Automobile ecu controller aging test abnormality determination method

Abstract

The application discloses an automobile ECU controller aging test abnormality determination method, and relates to the technical field of determination and analysis, which comprises the following steps: collecting a feedback jump voltage sequence corresponding to node state control data, calculating a feedback voltage change rate sequence, mapping the feedback jump voltage sequence and the feedback voltage change rate sequence into a second monitoring phase space trajectory; calculating a test area integral deviation value of the second monitoring phase space trajectory relative to a reference trajectory of a reference phase space; when the test area integral deviation value is not greater than a preset aging determination threshold value, writing a node identification code of a current target test node into an abnormal node record, and outputting determination result data that the target test node has an implicit driving ability degradation defect; when the test area integral deviation value is greater than the preset aging determination threshold value, reading a next node identification code in a node polling control table and continuing a node tracing and investigation cycle. The application has the effect of improving the aging test abnormality determination efficiency.

Description

Technical Field

[0001] This application relates to the field of judgment and analysis technology, and in particular to a method for judging abnormalities in aging tests of automotive ECU controllers. Background Technology

[0002] As the core unit of a vehicle's electronic control system, the automotive ECU controller undertakes critical functions such as power control, body control, safety control, and communication coordination. Its long-term reliability directly affects the overall vehicle's operational safety. Aging tests, as an important means of verifying the durability of ECU controllers, typically require continuous operation under environments such as high temperature, high humidity, and cyclic loads to expose potential electrical performance degradation that may occur during long-term service. Meanwhile, CAN, LIN, and other vehicle buses, as the main channels for data exchange between ECUs, play a fundamental role in the accurate transmission of vehicle control commands due to their signal edge quality, bus level recovery capability, and multi-node communication stability.

[0003] It is worth noting that aging failure of ECU controllers does not always manifest as communication interruptions, message loss, or dominant level errors. Some transceivers may only exhibit a decline in recessive level recovery capability in the early stages of aging, meaning insufficient pull-back capability against bus parasitic capacitances during the bus's transition from a dominant to a recessive state. This type of degradation is often not easily detected under single-node testing or light-load communication conditions, but it can cause local pauses in recessive transition edges, stepped ringing, and abnormal recovery times when multiple nodes communicate simultaneously, bus loads change, or specific node combinations are involved, thus creating intermittent communication vulnerabilities. Because this phenomenon is related to node topology, transmission timing, and bus load, its root cause is often difficult to pinpoint directly using conventional voltage thresholds.

[0004] However, existing ECU aging test methods primarily focus on statistical analysis of bit error rate, frame loss count, message response timeout, and whether bus voltage exceeds standard ranges, paying insufficient attention to the dynamic changes in physical layer edge morphology. Even when using oscilloscope acquisition, methods typically rely on single waveform amplitude, rise time, or overshoot for judgment, making it difficult to distinguish between normal ringing and implicit recovery drag caused by specific aging nodes. Furthermore, during simultaneous testing of multiple nodes within an aging chamber, manually disconnecting nodes one by one for troubleshooting is inefficient and easily disrupts the original load environment, causing the fault symptoms to disappear. Therefore, existing technologies struggle to detect early transceiver degradation before communication errors occur, and are even less effective at locating abnormal nodes when hardware connections are constantly being altered. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this application provides a method for determining abnormalities in the aging test of automotive ECU controllers.

[0006] This application provides a method for determining abnormalities in the aging test of an automotive ECU controller, comprising the following steps:

[0007] Acquire the dominant to recessive jump voltage time series data in the initial stage of aging, calculate and generate the initial voltage change rate sequence, and map it to the reference phase space reference trajectory according to the time sampling nodes;

[0008] The real-time implicit voltage jump time series data in the aging test cycle is obtained, the real-time voltage change rate sequence is calculated and generated, and the real-time implicit voltage jump time series data and the real-time voltage change rate sequence are mapped to the first monitoring phase space trajectory according to the same time sampling nodes.

[0009] Calculate the area integral deviation value of the first monitoring phase space trajectory relative to the reference phase space trajectory;

[0010] When the area integral deviation value is greater than the preset aging judgment threshold, the node polling control table is read to enter the node traceability and investigation loop.

[0011] In the node tracing and investigation loop, node state control data for a single target test node is generated sequentially, the feedback jump voltage sequence corresponding to the node state control data is collected, the feedback voltage change rate sequence is calculated, and the feedback jump voltage sequence and the feedback voltage change rate sequence are mapped to the second monitoring phase space trajectory.

[0012] Calculate the integral deviation of the test area of ​​the second monitoring phase space trajectory relative to the reference phase space trajectory;

[0013] When the integral deviation of the test area is not greater than the preset aging judgment threshold, the node identification code of the current target test node is written into the abnormal node record, and the judgment result data of the target test node having a hidden driving capability degradation defect is output.

[0014] When the test area integral deviation value is greater than the preset aging judgment threshold, the next node identifier code in the node polling control table is read and the node tracing and troubleshooting loop continues.

[0015] In summary, this application includes at least one of the following beneficial technical effects:

[0016] 1. This application provides a method for determining anomalies in the aging test of an automotive ECU controller. It acquires the timing data of explicit to implicit voltage jumps during the initial aging stage, calculates and generates an initial voltage change rate sequence, and then maps it to a reference phase space trajectory according to time sampling nodes. This transforms the implicit recovery characteristics of the ECU controller in its initial healthy state into a quantifiable and comparable dynamic trajectory reference. During the aging test cycle, it further acquires real-time implicit voltage jump timing data, calculates and generates a real-time voltage change rate sequence, and maps it together with the voltage timing data to a first monitoring phase space trajectory. Then, it calculates the first... By monitoring the area integral deviation of the phase space trajectory relative to the reference phase space trajectory, this application provides an overall measurement of subtle degradation phenomena such as voltage recovery speed, edge drag, and local steps during the transition from dominant to recessive phase space. Compared with traditional detection methods that rely solely on voltage amplitude, message error rate, or communication interruption results, this application can capture abnormal evolution of physical layer edge morphology before the communication function completely fails. This shifts aging testing from outcome-based judgment to process identification, significantly improving the early detection capability of recessive drive capability degradation defects and providing more refined data for ECU controller fault prediction and health management.

[0017] 2. When the area integral deviation value exceeds the preset aging judgment threshold, this application does not directly give a general abnormal conclusion. Instead, it reads the node polling control table and enters a node tracing and investigation loop. It generates node status control data for each individual target test node in turn, collects the corresponding feedback jump voltage sequence, calculates the feedback voltage change rate sequence, and maps it to the second monitoring phase space trajectory. Then, it uses the test area integral deviation value of the second monitoring phase space trajectory relative to the reference phase space trajectory as the tracing judgment basis. When the test area integral deviation value is not greater than the preset aging judgment threshold, it writes the node identification code of the current target test node into the abnormal node record and outputs that the target test node has a hidden drive. The system uses data to determine the performance degradation defect. When the integral deviation of the test area is still greater than the preset aging judgment threshold, the system continues to read the next node identification code to perform troubleshooting. Thus, this application combines anomaly detection and node location in the same phase space trajectory comparison mechanism, avoiding the problems of low efficiency, large environmental disturbances, and difficulty in fault reproduction caused by manually disconnecting nodes and repeatedly retesting in multi-node aging tests. It can complete the tracking of abnormal nodes while maintaining the original bus load environment as much as possible, thereby improving the accuracy, stability, and traceability of aging test judgment, and providing reliable support for subsequent maintenance screening, life assessment, and fault prediction and health management of the vehicle's electronic systems. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the 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.

[0019] Figure 1 This is a flowchart of the method for determining abnormalities in the aging test of an automotive ECU controller according to an embodiment of this application.

[0020] Figure 2 This is a schematic diagram of the implementation logic connection of the reference phase space reference trajectory in an embodiment of this application.

[0021] Figure 3 This is a schematic diagram of the logic connection for implementing the real-time voltage change rate sequence in an embodiment of this application.

[0022] Figure 4 This is a schematic diagram illustrating the logical connection for implementing the area integral deviation value in an embodiment of this application. Detailed Implementation

[0023] The following description, in conjunction with the implementation of the present invention, is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the concept of the invention, and all such modifications and additions should fall within the protection scope of the present invention.

[0024] Example

[0025] This application applies to the identification of bus physical layer anomalies in automotive ECU controllers during aging tests, particularly for identifying latent drive capability degradation defects that have not yet manifested as obvious frame drops, bit errors, or communication interruptions. In this embodiment, the aging test determination processor obtains bus voltage data, message time data, and node status data from the aging test acquisition system. Instead of directly using whether a single voltage amplitude exceeds the limit as the basis for anomalies, it compares the voltage sampling values ​​and voltage change rates during the dominant-to-latent transition process within the same phase space. After a trajectory deviation occurs, it uses a node polling control table to perform a cyclical node tracing and troubleshooting process. This allows for the location of target test nodes with deteriorating latent level recovery capabilities while bus communication can still be maintained.

[0026] This application discloses a method for determining abnormalities in the aging test of an automotive ECU controller.

[0027] The degradation of the recessive drive capability of the ECU node will change the equivalent output impedance, parasitic capacitance charging and discharging path, and transceiver recovery speed during the process of the bus releasing from the dominant level to the recessive level. This will cause the feedback transition voltage sequence to only show slight delay, local step, slope reduction, or ringing attenuation changes in the time domain, which is difficult to identify stably by traditional indicators such as single peak value, rise time, and spectral amplitude. Therefore, this application converts the collected feedback transition voltage sequence into a phase space trajectory composed of voltage coordinates and voltage change rate coordinates, mapping the original one-dimensional time voltage change process into a two-dimensional geometric shape. The segmented deviation area or area integral deviation value of the trajectory in the common voltage coverage area is used to characterize the degree of overall shape distortion of the monitoring trajectory relative to the aging initial reference trajectory. When a node exhibits a latent drive capability degradation, the recovery edge after releasing the bus will show a rate of change collapse, step disappearance, or local dragging within a specific voltage range, leading to a stable change in the area relationship between the phase space trajectory and the reference phase space trajectory. This invention transforms the difficult-to-determine edge detail differences into quantifiable, sortable, and traceable health indicators through this voltage-voltage rate of change two-dimensional trajectory area comparison. Combined with node polling control and repeated confirmation mechanisms, abnormal nodes are identified, thereby achieving early identification of ECU latent drive capability degradation defects.

[0028] Reference Figure 1-4 A method for determining abnormalities in the aging test of an automotive ECU controller, comprising the following steps:

[0029] Acquire the dominant to recessive jump voltage time series data in the initial stage of aging, calculate and generate the initial voltage change rate sequence, and map it to the reference phase space reference trajectory according to the time sampling nodes;

[0030] Furthermore, the dominant-to-recessive voltage transition time-series data of the initial aging stage are acquired, and the initial voltage change rate sequence is calculated and mapped to a reference phase space trajectory according to the time sampling nodes, including:

[0031] Extract multiple initial transition segments from a dominant state to a recessive state from the bus sampling data during the initial aging phase;

[0032] The initial jump segments are length-aligned according to the time sampling nodes to obtain the initial jump segment set;

[0033] The average voltage is calculated for the sampling points with the same index in the initial transition segment set to generate dominant to recessive transition voltage time series data;

[0034] Calculate the initial voltage change rate sequence based on the voltage difference between adjacent sampling points and the sampling time interval;

[0035] The voltage sample values ​​in the dominant-to-recessive voltage transition time series data are used as the horizontal coordinate data, and the initial voltage change rate sequence of the corresponding time sampling node is used as the vertical coordinate data to generate a reference phase space trajectory.

[0036] In one specific implementation, the aging test determination processor includes a data acquisition interface, a timing processing module, a phase space mapping module, a deviation calculation module, a node tracing module, and a result recording module.

[0037] The data acquisition interface receives bus sampling data output from the aging test acquisition system. This bus sampling data can be in a chronologically ordered record format, with each record including at least a time sampling node, a voltage sampling value, and a level status marker. For example, the record might be... ,in This represents the i-th time sampling node, and the unit can be microseconds. This indicates the voltage sampling value corresponding to this sampling time node, in volts. The level status flag corresponding to the sampling time node is indicated. The level status flag includes a dominant state and a recessive state. In one embodiment, the bus sampling frequency is 10MHz to 100MHz. The initial aging stage can be selected as the first 3 to 10 minutes after the start of the aging test. During this stage, the ECU controller has not yet experienced significant thermal stress accumulation, and the collected transition edges can be used as a reference basis for subsequent comparisons.

[0038] Specifically, the timing processing module first acquires the dominant-to-recessive transition voltage timing data during the initial aging stage. This data comes from the bus sampling data during the initial aging stage. The timing processing module searches for the position in the continuous records where the level state marker changes from a dominant state to a recessive state, and uses this position as the transition start point. It then extracts voltage segments from a number of sampling nodes before the transition start point to a number of sampling nodes after the transition start point. If the sampling frequency is 20MHz and the sampling time interval is 0.05 microseconds, it can extract 20 sampling points before the transition start point and 120 sampling points after the transition start point, forming an initial transition segment with a length of 141 sampling points. To avoid the influence of accidental noise, the timing processing module extracts multiple initial transition segments during the initial aging stage, preferably no less than 30 initial transition segments.

[0039] Because the trigger positions and effective lengths of different initial transition segments may vary slightly, the timing processing module aligns the lengths of multiple initial transition segments according to the time sampling nodes. During alignment, the time sampling node where the level state marker first changes from a dominant state to a recessive state is used as the zero point. Each initial transition segment is reprojected onto a unified relative time axis. If the actual sampling point of a certain initial transition segment does not completely coincide with the unified relative time axis, the timing processing module uses linear interpolation to supplement the corresponding voltage sampling value, thus obtaining a set of initial transition segments. Each initial transition segment in the set of initial transition segments has the same number of sampling points, and sampling points with the same sequence number represent the same relative time position.

[0040] Subsequently, the timing processing module calculates the average voltage of sampling points with the same index in the initial transition segment set, generating dominant to recessive transition voltage timing data. For example, suppose there are a total of The initial jump segment, the first The initial jump segment in the first The voltage sampling values ​​at each sampling point are Then, in the dominant-to-recessive transition voltage timing data, the first... Voltage sampling values ​​at each sampling point It can be represented as ,in, This indicates the number of initial transition segments involved in the mean calculation. This process weakens random ringing and sampling spikes in single edges, while preserving the stable pattern of the dominant-to-latent transition process.

[0041] After obtaining the dominant-to-recessive voltage transition time series data, the time series processing module calculates the initial voltage change rate sequence based on the voltage difference between adjacent sampling points and the sampling time interval. Specifically, for the first... Each time sampling node can be used The corresponding initial voltage change rate is obtained, where, Indicates the first The initial voltage change rate at each time sampling node, and These represent the voltage sample values ​​at two adjacent time sampling nodes in the dominant-to-recessive transition voltage time series data, respectively. This indicates the sampling time interval. Since the first time sampling node does not have a previous sampling point, the initial voltage change rate of the second time sampling node can be used to fill the gap, so that the initial voltage change rate sequence maintains the same length as the dominant to recessive voltage transition time series data.

[0042] The phase space mapping module uses the voltage sample values ​​in the dominant-to-recessive transition voltage time series data as the horizontal coordinate data and the initial voltage change rate sequence of the corresponding time sampling node as the vertical coordinate data to generate a reference phase space trajectory. In other words, the time sampling node... Corresponding to a two-dimensional coordinate point Multiple two-dimensional coordinate points are connected in the order of time sampling nodes to form a reference phase space trajectory. This reference phase space trajectory is not an ordinary voltage-time curve, but a trajectory that describes the dynamic relationship between the voltage sampling value and the voltage change rate, and can reflect the edge speed, local pauses and ringing patterns in the recessive level recovery process.

[0043] The real-time implicit voltage jump time series data in the aging test cycle is obtained, the real-time voltage change rate sequence is calculated and generated, and the real-time implicit voltage jump time series data and the real-time voltage change rate sequence are mapped to the first monitoring phase space trajectory according to the same time sampling nodes.

[0044] Furthermore, real-time implicit voltage transition timing data during the aging test cycle is acquired, and a real-time voltage change rate sequence is calculated, including:

[0045] Read the bus voltage sampling data, bus message timestamp data, and node communication status data in the aging test cycle;

[0046] Determine the candidate transition time window based on the bus message timestamp data;

[0047] Within the candidate transition time window, extract the voltage segment from the voltage sampling data that transitions from the dominant level region to the recessive level region to generate real-time recessive transition voltage timing data;

[0048] Based on the node communication status data, the real-time implicit voltage transition timing data is supplemented with the sending node identification code and the participating node identification code;

[0049] Differential calculations are performed on adjacent sampling points in the real-time implicit voltage transition time series data to generate a real-time voltage change rate sequence.

[0050] Furthermore, the real-time implicit voltage transition time series data and the real-time voltage change rate sequence are mapped to the first monitoring phase space trajectory at the same time sampling nodes, including:

[0051] Based on the same time sampling node, real-time implicit voltage transition time series data and real-time voltage change rate sequence are aligned and matched point by point;

[0052] Extract the single voltage sampling value under the same sampling node at the same time as the horizontal coordinate value, and use the corresponding single voltage change rate sampling value as the vertical coordinate value to construct a two-dimensional spatial coordinate point set with time sequence marking;

[0053] The spatial coordinate points in the two-dimensional spatial coordinate point set are connected sequentially according to the time sequence marking to generate the initial discrete phase space mapping curve;

[0054] The initial discrete phase space mapping curve is subjected to high-frequency noise filtering to generate the first monitoring phase space trajectory.

[0055] In one specific implementation, after entering the aging test cycle, the data acquisition interface continuously reads the bus voltage sampling data, bus message timestamp data, and node communication status data from the aging test cycle. The bus voltage sampling data is still stored in the form of time sampling nodes and voltage sampling values. The bus message timestamp data includes the start time, end time, and bit boundary time of each message frame. The node communication status data includes the sending node identification code, the participating node identification code, and the node's communication status at that moment. The timing processing module determines candidate transition time windows based on the bus message timestamp data. For example, it takes two sampling points forward and 80 sampling points backward, centered on the end time of the dominant bit in the message, to form a candidate transition time window. The width of the candidate transition time window should not be too narrow, otherwise it will truncate the later part of the implicit recovery; nor should it be too wide, otherwise it will mix in subsequent bit changes.

[0056] Within the candidate transition time window, the timing processing module extracts the voltage segment from the voltage sampling data that transitions from the dominant level region to the recessive level region, generating real-time recessive transition voltage timing data. The dominant and recessive level regions can be obtained based on the level state markers already calibrated in the test system, or they can be confirmed by combining the message decoding results with the voltage sampling values. To facilitate subsequent traceability, the timing processing module appends a sending node identifier code and a participating node identifier code to the real-time recessive transition voltage timing data based on the node communication status data. The sending node identifier code indicates the source of the message corresponding to the candidate transition time window, and the participating node identifier code indicates the set of nodes that are in the same bus communication environment and participate in the load formation during this aging test cycle.

[0057] Subsequently, the timing processing module performs differential calculations on adjacent sampling points in the real-time implicit voltage transition timing data to generate a real-time voltage change rate sequence. Each real-time voltage change rate in the real-time voltage change rate sequence corresponds to a time sampling node in the real-time implicit voltage transition timing data. The calculation method is consistent with the initial voltage change rate sequence, so the two can be directly compared in terms of dimensions and physical meaning.

[0058] The phase space mapping module maps real-time implicit voltage transition time-series data and real-time voltage change rate sequences to the first monitoring phase space trajectory based on the same time sampling nodes. During implementation, the real-time implicit voltage transition time-series data and the real-time voltage change rate sequence are first aligned and matched point-by-point based on the same time sampling nodes. If the number of sampling points in the truncated real-time implicit voltage transition time-series data differs from the reference phase space trajectory, resampling is performed while maintaining consistency in the initial transition point, ensuring that each real-time voltage sample value can find a corresponding real-time voltage change rate.

[0059] Subsequently, single voltage sampling values ​​at the same sampling node are extracted as horizontal coordinate values, and the corresponding single voltage change rate sampling values ​​are used as vertical coordinate values ​​to construct a two-dimensional spatial coordinate point set with time sequence markings. Each spatial coordinate point in the two-dimensional spatial coordinate point set retains both voltage amplitude information and the rate of change information near that voltage amplitude.

[0060] The phase space mapping module connects the spatial coordinate points within the two-dimensional spatial coordinate point set according to the time sequence markings to generate an initial discrete phase space mapping curve. Considering the possibility of short-term electromagnetic interference at the aging test site, the initial discrete phase space mapping curve may exhibit individual sharp points. The phase space mapping module performs high-frequency noise filtering on the initial discrete phase space mapping curve to generate the first monitoring phase space trajectory. The high-frequency noise filtering can be accomplished using a combination of three-point median filtering and five-point moving average. The three-point median filtering is used to remove isolated sharp points, while the five-point moving average is used to smooth out minor jitters between adjacent points. The coordinate points before and after filtering maintain the same time sequence markings to ensure that data misalignment does not occur in subsequent deviation calculations.

[0061] Calculate the area integral deviation value of the first monitoring phase space trajectory relative to the reference phase space trajectory;

[0062] Further, the area integral deviation value of the first monitoring phase space trajectory relative to the reference phase space trajectory is calculated, including:

[0063] According to the same voltage coordinate segmentation rule, effective trajectory segments are extracted from the reference phase space trajectory and the first monitoring phase space trajectory, respectively.

[0064] Within each voltage coordinate segment, the reference rate of change coordinates of the reference phase space reference trajectory and the monitoring rate of change coordinates of the first monitoring phase space trajectory are read.

[0065] Calculate the absolute difference between the reference rate of change coordinate and the monitored rate of change coordinate, and multiply the absolute difference by the corresponding voltage coordinate segment width to obtain the segment deviation area;

[0066] Sum the areas of all segmented deviations to obtain the area integral deviation value;

[0067] Write the area integral deviation value into the trajectory deviation record table corresponding to the sending node identification code and the participating node identification code.

[0068] In one specific implementation, the deviation calculation module calculates the area integral deviation value of the first monitoring phase space trajectory relative to the reference phase space trajectory. During calculation, the deviation calculation module extracts effective trajectory segments from both the reference phase space trajectory and the first monitoring phase space trajectory according to the same voltage coordinate segmentation rules. An effective trajectory segment can be a voltage range commonly covered by the horizontal coordinate data of both trajectories. For example, if the reference phase space trajectory covers 1.1V to 2.6V and the first monitoring phase space trajectory covers 1.05V to 2.55V, then the common coverage range is 1.1V to 2.55V. The deviation calculation module divides the common coverage range into several voltage coordinate segments, each with a width of 0.005V to 0.05V. Within each voltage coordinate segment, it reads the reference rate of change coordinates of the reference phase space trajectory and the monitoring rate of change coordinates of the first monitoring phase space trajectory. If multiple trajectory points exist within the same voltage coordinate segment, the deviation calculation module takes the average of the vertical coordinate data within that segment as the corresponding rate of change coordinate. If a segment lacks trajectory points, linear interpolation is performed based on adjacent segments.

[0069] Furthermore, the method for selecting the voltage coordinate segment width is supplemented as follows: After determining the common coverage area of ​​the reference phase space trajectory and the first monitoring phase space trajectory, the deviation calculation module first calculates the width of the common coverage area. ,in, These are represented as the maximum and minimum voltage values, respectively, and the voltage quantization resolution of the data acquisition interface is read. Meanwhile, in the reference phase space reference trajectory during the initial aging stage, after removing the sampling points corresponding to the ringing spikes at preset ratios, the median of the difference between adjacent transverse voltage coordinates is calculated. ;

[0070] Subsequently, the voltage coordinate segment width is determined jointly based on the target segment number constraint and the sampling resolution constraint, i.e., candidate widths are selected first. ,in, This is used to divide the common coverage area into approximately 100 voltage segments. This is used to avoid the segment width being close to the sampling quantization noise. This is used to ensure that each segment contains at least two adjacent trajectory sampling points; then... Limiting the voltage to between 0.005V and 0.05V, we obtain... and will according to Round down to the nearest integer multiple of the integer value. The final voltage coordinate segment width is determined by multiples of the given value. Therefore, each test does not use arbitrary values ​​within the range of 0.005V to 0.05V, but rather automatically determines a unique segment width based on the width of the common coverage area, the sampling voltage resolution, and the spacing between voltage points on the reference trajectory. Once this final voltage coordinate segment width is determined, it is written into the segment configuration record and remains consistent throughout the calibration process of the area integration deviation value, the test area integration deviation value, and the preset aging judgment threshold, thus avoiding the problem of incomparable deviation results due to different segment widths.

[0071] The linear interpolation is performed according to the voltage coordinate, not the time coordinate, because the area integral deviation value is calculated based on the correspondence between the voltage coordinate and the voltage change rate coordinate in the phase space trajectory. Specifically, the deviation calculation module records the center voltage of each voltage coordinate segment as... When the first The rate of change coordinates cannot be directly obtained because the lack of trajectory points within each voltage coordinate segment makes it impossible to directly obtain the coordinates within the segment. At that time, within the same phase space trajectory, search for the voltage coordinate segment that is closest to the nearest effective rate of change coordinate and is already present, to the left and right respectively. Let the center voltage of the effective segment on the left be... The rate of change coordinate is The effective segment center voltage on the right is The rate of change coordinate is ,and Then according to the formula Calculate the rate of change coordinates of the missing segments;

[0072] In the actual processing of phase space trajectories, voltage coordinates are not required to change monotonically with time. Therefore, searching to the left and right does not refer to searching forward or backward along the time sampling sequence, nor to searching along the trajectory drawing direction, but rather to searching in the direction of the segment number after voltage coordinate segmentation. Specifically, the deviation calculation module first divides all valid trajectory points corresponding to the dominant to recessive transitions into corresponding voltage coordinate segments according to their voltage coordinates. Then, it performs a unified representative value calculation on multiple rate-of-change coordinates within the same segment, which may be caused by non-monotonic ringing, local oscillation, or repeated passage through the same voltage value. For example, it uses the arithmetic mean, truncated mean, or median of the rate-of-change coordinates within the segment to obtain a unique representative rate-of-change coordinate for each voltage segment. ;

[0073] Following this step, the multi-valued relationship of multiple rates of change corresponding to the same voltage in the original trajectory has been transformed into a single-valued discrete function representing the rate of change for each voltage segment. If the... If a voltage coordinate segment lacks a trajectory point, then the center voltage of the segment is used. Based on this, in the segment sequence number arranged from smallest to largest voltage, searching to the left means searching for a segment whose center voltage is less than... and distance The most recent effective segment, searching to the right means searching for a segment whose center voltage is greater than... and distance The nearest valid segment; let the left valid segment obtained by the search be... The effective segmentation on the right is Then it will still be treated as follows Interpolation is performed. As a result, the interpolation direction is uniquely determined by the magnitude relationship of the center values ​​of the voltage segments, rather than by the temporal relationship of the non-monotonic trajectories; multiple rate-of-change points within the same segment are also compressed into a unique representative value through preset statistical rules. Therefore, each missing segment can only obtain a definite interpolation result, thus ensuring the consistency and reproducibility of the area integral deviation value calculation.

[0074] If multiple consecutive segments lack trajectory points, then each missing segment is determined by taking the valid segments on the left and right sides as endpoints and calculating their respective center voltages. The interpolation results are calculated separately. This interpolation process is performed independently for the reference phase space trajectory, the first monitoring phase space trajectory, or the second monitoring phase space trajectory, without using the rate of change coordinates of another trajectory to replace the current trajectory data. If the missing segment is located at the edge of the common coverage area and it is not possible to obtain effective segments on both the left and right sides at the same time, the deviation calculation module shortens the common coverage area to a voltage range that can be effectively clamped, or marks the edge segment as an invalid segment and does not participate in the area integral summation, thereby ensuring that the source of the interpolation results is clear, the coordinate reference is consistent, and avoids phase space comparison misalignment caused by interpolation based on the time coordinate.

[0075] The same voltage coordinate segmentation rules can be uniformly confirmed and written into the segmentation configuration record by the deviation calculation module after generating the reference phase space reference trajectory. Specifically, the deviation calculation module first reads the lateral coordinate data in the reference phase space reference trajectory to determine the reference voltage coverage range that is stably traversed during the dominant to recessive transition process. Then, combining the lateral coordinate data of multiple initial transition segments in the initial aging stage, it removes extreme voltage sampling values ​​formed only by ringing spikes to obtain the effective voltage coverage range for comparison. Subsequently, it determines the voltage coordinate segment width according to the voltage resolution of the data acquisition interface and the average spacing of adjacent lateral coordinate data in the reference phase space reference trajectory. The voltage coordinate segment width is greater than a preset multiple of the voltage resolution and less than the voltage at which the dominant to recessive transition occurs. The preset ratio of the effective voltage coverage interval width, for example, can be 0.005V to 0.05V, to ensure that each voltage coordinate segment can accommodate sufficient trajectory points without obscuring local recovery drag features. After determining the lower limit, upper limit, and voltage coordinate segment width of the effective voltage coverage interval, the deviation calculation module continuously divides the voltage coordinate segment from the lower limit value according to the fixed voltage coordinate segment width. Parts that are less than a complete segment at the end are discarded or merged into the previous voltage coordinate segment. The lower limit value, upper limit value, voltage coordinate segment width, number of segments, and corresponding reference phase space reference trajectory identifier are written into the segment configuration record. When calculating the area integral deviation value and the test area integral deviation value, the deviation calculation module does not re-divide the voltage coordinate segments, but directly reads the segment configuration record and extracts the effective trajectory segment from the part jointly covered by the horizontal coordinate data of the reference phase space reference trajectory, the first monitoring phase space trajectory, and the second monitoring phase space trajectory. This ensures that the reference rate of change coordinates, monitoring rate of change coordinates, and feedback rate of change coordinates between different trajectories all fall within the same set of voltage coordinate segments for comparison.

[0076] The calculation of the area integral deviation value follows the same dimensional rule. For the th Each voltage coordinate segment is denoted as the reference rate of change coordinate. The coordinates of the monitored rate of change are as follows The voltage coordinate segment width is denoted as Then the area of ​​the segmented deviation It can be represented as ,in, This represents the absolute difference between the baseline rate of change coordinate and the monitored rate of change coordinate, expressed in volts per second. The unit is volts, therefore the unit of the segmented deviation area is volt squared per second. The deviation calculation module sums the areas of all segmented deviations to obtain the area integral deviation value A, expressed as... ,in, This indicates the number of voltage coordinate segments. The larger the area integral deviation value, the more significant the difference in dynamic recovery between the first monitoring phase space trajectory and the reference phase space trajectory. The deviation calculation module writes the area integral deviation value into the trajectory deviation record table corresponding to the sending node identification code and the participating node identification code. The trajectory deviation record table also saves the acquisition time stamp, aging cycle number, reference phase space trajectory identifier, and first monitoring phase space trajectory identifier for cyclical use in node traceability and troubleshooting.

[0077] The unit of the segmented deviation area is volts squared per second. Its engineering meaning is not the physical area, energy, or power in an actual circuit, but rather a quantitative representation of the deviation in the edge recovery speed within a specific voltage segment, within the voltage coordinate-voltage rate of change phase space. Specifically, the difference between the reference rate of change coordinate and the monitored rate of change coordinate, in volts squared per second, represents the difference in the speed of the real-time explicit to implicit recovery edge relative to the initial reference edge of aging near that voltage amplitude. Multiplying this by the voltage segment width yields volts squared per second, representing the dynamic distortion intensity of the rate of change deviation accumulated over the voltage range within that voltage interval.

[0078] Therefore, the larger the area of ​​deviation in a certain segment, the more significant the difference between the implicit recovery speed of the bus voltage in that segment and the reference state, which may manifest as slower recovery, local dragging, step-like pauses, or changes in ringing patterns. The area integral deviation value obtained by summing the areas of deviation in all segments is used to measure the comprehensive dynamic deviation of the phase space trajectory from the reference trajectory during the entire dominant-to-implicit transition process. In other words, volts squared per second is used in this application as an engineering criterion dimension for comparing the changes in edge recovery capability under different aging cycles and different node control states, rather than directly corresponding to the energy consumption parameter of a single electrical component.

[0079] When the area integral deviation value is greater than the preset aging judgment threshold, the node polling control table is read to enter the node traceability and investigation loop.

[0080] Furthermore, when the area integral deviation value exceeds the preset aging judgment threshold, the node polling control table is read to enter the node traceability and troubleshooting loop, including:

[0081] Read the sending node identifier code, participating node identifier code, and acquisition time stamp corresponding to the area integral deviation value from the trajectory deviation record table;

[0082] A set of nodes to be investigated is generated based on the identification codes of the participating nodes;

[0083] The set of nodes to be investigated is sorted according to the bus load level corresponding to the data collection time stamp, forming a node polling control table;

[0084] Node status control data is generated starting from the first node identifier code in the node polling control table;

[0085] Write the node status control data, the node identification code of the target test node, and the current polling sequence number into the source tracing and investigation record table.

[0086] In one specific implementation, the preset aging judgment threshold can be given by calibration data before the aging test begins, or it can be determined by the fluctuation range of multiple sets of reference phase space trajectories during the initial aging stage. In one embodiment, the average area deviation between normal transition segments during the initial aging stage is taken. and standard deviation And set the preset aging judgment threshold to This method of setting values ​​ensures that the preset aging threshold matches the actual test harness, sampling frequency, and bus load.

[0087] The deviation calculation module compares the area integral deviation value with the preset aging judgment threshold. When the area integral deviation value does not exceed the preset aging judgment threshold, the result recording module only saves the trajectory deviation record table for this aging test cycle. When the area integral deviation value is greater than the preset aging judgment threshold, the node tracing module reads the node polling control table and enters the node tracing and investigation cycle.

[0088] The node tracing and troubleshooting loop does not simply process nodes one by one in a fixed order. Instead, it uses data already written to the trajectory deviation record table to determine the scope of the investigation. The node tracing module reads the sending node identifier code, participating node identifier code, and acquisition time stamp corresponding to the area integral deviation value from the trajectory deviation record table. The participating node identifier code can be a list of nodes, such as N01, N03, N05, and N08. The node tracing module generates a set of nodes to be investigated based on the participating node identifier codes. The acquisition time stamp is used to associate with the bus load level at that time. The bus load level can be determined by the number of packets per unit time, the average frame interval, and the number of participating nodes.

[0089] The node tracing module sorts the set of nodes to be investigated according to the bus load level corresponding to the data collection timestamp, forming a node polling control table. During sorting, under heavy load levels, the sending node identifier code and the participating node identifier codes with high message density are prioritized; under light load levels, the participating node identifier codes that are closer to the time interval of the abnormal transition segment are prioritized. This generated node polling control table more closely reflects the communication environment at the time of the anomaly, reducing the number of investigations involving irrelevant nodes.

[0090] The method for confirming the bus load level is set at the time mark corresponding to the collection time mark of each node to be investigated. A fixed statistical window, such as 100ms, 500ms or 1s, is selected before or centered on the collection time mark. The actual occupancy time of all valid communication frames on the bus within the window is counted. The actual occupancy time includes frame start, arbitration segment, control segment, data segment, CRC segment, ACK segment, frame end, frame interval and the equivalent transmission time caused by bit stuffing or protocol overhead.

[0091] Subsequently, the bus load rate corresponding to the acquisition time marker is calculated according to the formula: Bus load rate = Sum of actual occupied times of all valid communication frames within the statistical window / Statistical window duration × 100%. This bus load rate is used as the basis for confirming the load level when sorting the nodes to be investigated. If the system can directly read the real-time load rate from the bus analysis module, the measured load rate output by the bus analysis module is used first. If it cannot be read directly, the equivalent load rate is calculated based on the number of messages, message length, baud rate, and protocol overhead, thereby ensuring that each node to be investigated can obtain a quantifiable bus load level corresponding to its acquisition time marker.

[0092] The quantization range of bus load level can be publicly set as follows: when the bus load rate is greater than or equal to 0% and less than 30%, the corresponding acquisition time marker is confirmed to be in the light load level; when the bus load rate is greater than or equal to 30% and less than 70%, the corresponding acquisition time marker is confirmed to be in the medium load level; when the bus load rate is greater than or equal to 70% and less than or equal to 100%, the corresponding acquisition time marker is confirmed to be in the heavy load level. When forming the node polling control table, the set of nodes to be investigated can be sorted according to the priority order of heavy load level, medium load level, and light load level, so that nodes that are more likely to expose the hidden drive capability degradation defect under high bus occupancy state will enter the source tracing and investigation process first. For multiple nodes to be investigated at the same load level, they can be further sorted from largest to smallest according to the area integral deviation value. If the area integral deviation values ​​are the same, they can be sorted according to the order of acquisition time markers or node identification code order, thus forming a node polling control table with clear sorting rules, reproducibility, and easy execution.

[0093] After the node polling control table is formed, the node tracing module generates node status control data starting from the first node identifier code in the table. Node status control data is a set of data records for the aging test scheduler to identify, including the target test node's node identifier code, control activation time, control duration, control status field, and feedback sampling window identifier. For example, when the target test node is N03, the node status control data record would be node identifier code N03, control activation time... Control hold duration 200ms, control status field is isolated participation in communication status, feedback sampling window identifier. The node tracing module writes the node status control data, the node identification code of the target test node, and the current polling sequence number into the tracing and investigation record table. The current polling sequence number indicates the investigation position of the target test node in the node polling control table. The aging test scheduler adjusts the communication participation status of the target test node in the current investigation phase based on the node status control data. Subsequently, the data acquisition interface collects the feedback switching voltage sequence corresponding to the node status control data.

[0094] In the node tracing and investigation loop, node state control data for a single target test node is generated sequentially, the feedback jump voltage sequence corresponding to the node state control data is collected, the feedback voltage change rate sequence is calculated, and the feedback jump voltage sequence and the feedback voltage change rate sequence are mapped to the second monitoring phase space trajectory.

[0095] Furthermore, the feedback switching voltage sequence corresponding to the node status control data is collected, the feedback voltage change rate sequence is calculated, and the feedback switching voltage sequence and the feedback voltage change rate sequence are mapped to the second monitoring phase space trajectory, including:

[0096] Read the node status control data and the current polling sequence number from the source tracing and investigation record table;

[0097] Extract bus voltage sampling data within the feedback sampling window corresponding to the current polling sequence number to obtain the feedback switching voltage sequence;

[0098] The time sampling node of the feedback jump voltage sequence is calibrated, and the feedback voltage change rate sequence is calculated based on the calibrated feedback jump voltage sequence.

[0099] The calibrated feedback jump voltage sequence is used as the horizontal coordinate data, and the feedback voltage change rate sequence is used as the vertical coordinate data. The second monitoring phase space trajectory is generated by mapping them in time order.

[0100] In one specific implementation, when acquiring the feedback transition voltage sequence, the data acquisition interface and timing processing module read the node status control data and the current polling sequence number from the traceability record table, and determine the feedback sampling window corresponding to the current polling sequence number based on the feedback sampling window identifier in the node status control data. The start point of the feedback sampling window is usually later than the control effective time by a stabilization delay, which can be 5ms to 50ms, to avoid unstable waveforms at the moment of state switching. Within the feedback sampling window corresponding to the current polling sequence number, the timing processing module extracts the bus voltage sampling data to obtain the feedback transition voltage sequence. The feedback transition voltage sequence is still taken from the dominant to recessive transition process, and its truncation length and trigger position are consistent with the real-time recessive transition voltage timing data, facilitating comparison with the reference phase space reference trajectory.

[0101] To prevent minor deviations in the sampling start point from affecting the phase space trajectory, the timing processing module performs time sampling node calibration on the feedback jump voltage sequence. The time sampling node calibration uses the jump start point as the alignment reference, remapping the feedback jump voltage sequence to the same time sampling node as the reference phase space trajectory. If the original sampling time interval of the feedback jump voltage sequence is... The sampling time interval corresponding to the reference phase space trajectory is Then, a calibrated feedback switching voltage sequence is generated through linear interpolation, ensuring that the number of sampling points and their temporal positions in the calibrated feedback switching voltage sequence are consistent with the reference phase space trajectory. Next, the timing processing module calculates the feedback voltage change rate sequence based on the calibrated feedback switching voltage sequence, using the voltage difference between adjacent sampling points divided by the sampling time interval. The phase space mapping module uses the calibrated feedback switching voltage sequence as the horizontal coordinate data and the feedback voltage change rate sequence as the vertical coordinate data, mapping them sequentially to generate the second monitoring phase space trajectory.

[0102] Calculate the integral deviation of the test area of ​​the second monitoring phase space trajectory relative to the reference phase space trajectory;

[0103] In one specific implementation, the deviation calculation module continues to use the same calculation rules as the area integral deviation value to calculate the test area integral deviation value of the second monitoring phase space trajectory relative to the reference phase space trajectory. Since the second monitoring phase space trajectory is obtained after the node status control data corresponding to a certain target test node takes effect, the test area integral deviation value reflects the difference in bus edge recovery after the current target test node is included in the troubleshooting state. If a target test node has a latent drive capability degradation defect, when the communication participation state of the target test node is changed, the latent recovery drag caused by it will be weakened, the second monitoring phase space trajectory will return to the reference phase space trajectory, and the test area integral deviation value will decrease accordingly; if the current target test node is not the source of the anomaly, the trajectory deviation is still retained, and the test area integral deviation value is still greater than the preset aging judgment threshold.

[0104] When the integral deviation of the test area is not greater than the preset aging judgment threshold, the node identification code of the current target test node is written into the abnormal node record, and the judgment result data of the target test node having a hidden driving capability degradation defect is output.

[0105] Furthermore, when the test area integral deviation value is not greater than the preset aging judgment threshold, the node identifier code of the current target test node is written into the abnormal node record, and the judgment result data of the target test node having a hidden driving capability degradation defect is output, including:

[0106] Read the test area integral deviation value, node status control data and current polling sequence number corresponding to the current target test node;

[0107] The comparison result of the test area integral deviation value not exceeding the preset aging judgment threshold is converted into a step disappearance mark;

[0108] Bind the step disappearance marker to the node identification code of the current target test node to generate single-time source tracing evidence data;

[0109] Count the number of single source tracing evidence data corresponding to the same node identifier code within the same aging cycle;

[0110] When the number of traceability evidence data reaches the preset number of confirmations, the node identification code is written into the abnormal node record, and the judgment result data of the hidden driving capability decay defect is output.

[0111] In one specific implementation, when the test area integral deviation value is not greater than a preset aging judgment threshold, the result recording module reads the test area integral deviation value, node status control data, and current polling sequence number corresponding to the current target test node, and converts the comparison result of the test area integral deviation value not being greater than the preset aging judgment threshold into a step disappearance marker. This step disappearance marker indicates that the abnormal deviation in the first monitoring phase space trajectory relative to the reference phase space trajectory has disappeared or fallen back to an acceptable range under the investigation status corresponding to the current target test node. The result recording module binds the step disappearance marker to the node identification code of the current target test node to generate single-time traceability evidence data. This single-time traceability evidence data may include the node identification code, step disappearance marker, test area integral deviation value, node status control data, current polling sequence number, and aging cycle number.

[0112] To avoid misjudgment caused by occasional noise, the result recording module counts the number of single traceability evidence data corresponding to the same node identification code within the same aging cycle. When the number of single traceability evidence data reaches a preset number of confirmations, the node identification code is written into the abnormal node record, and the judgment result data of the latent driving capability decay defect is output. The preset number of confirmations can be 2 to 5 times, and in one embodiment, it is 3 times, that is, the target test node is confirmed to have a latent driving capability decay defect only after the same node identification code shows three consecutive or intermittent step disappearance markers within the same aging cycle. The judgment result data includes at least the node identification code of the target test node, the defect type, the triggered test area integral deviation value, the corresponding preset aging judgment threshold, and the aging cycle number.

[0113] Furthermore, to ensure the feasibility of the preset confirmation count, in the node tracing and investigation loop, when the test area integral deviation value corresponding to a current target test node is not greater than the preset aging judgment threshold and a step disappearance mark is generated, the node tracing module does not immediately read the next node identification code in the node polling control table. Instead, it writes the current target test node into the confirmation candidate state and keeps the node polling pointer at the node identification code, entering the same node repeated confirmation sub-loop. In the same node repeated confirmation sub-loop, the aging test scheduler first restores the target test node to the default communication state and waits for the preset recovery stabilization time. Then, it regenerates the node status control data with the repeated test sequence number, reopens the new feedback sampling window, re-collects the feedback jump voltage sequence, calculates the feedback voltage change rate sequence, maps the second monitoring phase space trajectory, and calculates the new test area integral deviation value. Each time the comparison result under the independent feedback sampling window is completed, a new single tracing evidence data is generated and the evidence count of the node identification code in the same aging loop is updated.

[0114] When the number of step disappearance markers for a node's identifier code reaches a preset number of confirmations, such as three times, the result recording module writes the node identifier code into the abnormal node record, outputs the judgment result data for the implicit drive capability degradation defect, and sends a loop termination command to the node tracing module to stop traversing the remaining nodes in the node polling control table. If the test area integral deviation value obtained from any repeated test in the same node repeated confirmation sub-loop is greater than the preset aging judgment threshold again, the result is recorded as a step retention marker, and the same node repeated confirmation sub-loop for that target test node is terminated. Subsequently, the node tracing module reads the next node identifier code in the node polling control table to continue the investigation. Thus, the preset number of confirmations within the same aging loop is not generated naturally by a single node polling table traversal, but is obtained by setting independent, intervald, and reproducible repeated feedback sampling windows for suspected nodes. This avoids misjudgment caused by single sporadic noise and avoids meaningless repeated traversal of all remaining nodes, thereby improving the efficiency and stability of tracing confirmation.

[0115] Furthermore, when the test area integral deviation value is not greater than the preset aging judgment threshold, and the node identification code of the current target test node has been written into the abnormal node record, and the judgment result data of the target test node having a hidden driving capability decay defect has been output, the node tracing module updates the current node tracing and investigation loop status to the located status, and sends a node tracing and investigation loop termination instruction to the aging test scheduler to stop reading the remaining node identification codes in the node polling control table.

[0116] After receiving the node tracing and troubleshooting loop termination command, the aging test scheduler cancels any unexecuted node status control data generation tasks and restores the already effective node status control data to the default communication state of the aging test loop. Subsequently, the result recording module writes the current polling sequence number, located status, abnormal node records, and judgment result data into the tracing and troubleshooting record table. In this way, if the current target test node has been confirmed to have a hidden drive capability degradation defect, the remaining nodes in the node polling control table are no longer traversed. This avoids meaningless feedback jump voltage sequence acquisition and second monitoring phase space trajectory calculation, and prevents continuous switching of multiple node status control data from interfering with the bus load environment, thus ensuring good timing consistency and test stability in the judgment results of the node tracing and troubleshooting loop.

[0117] Specifically, the confirmation of the preset aging judgment threshold can be completed by the deviation calculation module in the initial stage of aging: After generating the reference phase space reference trajectory, the deviation calculation module selects multiple sets of normal jump segments from the initial jump segment set that has been length aligned in the initial stage of aging, generates multiple initial phase space verification trajectories according to the aforementioned phase space mapping method, and calculates the initial area integral deviation value of each initial phase space verification trajectory relative to the reference phase space reference trajectory; then, outlier initial area integral deviation values ​​caused by sampling spikes are removed, and the mean deviation value is calculated for the remaining initial area integral deviation values. Sum of deviations and standard deviations Among them, the mean deviation This represents the average fluctuation level of the normal transition edge in the initial stage of aging relative to the reference trajectory in the reference phase space, and the standard deviation of the deviation. This indicates the degree of dispersion of the average fluctuation level; finally, the mean deviation is calculated. Standard deviation from the predetermined number of digits The preset aging threshold is obtained by adding the values. For example, when the preset number of bits is 3, the preset aging threshold is: The preset aging judgment threshold, along with the corresponding bus speed, bus load level, test temperature, and reference phase space reference trajectory identifier, is written into the threshold configuration record so that the subsequent area integral deviation value and test area integral deviation value are compared using the same preset aging judgment threshold under the same test conditions.

[0118] When the test area integral deviation value is greater than the preset aging judgment threshold, the next node identifier code in the node polling control table is read and the node tracing and troubleshooting loop continues.

[0119] Furthermore, when the test area integral deviation value exceeds the preset aging judgment threshold, the next node identifier code in the node polling control table is read and the node tracing and troubleshooting loop continues, including:

[0120] The comparison results of the test area integral deviation value being greater than the preset aging judgment threshold are converted into step retention marks;

[0121] Write the step retention mark, the node identification code of the current target test node, and the current polling sequence number into the source tracing and investigation record table;

[0122] Read the next node identifier code from the node polling control table based on the current polling sequence number;

[0123] When the next node identification code exists, generate the node status control data corresponding to the next node identification code, and collect the feedback switching voltage sequence corresponding to the node status control data;

[0124] If the next node identifier code does not exist, write all test area integral deviation values ​​of this round of node tracing and troubleshooting cycle into the unlocated anomaly record.

[0125] In one specific implementation, if the test area integral deviation value is greater than a preset aging judgment threshold, the result recording module converts the comparison result of the test area integral deviation value being greater than the preset aging judgment threshold into a step retention mark. The step retention mark indicates that after the node state control data corresponding to the current target test node takes effect, the abnormal deviation of the second monitoring phase space trajectory relative to the reference phase space trajectory still exists. The result recording module writes the step retention mark, the node identification code of the current target test node, and the current polling sequence number into the source tracing and investigation record table, so that the target test node is marked as a non-priority anomaly source in this round of node source tracing and investigation cycle.

[0126] The node tracing module then reads the next node identifier code from the node polling control table based on the current polling sequence number. If the next node identifier code exists, the module generates the node status control data corresponding to that code, and the data acquisition interface collects the feedback jump voltage sequence corresponding to the node status control data. Subsequent steps include calculating the feedback voltage change rate sequence, mapping the second monitoring phase space trajectory, and calculating the test area integral deviation. The entire node tracing and troubleshooting loop follows the node identifier codes in the node polling control table in sequence until an abnormal node record that meets the confirmation criteria is found, or until the node polling control table has been completely read.

[0127] If the next node identifier code does not exist, it means that the current node tracing and investigation cycle has covered all node identifier codes in the set of nodes to be investigated, but has not yet obtained the step disappearance mark that meets the preset number of confirmations.

[0128] At this point, the result recording module writes all test area integral deviation values ​​from this round of node tracing and troubleshooting loops into the unlocated anomaly record. The unlocated anomaly record includes the node identifier code, current polling sequence number, test area integral deviation value, step retention mark, and corresponding node status control data for each target test node. This unlocated anomaly record continues to participate in comparison in the next aging cycle. When the same combination of participating node identifier codes again shows an area integral deviation value greater than the preset aging judgment threshold, the node tracing module can prioritize reading the node identifier code in the unlocated anomaly record with a larger decrease in test area integral deviation value, thereby improving subsequent location efficiency.

[0129] Furthermore, after outputting the judgment results data for the latent driving capability decay defect, it also includes:

[0130] Read the node identification code, the number of single source tracing evidence data, and the corresponding test area integral deviation value from the abnormal node record;

[0131] Read the area integral deviation value corresponding to the node identification code in different aging cycles from the trajectory deviation record table;

[0132] The area integral deviation values ​​are grouped according to the aging cycle order to generate a node degradation sequence;

[0133] Defect level data is determined based on the deviation distribution of each aging cycle in the node degradation sequence;

[0134] The node identification code, defect level data, and corresponding baseline phase space reference trajectory are written together into the aging anomaly report data.

[0135] In one specific implementation, after outputting the judgment result data of the latent driving capability degradation defect, the result recording module can also generate aging anomaly report data. The result recording module reads the node identification code, the number of single-time traceability evidence data, and the corresponding test area integral deviation value from the anomaly node record, and reads the area integral deviation value corresponding to the node identification code in different aging cycles from the trajectory deviation record table. Since the same node identification code may be collected in multiple aging cycles, the result recording module groups the area integral deviation values ​​according to the aging cycle order to generate a node degradation sequence. The node degradation sequence can be represented as... ,in, Indicates the number of aging cycles. Indicates the first The area integral deviation value corresponding to the node identifier code in each aging cycle. If there are multiple area integral deviation values ​​in a certain aging cycle, the result recording module can take the median of the area integral deviation values ​​in that aging cycle as the result. This is to reduce occasional sampling interference.

[0136] After generating the node degradation sequence, the result recording module determines the defect level data based on the deviation distribution of each aging cycle in the node degradation sequence. The defect level data can be determined by both the deviation amplitude and the deviation stability. The deviation amplitude can be represented as a multiple of the area integral deviation value in the node degradation sequence relative to a preset aging judgment threshold, for example, a multiple. ,in, This indicates the preset aging threshold. Indicates the first The deviation ratio of each aging cycle, and the deviation stability can be represented by the proportion of aging cycles exceeding a preset aging judgment threshold, for example, the proportion. Where n represents the number of aging cycles in the node degradation sequence where the area integral deviation value is greater than the preset aging judgment threshold. This indicates the total number of aging cycles.

[0137] The results recording module combines the deviation ratio and percentage to determine the defect level data. For example, when the median deviation ratio is low and the percentage is low, the defect level data indicates early degradation; when the median deviation ratio is high and the percentage reaches a high level, the defect level data indicates stable degradation; when the area integral deviation value exceeds the preset aging judgment threshold in multiple consecutive aging cycles and the test area integral deviation value repeatedly falls back in the node tracing and investigation cycle, the defect level data indicates high-risk degradation. Finally, the results recording module writes the node identification code, defect level data, and corresponding reference phase space trajectory into the aging anomaly report data. The aging anomaly report data can also save abnormal node records, node degradation sequences, the number of single tracing evidence data, and the corresponding test area integral deviation value, allowing testers to review the physical layer degradation process of the ECU controller during aging testing.

[0138] In practical applications, the reference phase space trajectory can be established separately for different bus speeds, load levels, and test temperatures. If the aging test is conducted at three temperature stages: 85℃, 105℃, and 125℃, the aging test judgment processor can acquire the initial aging stage data at the beginning of each temperature stage and generate the corresponding reference phase space trajectory. Subsequently, the first and second monitoring phase space trajectories are compared with the reference phase space trajectory at the same temperature stage. After this processing, normal electrical parameter drift caused by temperature will not be mistaken for a hidden drive capability degradation defect, while local recovery anomalies caused by the aging of the target test node will still be reflected in the area integral deviation value and the test area integral deviation value. In the overall process, each type of data comes from the previous processing stage or the aging test acquisition system. The explicit to implicit jump voltage time series data is used to generate the initial voltage change rate sequence and the reference phase space trajectory. The real-time implicit jump voltage time series data is used to generate the real-time voltage change rate sequence and the first monitoring phase space trajectory. The area integral deviation value is used to trigger the node tracing and troubleshooting cycle. The feedback jump voltage sequence is used to generate the feedback voltage change rate sequence and the second monitoring phase space trajectory. The test area integral deviation value is finally used to form abnormal node records, unlocated abnormal records and aging abnormal report data. A continuous data processing link is formed between the data.

[0139] The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention, they should all fall within the protection scope of the present invention.

[0140] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0141] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to any specific implementation. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention.

Claims

1. A method for determining abnormalities in the aging test of an automotive ECU controller, characterized in that, Includes the following steps: Acquire the dominant to recessive jump voltage time series data in the initial stage of aging, calculate and generate the initial voltage change rate sequence, and map it to the reference phase space reference trajectory according to the time sampling nodes; The real-time implicit voltage jump time series data in the aging test cycle is obtained, the real-time voltage change rate sequence is calculated and generated, and the real-time implicit voltage jump time series data and the real-time voltage change rate sequence are mapped to the first monitoring phase space trajectory according to the same time sampling nodes. Calculate the area integral deviation value of the first monitoring phase space trajectory relative to the reference phase space trajectory; When the area integral deviation value is greater than the preset aging judgment threshold, the node polling control table is read to enter the node traceability and investigation loop. In the node tracing and investigation loop, node state control data for a single target test node is generated sequentially, the feedback jump voltage sequence corresponding to the node state control data is collected, the feedback voltage change rate sequence is calculated, and the feedback jump voltage sequence and the feedback voltage change rate sequence are mapped to the second monitoring phase space trajectory. Calculate the integral deviation of the test area of ​​the second monitoring phase space trajectory relative to the reference phase space trajectory; When the integral deviation of the test area is not greater than the preset aging judgment threshold, the node identification code of the current target test node is written into the abnormal node record, and the judgment result data of the target test node having a hidden driving capability degradation defect is output. When the test area integral deviation value is greater than the preset aging judgment threshold, the next node identifier code in the node polling control table is read and the node tracing and troubleshooting loop continues.

2. The method for determining abnormalities in the aging test of an automotive ECU controller according to claim 1, characterized in that, Acquire dominant-to-recessive voltage transition time-series data during the initial aging stage, calculate and generate an initial voltage change rate sequence, and map it to a reference phase space trajectory according to the time sampling nodes, including: Extract multiple initial transition segments from a dominant state to a recessive state from the bus sampling data during the initial aging phase; The initial jump segments are length-aligned according to the time sampling nodes to obtain the initial jump segment set; The average voltage is calculated for the sampling points with the same index in the initial transition segment set to generate dominant to recessive transition voltage time series data; Calculate the initial voltage change rate sequence based on the voltage difference between adjacent sampling points and the sampling time interval; The voltage sample values ​​in the dominant-to-recessive voltage transition time series data are used as the horizontal coordinate data, and the initial voltage change rate sequence of the corresponding time sampling node is used as the vertical coordinate data to generate a reference phase space trajectory.

3. The method for determining abnormalities in the aging test of an automotive ECU controller according to claim 1, characterized in that, Acquire real-time implicit voltage transition timing data during the aging test cycle, and calculate and generate a real-time voltage change rate sequence, including: Read the bus voltage sampling data, bus message timestamp data, and node communication status data in the aging test cycle; Determine the candidate transition time window based on the bus message timestamp data; Within the candidate transition time window, extract the voltage segment from the voltage sampling data that transitions from the dominant level region to the recessive level region to generate real-time recessive transition voltage timing data; Based on the node communication status data, the real-time implicit voltage transition timing data is supplemented with the sending node identification code and the participating node identification code; Differential calculations are performed on adjacent sampling points in the real-time implicit voltage transition time series data to generate a real-time voltage change rate sequence.

4. The method for determining abnormalities in the aging test of an automotive ECU controller according to claim 1, characterized in that, The real-time implicit voltage transition time series data and the real-time voltage change rate sequence are mapped to the first monitoring phase space trajectory according to the same time sampling nodes, including: Based on the same time sampling node, real-time implicit voltage transition time series data and real-time voltage change rate sequence are aligned and matched point by point; Extract the single voltage sampling value under the same sampling node at the same time as the horizontal coordinate value, and use the corresponding single voltage change rate sampling value as the vertical coordinate value to construct a two-dimensional spatial coordinate point set with time sequence marking; The spatial coordinate points in the two-dimensional spatial coordinate point set are connected sequentially according to the time sequence marking to generate the initial discrete phase space mapping curve; The initial discrete phase space mapping curve is subjected to high-frequency noise filtering to generate the first monitoring phase space trajectory.

5. The method for determining abnormalities in the aging test of an automotive ECU controller according to claim 1, characterized in that, Calculate the area integral deviation of the first monitoring phase space trajectory relative to the reference phase space trajectory, including: According to the same voltage coordinate segmentation rule, effective trajectory segments are extracted from the reference phase space trajectory and the first monitoring phase space trajectory, respectively. Within each voltage coordinate segment, the reference rate of change coordinates of the reference phase space reference trajectory and the monitoring rate of change coordinates of the first monitoring phase space trajectory are read. Calculate the absolute difference between the reference rate of change coordinate and the monitored rate of change coordinate, and multiply the absolute difference by the corresponding voltage coordinate segment width to obtain the segment deviation area; Sum the areas of all segmented deviations to obtain the area integral deviation value; Write the area integral deviation value into the trajectory deviation record table corresponding to the sending node identification code and the participating node identification code.

6. The method for determining abnormalities in the aging test of an automotive ECU controller according to claim 5, characterized in that, When the area integral deviation value exceeds the preset aging judgment threshold, the node polling control table is read to enter the node tracing and troubleshooting loop, including: Read the sending node identifier code, participating node identifier code, and acquisition time stamp corresponding to the area integral deviation value from the trajectory deviation record table; A set of nodes to be investigated is generated based on the identification codes of the participating nodes; The set of nodes to be investigated is sorted according to the bus load level corresponding to the data collection time stamp, forming a node polling control table; Node status control data is generated starting from the first node identifier code in the node polling control table; Write the node status control data, the node identification code of the target test node, and the current polling sequence number into the source tracing and investigation record table.

7. The method for determining abnormalities in the aging test of an automotive ECU controller according to claim 6, characterized in that, The system collects the feedback jump voltage sequence corresponding to the node status control data, calculates the feedback voltage change rate sequence, and maps the feedback jump voltage sequence and the feedback voltage change rate sequence to a second monitoring phase space trajectory, including: Read the node status control data and the current polling sequence number from the source tracing and investigation record table; Extract bus voltage sampling data within the feedback sampling window corresponding to the current polling sequence number to obtain the feedback switching voltage sequence; The time sampling node of the feedback jump voltage sequence is calibrated, and the feedback voltage change rate sequence is calculated based on the calibrated feedback jump voltage sequence. The calibrated feedback jump voltage sequence is used as the horizontal coordinate data, and the feedback voltage change rate sequence is used as the vertical coordinate data. The second monitoring phase space trajectory is generated by mapping them in time order.

8. The method for determining abnormalities in the aging test of an automotive ECU controller according to claim 7, characterized in that, When the test area integral deviation value is not greater than the preset aging judgment threshold, the node identifier code of the current target test node is written into the abnormal node record, and the judgment result data of the target test node having a hidden driving capability degradation defect is output, including: Read the test area integral deviation value, node status control data and current polling sequence number corresponding to the current target test node; The comparison result of the test area integral deviation value not exceeding the preset aging judgment threshold is converted into a step disappearance mark; Bind the step disappearance marker to the node identification code of the current target test node to generate single-time source tracing evidence data; Count the number of single source tracing evidence data corresponding to the same node identifier code within the same aging cycle; When the number of traceability evidence data reaches the preset number of confirmations, the node identification code is written into the abnormal node record, and the judgment result data of the hidden driving capability decay defect is output.

9. The method for determining abnormalities in the aging test of an automotive ECU controller according to claim 8, characterized in that, When the test area integral deviation value exceeds the preset aging judgment threshold, the next node identifier code in the node polling control table is read and the node tracing and troubleshooting loop continues, including: The comparison results of the test area integral deviation value being greater than the preset aging judgment threshold are converted into step retention marks; Write the step retention mark, the node identification code of the current target test node, and the current polling sequence number into the source tracing and investigation record table; Read the next node identifier code from the node polling control table based on the current polling sequence number; When the next node identification code exists, generate the node status control data corresponding to the next node identification code, and collect the feedback switching voltage sequence corresponding to the node status control data; If the next node identifier code does not exist, write all test area integral deviation values ​​of this round of node tracing and troubleshooting cycle into the unlocated anomaly record.

10. The method for determining abnormalities in the aging test of an automotive ECU controller according to claim 8, characterized in that, After outputting the judgment results data for the latent drive capability decay defect, it also includes: Read the node identification code, the number of single source tracing evidence data, and the corresponding test area integral deviation value from the abnormal node record; Read the area integral deviation value corresponding to the node identification code in different aging cycles from the trajectory deviation record table; The area integral deviation values ​​are grouped according to the aging cycle order to generate a node degradation sequence; Defect level data is determined based on the deviation distribution of each aging cycle in the node degradation sequence; The node identification code, defect level data, and corresponding baseline phase space reference trajectory are written together into the aging anomaly report data.

Images