A CAN bus fault node intelligent location and communication self-healing method based on signal feature analysis
By acquiring and analyzing the physical layer signal characteristics of the CAN bus in real time, and combining the fault feature library and timing correlation analysis, intelligent location and self-healing of CAN bus fault nodes are realized, solving the problem of low fault response efficiency in existing technologies and ensuring rapid fault recovery and communication continuity of the ship's power monitoring system.
Patent Information
- Application Number
- CN202611333067.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-31
- Publication Date
- 2026-11-10
AI Technical Summary
In existing ship power monitoring systems, CAN bus fault location is difficult and lacks self-healing capabilities, resulting in low fault response efficiency. It is unable to quickly and accurately locate fault nodes and automatically isolate or switch communication paths in long-distance wiring scenarios, and is prone to misjudgment and false alarms, especially in environments with strong electromagnetic interference.
By acquiring voltage waveform data of CAN_H and CAN_L in real time, physical layer features such as common-mode voltage, differential voltage amplitude, signal edge rate, and bit time deviation are extracted. Combined with fault feature library and timing correlation analysis, fault nodes are intelligently identified and hierarchical self-healing operations are performed, such as communication parameter adjustment, node isolation, and redundant channel switching.
It enables precise location and second-level automatic recovery of CAN bus fault nodes, improves the communication resilience and continuity of the ship's power monitoring system under harsh conditions, and avoids the spread of single-point faults.
Smart Images

Figure SMS_1 
Figure SMS_2
Abstract
Description
Technical Field
[0001] This invention belongs to the field of marine electronics and automatic control technology, specifically a method for intelligent location and communication self-healing of CAN bus fault nodes based on signal feature analysis. Background Technology
[0002] In shipboard power monitoring systems, the CAN bus typically employs a bus topology, where multiple device nodes (such as generator controllers, switchboard monitoring units, and circuit breaker intelligent trip units) share the same physical bus. When communication anomalies occur on the bus (such as a surge in error frames, bus shutdown, or message loss), existing technologies have the following limitations:
[0003] Fault localization is difficult: Traditional methods rely solely on the error counters of the CAN controller (transmit error count TEC, receive error count REC) to determine the health status of nodes, but they cannot distinguish whether the fault is caused by a specific node (such as a damaged CAN transceiver or a failed terminating resistor) or a problem at the bus physical layer (such as cable damage or electromagnetic interference). In multi-node, long-distance cabling scenarios on ships, troubleshooting node by node is time-consuming and labor-intensive, severely impacting fault response efficiency.
[0004] Lack of self-healing ability: Once the existing system detects a bus fault, it usually adopts a passive mode of "alarm + manual investigation", which cannot automatically isolate the faulty node or switch the communication path. This may cause a single point of failure to spread into a systemic communication interruption, affecting the real-time and continuity of power monitoring.
[0005] Insufficient utilization of signal characteristics: The physical layer characteristics of the CAN bus differential signals (CAN_H / CAN_L) (such as common-mode voltage offset, differential voltage amplitude, signal edge symmetry, bit time deviation, etc.) contain rich fault information, but existing ship power monitoring systems rarely collect and analyze physical layer signals in real time, missing the opportunity to accurately locate the root cause of faults in the early stages.
[0006] Among existing related patent technologies, such as the patent document (CN121187275A) which discloses a fault diagnosis method and system for CAN bus communication failure, in standard CAN bus networks (such as traditional automotive ECUs, industrial sensors, or ship monitoring nodes), nodes typically only have CAN transceivers and microcontrollers, lacking hardware interfaces for remote "dynamic power-off" or "transceiver enable / disable" control by an external master. To achieve the isolation verification described in this patent, hardware modifications must be made to each node in the network (such as adding intelligent power switches or controllable isolation relays). This is extremely costly to modify in existing systems and almost impossible to implement in large networks (such as ships), rendering this "deep verification" step a theoretical concept. The physical layer feature extraction in this patent (such as measuring terminal resistance and voltage to ground) is essentially a global measurement at the bus level. When the bus has dozens of nodes and the cable is long, if a "CAN-H short circuit to ground" or "cable damage" occurs, the electrical characteristics can only indicate that "a short circuit exists on the bus," but cannot directly pinpoint the specific physical node or cable segment. Without relying on the aforementioned difficult-to-implement "node-by-node dynamic isolation," the system can only provide a vague fault range, and maintenance personnel still need to manually troubleshoot segment by segment, failing to truly solve the pain point of "time-consuming and labor-intensive node-by-node troubleshooting in long-distance cabling scenarios." In environments with strong electromagnetic interference, such as ships and heavy machinery, transient electromagnetic pulses (EMI) can cause brief ringing, edge slowing, or common-mode voltage shift in the CAN signal. The electrical characteristic vectors of these transient interferences are highly similar to the characteristics of permanent physical faults such as "missing terminating resistors" or "poor contact." This patent does not propose a mechanism (such as time window statistics or fault duration determination) to distinguish between "transient intermittent interference" and "permanent hardware damage." This can easily lead to frequent false alarms in the system, or even erroneously trigger "dynamic isolation," misjudging healthy nodes as fault sources and kicking them off the network. If the system adds misjudged fault features as "new samples" to the knowledge base when the confidence level of the "preliminary diagnostic conclusion" is low, it will pollute the knowledge base and cause the accuracy of subsequent diagnoses to drop exponentially. This patent does not describe the "confidence verification" and "manual / advanced logic confirmation" mechanisms during online updates. The patent emphasizes that these updates are "executed by an embedded processor." Embedded MCUs have limited storage and computing power; continuously adding new sample vectors will cause the knowledge base to expand excessively, eventually leading to memory overflow or excessively long matching and retrieval times, thus violating the original intention of "rapid localization."
[0007] In summary, although some studies have mentioned CAN bus fault diagnosis methods (such as node troubleshooting based on error frame statistics and offline detection based on impedance measurement), there is a lack of a comprehensive method that can acquire physical layer signal characteristics online in real time, intelligently locate faulty nodes, and automatically perform communication self-healing in the scenario of ship power monitoring with high real-time and high reliability requirements. Summary of the Invention
[0008] This invention proposes a method for intelligent location and communication self-healing of CAN bus fault nodes based on signal feature analysis. The problem to be solved is to realize real-time online monitoring of the signal characteristics of the CAN bus physical layer, intelligently locate fault nodes through signal feature pattern recognition, and automatically perform self-healing operations such as fault isolation and communication path switching. This transforms "passive alarm + manual investigation" into "active perception + intelligent location + automatic recovery", minimizing the impact time of faults and ensuring the communication continuity and reliability of the ship's power monitoring system.
[0009] To achieve the above objectives, the present invention adopts the following technical solution: A method for intelligent location and communication self-healing of faulty nodes on a CAN bus based on signal feature analysis, comprising the following steps: (1) Real-time acquisition of physical layer signals: The voltage waveform data of CAN_H and CAN_L are acquired in real time by a high-precision signal acquisition module connected to the CAN bus; (2) Signal feature extraction: Extract key physical layer feature parameters from waveform data, including common-mode voltage, differential voltage amplitude, signal edge rate, bit time deviation, and signal symmetry index; (3) Fault mode identification and node location: The extracted signal features are matched with the preset fault feature library to identify the fault type, and the fault node ID is located by combining the message timing correlation analysis. (4) Self-healing strategy execution: Based on the fault type and severity, the system automatically performs graded self-healing operations, including: adaptive adjustment of communication parameters, logical isolation of faulty nodes, and switching of redundant communication channels; (5) Status recording and feedback: Record fault events, location results and self-healing operation logs, and report fault diagnosis reports to the monitoring system.
[0010] Furthermore, in step (1), the real-time acquisition of physical layer signals includes: deploying signal acquisition probes at key locations on the CAN bus backbone, using a high-speed ADC to synchronously sample CAN_H and CAN_L, and connecting the acquisition module to the main monitoring system via an isolated SPI or Ethernet interface to ensure that the acquisition process does not affect normal bus communication.
[0011] Furthermore, key locations on the CAN bus backbone include both ends of the bus and the main branch nodes; the high-speed ADC sampling rate is no less than 1MS / s.
[0012] Furthermore, in step (2), the key physical layer characteristic parameters are defined and calculated: Common-mode voltage (Vcm): Vcm = (CAN_H + CAN_L) / 2, normal range: 2.0V ~ 3.0V. Deviation from this range indicates poor grounding, power supply interference, or abnormal node drive. Differential voltage amplitude (Vdiff): Vdiff = CAN_H - CAN_L, normal range for dominant bit: 1.5V ~ 3.0V; normal range for recessive bit: -0.5V ~ 0.05V. Amplitude attenuation indicates that the cable is too long, the terminating resistor is faulty, or the node driving capability is degraded. Signal edge rate (Slew Rate): Calculates the rate of change of voltage dV / dt between the dominant bit transition edge and the recessive-dominant and dominant-recessive transition edges. A slow edge indicates aging of the node transceiver, excessive cable capacitance, or mismatched termination resistor; a slow edge may be accompanied by signal ringing. Bit Time Deviation (BTD): The ratio of the deviation between the actual sampling point position and the standard bit time: BTD = |T_sample - T_nominal| / T_nominal × 100%. A deviation exceeding ±5% indicates node crystal drift or baud rate configuration error. Signal symmetry index (SI): SI = |Vdiff_dominant + Vdiff_recessive| / Vdiff_dominant. This index reflects the degree of symmetry between CAN_H and CAN_L. The closer the value is to 0, the better the symmetry. An increase in SI indicates cable asymmetry, poor connector contact, or imbalance of internal circuitry in the node.
[0013] Furthermore, in step (3), fault mode identification and node localization specifically include: (1) Fault feature library construction: Based on common fault scenarios of ship CAN bus, a fault feature library is established; (2) Node localization algorithm: The algorithm adopts a two-dimensional localization method of "temporal correlation + feature matching", wherein: Temporal correlation analysis: Record the timestamp of each error frame and correlate it with the node ID that recently sent the message. If the probability of an error frame appearing after a certain node ID sends a message is significantly higher than that of other nodes, then that node is a suspect node.
[0014] Feature matching verification: Extract the physical layer signal features when the suspect node sends messages and match them with the fault feature database. Weighted Euclidean distance is used to calculate similarity. Similarity = 1 / (1 + √Σ(w_i × (F_measured_i - F_fault_i)²)) Where F_measured_i is the measured feature value, F_fault_i is the standard value of the i-th feature in the fault feature library, and w_i is the feature weight, which is dynamically adjusted according to the fault type. When the similarity exceeds the threshold (e.g., 0.85), the fault type and the suspected node are confirmed.
[0015] Furthermore, the fault types include node drive capability degradation, abnormal terminating resistance, cable impedance mismatch, and electromagnetic interference injection.
[0016] Furthermore, in step (4), the implementation of the graded self-healing strategy specifically includes: (1) Perform a three-level self-healing operation based on the severity and type of the fault: Level 1 is parameter adaptive. The trigger condition is a slight deviation of bit time exceeding the standard by 5%~10% or slight electromagnetic interference. The self-healing operation automatically fine-tunes the sampling point position from 75% to 80%, or temporarily reduces the baud rate from 250Kbps to 125Kbps. Level 2 is node isolation. The trigger condition is that a node's drive declines or it continuously sends erroneous frames, and the error count is >128. The self-healing operation disconnects the CAN transceiver power supply of the node through a relay or MOSFET switch, or sends a "node sleep" command to make it actively exit the bus. Level 3 is channel switching, triggered by severe damage to the bus physical layer: cable breakage, multiple grounding short circuits, or failure of Level 1 / Level 2 self-healing. The self-healing operation switches to a redundant CAN channel (such as the CAN-B channel), and the main monitoring system continues to communicate with each node through the redundant gateway.
[0017] (2) Self-healing decision-making process: Anomaly detected → Feature extraction and fault identification performed → Fault type and severity determined → If parameters can be repaired, Level 1 self-healing is performed → If a node-level fault occurs, Level 2 self-healing is performed → If a bus-level fault occurs, Level 3 self-healing is performed.
[0018] Furthermore, in the self-healing decision-making process, Level 1 self-healing continuously monitors the process; if recovery occurs, logs are recorded; if recovery fails, the process is upgraded to Level 2, isolating the node and allowing redundant channels to take over. Level 3 self-healing involves channel switching and marking the original channel as pending repair.
[0019] Furthermore, in step (5), the status recording and feedback specifically include: (1) Fault event log: Records the fault occurrence time, duration, fault type, location node ID, trigger signal characteristic value, self-healing operation performed and result; (2) Diagnostic report generation: After the self-healing operation is completed, a structured diagnostic report is automatically generated, which includes root cause analysis of the failure and recommended maintenance measures; (3) Human-machine interface presentation: The location of fault nodes is visualized in the form of a topology map in the power monitoring host computer interface, with highlighted display, fault type labels and self-healing status, supporting maintenance personnel to view detailed waveforms and characteristics with one click.
[0020] Further recommended maintenance measures include "replacing the CAN transceiver of node X" and "checking the terminating resistor of bus segment Y".
[0021] Compared with the prior art, the present invention has the following significant advantages: 1. It breaks through the limitation of existing technologies that rely solely on protocol layer error counting for fault diagnosis, and introduces physical layer signal feature analysis to achieve accurate location of fault root causes, with positioning accuracy reaching the node level; 2. The fault response time is reduced from "hour-level manual troubleshooting" to "second-level automatic processing", significantly improving the communication resilience of the ship's power monitoring system under harsh conditions; 3. By employing a hierarchical self-healing strategy, faulty nodes can be isolated or communication paths can be switched without affecting normal node communication, thus preventing the spread of single-point failures and ensuring continuous system operation. Detailed Implementation
[0022] The present invention proposes a method for intelligent location and communication self-healing of CAN bus fault nodes based on signal feature analysis, the execution steps of which are as follows: (1) Real-time acquisition of physical layer signals: The voltage waveform data of CAN_H and CAN_L are acquired in real time by a high-precision signal acquisition module connected to the CAN bus; (2) Signal feature extraction: Extract key physical layer feature parameters from waveform data, including common-mode voltage, differential voltage amplitude, signal edge rate, bit time deviation, and signal symmetry index; (3) Fault mode identification and node location: The extracted signal features are matched with the preset fault feature library to identify the fault type (such as node drive capability degradation, abnormal terminal resistance, cable impedance mismatch, electromagnetic interference injection), and the fault node ID is located by combining message timing correlation analysis. (4) Self-healing strategy execution: Based on the fault type and severity, the system automatically performs graded self-healing operations, including: adaptive adjustment of communication parameters, logical isolation of faulty nodes, and switching of redundant communication channels; (5) Status recording and feedback: Record fault events, location results and self-healing operation logs, and report fault diagnosis reports to the monitoring system.
[0023] Example: 1. Real-time acquisition and feature extraction of physical layer signals (1) Signal acquisition method: Signal acquisition probes are deployed at key locations on the CAN bus backbone (such as both ends of the bus and branch nodes), and high-speed ADCs (sampling rate not less than 1MS / s) are used to synchronously sample CAN_H and CAN_L. The acquisition module is connected to the main monitoring system via an isolated SPI or Ethernet interface to ensure that the acquisition process does not affect normal bus communication.
[0024] (2) Definition and calculation of key feature parameters: Common-mode voltage (Vcm): Vcm = (CAN_H + CAN_L) / 2, normal range: 2.0V ~ 3.0V (ISO11898-2 standard). Deviation from this range indicates poor grounding, power supply interference, or abnormal node drive.
[0025] Differential voltage amplitude (Vdiff): Vdiff = CAN_H - CAN_L, normal range for dominant bit: 1.5V ~ 3.0V; normal range for recessive bit: -0.5V ~ 0.05V. Amplitude attenuation indicates excessive cable length, failure of terminating resistor, or deterioration of node driving capability.
[0026] Signal edge rate (Slew Rate): Calculates the rate of voltage change (dV / dt) at the dominant bit transition edge (recessive→dominant, dominant→recessive). A slow edge indicates aging of the node transceiver, excessive cable capacitance, or mismatched terminating resistors; a slow edge may be accompanied by signal ringing.
[0027] Bit Time Deviation (BTD): The ratio of the deviation between the actual sampling point position and the standard bit time: BTD = |T_sample - T_nominal| / T_nominal × 100%. A deviation exceeding ±5% indicates node crystal drift or baud rate configuration error.
[0028] Signal symmetry index (SI): SI = |Vdiff_dominant + Vdiff_recessive| / Vdiff_dominant. This index reflects the degree of symmetry between CAN_H and CAN_L. The closer the value is to 0, the better the symmetry. An increased SI indicates cable asymmetry, poor connector contact, or imbalance in the internal circuitry of the node.
[0029] 2. Fault Mode Identification and Node Location (1) Construction of the fault feature library: A fault feature library was established based on common fault scenarios of ship CAN bus, as shown in the following example:
[0030] (2) Node localization algorithm: Employing a two-dimensional localization approach combining temporal correlation and feature matching: Temporal correlation analysis: Record the timestamp of each erroneous frame and correlate it with the node ID that most recently sent the message. If the probability of an erroneous frame appearing after a certain node ID sends a message is significantly higher than that of other nodes, then that node is a suspect node.
[0031] Feature matching verification: Extract the physical layer signal features when the suspect node sends messages and match them with the fault feature database. Weighted Euclidean distance is used to calculate similarity. Similarity = 1 / (1 + √Σ(w_i × (F_measured_i - F_fault_i)²)) Where F_measured_i is the measured feature value, F_fault_i is the standard value of the i-th feature in the fault feature library, and w_i is the feature weight (dynamically adjusted according to the fault type). When the similarity exceeds the threshold (e.g., 0.85), the fault type and suspected node are confirmed.
[0032] 3. Implementation of a graded self-healing strategy (1) Perform a three-level self-healing operation based on the severity and type of the fault:
[0033] (2) Self-healing decision-making process: Anomaly detected → Feature extraction and fault identification performed → Fault type and severity determined → If parameters can be repaired, Level 1 self-healing is performed (continuous monitoring; logs are recorded if recovery occurs; otherwise, the process is escalated to Level 2) → For node-level faults, Level 2 self-healing is performed (isolate the node and take over the redundant channel) → For bus-level faults, Level 3 self-healing is performed (channel switching; the original channel is marked as pending repair).
[0034] 4. Status recording and feedback (1) Fault event log: Records the time of occurrence of the fault, duration, fault type, location node ID, trigger signal characteristic value, self-healing operation performed and results.
[0035] (2) Diagnostic report generation: After the self-healing operation is completed, a structured diagnostic report is automatically generated, which includes root cause analysis of the fault and recommended maintenance measures (such as "replace the CAN transceiver of node X" and "check the termination resistor of bus Y segment").
[0036] (3) Human-machine interface presentation: The location of the fault node (highlighted), fault type label and self-healing status are presented in the form of a topology map in the power monitoring host computer interface, and the operation and maintenance personnel can view detailed waveforms and features with one click.
[0037] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the invention. Those skilled in the art can make various changes and modifications without departing from the concept and scope of the present invention. Therefore, all technical solutions obtained by equivalent substitution or equivalent transformation fall within the protection scope of the present invention.
Claims
1. A method for intelligent location and communication self-healing of faulty nodes in a CAN bus based on signal feature analysis, characterized in that, The execution steps are as follows: (1) Real-time acquisition of physical layer signals: The voltage waveform data of CAN_H and CAN_L are acquired in real time by a high-precision signal acquisition module connected to the CAN bus; (2) Signal feature extraction: Extract key physical layer feature parameters from waveform data, including common-mode voltage, differential voltage amplitude, signal edge rate, bit time deviation, and signal symmetry index; (3) Fault mode identification and node location: The extracted signal features are matched with the preset fault feature library to identify the fault type, and the fault node ID is located by combining the message timing correlation analysis. (4) Self-healing strategy execution: Based on the fault type and severity, the system automatically performs graded self-healing operations, including: adaptive adjustment of communication parameters, logical isolation of faulty nodes, and switching of redundant communication channels; (5) Status recording and feedback: Record fault events, location results and self-healing operation logs, and report fault diagnosis reports to the monitoring system.
2. The method according to claim 1, characterized in that, In step (1), the real-time acquisition of physical layer signals includes: deploying signal acquisition probes at key locations on the CAN bus backbone, using a high-speed ADC to synchronously sample CAN_H and CAN_L, and connecting the acquisition module to the main monitoring system via an isolated SPI or Ethernet interface to ensure that the acquisition process does not affect normal bus communication.
3. The method according to claim 2, characterized in that, Key locations on the CAN bus backbone include both ends of the bus and the main branch nodes; the high-speed ADC sampling rate should not be less than 1MS / s.
4. The method according to claim 1, characterized in that, In step (2), the definition and calculation of key physical layer characteristic parameters include: Common-mode voltage (Vcm): Vcm = (CAN_H + CAN_L) / 2, normal range: 2.0V ~ 3.0V. Deviation from this range indicates poor grounding, power supply interference, or abnormal node drive. Differential voltage amplitude (Vdiff): Vdiff = CAN_H - CAN_L, normal range for dominant bit: 1.5V ~ 3.0V; normal range for recessive bit: -0.5V ~ 0.05V. Amplitude attenuation indicates that the cable is too long, the terminating resistor is faulty, or the node driving capability is degraded. Signal edge rate (Slew Rate): Calculates the rate of change of voltage dV / dt between the dominant bit transition edge and the recessive-dominant and dominant-recessive transition edges. A slow edge indicates aging of the node transceiver, excessive cable capacitance, or mismatched termination resistor; a slow edge may be accompanied by signal ringing. Bit Time Deviation (BTD): The ratio of the deviation between the actual sampling point position and the standard bit time: BTD = |T_sample - T_nominal| / T_nominal × 100%. A deviation exceeding ±5% indicates node crystal drift or baud rate configuration error. Signal symmetry index (SI): SI = |Vdiff_dominant + Vdiff_recessive| / Vdiff_dominant. This index reflects the degree of symmetry between CAN_H and CAN_L. The closer the value is to 0, the better the symmetry. An increase in SI indicates cable asymmetry, poor connector contact, or imbalance of internal circuitry in the node.
5. The method according to claim 1, characterized in that, In step (3), fault mode identification and node localization specifically include: (1) Fault feature library construction: Based on common fault scenarios of ship CAN bus, a fault feature library is established; (2) Node localization algorithm: adopts a two-dimensional localization method of "temporal correlation + feature matching", wherein: Temporal correlation analysis: Record the timestamp of each error frame and correlate it with the node ID of the most recently sent message. If the probability of an error frame appearing after a certain node ID sends a message is significantly higher than that of other nodes, then that node is a suspect node. Feature matching verification: Extract the physical layer signal features when the suspect node sends messages and match them with the fault feature database. Weighted Euclidean distance is used to calculate similarity. Similarity = 1 / (1 + √Σ(w_i × (F_measured_i - F_fault_i)²)) Where F_measured_i is the measured feature value, F_fault_i is the standard value of the i-th feature in the fault feature library, and w_i is the feature weight, which is dynamically adjusted according to the fault type. When the similarity exceeds the threshold (e.g., 0.85), the fault type and the suspected node are confirmed.
6. The method according to claim 1, characterized in that, Fault types include node drive capability degradation, abnormal terminating resistance, cable impedance mismatch, and electromagnetic interference injection.
7. The method according to claim 1, characterized in that, In step (4), the implementation of the graded self-healing strategy specifically includes: (1) Perform a three-level self-healing operation based on the severity and type of the fault: Level 1 is parameter adaptive. The trigger condition is a slight deviation of bit time exceeding the standard by 5%~10% or slight electromagnetic interference. The self-healing operation automatically fine-tunes the sampling point position from 75% to 80%, or temporarily reduces the baud rate from 250Kbps to 125Kbps. Level 2 is node isolation. The trigger condition is that a node's drive declines or it continuously sends erroneous frames, and the error count is >128. The self-healing operation disconnects the CAN transceiver power supply of the node through a relay or MOSFET switch, or sends a "node sleep" command to make it actively exit the bus. Level 3 is channel switching, triggered by severe damage to the bus physical layer: cable breakage, multiple grounding short circuits, or failure of Level 1 / Level 2 self-healing. The self-healing operation switches to a redundant CAN channel (such as CAN-B channel), and the main monitoring system continues to communicate with each node through the redundant gateway. (2) Self-healing decision-making process: Anomaly detected → Feature extraction and fault identification performed → Fault type and severity determined → If parameters can be repaired, Level 1 self-healing is performed → If a node-level fault occurs, Level 2 self-healing is performed → If a bus-level fault occurs, Level 3 self-healing is performed.
8. The method according to claim 1, characterized in that, In the self-healing decision-making process, Level 1 self-healing continuously monitors the process. If the process recovers, it logs the results. If the process fails to recover, it is upgraded to Level 2, which isolates the node and takes over the redundant channel. Level 3 self-healing involves switching the channel and marking the original channel as pending repair.
9. The method according to claim 1, characterized in that, In step (5), the status recording and feedback specifically include: (1) Fault event log: Records the fault occurrence time, duration, fault type, location node ID, trigger signal characteristic value, self-healing operation performed and result; (2) Diagnostic report generation: After the self-healing operation is completed, a structured diagnostic report is automatically generated, which includes root cause analysis of the failure and recommended maintenance measures; (3) Human-machine interface presentation: The location of fault nodes is visualized in the form of a topology map in the power monitoring host computer interface, with highlighted display, fault type labels and self-healing status, supporting maintenance personnel to view detailed waveforms and characteristics with one click.
10. The method according to claim 9, characterized in that, Recommended maintenance measures include "replacing the CAN transceiver of node X" and "checking the terminating resistor of bus segment Y".
Citation Information
Patent Citations
Fault diagnosis method and fault diagnosis system for CAN bus communication failure
CN121187275A
