CN-121600616-B - Intelligent automobile fault diagnosis system

CN121600616BCN 121600616 BCN121600616 BCN 121600616BCN-121600616-B

Abstract

The invention relates to the technical field of fault diagnosis, in particular to an intelligent automobile fault diagnosis system which comprises an acquisition sequence checking module, an intensity abnormal grouping module, a trend path reconstruction module, a frequency priority weighting module and a reverse position estimation judging module. According to the invention, the sequence of signals and the abrupt change of intensity are checked through time labels and sampling point serial numbers, the discrimination and clustering of the acquisition time sequence disorder phenomenon are realized, the grouping identification of jump type abnormal signals is completed by combining the connection path sequence of signal nodes in a wiring structure diagram, the signal evolution paths are screened by means of trend change consistency in the time sequence and are mapped with the running state, the triggering frequency of each path in different time periods is subjected to stepwise retrieval and incremental trend judgment, the tracing point positions of the signals are reversely positioned through path spatial distribution, the accurate identification capability of potential fault source points can be improved, and the problem that the diagnosis accuracy is easy to drop under the condition that the fault nodes cross space-time dislocation is solved.

Inventors

Yiliyazi Rehemann River
ZHAO DEWU
YANG YI
WANG YONGXIANG
XU MING

Assignees

贵州中扬醇动力科技有限公司

Dates

Publication Date: 20260508
Application Date: 20260130

Claims (7)

1. An intelligent diagnostic system for an automobile fault, the system comprising: The acquisition sequence verification module acquires a fault code time tag and a sampling point number sequence uploaded by a sensing node in an automobile, judges whether signal items with staggered sequence and abrupt strength change exist between sampling points according to the time tag sequence, and generates an acquisition sequence abnormal signal set by dividing coincidence conditions into a staggered signal set; The strength anomaly grouping module calls an automobile wiring structure diagram based on the acquisition sequence anomaly signal set, extracts the connection path sequence of signal nodes in the structure diagram, carries out connection continuity judgment with the previous signal path, and generates a path jump association fault group; The trend path reconstruction module screens trend change sequences with consistent directions in continuous segments according to the time sequence of the signals in the path jump association fault group, classifies the trend change sequences as the same signal evolution path number, and performs mapping binding with the running state of the automobile to generate a trend linkage path identification set; The frequency priority weighting module calls the path numbers in the trend linkage path identification set, searches the triggering conditions in the automobile running state recording section, counts the repeated triggering frequency of signals corresponding to each group of path numbers according to the running stage group, generates a key path priority response table, and quantitatively sorts the frequency and the criticality of automobile fault diagnosis; The reverse position estimation judging module extracts the spatial distribution sequence of the signal nodes corresponding to the paths in the structural diagram by utilizing the priority path numbers in the key path priority response table, carries out reverse tracking on the nodes with position jump, marks the previous signal nodes as source points if no continuous connection path exists between the nodes and the previous signal nodes, and generates a fault source point estimation result; The fault source point presumption result comprises a potential source node number, a corresponding space position and a path disconnection mark.
2. The intelligent diagnosis system of claim 1, wherein the collection sequence abnormal signal set includes a dislocation signal item number, a time tag difference value and an intensity mutation index, the path jump association fault group includes a jump node number, a jump segment signal value and a front-back path number difference, the trend linkage path identification set includes a trend path number, a direction consistent change sequence and running state mapping information, and the critical path priority response table includes a path number, an affiliated running stage and an increasing frequency identification.
3. The intelligent diagnostic system for an automotive fault of claim 1, wherein the acquisition sequence verification module comprises: The fault code sequencing submodule acquires a fault code time tag and a sampling point number sequence uploaded by a sensing node in an automobile, extracts sampling point numbers corresponding to each group of time tags, performs ascending sequence arrangement according to the time tags, calculates index difference values of adjacent numbers in an original sequence and a sequencing sequence, and generates a sampling sequence offset value sequence; the dislocation judging submodule calls the sampling sequence offset value sequence, extracts continuous homodromous offset fragments according to the offset direction change of adjacent sampling points, calculates a dislocation signal change rate value by analyzing the intensity change amplitude, and combines and compares the dislocation signal change rate value with the fragment number change degree to generate a dislocation signal fragment index group; And the abnormal signal extraction sub-module extracts sampling numbers and time labels corresponding to the fragments according to the fragment information positioned by the dislocation signal fragment index group, completes segment reconstruction and sampling sequence mark updating, and acquires an acquisition sequence abnormal signal set.
4. The intelligent diagnostic system for an automotive fault of claim 3, wherein the intensity anomaly grouping module comprises: The connection path extraction submodule calls signal node connection information in an automobile wiring structure diagram based on the acquisition sequence abnormal signal set, positions the connection relation of each pair of adjacent abnormal signals in the structure diagram, extracts real-time path sequence numbers, arranges node indexes according to the signal acquisition sequence, and generates a path index sequence value; The path jump judging submodule calls the path index sequence value, compares the serial numbers of the structural connection sequences of any two continuous signals according to the signal acquisition sequence, judges that the two nodes do not form a continuous path in the wiring structure diagram, counts the occurrence times of structural jump, and generates a jump signal pair frequency value; And the jump fault marking submodule calls the jump signal pair frequency values, sequentially extracts signal acquisition positions according to real-time signal numbers, calculates path jump intensity distribution values, extracts signal point pair combinations with prominent numerical proportion according to the path jump intensity distribution values, and generates a path jump associated fault group.
5. The vehicle fault intelligent diagnostic system of claim 4, wherein the trend path reconstruction module comprises: The jump signal screening submodule sequentially detects the time segment continuity in the time sequence based on the time sequence of the signals in the path jump association fault group, calculates the value amplification of adjacent time nodes, judges whether trend changes exist in the change direction, screens signal segments with continuous and same amplification direction, and generates a jump signal segment set with consistent trend; The direction sequence classifying sub-module invokes the trend consistent jump signal segment set, analyzes the change direction marks of the signal segments, classifies according to the trend direction, identifies the corresponding relation between the classifying number and the signal path position, and obtains a trend classifying signal path sequence; The state path binding sub-module classifies the signal path sequence based on the trend, extracts time period and automobile running state data, constructs a mapping combination of the path sequence and the state data, calculates trend linkage intensity values of each group of path state combinations, screens path state combinations with outstanding linkage degree, and obtains a trend linkage path identification set.
6. The intelligent diagnostic system for an automotive fault of claim 5, wherein the frequency priority weighting module comprises: The path triggering statistics sub-module calls the path numbers in the trend linkage path identification set, retrieves signal time stamps according to the path number indexes in the running state recording section, groups the time stamps according to the running stage, and counts the signal triggering times of each group of path numbers in the corresponding stage group by group to generate a stage signal triggering frequency value; the stage trend identifying sub-module extracts a frequency change sequence corresponding to the path number in the differentiation stage according to the stage signal triggering frequency value and the path number arrangement, judges whether any path number has an increasing frequency value in the continuous operation stage, screens a path number set meeting trend requirements, and generates a path frequency increasing trend sequence; The critical path extraction submodule calls a path number set in the path frequency increasing trend sequence, extracts an operation stage label corresponding to the path number, sorts the mapping structure of the path number and the stage label according to the stage number sequence, and generates a critical path priority response table.
7. The intelligent diagnosis system of claim 1, wherein the reverse position estimation determination module comprises: The space sequence extraction submodule calls the priority path number in the key path priority response table, retrieves the signal node index associated with the path number, acquires the corresponding space coordinate position according to the node number in the structure diagram, constructs a signal node space arrangement array according to the path number sequence, and generates a path signal space sequence list; the connectivity investigation submodule extracts the connection relation of two adjacent signal nodes in the structure diagram based on the path signal space sequence list, judges a connection path, records a node pair set which is not connected, marks the node pair set as a path interruption node and generates a signal jump node index group; The source point tracking and identifying submodule calls the signal jumping node index group, performs reverse path index on a previous signal node of the jumping node in the structure diagram, judges whether the previous node is a path terminal or a non-connected node, and if the judgment result is a path breakpoint, records to generate a fault source point presumption result.

Description

Intelligent automobile fault diagnosis system Technical Field The invention relates to the technical field of fault diagnosis, in particular to an intelligent automobile fault diagnosis system. Background The technical field of fault diagnosis relates to the identification, positioning and diagnosis of abnormal states of various devices and systems in the operation process, and core matters comprise the acquisition, identification and judgment of fault information, the detection and classification of faults are realized according to the operation parameters or state variables of the devices, and the analysis of the development trend of the faults is realized. The technical field is widely applied to a plurality of industries such as manufacturing, electric power, transportation and the like, particularly in the transportation field, the fault diagnosis technology is used as an important means for guaranteeing equipment safety and improving system reliability and maintenance efficiency, is gradually changed from manual experience judgment to intelligent analysis, covers key contents such as sensor data acquisition, fault pattern recognition, feature extraction, system diagnosis rule construction and the like, and forms a complete flow for realizing fault detection and positioning based on an electric digital data processing system. The traditional intelligent automobile fault diagnosis system is characterized in that a pointer is used for indicating fault symptoms of key components such as an automobile electric control system, an engine, a transmission and the like in the using process, vehicle running data are collected through arranging vehicle-mounted sensors, judgment of abnormal states and preliminary positioning of fault positions are achieved by using a threshold comparison mode based on logic judgment or experience rules, a controller local area network taking an engine control unit as a core is used for receiving running parameter information transmitted by the vehicle-mounted sensors, the vehicle states are compared item by item through set fault codes and corresponding relation tables, and specific fault types existing in the system are identified. In the prior art, when fault judgment is carried out, a preset threshold value and a fixed logic rule are relied on, recognition and analysis of a correlation path among multiple signals are lacked, especially when errors exist in a sensor sampling time sequence or delay exists in data transmission, dislocation exists between fault information and an actual source point, accuracy of a positioning result is affected, the traditional mode cannot be combined with operation trend, node distribution and frequency change to carry out comprehensive judgment, so that response to phenomena such as path jump and signal evolution in a complex scene is delayed, source point missed judgment in certain cross wiring or nonlinear conduction structures is easy to cause, and especially when the system structure is complex or multiple node overlapping areas, misjudgment or positioning deviation is easy to occur, and maintenance decision and subsequent fault elimination efficiency are affected. Disclosure of Invention The invention aims to solve the defects in the prior art and provides an intelligent automobile fault diagnosis system. In order to achieve the above purpose, the present invention adopts the following technical scheme, and an intelligent diagnosis system for automobile faults comprises: The acquisition sequence verification module acquires a fault code time tag and a sampling point number sequence uploaded by a sensing node in an automobile, judges whether signal items with staggered sequence and abrupt strength change exist between sampling points according to the time tag sequence, and generates an acquisition sequence abnormal signal set by dividing coincidence conditions into a staggered signal set; The strength anomaly grouping module calls an automobile wiring structure diagram based on the acquisition sequence anomaly signal set, extracts the connection path sequence of signal nodes in the structure diagram, carries out connection continuity judgment with the previous signal path, and generates a path jump association fault group; The trend path reconstruction module screens trend change sequences with consistent directions in continuous segments according to the time sequence of the signals in the path jump association fault group, classifies the trend change sequences as the same signal evolution path number, and performs mapping binding with the running state of the automobile to generate a trend linkage path identification set; And the frequency priority weighting module calls the path numbers in the trend linkage path identification set, searches the triggering condition in the automobile running state recording section, counts the repeated triggering frequency of signals corresponding to each group of path numbers according to the running stage