Search

CN-122020487-A - Intelligent network-connected automobile fault detection method, equipment and medium

CN122020487ACN 122020487 ACN122020487 ACN 122020487ACN-122020487-A

Abstract

The invention discloses a method, equipment and medium for detecting faults of an intelligent network-connected automobile, and relates to the technical field of fault diagnosis, comprising the steps of acquiring control link data and network-connected state information of the intelligent network-connected automobile, segmenting the control link data according to a preset time window, and generating windowed data segments; and when the drift index reaches the updating judgment threshold value, generating an inspection parameter updating packet, and applying the inspection parameter updating packet to update the closed-loop consistency constraint and the observation enhancement strategy. The invention updates the closed-loop consistency constraint and the observation enhancement strategy by applying the inspection parameter update package, thereby enhancing the reliability and continuity of fault diagnosis of the intelligent network-connected automobile.

Inventors

  • BAI CAISHENG

Assignees

  • 兰州现代职业学院

Dates

Publication Date
20260512
Application Date
20260401

Claims (10)

  1. 1. An intelligent network connection automobile fault detection method is characterized by comprising the following steps of, Acquiring control link data and network connection state information of an intelligent network connection automobile, segmenting the control link data according to a preset time window, and generating windowed data segments; Performing consistency evaluation on the windowed data segments according to preset closed-loop consistency constraints, generating residual vectors, mapping the residual vectors into suspected fault sets, and calculating observability indexes based on the suspected fault sets, the residual vectors and networking state information; Selecting and configuring an observation enhancement strategy according to the observability index, sampling and fragment caching control link data through the observation enhancement strategy to obtain a fault event packet, correcting a residual vector based on the fault event packet, and outputting a corrected residual vector; performing fault discrimination on the corrected residual vector, outputting a fault diagnosis result, caching the corrected residual vector according to a preset time window to form a residual vector sequence, and calculating a drift index based on the fault diagnosis result and the residual vector sequence; When the drift index reaches the update decision threshold, an inspection parameter update package is generated and applied to update the closed-loop consistency constraint and the observation enhancement strategy.
  2. 2. The method for detecting intelligent network-connected automobile faults according to claim 1, wherein the step of generating windowed data segments comprises the following steps, Acquiring control link data and network connection state information of an intelligent network connection automobile, unifying time stamps of the control link data, and segmenting the control link data after unifying the time stamps according to a preset time window to obtain a plurality of window segments; and binding the window segments with the network connection state information one by one to generate windowed data segments.
  3. 3. The intelligent network connection automobile fault detection method of claim 2, wherein the generating of the residual vector comprises the following steps, Extracting a control instruction sequence, a feedback measurement sequence, a state estimation sequence and an actuator constraint state from the windowed data segment to serve as closed-loop consistency evaluation data; Setting closed-loop consistency constraints according to vehicle control logic and actuator constraint logic, wherein the closed-loop consistency constraints comprise control time sequence constraints, state consistency constraints and actuator consistency constraints; Extracting a trigger time from a control instruction sequence, extracting an arrival time from a feedback measurement sequence, calculating a time difference between the trigger time and the arrival time, comparing the time difference with an allowable delay range limited by control time sequence constraint, and generating control time sequence consistency deviation; Time alignment is carried out on the state estimation sequence and the feedback measurement sequence, differential value calculation is carried out, a deviation sequence is output, the deviation sequence is compared with an allowable deviation range limited by the state consistency constraint, and state consistency deviation is generated; extracting a limiting interval and a saturation interval from the constraint state of the actuator, intercepting fragments in the limiting interval and the saturation interval in a control instruction sequence and a feedback measurement sequence, outputting a control instruction subsequence and a feedback measurement subsequence, performing differential value calculation to obtain an actuator deviation sequence, and comparing the actuator deviation sequence with an allowable deviation range limited by actuator consistency constraint to generate actuator consistency deviation; normalizing and combining the control time sequence consistency deviation, the state consistency deviation and the actuator consistency deviation to generate a residual vector.
  4. 4. The intelligent network connection automobile fault detection method according to claim 3, wherein the calculating the observability index based on the suspected fault set, the residual vector and the network connection state information comprises the following specific steps of, Carrying out component analysis on the residual vector to generate residual characteristics; Matching the residual characteristics with a preset fault mode library to generate a suspected fault set; respectively associating and corresponding each candidate fault type in the suspected fault set with a residual vector to generate a fault residual association table; And configuring an observability evaluation rule according to the network connection state information, and carrying out observability evaluation on the fault residual error association table through the observability evaluation rule to generate an observability index.
  5. 5. The method for detecting intelligent network-connected automobile faults as claimed in claim 4, wherein the step of obtaining the fault event package is as follows, Sequencing the suspected fault sets through observability indexes to form a priority diagnosis sequence; Carrying out hierarchical sampling scheduling on the priority diagnosis sequence, and configuring a fragment caching rule through networking state information to form an observation enhancement strategy; Sampling control link data through an observation enhancement strategy to generate a high-observability data segment; and caching and packaging the high observability data fragments according to fragment caching rules to obtain a fault event packet.
  6. 6. The method for intelligent network-connected automobile fault detection according to claim 5, wherein the outputting of the correction residual vector means performing consistency evaluation on the high observability data section in the fault event package to obtain a correction deviation component, updating the residual vector by the correction deviation component, and outputting the correction residual vector.
  7. 7. The method for intelligent network-connected automobile fault detection according to claim 6, wherein the forming of the residual vector sequence comprises the following specific steps, Correspondingly matching the correction residual vector with the candidate fault type, generating fault similarity, normalizing the fault similarity, and outputting a discrimination score; Determining the fault type and the fault component through the discrimination score, and outputting a fault diagnosis result; Binding the fault diagnosis result with the correction residual vector one by one and marking a time stamp to generate a correction residual time sequence record; And caching and collecting the corrected residual vector records according to a preset time window to form a residual vector sequence.
  8. 8. The method for intelligent network-connected automobile fault detection according to claim 7, wherein the application of the inspection parameter update package updates the closed-loop consistency constraint and the observation enhancement strategy by the following steps, Comparing the drift index with an update judgment threshold value to obtain an update trigger judgment result; When the update triggering judgment result represents that updating is needed, parameters of closed loop consistency constraint and parameters of observation enhancement strategy are adjusted through drift indexes, and the parameters are packaged into an inspection parameter update package; and carrying out parameter replacement on the closed-loop consistency constraint and the observation enhancement strategy by checking the parameter updating packet, and outputting the updated closed-loop consistency constraint and the updated observation enhancement strategy.
  9. 9. A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the computer program is characterized in that the processor realizes the steps of the intelligent network connection automobile fault detection method according to any one of claims 1-8 when executing the computer program.
  10. 10. A computer readable storage medium having a computer program stored thereon, wherein the computer program when executed by a processor implements the steps of the intelligent network connection automobile fault detection method of any one of claims 1 to 8.

Description

Intelligent network-connected automobile fault detection method, equipment and medium Technical Field The invention relates to the technical field of fault diagnosis, in particular to an intelligent network connection automobile fault detection method, equipment and medium. Background Along with the evolution of the electronic and electric architecture of the automobile from the centralized and cross-domain fusion type of the distributed domain, the intelligent network-connected automobile gradually forms a complex control closed loop of sensing-decision-execution-communication deep coupling, and a key execution link in the running process of the automobile is not only dependent on real-time coordination among vehicle-mounted controllers, but also is jointly influenced by the communication state of the automobile road cloud, the constraint boundary of an executor and the state estimation precision, and particularly in the scene of the intelligent network-connected automobile, the control link fault is not limited to the traditional hardware failure any more, and also comprises communication disturbance, closed loop mismatch, state drift, constraint conflict and other complex anomalies. However, the prior art still has the defects that the detection is only carried out aiming at single fault characteristics, so that the identification capability of time sequence faults, slow-release faults and coupling faults is insufficient, the existing fault detection strategies often adopt fixed constraint parameters and fixed observation strategies, and the parameter drift, working condition migration and control link degradation in long-term running of the vehicle are difficult to cope with. Disclosure of Invention The present invention has been made in view of the above-described problems occurring in the prior art. Therefore, the invention provides an intelligent network connection automobile fault detection method which solves the problem that parameter drift, working condition migration and control link degradation in long-term running of a vehicle are difficult to deal with. In order to solve the technical problems, the invention provides the following technical scheme: in a first aspect, the present invention provides a method for detecting faults in an intelligent network-connected vehicle, comprising, Acquiring control link data and network connection state information of an intelligent network connection automobile, segmenting the control link data according to a preset time window, and generating windowed data segments; Performing consistency evaluation on the windowed data segments according to preset closed-loop consistency constraints, generating residual vectors, mapping the residual vectors into suspected fault sets, and calculating observability indexes based on the suspected fault sets, the residual vectors and networking state information; Selecting and configuring an observation enhancement strategy according to the observability index, sampling and fragment caching control link data through the observation enhancement strategy to obtain a fault event packet, correcting a residual vector based on the fault event packet, and outputting a corrected residual vector; performing fault discrimination on the corrected residual vector, outputting a fault diagnosis result, caching the corrected residual vector according to a preset time window to form a residual vector sequence, and calculating a drift index based on the fault diagnosis result and the residual vector sequence; When the drift index reaches the update decision threshold, an inspection parameter update package is generated and applied to update the closed-loop consistency constraint and the observation enhancement strategy. As an optimal scheme of the intelligent network connection automobile fault detection method, the method comprises the following steps of generating windowed data segments, Acquiring control link data and network connection state information of an intelligent network connection automobile, unifying time stamps of the control link data, and segmenting the control link data after unifying the time stamps according to a preset time window to obtain a plurality of window segments; and binding the window segments with the network connection state information one by one to generate windowed data segments. As an optimal scheme of the intelligent network connection automobile fault detection method, the method comprises the following specific steps of, Extracting a control instruction sequence, a feedback measurement sequence, a state estimation sequence and an actuator constraint state from the windowed data segment to serve as closed-loop consistency evaluation data; Setting closed-loop consistency constraints according to vehicle control logic and actuator constraint logic, wherein the closed-loop consistency constraints comprise control time sequence constraints, state consistency constraints and actuator consistency constraints;