CN116859236A - Vehicle motor fault diagnosis method and device and vehicle networking platform - Google Patents
Vehicle motor fault diagnosis method and device and vehicle networking platform Download PDFInfo
- Publication number
- CN116859236A CN116859236A CN202310786954.9A CN202310786954A CN116859236A CN 116859236 A CN116859236 A CN 116859236A CN 202310786954 A CN202310786954 A CN 202310786954A CN 116859236 A CN116859236 A CN 116859236A
- Authority
- CN
- China
- Prior art keywords
- frequency
- fault
- data
- low
- motor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000003745 diagnosis Methods 0.000 title claims abstract description 160
- 238000000034 method Methods 0.000 title claims abstract description 45
- 230000006855 networking Effects 0.000 title claims abstract description 24
- 239000012634 fragment Substances 0.000 claims abstract description 118
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 74
- 230000002159 abnormal effect Effects 0.000 claims abstract description 27
- 238000012216 screening Methods 0.000 claims abstract description 21
- 238000007619 statistical method Methods 0.000 claims description 24
- 230000008859 change Effects 0.000 claims description 20
- 238000010845 search algorithm Methods 0.000 claims description 13
- 230000002776 aggregation Effects 0.000 claims description 11
- 238000004220 aggregation Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 9
- 230000000007 visual effect Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 description 16
- 238000003860 storage Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 9
- 238000011144 upstream manufacturing Methods 0.000 description 7
- 238000004458 analytical method Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 230000006978 adaptation Effects 0.000 description 4
- 238000012512 characterization method Methods 0.000 description 4
- 238000009826 distribution Methods 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 230000015572 biosynthetic process Effects 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000005856 abnormality Effects 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000003111 delayed effect Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- 238000010923 batch production Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000004171 remote diagnosis Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/251—Fusion techniques of input or preprocessed data
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/005—Testing of electric installations on transport means
- G01R31/006—Testing of electric installations on transport means on road vehicles, e.g. automobiles or trucks
- G01R31/007—Testing of electric installations on transport means on road vehicles, e.g. automobiles or trucks using microprocessors or computers
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/34—Testing dynamo-electric machines
- G01R31/343—Testing dynamo-electric machines in operation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Hardware Design (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The embodiment of the disclosure provides a motor fault diagnosis method and device for a vehicle and a vehicle networking platform, wherein the motor fault diagnosis method for the vehicle comprises the following steps: acquiring vehicle networking buried point signal data of a motor, wherein the vehicle networking buried point signal data comprises low-frequency data and high-frequency data; diagnosing the low-frequency data based on a first-stage dynamic sliding window algorithm, and marking abnormal motor information in the low-frequency data by using a first-stage tag to obtain a low-frequency fault information fragment; expanding a diagnosis range of high-frequency data according to the first-stage tag, diagnosing the high-frequency data corresponding to the diagnosis range based on a second-stage dynamic sliding window, and screening high-frequency fault information fragments from the high-frequency data; and outputting a corresponding motor fault diagnosis result based on the low-frequency fault information fragment and the high-frequency fault information fragment.
Description
Technical Field
The disclosure relates to the technical field of vehicle motor fault diagnosis, in particular to a vehicle motor fault diagnosis method and device and a vehicle networking platform.
Background
The signals usually returned by the vehicle-end motor of the pure electric vehicle comprise low-frequency buried point signal data and high-frequency buried point signal data so as to meet the related service requirements of data transmission, storage and the like of the vehicle-end motor of the national and enterprise monitoring platform. The low-frequency buried point signal data are second-level signal data with less total quantity and small volume, and the high-frequency buried point signal data are millisecond-level signal data with more total quantity, large volume and rich details.
The motor fault diagnosis related method based on the low-frequency buried point signal data has the problems that the details of the fault moment signal are not complete, the specific reasons of motor faults are difficult to locate, the fault time range is inaccurate, the fault level representation is not obvious and the like; the motor fault diagnosis related method based on the high-frequency buried point signal data has the problems of low data pulling speed, large statistical analysis scale, long time consumption, high fault analysis error rate and the like caused by the limitation of a scheduling algorithm to the total data amount and fault rechecking positioning.
In summary, when the high-frequency buried point signal data and the low-frequency buried point signal data are used for diagnosing the motor faults in the prior art, the problems of insufficient utilization of the high-frequency buried point signal data and the low-frequency buried point signal data, low value density, poor value contribution degree and the like exist, and the problems of low fault diagnosis accuracy, poor diagnosis efficiency and the like are caused.
Disclosure of Invention
The embodiment of the disclosure aims to provide a vehicle motor fault diagnosis method and device and a vehicle networking platform, which can solve the technical problems of insufficient utilization of high-frequency buried point signal data and low-frequency buried point signal data, low value density, poor value contribution and the like in the prior art when motor fault diagnosis is carried out by utilizing the high-frequency buried point signal data and the low-frequency buried point signal data.
The embodiment of the disclosure adopts the following technical scheme:
a motor fault diagnosis method for a vehicle, comprising:
acquiring vehicle networking buried point signal data of a motor, wherein the vehicle networking buried point signal data comprises low-frequency data and high-frequency data;
diagnosing the low-frequency data based on a first-stage dynamic sliding window algorithm, and marking abnormal motor information in the low-frequency data by using a first-stage tag to obtain a low-frequency fault information fragment;
expanding a diagnosis range of high-frequency data according to the first-stage tag, diagnosing the high-frequency data corresponding to the diagnosis range based on a second-stage dynamic sliding window, and screening high-frequency fault information fragments from the high-frequency data;
and outputting a corresponding motor fault diagnosis result based on the low-frequency fault information fragment and the high-frequency fault information fragment.
In some embodiments, diagnosing the low frequency data based on a first-stage dynamic sliding window algorithm, and marking motor abnormality information in the low frequency data with a first-stage tag to obtain a low frequency fault information fragment, including:
based on a preset data signal name, a diagnosis strategy and the first-stage dynamic sliding window algorithm, carrying out data feature recognition on the low-frequency data, and recognizing a low-frequency signal quantity state change point and a low-frequency signal abnormal point;
and marking the low-frequency signal quantity state change point and the low-frequency signal abnormal point by the first-stage tag.
In some embodiments, expanding a diagnosis range of high-frequency data according to the first-stage tag, diagnosing high-frequency data corresponding to the diagnosis range based on a second-stage dynamic sliding window, and screening high-frequency fault information fragments from the high-frequency data, including:
screening corresponding motor high-frequency data fragments from the high-frequency data according to the first-stage tag, and performing data expansion on the motor high-frequency data fragments according to a time sequence to obtain priori high-frequency key information fragments;
and diagnosing the priori high-frequency key information fragments based on a second-stage dynamic sliding window algorithm and a binary search algorithm, and screening the high-frequency fault information fragments from the priori high-frequency key information fragments.
In some embodiments, diagnosing the a priori high frequency critical information segments based on a second level dynamic sliding window algorithm and a binary search algorithm, screening high frequency fault information segments from the a priori high frequency critical information segments, comprising:
based on a preset data signal name, a second diagnosis strategy, the second-stage dynamic sliding window algorithm and a binary search algorithm, carrying out data feature recognition on the priori high-frequency key information fragment, and recognizing state change points and abnormal points of a second semaphore to obtain a preliminary high-frequency fault information fragment;
and carrying out statistical analysis on the fault data of the preliminary high-frequency fault information fragments, and marking the statistical analysis characteristics of the fault data by using a second-stage label.
In some embodiments, the method further comprises:
diagnosing the first-level fault reason of the high-frequency fault information fragment according to the typical characteristics accumulated in the electric drive working condition library;
and selecting a corresponding rechecking algorithm from a preset algorithm library based on the statistical analysis characteristics of the second-level tag marks and the first-level fault reasons, and rechecking the faults of the high-frequency fault information fragments.
In some embodiments, fault rechecking the high frequency fault information segment includes:
Fitting and forming an ideal signal quantity curve according to the forming parameters of the primary fault reasons and the rechecking algorithm;
and comparing the ideal signal quantity curve with the actual signal quantity curve of the high-frequency fault information fragment, and checking the fault diagnosis accuracy.
In some embodiments, outputting a corresponding motor fault diagnosis result based on the low frequency fault information piece and the high frequency fault information piece includes:
and carrying out full-flow characteristic information aggregation on the low-frequency fault information fragments and the high-frequency fault information fragments to obtain a motor fault diagnosis result.
In some embodiments, the method further comprises:
and storing the motor fault diagnosis result and/or outputting the motor fault diagnosis result in a visual mode.
The embodiment of the disclosure also provides a motor fault diagnosis device for a vehicle, comprising:
the acquisition module is configured to acquire the vehicle networking buried point signal data of the motor, wherein the vehicle networking buried point signal data comprises low-frequency data and high-frequency data;
the first diagnosis module is configured to diagnose the low-frequency data based on a first-stage dynamic sliding window algorithm, and mark abnormal motor information in the low-frequency data by a first-stage label to obtain a low-frequency fault information fragment;
The second diagnosis module is configured to expand the diagnosis range of the high-frequency data according to the first-stage tag, diagnose the high-frequency data corresponding to the diagnosis range based on a second-stage dynamic sliding window, and screen high-frequency fault information fragments from the high-frequency data;
and the output module is configured to output a corresponding motor fault diagnosis result based on the low-frequency fault information fragment and the high-frequency fault information fragment.
The embodiment of the disclosure also provides a vehicle networking platform, which at least comprises a memory and a processor, wherein the memory stores a computer program, and the processor realizes the steps of the method when executing the computer program on the memory.
According to the vehicle motor fault diagnosis method, device and vehicle networking platform, after vehicle networking buried point signal data of a motor are obtained, the low frequency data are diagnosed based on a first-stage dynamic sliding window algorithm, motor abnormal information in the low frequency data is marked by a first-stage label to obtain a low frequency fault information fragment, then the diagnosis range of high frequency data is expanded according to the first-stage label, the high frequency data corresponding to the diagnosis range are diagnosed based on a second-stage dynamic sliding window, the high frequency fault information fragment is screened from the high frequency data, and then the corresponding motor fault diagnosis result is output based on the low frequency fault information fragment and the high frequency fault information fragment, a fault sign can be firstly screened out by using the low frequency data, then the fault sign is used for obtaining detailed motor fault feature details from the high frequency data, and then the low frequency fault information fragment and the high frequency fault information fragment are fused for diagnosis, so that a final motor fault diagnosis result is obtained, the characteristics of the low frequency data are convenient to feature and the high frequency data, the characteristics of the high frequency fault information fragment and the high frequency fault information fragment are combined with the characteristics of the high frequency data, the high-speed statistics data and the high-speed fault diagnosis data are analyzed, the vehicle fault diagnosis is improved, the vehicle fault diagnosis is accurate, and the vehicle fault diagnosis is high-scale is analyzed, and the vehicle fault diagnosis is high-efficient, and the vehicle fault diagnosis is high-level is analyzed, and the vehicle fault diagnosis is has high quality and high quality fault diagnosis accuracy. In addition, the motor fault diagnosis method for the vehicle can provide guarantee and guidance for motor fault problem positioning, after-sales maintenance and the like, and enrich the value of big data of the motor for the vehicle and the availability of data dimension.
Drawings
In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments described in the present disclosure, and other drawings may be obtained according to these drawings without inventive effort to a person of ordinary skill in the art.
FIG. 1 is a flow chart of a motor fault diagnosis method for a vehicle in accordance with an embodiment of the present disclosure;
fig. 2 is a diagnostic architecture diagram of a vehicle networking platform of the vehicle motor fault diagnosis method according to an embodiment of the present disclosure;
FIG. 3 is another flow chart of a motor fault diagnosis method for a vehicle in accordance with an embodiment of the present disclosure;
FIG. 4 is a schematic diagram of an algorithm library of a central operation scheduling server cluster according to an embodiment of the disclosure;
FIG. 5 is a schematic diagram of a first level fault marking unit according to an embodiment of the present disclosure;
FIG. 6 is a schematic diagram of a data segment processing unit according to an embodiment of the disclosure;
FIG. 7 is a schematic diagram of a second level fault marking unit according to an embodiment of the present disclosure;
fig. 8 is a schematic structural diagram of an algorithm adaptation unit according to an embodiment of the disclosure;
FIG. 9 is a schematic diagram of a diagnosis result processing unit according to an embodiment of the present disclosure;
fig. 10 is a schematic structural diagram of a motor fault diagnosis device for a vehicle according to an embodiment of the present disclosure.
Detailed Description
Various aspects and features of the disclosure are described herein with reference to the drawings.
It should be understood that various modifications may be made to the embodiments of the application herein. Therefore, the above description should not be taken as limiting, but merely as exemplification of the embodiments. Other modifications within the scope and spirit of this disclosure will occur to persons of ordinary skill in the art.
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and, together with a general description of the disclosure given above and the detailed description of the embodiments given below, serve to explain the principles of the disclosure.
These and other characteristics of the present disclosure will become apparent from the following description of a preferred form of embodiment, given as a non-limiting example, with reference to the accompanying drawings.
It should also be understood that, although the present disclosure has been described with reference to some specific examples, a person skilled in the art will certainly be able to achieve many other equivalent forms of the present disclosure, having the characteristics as set forth in the claims and hence all coming within the field of protection defined thereby.
The above and other aspects, features and advantages of the present disclosure will become more apparent in light of the following detailed description when taken in conjunction with the accompanying drawings.
Specific embodiments of the present disclosure will be described hereinafter with reference to the accompanying drawings; however, it is to be understood that the disclosed embodiments are merely examples of the disclosure, which may be embodied in various forms. Well-known and/or repeated functions and constructions are not described in detail to avoid obscuring the disclosure in unnecessary or unnecessary detail. Therefore, specific structural and functional details disclosed herein are not intended to be limiting, but merely as a basis for the claims and as a representative basis for teaching one skilled in the art to variously employ the present disclosure in virtually any appropriately detailed structure.
The specification may use the word "in one embodiment," "in another embodiment," "in yet another embodiment," or "in other embodiments," which may each refer to one or more of the same or different embodiments in accordance with the disclosure.
Example 1
Fig. 1 shows a flowchart of a motor failure diagnosis method for a vehicle according to a first embodiment of the present disclosure. As shown in fig. 1, a first embodiment of the present disclosure provides a motor fault diagnosis method for a vehicle, including:
S101: and acquiring the vehicle networking buried point signal data of the motor, wherein the vehicle networking buried point signal data comprises low-frequency data and high-frequency data.
Along with the development of the internet of vehicles, a user can generate a large amount of data in the driving process, and the internet of vehicles embedded data is data of the real use condition of the vehicle by the related vehicle end user, is a specific record generated when the user uses the internet of vehicles function of the vehicle, and is reported back to a cloud server (internet of vehicles platform) of the internet of vehicles through a mobile phone network or an internet of vehicles.
The internet of vehicles platform can be the cloud platform of the vehicle enterprise, and also can be the cloud platform of the national level. According to the state flag of GB32960.3, new energy vehicles need to report vehicle data to the state and enterprise monitoring platforms, and these data generally include: (1) conventional vehicle state: conventional vehicle state information including speed, mileage, position, vehicle state (parking or driving state, etc.), SOC, etc.; (2) three electrical component operation related status information: such as voltage, current, motor speed, motor torque, charge related information, etc.; (3) Alarm information for a preset controller diagnosis of a vehicle component: and the early warning grades of different components, fault code related information generated by related vehicle end diagnoses and the like.
The high-frequency data and the low-frequency data refer to the vehicle networking buried point signal data with different signal frequencies, different signal magnitudes and different return times when the electric vehicle runs. In a specific Internet of vehicles architecture, high-frequency data and low-frequency data are defined according to the acquisition frequency of a preset buried point signal, and the low-frequency data are reported back in real time generally in a second level; the high-frequency data takes millisecond level as a unit, has large capacity and high reporting flow consumption, and generally adopts a data packet form to go off-line or delay back.
After the vehicle is connected to the network of the cloud server, the cloud server can receive the low-frequency data (low-frequency buried point signal data) of the motor reported back by the vehicle in real time, and can also receive the high-frequency data (high-frequency buried point signal data) of the motor offline or in a delayed manner, so that the vehicle motor fault remote diagnosis is carried out according to the obtained high-frequency data and low-frequency data. In this embodiment, the cloud server does not synchronize the acquisition of the high-frequency data and the low-frequency data, the high-frequency data may be historical high-frequency data acquired in advance, and the historical high-frequency data is stored in the high-frequency database in advance as a high-frequency data source, and when needed, the required high-frequency data is called. The low frequency data may be stored in real time in a low frequency database as a low frequency data source.
In addition, in the embodiment, the internet of vehicles platform receives internet of vehicles buried point signal data of various motors of the whole vehicle user.
In this embodiment, a dedicated low frequency data server cluster and high frequency data server cluster may be provided to store low frequency data and high frequency data, respectively.
S102: and diagnosing the low-frequency data based on a first-stage dynamic sliding window algorithm, and marking the motor abnormality information in the low-frequency data by using a first-stage tag to obtain a low-frequency fault information fragment.
The first-stage dynamic sliding window algorithm is a sliding window algorithm combined with self-adaptive thread acceleration, and the number of threads of the first-stage dynamic sliding window is determined according to the number of driving cycles and the duration length of a single driving cycle. Specifically, for a single driving cycle of 30min or less, the algorithm matches it with a single sliding window; for the driving cycle of > 30min, the algorithm continuously divides the driving cycle time of a single time to meet the condition, and finally, a sliding window is matched for the divided single driving cycle to perform state search and feature statistics.
In the step, after a certain number of low frequency data are obtained, state change can be carried out on the motor fault position buried point signal quantity of the full-quantity user vehicle based on a first-level dynamic sliding window algorithm, vehicle information with abnormal motor state is obtained through a first-level tag, and a low frequency fault information fragment is formed according to information such as a fault time stamp of the vehicle.
The low-frequency fault information fragment comprises at least one of information such as a vehicle unique identifier, a fault starting time stamp, a fault ending time stamp, a vehicle speed, oil consumption, mileage and the like. The first-level tag may be a total tag of the low-frequency fault information fragment, or may be a tag corresponding to each information in the low-frequency fault information fragment, for example, the first-level tag may include a vehicle unique identification tag, a fault start time stamp tag, and the like.
S103: and expanding the diagnosis range of the high-frequency data according to the first-stage tag, diagnosing the high-frequency data corresponding to the diagnosis range based on a second-stage dynamic sliding window, and screening high-frequency fault information fragments from the high-frequency data.
After the low-frequency fault information fragments are obtained, the high-frequency data fragments of the motor can be directionally and extendedly searched in the high-frequency data source according to the first-stage labels corresponding to the low-frequency fault information fragments, the diagnosis range is extended, then, the high-frequency data in the diagnosis range is diagnosed based on the second-stage dynamic sliding window, and the high-frequency fault information fragments corresponding to the first-stage labels are screened. The high-frequency fault information fragment contains detailed fault characteristic information such as fault type, fault code and the like.
For example, a motor high frequency data segment of a target vehicle corresponding to a vehicle unique identifier can be located in a high frequency data source with two first-stage tags, namely a vehicle unique identifier (e.g. a frame number, a license plate number, etc.), a multi-fault start time stamp, and the high frequency fault information segment can be screened from the motor high frequency data segment based on a second-stage dynamic sliding window.
S104: and outputting a corresponding motor fault diagnosis result based on the low-frequency fault information fragment and the high-frequency fault information fragment.
After the low-frequency fault information fragment and the high-frequency fault information fragment are obtained, the low-frequency fault information fragment and the high-frequency fault information fragment can be fused and diagnosed, the motor fault is comprehensively judged, and then a corresponding motor fault diagnosis result is output.
According to the vehicle motor fault diagnosis method provided by the embodiment of the disclosure, after the vehicle networking buried point signal data of the motor is obtained, the low-frequency data is diagnosed based on a first-stage dynamic sliding window algorithm, the motor abnormal information in the low-frequency data is marked by a first-stage label to obtain a low-frequency fault information fragment, then the diagnosis range of the high-frequency data is expanded according to the first-stage label, the high-frequency data corresponding to the diagnosis range is diagnosed based on a second-stage dynamic sliding window, the high-frequency fault information fragment is screened from the high-frequency data, the corresponding motor fault diagnosis result is output based on the low-frequency fault information fragment and the high-frequency fault information fragment, the fault mark can be firstly and rapidly screened by using the low-frequency data, then detailed motor fault characteristic details are obtained from the high-frequency data by using the fault mark, the low-frequency fault information fragment and the high-frequency fault information fragment are fused to be diagnosed, the final motor fault diagnosis result is obtained, the characteristics of the low-frequency data that are convenient for characteristic statistics and rapid analysis and the characteristics of the high-frequency data signal characterization and the motor fault specification are obviously combined, the fault diagnosis is high, the vehicle fault diagnosis is effective, and the vehicle fault diagnosis is carried out is large in a large-scale, and the user is high in the fault quantity is effectively analyzed, and the fault quantity is high, and the vehicle fault is effectively is analyzed. In addition, the motor fault diagnosis method for the vehicle can provide guarantee and guidance for motor fault problem positioning, after-sales maintenance and the like, and enrich the value of big data of the motor for the vehicle and the availability of data dimension.
Optionally, as shown in fig. 2, in this embodiment, the diagnostic architecture of the internet of vehicles platform includes a central operation scheduling server cluster, so as to execute the steps S101 to S104, where the central operation scheduling server cluster includes an algorithm library, and the algorithm library stores the execution logic algorithm of the steps. The central operation scheduling server cluster can call the low-frequency data and the high-frequency data from the low-frequency data server cluster and the high-frequency data server cluster to perform fault diagnosis.
In some embodiments, in step S102, the low frequency data is diagnosed based on a first-stage dynamic sliding window algorithm, and motor anomaly information in the low frequency data is marked with a first-stage tag, so as to obtain a low frequency fault information segment, which includes:
s1021: based on a preset data signal name, a first diagnosis strategy and the first-stage dynamic sliding window algorithm, carrying out data feature identification on the low-frequency data, and identifying a state change point and an abnormal point of a first semaphore;
s1022: and marking the state change point and the abnormal point of the first semaphore by the first-stage tag.
The vehicle networking platform can perform feature recognition on motor fault position buried point semaphores in low-frequency data based on a first-stage dynamic sliding window algorithm, perform diagnosis and screening on signal state change based on a preset first diagnosis strategy, recognize state change points of the semaphores and low-frequency signal abnormal points, mark the state change points and the low-frequency signal abnormal points by a first-stage label, and output relevant fault data containing the state change points and the abnormal points of the first-stage label as low-frequency fault information fragments.
Optionally, as shown in fig. 4 and fig. 5, the algorithm library of the central operation scheduling server cluster includes a first-stage fault marking unit 10, where the first-stage fault marking unit 10 specifically includes a first feature identification module 101 and a first segment output module 102, the first feature identification module 101 performs data feature identification on the obtained low-frequency data according to a preset data signal name, a first diagnosis policy and a first-stage dynamic window, identifies a state change point and an abnormal point of a first signal quantity from the data feature identification, and marks a first-stage label; the first segment output module 102 is configured to batch-process and format the low frequency data marked with the first-stage tag according to a preset format, and output a normalized low frequency fault information segment, so that the low frequency fault information segment can be conveniently extracted and analyzed by other subsequent units or modules, and the motor data segment can be conveniently circulated between different modules and between units.
In some embodiments, after the low frequency data is acquired and before step S1021 is performed, the method further includes preprocessing the low frequency data, specifically including:
s1023: classifying the low frequency data; and/or
S1024: the low frequency data is ordered based on a time series.
Specifically, as shown in fig. 5, the first-stage fault marking unit 10 further includes a first classifier module 103 and a first sequencer module 104, where the first classifier module 103 is configured to classify different low-frequency data according to the unique identifier of the vehicle, that is, the low-frequency data may be classified by the first classifier module 103, so as to avoid data confusion of different motors.
The embedded point signal data of the internet of vehicles is transmitted back online or offline, the high frequency data and the low frequency data are stored in different server clusters, and because the network is interrupted or delayed, the data are not strictly ordered according to time sequence, therefore, in the embodiment, the low frequency data are checked and adjusted to the position sequence of the data item according to the time stamp sequence by the first sequencer module 104 before the step S1021 is executed, so that a data source with a standard time sequence is formed, and the subsequent screening by using the first-stage dynamic window is facilitated.
In some embodiments, in step S103, extending a diagnosis range of the high-frequency data according to the first-stage tag, diagnosing the high-frequency data corresponding to the diagnosis range based on a second-stage dynamic sliding window, and screening a high-frequency fault information fragment from the high-frequency data, including:
S1031: screening corresponding motor high-frequency data fragments from the high-frequency data according to the first-stage tag, and performing data expansion on the motor high-frequency data fragments according to a time sequence to obtain priori high-frequency key information fragments;
s1032: and diagnosing the priori high-frequency key information fragments based on a second-stage dynamic sliding window algorithm and a binary search algorithm, and screening the high-frequency fault information fragments from the priori high-frequency key information fragments.
In this embodiment, after locating and searching the corresponding first high frequency data (high frequency fault data) according to the first-level tag, through step S1031, the diagnosis range of the high frequency data is extended by searching the second high frequency data within a certain time range before and after the time point of the first high frequency data, and then the high frequency data after extending the diagnosis range is preprocessed to form structured priori high frequency key information segments arranged according to the time sequence, where the priori high frequency key information segments include all signal amounts of the high frequency buried points of the motor.
Optionally, as shown in fig. 4, the algorithm library of the central operation scheduling server cluster includes a data segment processing unit 20, where the data segment processing unit 20 firstly extracts a unique vehicle identifier, a fault start timestamp and the like according to a specific first-stage tag in the low-frequency key information segments output by the first-stage fault marking unit 10 to form a positioning identifier, and according to the positioning identifier, searches for a motor high-frequency data segment for positioning a target vehicle corresponding to the unique vehicle identifier in the high-frequency data source, and processes the high-frequency fault data searched in a front-back expansion manner to form a priori high-frequency information segment.
As shown in fig. 6, the data fragment processing unit 20 may specifically include a tag identification module 201, a fragment positioning module 202, and a second fragment output module 203. The tag identification module 201 is configured to extract a tag existing in the low frequency data source according to a preset extraction policy, and output a positioning identifier such as a unique vehicle identifier through selective operations such as certain combination transformation, etc., so as to be used by other modules or units in the downstream. The segment locating module 202 is used for locating, searching and cutting the high frequency data source according to the locating identification to form a new data source. The second segment output module 203 is configured to output the a priori high frequency information segment in a preset format, so as to facilitate use by other modules or units.
For the priori high-frequency key information segments of all motors selected by directional screening, in step S1032, data fine-combing and statistical analysis can be performed on all priori high-frequency key segments by using a second-stage dynamic window and a binary search algorithm, for example, statistical analysis can be performed on the states of motor temperature, voltage, current and the like and the buried point signal zone bit data of motor fault codes and the like, the duration of faults and each duration are output, the first-stage fault cause is identified, the fault details are filled, high-frequency fault information segment output is formed, and in the embodiment, the high-frequency fault information segment containing all state information of the motors within each fault duration can also be output.
In some embodiments, in step S1032, diagnosing the a priori high frequency critical information segments based on a second level dynamic sliding window algorithm and a binary search algorithm, and screening high frequency fault information segments from the a priori high frequency critical information segments includes:
s201: based on a preset data signal name, a second diagnosis strategy, the second-stage dynamic sliding window algorithm and a binary search algorithm, carrying out data feature recognition on the priori high-frequency key information fragment, and recognizing state change points and abnormal points of a second semaphore to obtain a preliminary high-frequency fault information fragment;
s202: and carrying out statistical analysis on the fault data of the preliminary high-frequency fault information fragments, and marking the statistical analysis characteristics of the fault data by using a second-stage label.
Similar to the diagnosis of the low frequency data source, in this embodiment, the data feature recognition may be performed on the priori high frequency key information segment based on the preset data signal name, the second diagnosis policy, the second level dynamic sliding window algorithm, and the binary search algorithm, to identify the state change point and the abnormal point of the second signal quantity, and tag the state change point and the abnormal point with a tag (for example, tag a fault tag such as a fault code DTC); and then, acquiring relevant fault data of the state change point and the abnormal point according to the label, carrying out statistical analysis, finely calibrating the duration of the fault and the duration of each fault by using the second-stage label, identifying the first-stage fault cause, filling fault details, and highlighting the characterization state of the fault.
Optionally, as shown in fig. 4 and fig. 7, the algorithm library of the central operation scheduling server cluster includes a second-level fault marking unit 30, where the second-level fault marking unit 30 includes a second feature identification module 301, a statistical analysis module 302, a feature reconstruction module 303, a feature aggregation module 304, and a third segment output module 305. The second feature recognition module 301 is configured to perform the above step S201, recognize a state change point and an abnormal point of the second signal, and mark a tag, to obtain a preliminary high-frequency fault information fragment. The statistical analysis module 302 is configured to perform statistical analysis on the high-frequency signal feature trend, the maximum value, the average value, and other statistical quantities within the high-frequency data range after feature recognition, and identify a primary failure cause. The primary fault cause refers to a range class to which the fault is approximately included, and in this embodiment, various faults may be reasonably classified in advance, and the fault cause may be classified into multiple stages. When the first-stage fault cause is identified, the fault characteristics of the state change point and the abnormal point of the second signal quantity can be compared with the typical characteristics accumulated in the electric drive working condition library, and the first-stage fault cause is identified.
As shown in fig. 3, in this embodiment, an independent electric driving knowledge experience library server cluster may be configured to store an electric driving condition library including information such as a motor status segment.
The feature reconstruction module 303 is configured to reconstruct fault features such as a feature of a fault characterization state according to the statistical analysis feature, for example, extract key features, delete redundant features, and perform permutation and combination on different features. The feature aggregation module 304 is configured to integrate the intermediate conclusion of the fault analysis and diagnosis given by the upstream module according to a preset diagnosis policy and a fault segment start time stamp, screen out possible derivative faults, obtain a high-frequency diagnosis result in the full period of each motor, and output the result through the third segment output module 305.
Similar to the processing of low frequency data described above, in this embodiment, the second level fault marking unit 30 may further include a second classifier module 306 and a second sequencer module 307 to pre-process the high frequency data source prior to feature recognition.
The second-stage dynamic sliding window algorithm is a sliding window algorithm combined with self-adaptive thread acceleration, the second-stage dynamic window is determined based on the duration time length of fault information fragments in multiple priori high-frequency key information fragments identified by the first-stage tag, specifically, for the fault information fragments smaller than 5min, the algorithm matches a single sliding window with the second-stage dynamic window algorithm, for the fault information fragments larger than 5min, the algorithm continuously divides the single sliding window to enable the single sliding window to meet the conditions, and finally, each divided fault information fragment is matched with one sliding window to perform state search and feature statistics.
The fault code output of the second stage dynamic window satisfies the following rule: firstly, sorting the statistical fault codes of all windows according to time stamps, then normalizing standard deviation and calculating the out-of-value kurtosis K of data distribution,where μ is the mean and σ is the standard deviation).
When K is greater than 0, the data distribution is gathered and the peak exists, and the marked X with the largest number of occurrences of the value is obtained m ，X m The final fault code is obtained; otherwise, when K is less than or equal to 0, taking the value X appearing for the first time st As a final fault code.
X m Computing complianceI.e. the maximum value of the sum of the frequency of the statistics of the code values of each window.
X GZ The calculation formula of (2) is summarized as follows:
the feature reconstruction module 303 is configured to reconstruct fault features such as a feature of a fault characterization state according to the statistical analysis feature, for example, extract key features, delete redundant features, and perform permutation and combination on different features. The feature aggregation module 304 is configured to integrate the intermediate conclusions of the fault analysis and diagnosis provided by the upstream module according to a preset diagnosis strategy and a start time stamp of a fault segment, and screen out possible derivative faults to obtain a high-frequency diagnosis result in the full period of each motor.
In some embodiments, the method further comprises:
S105: diagnosing the first-level fault reason of the high-frequency fault information fragment according to the typical characteristics accumulated in the electric drive working condition library;
s106: and selecting a corresponding rechecking algorithm from a preset algorithm library based on the statistical analysis characteristics of the second-level tag marks and the first-level fault reasons, and rechecking the faults of the high-frequency fault information fragments.
In this embodiment, after the primary diagnosis result of the primary failure cause is identified by using the high-frequency failure information segment, the failure cause of the high-frequency failure information segment is rechecked by selecting a corresponding rechecking algorithm in a preset algorithm library in combination with the statistical analysis feature of the fine calibration of the upstream second-stage failure marking unit 30, and the failure detail of the high-frequency failure information segment is classified and positioned by the machine learning algorithm of the corresponding label, so as to ensure the accuracy of failure diagnosis. The first-stage fault cause and the statistical analysis feature marked by the second-stage label are both intermediate results of fault diagnosis.
In some embodiments, in step S106, performing fault review on the high-frequency fault information segment includes:
s301: fitting and forming an ideal signal quantity curve according to the forming parameters of the primary fault reasons and the rechecking algorithm;
S302: and comparing the ideal signal quantity curve with the actual signal quantity curve of the high-frequency fault information fragment, and checking the fault diagnosis accuracy.
In this embodiment, according to the first-level fault cause and the statistical analysis feature of the fault, the forming parameters and the algorithm tree formed by the first-level fault cause may be selected in the algorithm library, a virtual ideal signal amount curve may be formed by fitting, and then the actual signal amount curve of the high-frequency fault information segment may be compared with the virtual ideal signal amount curve to obtain a comparison result, and the comparison result may be identified in the form of a label, for example, an outlier range of the actual signal amount curve may be identified. The formation parameters formed by the primary fault cause can be the formation parameters stored in an algorithm library, and can also be recommended parameters determined according to the statistical analysis characteristics.
As shown in fig. 4, the algorithm library of the central operation scheduling server cluster may include an algorithm adaptation unit 40 and a fault cause rechecking unit 50 to screen a rechecking algorithm and perform fault rechecking on the fault cause diagnosed based on the high frequency fault information fragment according to the rechecking algorithm.
As shown in fig. 8, the algorithm adaptation unit 40 specifically includes a first tag identification module 401, a parameter fitting module 402, a feature discrimination module 403, and a queue scheduling module 404.
The first tag identification module 401 is configured to identify a second level tag in the high frequency fault information piece. The parameter fitting module 402 mainly provides recommended parameters for adapting the current fault segment characteristics and volumes and/or formation parameters for adapting the primary fault cause, and provides a rechecking algorithm for adapting the primary fault cause. The feature discrimination module 403 forms a virtual ideal semaphore curve for comparison with the actual semaphore curve within the current fault segment by fitting parameters and algorithms provided upstream, forming a comparison result tag. The queue scheduling module 404 is used for controlling the machine kernel resources, avoiding running blockage and preventing each resource from being robbed and downtime.
The fault cause rechecking unit 50 specifically includes a feature comparing module 501 and a fourth segment output module 502, where the feature comparing module 501 compares all intermediate results output by the upstream unit, and finally determines fault cause details of the fault segment based on the historical fault diagnosis result and the knowledge experience base, forms a final fault detail identifier, refines the rechecking result, and outputs the rechecking result through the fourth segment output module 502.
In some embodiments, in step S104, outputting a corresponding motor fault diagnosis result based on the low frequency fault information segment and the high frequency fault information segment includes:
S1041: and carrying out full-flow characteristic information aggregation on the low-frequency fault information fragments and the high-frequency fault information fragments to obtain a motor fault diagnosis result.
As shown in fig. 4, the algorithm library of the central operation scheduling server cluster may further include a diagnosis result processing unit 60, which may aggregate all feature information in the whole diagnosis process according to the upstream low-frequency and high-frequency fault information segments and the statistical analysis features, labels and details generated by the processing of the units of each stage, to obtain a diagnosis result aggregate table.
It should be noted that the above aggregation process does not involve modification of the label, but only includes selectively forming motor fault diagnosis result reports or data reports corresponding to different specific requirements.
As shown in fig. 9, the diagnostic result processing unit 60 specifically includes a second tag identification module 601, a third classifier module 602, and a feature aggregation module 603, where the second tag identification module 601 is configured to identify the first-stage tag, the second-stage tag, and other tags; the third classifier module 602 is configured to classify the low-frequency fault information segments according to the low-frequency fault information segments to obtain final fault diagnosis features; the feature aggregation module 603 is configured to aggregate feature information within a full diagnosis flow range, so as to obtain a final motor fault diagnosis result.
In some embodiments, the method further comprises:
s1042: and storing the motor fault diagnosis result and/or outputting the motor fault diagnosis result in a visual mode.
The diagnosis result processing unit 60 further comprises a distributor module 604, a memory module 605 and a visualization module 606, wherein the distributor module 604 distributes various diagnosis reports and data reports formed by feature aggregation according to defined distribution strategies and authorities; the memory module 604 outputs the diagnosis result detail or the low-frequency and high-frequency key information fragments to the local for structural storage in a preset output format. The visualization module 606 is configured to visually display the details of the diagnosis result or the signal quantity corresponding to the diagnosis result, so that a user can conveniently and timely take a corresponding policy according to the diagnosis result of the motor.
As shown in fig. 2 and 3, a knowledge experience base can be formed based on motor fault diagnosis results, fault conditions of the whole life cycle of the motor can be accumulated, and a certain warning prompt can be given in early stage of motor failure so as to avoid more serious fault problems.
As shown in fig. 2, the internet of vehicles platform sets an independent electric drive intelligent system server cluster containing a motor fault diagnosis detail database for vehicles to interact with the above-mentioned low frequency data server cluster for storing low frequency data sources, high frequency data server cluster for storing high frequency data sources, electric drive knowledge experience base server cluster for storing electric drive working condition base containing motor state segments and the like, and central operation scheduling server cluster containing algorithm base.
In summary, the embodiment of the disclosure forms a first-stage tag for preliminary screening diagnosis of low-frequency data fragments based on a first-stage dynamic sliding window, and expands the diagnosis range of high-frequency data fragments by directional positioning of the first-stage tag, then performs statistical analysis on the high-frequency fault information fragments by using a second-stage dynamic sliding window and a binary search algorithm, classifies and positions fault details of the high-frequency fault information fragments by using a machine learning algorithm corresponding to the tag, rechecks the high-frequency fault information fragments to fill the diagnosis details, finally aggregates and outputs motor fault diagnosis results for full-flow characteristic information, and the full-flow prominently reduces the time for screening motor fault marks in low-frequency data and prominently characterizes the detailed state of operation during motor fault period in high-frequency data, thereby effectively avoiding the problems of unreliability of low-frequency data information and long time consumption of full-scale retrieval of high-frequency data, rapidly analyzing motor faults of large-scale and full-scale vehicle users, positioning detailed reasons of motor faults, effectively improving the rule quantity and accuracy of fault analysis diagnosis of a vehicle motor, reducing diagnosis time consumption, and providing high-quality and high-efficiency diagnosis results.
In addition, in the embodiment, different data are stored in different independent server clusters, and are distributed and scheduled through a central unified algorithm, so that the capacity expansion is facilitated; the algorithm operation is based on unit construction, so that modification and sequencing combination are facilitated, and batch processing is automated; the unit is composed of different modules, is convenient for multiplexing, and is convenient for adding, deleting and changing corresponding modules according to actual diagnosis requirements. The fault information fragments are circulated in a basic form of the data fragments, the upstream and downstream formats are unified, distribution and aggregation are convenient, and safe transmission and storage of the data are facilitated.
Example two
Fig. 10 is a schematic structural diagram of a motor failure diagnosis device for a vehicle according to a second embodiment of the present disclosure, and as shown in fig. 10, the motor failure diagnosis device for a vehicle according to the second embodiment of the present disclosure includes:
an acquisition module 100 configured to acquire internet of vehicles buried point signal data of a motor, wherein the internet of vehicles buried point signal data includes low frequency data and high frequency data;
the first diagnosis module 200 is configured to diagnose the low-frequency data based on a first-stage dynamic sliding window algorithm, and mark abnormal motor information in the low-frequency data by a first-stage tag to obtain a low-frequency fault information fragment;
the second diagnosis module 300 is configured to expand the diagnosis range of the high-frequency data according to the first-stage tag, diagnose the high-frequency data corresponding to the diagnosis range based on a second-stage dynamic sliding window, and screen high-frequency fault information fragments from the high-frequency data;
and an output module 400 configured to output a corresponding motor fault diagnosis result based on the low frequency fault information piece and the high frequency fault information piece.
Example III
The third embodiment of the present disclosure also provides a motor failure diagnosis system for a vehicle, including the motor failure diagnosis device for a vehicle described above.
As shown in fig. 3, the motor fault diagnosis system for a vehicle includes three layers of a low frequency data layer, a high frequency data layer and a data application layer, and the operation of the motor fault diagnosis system for a vehicle starts from the first-stage dynamic sliding window fault calibration of a full-scale user motor in the low frequency data layer, and this step outputs a low frequency key information fragment (low frequency key information). The low frequency data segments with labels are extracted to form positioning marks for positioning and structuring the high frequency data segments of the high frequency data source in the high frequency data layer, the formed priori high frequency information segments are calibrated and output to the data application layer through the second-stage dynamic sliding window of the motor fault of the directional user and the binary search algorithm, statistical characteristics in the segments and corresponding second-stage label flows are transferred to the data application layer, the algorithm adaptation fitting characteristics of the data application layer are utilized to generate comparison segments, and then the high frequency fault information (high frequency key information) after the rechecking is output through fault reason rechecking and secondary classification positioning comparison checking. The high-frequency and low-frequency fault information is aggregated, calculated and transferred to an electric drive knowledge experience base and an electric drive digital intelligent system of a data application layer through diagnosis results according to labels, feature details and the like in the high-frequency and low-frequency fault information, and the whole diagnosis flow of the motor fault diagnosis system is completed.
It should be noted that, the motor fault diagnosis device and system for a vehicle provided in the embodiments of the present disclosure correspond to the motor fault diagnosis method for a vehicle in the foregoing embodiments, and based on the motor fault diagnosis method for a vehicle, those skilled in the art can understand the specific implementation manner of the motor fault diagnosis device and system for a vehicle for distributing information in the embodiments of the present disclosure and various variations thereof, and any optional item in the motor fault diagnosis method for a vehicle embodiment is also applicable to the motor fault diagnosis device and system for a vehicle, which are not repeated herein.
Example IV
The fourth embodiment of the present disclosure further provides an internet of vehicles platform, at least comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method when executing the computer program on the memory.
In some embodiments, a processor executing a computer program may be a processing device including more than one general purpose processing device, such as a microprocessor, central Processing Unit (CPU), graphics Processing Unit (GPU), or the like. More specifically, the processor may be a Complex Instruction Set Computing (CISC) microprocessor, a Reduced Instruction Set Computing (RISC) microprocessor, a Very Long Instruction Word (VLIW) microprocessor, a processor running other instruction sets, or a processor running a combination of instruction sets. The processor may also be one or more special purpose processing devices such as an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA), a Digital Signal Processor (DSP), a system on a chip (SoC), or the like.
The memory may be read-only memory (ROM), random-access memory (RAM), phase-change random-access memory (PRAM), static random-access memory (SRAM), dynamic random-access memory (DRAM), electrically erasable programmable read-only memory (EEPROM), other types of random-access memory (RAM), flash memory disk or other forms of flash memory, cache, registers, static memory, compact disc read-only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, or other magnetic storage devices, or any other possible non-transitory medium which can be used to store information or instructions that can be accessed by a computer device, and the like.
The internet of vehicles platform of the disclosed embodiments may include a plurality of server clusters, for example, including a low frequency data server cluster for storing a low frequency data source, a high frequency data server cluster for storing a high frequency data source, an electric drive knowledge experience library server cluster for storing an electric drive working condition library including a motor state segment and the like, a central operation scheduling server cluster including an algorithm library, and an electric drive intelligent system server cluster including a vehicle motor fault diagnosis detail database, where each server cluster is connected through a network bus communication to implement information interaction.
Example five
A fifth of the embodiments of the present disclosure provides a computer-readable storage medium storing a computer program which, when executed by a processor, implements the steps of the above-described method.
The computer-readable storage media of the embodiments of the present disclosure may employ any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. In the disclosed embodiments, the computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device, such as the memory described above.
The computer programs of embodiments of the present disclosure may be organized into one or more computer-executable components or modules. Aspects of the disclosure may be implemented with any number and combination of such components or modules. For example, aspects of the present disclosure are not limited to the specific computer-executable instructions or the specific components or modules illustrated in the figures and described herein. Other embodiments may include different computer-executable instructions or components having more or less functionality than illustrated and described herein.
The foregoing description is only of the preferred embodiments of the present disclosure and description of the principles of the technology being employed. It will be appreciated by persons skilled in the art that the scope of the disclosure referred to in this disclosure is not limited to the specific combinations of features described above, but also covers other embodiments which may be formed by any combination of features described above or equivalents thereof without departing from the spirit of the disclosure. Such as those described above, are mutually substituted with the technical features having similar functions disclosed in the present disclosure (but not limited thereto).
Claims (10)
1. A motor failure diagnosis method for a vehicle, comprising:
acquiring vehicle networking buried point signal data of a motor, wherein the vehicle networking buried point signal data comprises low-frequency data and high-frequency data;
diagnosing the low-frequency data based on a first-stage dynamic sliding window algorithm, and marking abnormal motor information in the low-frequency data by using a first-stage tag to obtain a low-frequency fault information fragment;
expanding a diagnosis range of high-frequency data according to the first-stage tag, diagnosing the high-frequency data corresponding to the diagnosis range based on a second-stage dynamic sliding window, and screening high-frequency fault information fragments from the high-frequency data;
And outputting a corresponding motor fault diagnosis result based on the low-frequency fault information fragment and the high-frequency fault information fragment.
2. The method of claim 1, wherein diagnosing the low frequency data based on a first level dynamic sliding window algorithm and marking motor anomaly information in the low frequency data with a first level tag to obtain a low frequency fault information segment comprises:
based on a preset data signal name, a diagnosis strategy and the first-stage dynamic sliding window algorithm, carrying out data feature recognition on the low-frequency data, and recognizing a low-frequency signal quantity state change point and a low-frequency signal abnormal point;
and marking the low-frequency signal quantity state change point and the low-frequency signal abnormal point by the first-stage tag.
3. The method of claim 1, wherein expanding a diagnostic range of high frequency data according to the first level tag and diagnosing high frequency data corresponding to the diagnostic range based on a second level dynamic sliding window, and wherein screening high frequency fault information pieces from the high frequency data comprises:
screening corresponding motor high-frequency data fragments from the high-frequency data according to the first-stage tag, and performing data expansion on the motor high-frequency data fragments according to a time sequence to obtain priori high-frequency key information fragments;
And diagnosing the priori high-frequency key information fragments based on a second-stage dynamic sliding window algorithm and a binary search algorithm, and screening the high-frequency fault information fragments from the priori high-frequency key information fragments.
4. A method according to claim 3, wherein diagnosing the a priori high frequency critical information segments based on a second level dynamic sliding window algorithm and a binary search algorithm, screening high frequency fault information segments from the a priori high frequency critical information segments comprises:
based on a preset data signal name, a second diagnosis strategy, the second-stage dynamic sliding window algorithm and a binary search algorithm, carrying out data feature recognition on the priori high-frequency key information fragment, and recognizing state change points and abnormal points of a second semaphore to obtain a preliminary high-frequency fault information fragment;
and carrying out statistical analysis on the fault data of the preliminary high-frequency fault information fragments, and marking the statistical analysis characteristics of the fault data by using a second-stage label.
5. The method according to claim 1, wherein the method further comprises:
diagnosing the first-level fault reason of the high-frequency fault information fragment according to the typical characteristics accumulated in the electric drive working condition library;
And selecting a corresponding rechecking algorithm from a preset algorithm library based on the statistical analysis characteristics of the second-level tag marks and the first-level fault reasons, and rechecking the faults of the high-frequency fault information fragments.
6. The method of claim 5, wherein fault rechecking the high frequency fault information segment comprises:
fitting and forming an ideal signal quantity curve according to the forming parameters of the primary fault reasons and the rechecking algorithm;
and comparing the ideal signal quantity curve with the actual signal quantity curve of the high-frequency fault information fragment, and checking the fault diagnosis accuracy.
7. The method of claim 1, wherein outputting a corresponding motor fault diagnosis result based on the low frequency fault information piece and the high frequency fault information piece, comprises:
and carrying out full-flow characteristic information aggregation on the low-frequency fault information fragments and the high-frequency fault information fragments to obtain a motor fault diagnosis result.
8. The method according to claim 1, wherein the method further comprises:
and storing the motor fault diagnosis result and/or outputting the motor fault diagnosis result in a visual mode.
9. A motor failure diagnosis apparatus for a vehicle, comprising:
the acquisition module is configured to acquire the vehicle networking buried point signal data of the motor, wherein the vehicle networking buried point signal data comprises low-frequency data and high-frequency data;
the first diagnosis module is configured to diagnose the low-frequency data based on a first-stage dynamic sliding window algorithm, and mark abnormal motor information in the low-frequency data by a first-stage label to obtain a low-frequency fault information fragment;
the second diagnosis module is configured to expand the diagnosis range of the high-frequency data according to the first-stage tag, diagnose the high-frequency data corresponding to the diagnosis range based on a second-stage dynamic sliding window, and screen high-frequency fault information fragments from the high-frequency data;
and the output module is configured to output a corresponding motor fault diagnosis result based on the low-frequency fault information fragment and the high-frequency fault information fragment.
10. An internet of vehicles platform comprising at least a memory having a computer program stored thereon and a processor which, when executing the computer program on the memory, implements the steps of the method of any of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310786954.9A CN116859236A (en) | 2023-06-29 | 2023-06-29 | Vehicle motor fault diagnosis method and device and vehicle networking platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310786954.9A CN116859236A (en) | 2023-06-29 | 2023-06-29 | Vehicle motor fault diagnosis method and device and vehicle networking platform |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116859236A true CN116859236A (en) | 2023-10-10 |
Family
ID=88235107
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310786954.9A Pending CN116859236A (en) | 2023-06-29 | 2023-06-29 | Vehicle motor fault diagnosis method and device and vehicle networking platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116859236A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117708098A (en) * | 2024-02-05 | 2024-03-15 | 中国第一汽车股份有限公司 | Battery fault diagnosis method, device, electronic equipment and storage medium |
-
2023
- 2023-06-29 CN CN202310786954.9A patent/CN116859236A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117708098A (en) * | 2024-02-05 | 2024-03-15 | 中国第一汽车股份有限公司 | Battery fault diagnosis method, device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Manco et al. | Fault detection and explanation through big data analysis on sensor streams | |
| CN107948166B (en) | Deep learning-based traffic anomaly detection method and device | |
| CN106209405B (en) | Method for diagnosing faults and device | |
| CN109204389B (en) | Fault diagnosis and self-healing method and system for subway equipment | |
| CN109754110A (en) | A kind of method for early warning and system of traction converter failure | |
| CN106600115A (en) | Intelligent operation and maintenance analysis method for enterprise information system | |
| CN113887616A (en) | Real-time abnormity detection system and method for EPG (electronic program guide) connection number | |
| CN116859236A (en) | Vehicle motor fault diagnosis method and device and vehicle networking platform | |
| CN104750800A (en) | Motor vehicle clustering method based on travel time characteristic | |
| CN111445597B (en) | Data stitching and integration for machine learning | |
| CN111143429A (en) | Vehicle road test data evaluation method, device and system | |
| CN111666978B (en) | Intelligent fault early warning system for IT system operation and maintenance big data | |
| CN112836720A (en) | Building operation and maintenance equipment abnormity diagnosis method and system and computer readable storage medium | |
| CN115456107A (en) | Time series abnormity detection system and method | |
| CN115996249A (en) | Data transmission method and device based on grading | |
| CN111339052A (en) | Unstructured log data processing method and device | |
| CN106355314A (en) | A Vehicle Engine Air Valve Quality Streaming Data Processing Method and Device | |
| WO2021217626A1 (en) | Vehicle data processing method and apparatus, computer device, and storage medium | |
| CN107539038A (en) | A kind of vehicle tyre pressure method for monitoring state and device | |
| WO2023050967A1 (en) | System abnormality detection processing method and apparatus | |
| CN115062144A (en) | Log anomaly detection method and system based on knowledge base and integrated learning | |
| CN114817693A (en) | Equipment failure knowledge recommendation method and system | |
| CN113572628A (en) | Data association method and device, computing equipment and computer storage medium | |
| Fan et al. | Using machine learning to forecast patent quality–take “vehicle networking” industry for example | |
| CN117009105B (en) | Method for pre-alarming state of subway vehicle-mounted equipment based on storm flow calculation in real time |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |