be in a safe state. At this time, the relevant data of the automotive communication harness during the operation of the vehicle are continuously collected and the collected relevant data of the automotive communication harness are stored in the automotive quality inspection database.

[0100] If the waveform response offset coefficient Bx of the automotive communication harness is greater than the waveform response offset threshold BXyz, i.e., Bx > BXyz, then the automotive communication harness is determined to be in a risky state. At this time, the risk alarm mechanism is automatically triggered, and the communication waveform abnormal response identification process is carried out.

[0101] In this embodiment, by setting a waveform response offset threshold BXyz and comparing it with the extracted waveform response offset coefficient Bx, a data-driven preliminary safety status judgment mechanism for communication harnesses is established. This method can determine whether there is an abnormal waveform response offset in the automotive communication harness during normal vehicle operation without relying on an external testing platform or disassembly operation, thereby quickly identifying potential communication performance degradation trends. Compared with traditional detection methods based on physical damage or functional failure, this solution can predict and warn before the communication quality has seriously deteriorated, exhibiting stronger proactive and real-time capabilities. Furthermore, by incorporating communication data under safe conditions into the automotive quality inspection database, rich data support is provided for subsequent algorithm model optimization and risk identification modeling, thereby constructing a continuously evolving data closed-loop system and improving the intelligent level of communication health management of the entire vehicle electronic system.

[0102] Example 4

[0103] Please refer to Figure 1 , Figure 3 and Figure 4 Specifically: The specific steps of S3 include,

[0104] S31. Call the vehicle electronic wiring topology database, read the wiring path model and communication type of the vehicle communication harness, install differential sampling probes on the vehicle communication harness, collect the communication waveform electrical signal data of the vehicle communication harness in real time, and construct a differential waveform data set. The communication waveform electrical signal data includes the current difference I between conductor A and conductor B in the vehicle communication harness. AB (t), the voltage difference V between conductor A and conductor B AB (t) and voltage waveforms of conductors A and B in the automotive communication harness.

[0105] S3 specific steps also include,

[0106] S32. Based on the differential waveform data set, calculate the deviation between the actual differential transmission path of the automotive communication harness and the theoretical design impedance within the time window [tW, t], to obtain the impedance drift factor ΔZ. diff Among them, the impedance drift factor ΔZ diff (t) is obtained in the following way:

[0107]

[0108] In the formula, Z ref V represents the design reference impedance of the automotive communication harness. AB (τ) and I AB(τ) represents the voltage difference and current difference between conductor A and conductor B in the automotive communication harness at the data sampling time point τ, respectively. τ represents the index of the data sampling time point, τ∈{tW, t}, W represents the length of the time window [tW, t], and k represents the adjustment coefficient. The specific value of the adjustment coefficient k is obtained by the empirical parameter calibration method.

[0109] in, This is the impedance deviation term, used to reflect the difference between the actual impedance in the current automotive communication harness and the design reference impedance Z. ref The deviation between them The larger the value, the more obvious the structural deviation or material mismatch of the automotive communication wiring harness in this structural segment.

[0110] It is a dynamic adjustment term used to eliminate the impedance drift illusion caused by high-frequency common-mode oscillation or short-term spikes. In the impedance changes of automotive communication harnesses, the impedance changes caused by structural distortion are usually slow and trend-oriented, while non-structural distortion will cause the impedance to swing violently at high frequencies in a short period of time. Therefore, subtracting the dynamic adjustment term can suppress high-frequency swing interference and retain low-frequency structural abnormal components.

[0111] The adjustment coefficient k is a dynamic balance factor between the structural disturbance trend term and the static impedance offset term. It is used to normalize the dynamic adjustment change term to the same dimension as the impedance deviation term, balance the relative influence intensity of short-term fluctuation disturbance and trend drift, and thus suppress misjudgment caused by high-frequency spikes, making the impedance offset characteristics of structural induced degradation more stable.

[0112] Impedance drift factor ΔZ diff By synchronously measuring the differential-mode impedance shift and its derivative change intensity of the automotive communication harness within a time window, and eliminating high-frequency noise interference, the obtained true structural impedance disturbance index can effectively identify wiring structure-induced degradation phenomena caused by bending, clamping, shielding abnormalities, etc. It is a key feature index for identifying structural anomaly types and classifying communication degradation levels, and has good engineering adaptability, physical interpretability, and diagnostic reliability.

[0113] S33. Perform FFT analysis on the voltage waveforms of conductors A and B in the automotive communication wiring harness to extract the dominant frequency phase Φ of conductors A and B in the automotive communication wiring harness. A (t) and Φ B (t), and calculate the average absolute value of the phase difference between the main frequency components of conductors A and B in the automotive communication harness within the time window [tW, t], in order to obtain the phase symmetry factor Φ. CM Wherein, the phase symmetry factor Φ CM The specific method of obtaining it is as follows:

[0114]

[0115] In the formula, Φ A (t) represents the dominant frequency phase of conductor A in the automotive communication harness, Φ B (t) represents the dominant frequency phase of conductor B in the automotive communication harness.

[0116] FFT analysis is an algorithm that transforms a time-domain signal into a frequency-domain signal. Its core purpose is to decompose a signal into a set of sinusoidal components of different frequencies and extract the amplitude and phase information of each frequency.

[0117] Phase symmetry factor Φ CM This is used to analyze whether conductors A and B, which are used in constructing automotive communication harnesses, are phase-synchronized on the main frequency component, with a phase symmetry factor Φ. CM It can effectively reflect the phase asynchrony caused by uneven wiring structure, path disturbance or shield offset, and is an important parameter for judging the impact of structural degradation on frequency domain synchronization;

[0118] S3 specific steps also include,

[0119] S34. Based on the obtained impedance drift factor ΔZ diff and phase symmetry factor Φ CM A weighted summation calculation is performed to obtain the communication waveform heterogeneity index F of the automotive communication harness. DST Among them, the communication waveform distortion index F DST The specific method of obtaining it is as follows:

[0120]

[0121] In the formula, α and β represent the impedance drift factor ΔZ, respectively. diff and phase symmetry factor Φ CM The weighting coefficients, C represents the first correction constant, where the specific values ​​of the weighting coefficients α and β are set by the customer according to the actual situation, and 0 < α < 1, 0 < β < 1, α + β = 1;

[0122] S35. Preset the first communication waveform distortion threshold δ1 and the second communication waveform distortion threshold δ2, and set the communication waveform distortion index F DST The distortion level of the automotive communication harness is evaluated by comparing it with the first communication waveform distortion threshold δ1 and the second communication waveform distortion threshold δ2. The specific evaluation content is as follows:

[0123] If the communication waveform heterogeneity index F DST If the value is less than or equal to the first communication waveform distortion threshold δ1, the automotive communication harness is determined to be at the first distortion level. At this time, the automotive communication harness has no significant structural interference and is marked as “C1”.

[0124] If the communication waveform heterogeneity index FDST If the value is greater than the first communication waveform distortion threshold δ1 and less than the second communication waveform distortion threshold δ2, then the automotive communication harness is determined to be at the second distortion level. At this time, structural interference of the automotive communication harness initially appears, and the automotive communication harness is marked as "C2".

[0125] If the communication waveform heterogeneity index F DST If the value is greater than or equal to the second communication waveform distortion threshold δ2, the automotive communication harness is determined to be at the third distortion level. At this time, the automotive communication harness wiring structure is in a distortion state, automatically triggering the automotive wiring structure risk inspection process and marking the automotive communication harness as "C3".

[0126] In this embodiment, differential sampling probes are installed on the automotive communication harness. Combined with the wiring path model and communication type information provided by the electronic wiring topology database, differential waveform data sets are constructed in real time. The impedance offset and the main phase difference of the frequency within the time window [tW, t] are introduced as dual feature parameters to calculate two structurally sensitive feature indices: "impedance drift factor" and "phase symmetry factor". The two structurally sensitive indices are then weighted and fused to propose a communication waveform alienation index construction method. Combined with the set multi-level thresholds, the alienation level is accurately classified, which significantly improves the diagnostic capability for dynamic alienation interference caused by bending, clamping, or shielding structure offset of automotive communication harnesses. A highly robust and highly sensitive structurally induced electrical degradation classification system is constructed, which effectively supports the generation of subsequent wiring structure inspection and maintenance suggestions, and significantly improves the stability and maintainability of the vehicle's electronic network.

[0127] Example 5

[0128] Please refer to Figure 1 Specifically: The specific steps of S4 include,

[0129] S41. After determining that the automotive communication wiring harness is of the third level of alienation, read the CAD drawing of the wiring harness of the vehicle, divide the automotive communication wiring harness into N wiring harness segments according to a fixed physical length, and establish a virtual mapping model of the wiring harness segments. The virtual mapping model of the wiring harness segments specifically includes the spatial position of each wiring harness segment in the vehicle body, the minimum bending radius of the wiring harness segment, and the maximum torsion angle of the wiring harness segment.

[0130] S42. Inject a short-period synchronous communication pulse into one end of the automotive communication harness, and use a time-domain reflectometry (TDR) system to collect the communication waveform electrical signal data of the automotive communication harness in real time at the end of each harness segment, and obtain the communication waveform distortion index F of each harness segment. DST If the communication waveform distortion index F of the harness segment DST Greater than μ FDST +2σ FDSTIf the wire harness segment is marked as a "structure-induced degradation risk segment", recommended measures are generated, including adjusting the wire harness bending radius to 120% of the actual bending radius and adjusting the clamping point density to 80% of the actual clamping point density. If the communication waveform alienation index F of the wire harness segment is... DST Less than or equal to μ FDST +2σ FDST If so, no processing is required;

[0131] Where, μ FDST +2σ FDST The anomaly detection threshold representing the communication waveform distortion index is specifically the communication waveform distortion index F for all harness segments. DST The mean plus the communication waveform heterogeneity index F of all harness segments DST The standard deviation of μ is a simplified version of the 3σ rule based on the statistical principle of normal distribution; approximately 95.45% of the data falls within μ. FDST ±2σ FDST Between, exceeding μ FDST +2σ FDST Then it is judged as abnormal;

[0132] Among them, the time domain reflectance sampling system (TDR) is an analysis system used to detect the location and characteristics of structural defects, impedance discontinuities, or physical damage in cables, wire harnesses, and printed circuit boards. By injecting a fast rising edge pulse or a short period pulse into one end of the communication wire harness, when the pulse propagates along the wire harness, it will partially reflect when it encounters a point of impedance discontinuity in the wire harness, such as a bend, looseness, or poor crimping structure.

[0133] S43. For each "structural induced degradation risk segment", generate an anomaly analysis table and push the generated anomaly analysis table to the automotive MES system. The anomaly analysis table includes the complete radius, bending shape and communication distortion level of each "structural induced degradation risk segment". The anomaly analysis table is used for subsequent maintenance, factory upgrades and process modifications.

[0134] Among them, the automotive MES system is the core production management platform connecting the "design / planning layer" and the "equipment / execution layer," used for process execution management, material and component tracking, defect and anomaly tracing, and process adjustment synchronization.

[0135] In this embodiment, by introducing a "structure-induced degradation risk segment" identification mechanism, precise location and dynamic tracing of local abnormal areas in the wiring structure are achieved. This overcomes the technical limitations of traditional whole-segment detection or overall impedance judgment methods, such as excessively coarse granularity and large diagnostic errors. Through virtual mapping modeling based on the harness CAD drawing, physical structural features, such as minimum bending radius and maximum torsion angle, are associated with the actual wiring path. Combined with the TDR system, the communication waveform distortion index F of each harness segment is obtained. DSTIt can accurately identify local degradation sources without disassembling the wire harness. Furthermore, the system automatically generates an anomaly analysis table based on each "structurally induced degradation risk segment" and pushes it to the MES system, realizing closed-loop linkage and real-time adjustment with production planning and process design, effectively improving the structural consistency control and communication quality assurance capabilities of wire harness products in the mass manufacturing process.

[0136] Example 6

[0137] Please refer to Figure 1 Specifically: The S5 steps include,

[0138] S51. Receive and generate equipment maintenance suggestions based on the abnormal result analysis table, including increasing the wiring radius, removing clamping points, replacing the shielding structure, replacing the entire wiring harness, and encapsulating local stress buffers. Bind the equipment maintenance instructions to the "structural induced degradation risk segment" and send the equipment maintenance suggestions to the vehicle owner terminal and the maintenance personnel terminal simultaneously. During the execution of the maintenance task, collect relevant data of the "structural induced degradation risk segment" and provide data feedback.

[0139] The vehicle owner terminal is used to inform the owner of the risk level of the communication structure abnormality and the necessity of repair, while the repair personnel terminal is used to display the location of the structural sections, structural diagrams, and repair operation instructions.

[0140] In this embodiment, by binding maintenance instructions with "structurally induced degradation risk segments" and pushing them to the vehicle owner's terminal and the maintenance personnel's terminal respectively, not only is the accuracy and efficiency of maintenance execution improved, but users' understanding and responsiveness to the risk level of communication structure anomalies and the necessity of maintenance are also enhanced. At the same time, the data retrieval mechanism during the execution of maintenance tasks can realize quantitative feedback on the effect of fault repair, thereby supporting subsequent case archiving, model optimization, and closed-loop quality control. Overall, this step realizes a closed-loop maintenance management system of "identification-push-execution-feedback", improving the systematicness, timeliness, and engineering closure of communication structure anomaly governance.

[0141] Example 7

[0142] Please refer to Figure 2 Specifically: an automotive wiring harness quality inspection system based on multi-source heterogeneous data, including a data acquisition and edge calibration module, a preliminary early warning module, a heterogeneity analysis module, a fault source identification module, and a feedback module;

[0143] The data acquisition and edge calibration module is used to acquire relevant data of the vehicle communication harness in real time during vehicle operation using the high-speed ADC interface inside the vehicle ECU, and to perform edge calibration, and extract the waveform response offset coefficient Bx in the calibrated edge segment;

[0144] The preliminary warning module is used to preset the waveform response offset threshold BXyz, compare and analyze the waveform response offset coefficient Bx and the waveform response offset threshold BXyz, assess the safety status of the automotive communication harness, and trigger the risk alarm mechanism.

[0145] The heterogeneity analysis module is used to collect communication waveform electrical signal data in real time and perform feature extraction to obtain the impedance drift factor ΔZ. diff and phase symmetry factor Φ CM Then, a weighted fusion calculation is performed to obtain the communication waveform heterogeneity index F. DST Classify automotive communication harnesses according to their degree of differentiation;

[0146] The fault source identification module is used to divide the automotive communication harness of the third level of alienation into N harness segments, and combine this with the communication waveform alienation index F. DST It identifies structurally induced degradation risk segments and generates an anomaly analysis table;

[0147] The feedback module is used to generate repair suggestions for automotive communication wiring harnesses based on the abnormal result analysis table, and to collect relevant data on automotive communication wiring harnesses during the repair process in real time for feedback.

[0148] In this embodiment, by constructing a closed-loop structure covering the entire process from acquisition, analysis, classification, identification to feedback, the data acquisition and edge calibration module accurately identifies key areas of the waveform, enabling early detection of signal edge abrupt changes and structural offsets. The preliminary warning module can intervene immediately at the stage of minor anomalies to prevent the fault from escalating. The anomaly analysis module integrates impedance drift factor and phase symmetry factor to accurately characterize structural anomaly features. The fault source identification module enables accurate location of structural risk sections. The feedback module supports intelligent generation of maintenance suggestions and data feedback, forming a self-learning optimization mechanism, which improves the online diagnostic accuracy, response efficiency, and subsequent maintainability of automotive communication harnesses.

[0149] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for detecting quality of automobile wire harness based on multi-source heterogeneous data, characterized in that: Comprise the following steps, S1, using the high-speed ADC interface inside the vehicle ECU, real-time acquisition of vehicle running process related data of automobile communication wire harness, and edge calibration, and extract the waveform response offset coefficient Bx in the calibration edge section; S2, preset waveform response offset threshold BXyz, and compare and analyze the waveform response offset coefficient Bx and the waveform response offset threshold BXyz, evaluate the safety state of automobile communication wire harness, and trigger the risk alarm mechanism; S3, collect communication waveform signal data in real time, and perform feature extraction to obtain impedance drift factor ΔZ diff and phase symmetry factor Φ CM , and perform weighted fusion calculation to obtain communication waveform alienation index F DST , and perform automobile communication wire harness alienation level classification; S4, divide the third dissimilation level of automobile communication wire harness into N wire harness segments, and combine the communication waveform dissimilation index F DST , identify the structure-induced degradation risk segment, and generate an abnormal result analysis table; S5, according to the abnormal result analysis table, generate automobile communication wire harness maintenance suggestion, and collect automobile communication wire harness related data in real time during maintenance process for feedback.

2. The method for detecting the quality of automobile wire harness based on multi-source heterogeneous data according to claim 1, characterized in that: S1 specific steps include, S11, in the process of vehicle running, using vehicle gauge ADC collector and vehicle controller connection, continuous data sampling, real-time acquisition of automobile communication wire harness voltage waveform diagram in a period of time, and building automobile communication wire harness related data set H; S12, according to automobile communication wire harness related data set H, edge calibration is carried out, wherein the specific process of edge calibration is as follows: Setting a fixed time sampling window T w and extracting the automobile communication wire harness voltage waveform within the fixed time sampling window T w , and calculating the voltage difference ΔV(t i ) at each data sampling time point to generate a differential sequence, wherein the voltage difference ΔV(t i ) is specifically obtained in the following manner: AV(t) = V(t) - V(t-1) i ) = V(t i+1 )- V(t i ) In the formula, ΔV(t) i ) represents the data sampling time point t i The voltage difference of the automotive communication harness, V(t) i+1 ) and V(t i ) represent the data sampling time points t and t, respectively. i+1 and t i The voltage of the automotive communication harness at that time; Setting a judgment threshold ∈ s In the differential sequence, the judgment threshold ∈ s and the voltage difference ΔV(t i ) are compared and evaluated to analyze the voltage mutation state at the data sampling time point, where, taking the data sampling time point t i as an example, if ΔV(t i ) ≥ ∈ s , it indicates that the data sampling time point t i is a mutation point, at this time, the data sampling time point t i is marked as "1", if ΔV(t i ) < ∈ s , it indicates that the data sampling time point t i is a normal point, at this time, the data sampling time point t i is marked as "0", and all the mutation points are collected to form a mutation mark sequence; And in the mutation marker sequence, extract the data sampling time points of consecutive M markers as "1", construct a candidate edge section, obtain a plurality of candidate edge sections, and for each candidate edge section, calculate the mutation score PF of the candidate edge section, and select the candidate edge section with the maximum mutation score PF as the edge calibration section, wherein the mutation score PF of the candidate edge section j is calculated as follows: (j) The specific acquisition method is: wherein denotes the mean value of the voltage difference values at the M data sampling time points within the candidate edge section j, denotes the variance of the voltage difference values at the M data sampling time points within the candidate edge section j, denotes a perturbation adjustment constant, and ln denotes the natural logarithm function.

3. The method of claim 2, wherein: S1 specific steps also include, S13, according to automobile communication wire harness related data set H, extract the voltage waveform of automobile communication wire harness in the edge calibration section, and compare and analyze the voltage waveform of automobile communication wire harness in the edge calibration section with the calibration voltage waveform, to obtain the waveform response offset coefficient Bx, wherein the waveform response offset coefficient Bx is obtained as follows: In the formula, V sig (t m ) represents the automobile communication wire harness voltage at the data sampling time point t m (t ref ) represents the automobile communication wire harness standard voltage at the data sampling time point t m (t m ) represents the total number of data sampling time points in the edge calibration section, and m represents the index of the data sampling time point in the edge calibration section.

4. The method of claim 3, wherein: S2 specific steps include, S21, preset the waveform response offset threshold BXyz of automobile communication wire harness, and compare and analyze the waveform response offset threshold BXyz of automobile communication wire harness and the waveform response offset coefficient Bx, to evaluate the safety state of automobile communication wire harness, and the specific evaluation process is as follows: If the waveform response offset coefficient Bx of automobile communication wire harness is less than or equal to the waveform response offset threshold BXyz, that is, Bx≤BXyz, it is judged that the automobile communication wire harness is in safe state, at this time, the automobile communication wire harness related data in the process of automobile running is continuously collected, and the collected automobile communication wire harness related data is stored in automobile quality inspection database; If the waveform response offset coefficient Bx of automobile communication wire harness is greater than the waveform response offset threshold BXyz, that is, Bx>BXyz, it is judged that the automobile communication wire harness is in risk state, at this time, the risk alarm mechanism is triggered automatically, and the communication waveform alienation response identification process is carried out.

5. The method for detecting the quality of automobile wire harness based on multi-source heterogeneous data according to claim 4, characterized in that: S3 specific steps include, S31, call the vehicle electronic wiring topology database, read the wiring path model of the automobile communication wire harness and the communication type of the automobile communication wire harness, and install the differential sampling probe on the automobile communication wire harness, collect the communication waveform electrical signal data of the automobile communication wire harness in real time, and construct the differential waveform data set, wherein the communication waveform electrical signal data includes the current difference I AB (t), the voltage difference V of the conductor A and the conductor B AB (t) and the voltage waveform diagram of the conductor A and the conductor B of the automobile communication wire harness.

6. The method for detecting the quality of automobile wire harness based on multi-source heterogeneous data according to claim 5, characterized in that: S3 specific steps also include, S32、According to the differential waveform data set, calculate the deviation amount of the actual harness differential transmission path and the theoretical design impedance in the time window [t-W, t] of the automobile communication harness, to obtain the impedance drift factor ΔZ diff Wherein, the impedance drift factor ΔZ diff (t) The specific acquisition method is: In the formula, Z ref V represents the design reference impedance of the automotive communication harness. AB (τ) and I AB (τ) represents the voltage difference and current difference between conductor A and conductor B in the automotive communication harness at the data sampling time point τ, respectively. τ represents the index of the data sampling time point, τ∈{tW, t}, W represents the length of the time window [tW, t], and k represents the adjustment coefficient. S33, performing FFT analysis on the voltage waveforms of the conductor A and the conductor B in the automobile communication wire harness respectively to extract the main frequency phase Φ of the conductor A and the conductor B in the automobile communication wire harness A (t) and Φ B (t), and calculating the average absolute value of the phase difference of the main frequency components of the conductor A and the conductor B in the automobile communication wire harness in the time window [t-W, t] to obtain the phase symmetry factor Φ CM , wherein the phase symmetry factor Φ CM The specific acquisition method is: where Φ A (t) represents the fundamental frequency phase of the conductor A of the automotive communication wire harness B (t) represents the fundamental frequency phase of the conductor B of the automotive communication wire harness 7. The method of claim 6, wherein the method is based on multi-source heterogeneous data of automobile wiring harness quality detection. S3 specific steps also include, S34, according to the impedance drift factor ΔZ obtained diff and the phase symmetry factor Φ CM , weighted and aggregated to obtain the communication waveform alienation index F of the automobile communication wire harness DST , wherein the communication waveform alienation index F DST The specific acquisition method is: In the formula, α and β respectively represent weight coefficients of impedance drift factor ΔZ diff and phase symmetry factor Φ CM , and C represents a first correction constant. S35, preset the first communication waveform alienation threshold δ1 and the second communication waveform alienation threshold δ2, and compare the communication waveform alienation index F DST and the first communication waveform alienation threshold δ1 and the second communication waveform alienation threshold δ2 are compared and analyzed, and the alienation level of the automobile communication wire harness is evaluated, and the specific evaluation content is as follows: If the communication waveform alienation index F DST If the communication waveform alienation index F is less than or equal to the first communication waveform alienation threshold value δ1, it is determined that the automobile communication wire harness is of the first alienation level, at which time the automobile communication wire harness has no significant structural interference, and the automobile communication wire harness is marked as "C1"; If the communication waveform alienation index F DST If the communication waveform alienation index F is greater than the first communication waveform alienation threshold value δ1 and less than the second communication waveform alienation threshold value δ2, it is determined that the automobile communication wire harness is of a second alienation level, at which point the automobile communication wire harness structure interference initially appears, and the automobile communication wire harness is marked as "C2"; If the communication waveform alienation index F DST If the communication waveform alienation index F is greater than or equal to the second communication waveform alienation threshold δ2, it is determined that the automobile communication wiring harness is of a third alienation level, at which time the automobile communication wiring harness wiring structure is in an alienation state, an automobile wiring structure risk check process is automatically triggered, and the automobile communication wiring harness is marked as "C3".

8. The method for detecting the quality of automobile wire harness based on multi-source heterogeneous data according to claim 7, characterized in that: S4 specific steps include, S41, after judging that the automobile communication wire harness is the third alienation level, read the wire harness wiring CAD diagram of the automobile, and divide the automobile communication wire harness into N wire harness sections according to the fixed physical length, and establish the wire harness section virtual mapping model, wherein the wire harness section virtual mapping model specifically includes the spatial position of each wire harness section in the vehicle body, the minimum bending radius of the wire harness section and the maximum torsion angle of the wire harness section; S42. Inject a short-period synchronous communication pulse into one end of the automotive communication harness, and use a time-domain reflectometry (TDR) system to collect the communication waveform electrical signal data of the automotive communication harness in real time at the end of each harness segment, and obtain the communication waveform distortion index F of each harness segment. DST If the communication waveform distortion index F of the harness segment DST Greater than μ FDST +2σ FDST If the wire harness segment is marked as a "structure-induced degradation risk segment", recommended measures are generated, including adjusting the wire harness bending radius to 120% of the actual bending radius and adjusting the clamping point density to 80% of the actual clamping point density. If the communication waveform alienation index F of the wire harness segment is... DST Less than or equal to μ FDST +2σ FDST If so, no processing is required; S43, for each "structure-induced degradation risk section", generate an abnormal result analysis table, and push the generated abnormal result analysis table to the automobile MES system, wherein the abnormal result analysis table includes the complete radius, bending shape and communication alienation level of each "structure-induced degradation risk section".

9. The method of claim 8, wherein: S5, the specific steps include, S51, receiving and generating equipment maintenance suggestions according to the abnormal result analysis table, including increasing the wiring radius, removing the clamping point, replacing the shielding structure, replacing the wire harness as a whole, and locally buffering the package, binding the equipment maintenance instructions and the "structure-induced degradation risk section", and synchronously sending the equipment maintenance suggestions to the owner terminal and the maintenance personnel terminal, and collecting the related data of the "structure-induced degradation risk section" during the maintenance task execution process, and performing data feedback.

10. A multi-source heterogeneous data-based automobile wire harness quality detection system for implementing the multi-source heterogeneous data-based automobile wire harness quality detection method of any one of claims 1-9. It includes a data acquisition and edge calibration module, a preliminary warning module, an alienation analysis module, a fault source identification module and a feedback module; The data acquisition and edge calibration module is used to use the high-speed ADC interface inside the vehicle ECU to collect the related data of the automobile communication wire harness during the vehicle operation process in real time, and to calibrate the edge, and to extract the waveform response offset coefficient Bx in the calibrated edge section; The preliminary warning module is used to preset the waveform response offset threshold BXyz, and to compare and analyze the waveform response offset coefficient Bx and the waveform response offset threshold BXyz, to evaluate the safety state of the automobile communication wire harness, and to trigger the risk warning mechanism; The dissimilation analysis module is used for collecting communication waveform electric signal data in real time and performing feature extraction to obtain an impedance drift factor ΔZ diff and a phase symmetry factor Φ CM , and performing weighted fusion calculation to obtain a communication waveform dissimilation index F DST , and performing automobile communication wire harness dissimilation grade classification; The fault source identification module is used to divide the third alienation level automobile communication wire harness into N wire harness segments, and combine the communication waveform alienation index F DST to identify the structure-induced degradation risk segment and generate an abnormal result analysis table; The feedback module is used to generate automobile communication wire harness maintenance suggestions according to the abnormal result analysis table, and to collect the related data of the automobile communication wire harness during the maintenance process in real time for feedback.

Images