Self-cleaning VI flying probe detection device for motor train unit board card maintenance
By integrating the signal acquisition module, main control unit, and directional cleaning execution system, the problems of probe contamination and contact stability in the VI flying probe testing device were solved, achieving high-precision and high-efficiency testing for EMU board maintenance.
Patent Information
- Application Number
- CN202511341742.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-19
- Publication Date
- 2026-01-23
AI Technical Summary
Existing VI flying probe testing devices suffer from problems such as probe contamination, contact stability defects, and low operating efficiency. Especially in the maintenance of EMU circuit boards, the probes are easily affected by flux residues and dust, and vibration interference causes fluctuations in contact resistance. Furthermore, they cannot be adapted to different types of circuit boards.
The self-cleaning VI flying probe testing device integrates a signal acquisition module, a main control unit, and a directional cleaning execution system. Through signal processing, contamination feature recognition, and cleaning strategy generation, combined with a magnetic connector, a three-degree-of-freedom robotic arm, and a cleaning head, it achieves self-cleaning of the probe and high-precision testing.
It improved the accuracy of VI curve testing on EMU circuit boards, reduced the false judgment rate, improved the efficiency of VI flying probe testing, and ensured the stability and adaptability of the probe.
Smart Images
Figure CN121372979A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of EMU maintenance, in particular to a self-cleaning VI flying probe detection device for EMU board card maintenance. BACKGROUND
[0002] The EMU depot is responsible for checking the state of the vehicle. The EMU and the traditional train are very different in terms of locomotive structure and vehicle parts, and need to be equipped with an EMU depot to be responsible for the maintenance and maintenance of the EMU train. Daily safety inspection is necessary to ensure the daily safe operation and efficient transportation of the EMU. The EMU depot is usually set up in the passenger transport center and the area with large passenger flow, such as Beijing South EMU Depot and Xi'an North EMU Depot. The newly built EMU depot inspection warehouse should have at least 6 lines and 12 standard group positions, and the storage line and the temporary repair warehouse capacity should match the inspection warehouse capacity.
[0003] The EMU repair process is divided into five levels. The first and second level maintenance is the operation maintenance, which is carried out in the EMU depot. The third, fourth and fifth level maintenance is the senior repair, which is carried out in the repair unit with corresponding repair qualifications.
[0004] The device fault usually appears the change of impedance characteristics between pins. By comparing the VI curves between the same nodes of good and bad circuit boards (or devices), the node where the impedance characteristics change can be found, so as to determine the specific device that appears the fault. The process of directly observing or comparing the VI curves is called VI curve test.
[0005] In the field of EMU electronic board card maintenance, VI flying probe detection device is generally used to perform VI curve test. The existing VI flying probe detection device has three technical problems:
[0006] Probe contamination: the probe equipment is easily affected by the attachment of soldering agent residues and dust, and needs to be frequently stopped for manual cleaning;
[0007] Contact stability defect: vibration interference will cause serious contact resistance fluctuation, and multi-channel signal interference will cause high false judgment rate;
[0008] Low operation efficiency: the connector tooling is not modularized, and cannot adapt to the hundreds of board cards used by the existing EMU. Each type of board card to be tested needs to be disassembled and installed with the connector tooling. SUMMARY
[0009] The present application solves the technical problems of probe contamination, contact stability defect and low operation efficiency of the existing VI flying probe detection device, and provides a self-cleaning VI flying probe detection device for EMU board card maintenance.
[0010] In order to solve the above technical problems, the technical scheme of the present application is as follows:
[0011] A self-cleaning VI flying probe detection device for EMU board card maintenance, comprising, in sequence: a signal acquisition module, a main control unit, and a directional cleaning execution system.
[0012] Among them:
[0013] The signal acquisition module is used to collect the electrical signals of the test points of the board card to be tested in real time.
[0014] The main control unit is provided with an impedance preprocessing module, a feature extraction module, a pattern matching module, and a cleaning strategy generation module; the pattern matching module is connected with the impedance preprocessing module, the feature extraction module, and the cleaning strategy generation module in data; the impedance preprocessing module is connected with the feature extraction module in data; the impedance preprocessing module is used to process the original electrical signals collected by the signal acquisition module; the feature extraction module is used to perform convolution operation on the impedance signals and wavelet basis functions with localized characteristics in time-frequency domain, decompose the impedance signals into components of different frequency bands, and extract pollution feature vectors based on semantic recognition technology; the pattern matching module is used to calculate the Euclidean distance between the real-time extracted pollution feature vectors and the model parameter library, determine the pollution type matching degree, and output the optimal recognition result; the cleaning strategy generation module is used to calculate the required cleaning intensity and cleaning path parameters according to the matching result of the pollution type matching degree, and generate control instructions.
[0015] The directional cleaning execution system includes a magnetic connector module, a three-degree-of-freedom mechanical arm, and a wind power cleaning head and a laser cleaning head respectively arranged at the front end of the three-degree-of-freedom mechanical arm; wherein the magnetic connector module is used to realize quick disassembly adjustment through magnetic attraction according to different types of EMU board cards to be tested; the three-degree-of-freedom mechanical arm is used to control and adjust the specific position of its front end according to the control instructions; the wind power cleaning head is used to clean the particulate pollutants; and the laser cleaning head is used to clean the oxide layer.
[0016] In the above technical solution, the signal acquisition module includes a VI probe device and a probe base; the signal acquisition module collects the electrical signals of the test points of the board card to be tested on the probe base in real time through the VI probe device.
[0017] In the above technical solution, the processing of the original electrical signals collected by the signal acquisition module by the impedance preprocessing module includes AD conversion.
[0018] In the above technical solution, the processing of the original electrical signals collected by the signal acquisition module by the impedance preprocessing module includes filtering out interference signals.
[0019] In the above technical solution, the processing of the original electrical signals collected by the signal acquisition module by the impedance preprocessing module includes baseline calibration and noise suppression.
[0020] In the technical solution, the feature extraction module is provided with a pollution feature recognition model, and the pollution feature recognition model is provided with a training set of multiple pollution types.
[0021] In the technical solution, the feature extraction module is further used to establish an enhanced training subset for different materials and packaging processes of the to-be-tested board card, and the pollution feature recognition model is formed through data enhancement of the generative adversarial network algorithm.
[0022] In the technical solution, the control instruction generated by the cleaning strategy generation module includes: for the probe tip pollution of the VI probe device, 45° angle jet cleaning is adopted.
[0023] In the technical solution, the control instruction generated by the cleaning strategy generation module includes: for the probe sidewall pollution of the VI probe device, a surrounding cleaning is performed.
[0024] In the technical solution, the control instruction generated by the cleaning strategy generation module includes: after cleaning, impedance retesting is automatically performed to form a closed-loop quality control.
[0025] The present application has the following beneficial effects:
[0026] The motor train unit board card maintenance self-cleaning VI flying needle detection device can greatly improve the precision of the board card VI curve test, reduce the misjudgment rate, and improve the efficiency of the VI flying needle detection of the CRH380B platform motor train unit board card.
[0027] The motor train unit board card maintenance self-cleaning VI flying needle detection device uses the pollution feature recognition model to analyze the probe contact impedance obtained in real time; according to the historical database, the pollution type (solder residue / dust adhesion / oxide layer, etc.) is judged, and the key parameters (pollutant particle size, adhesion strength, etc.) are recognized according to the pollution feature; through the integration of the directional cleaning execution system on the VI probe device, the corresponding cleaning mode is dynamically triggered, so that high-precision testing and self-cleaning of the VI probe device are realized. BRIEF DESCRIPTION OF DRAWINGS
[0028] The present application will be further described in detail below in combination with the drawings and specific embodiments.
[0029] Figure 1 The figure is a schematic diagram of the architecture of the motor train unit board card maintenance self-cleaning VI flying needle detection device.
[0030] Figure 2 The figure is a schematic diagram of the architecture of the motor train unit board card maintenance self-cleaning VI flying needle detection device.
[0031] The reference signs in the figure are:
[0032] 1-signal acquisition module;
[0033] 2-master control unit;
[0034] 3-directional cleaning execution system;
[0035] 11-VI probe device;
[0036] 12-probe base;
[0037] 21-impedance preprocessing module;
[0038] 22-feature extraction module;
[0039] 23-pattern matching module;
[0040] 24-cleaning strategy generation module;
[0041] 221-pollution feature identification model;
[0042] 31-magnetic connector module;
[0043] 32-three-degree-of-freedom mechanical arm;
[0044] 33-wind cleaning head;
[0045] 34-laser cleaning head. DETAILED DESCRIPTION
[0046] The application will be described in detail below with reference to the accompanying drawings.
[0047] Example 1
[0048] As shown in Figure 1 and 2 The EMU board card maintenance self-cleaning VI flying probe detection device of the application comprises a signal acquisition module 1, a master control unit 2, and a directional cleaning execution system 3.
[0049] Among them:
[0050] The signal acquisition module 1 comprises a VI probe device 11 and a probe base 12; the signal acquisition module 1 is used to collect the electrical signals of the test points of the board card to be tested on the probe base 12 in real time through the VI probe device 11;
[0051] The master control unit 2 is respectively provided with an impedance preprocessing module 21, a feature extraction module 22, a pattern matching module 23, and a cleaning strategy generation module 24; the pattern matching module 23 is connected with the impedance preprocessing module 21, the feature extraction module 22, and the cleaning strategy generation module 24 in data; the impedance preprocessing module 21 is connected with the feature extraction module 22 in data;
[0052] The impedance pretreatment module 21 is used for processing the original electrical signal collected by the signal collection module 1, including AD conversion (sampling rate 1 MHz), filtering out interference signals other than 1 kHz-1 MHz, and baseline calibration and noise suppression according to impedance characteristic parameters; in this embodiment, the sampling rate of AD conversion is 1 MHz.
[0053] The feature extraction module 22 is used for performing convolution operation on the impedance signal and a wavelet basis function having a localized characteristic in the time-frequency domain, decomposing the impedance signal into components of different frequency bands, and extracting a pollution feature vector based on semantic recognition technology; the feature extraction module 22 is provided with a pollution feature recognition model 221, which is provided with training sets of multiple pollution types such as flux residue, dust deposition, and metal oxidation; the feature extraction module 22 can establish an enhanced training subset for different materials and packaging processes of the to-be-tested board card, perform data enhancement through a generative adversarial network (GAN) algorithm, and form a high-precision pollution feature recognition model 221; in this embodiment, the material of the to-be-tested board card is FR4, and the pollution feature vector is an organic matter adhesion index and a particle size distribution spectrum of particulate matter.
[0054] The pattern matching module 23 is used for performing Euclidean distance calculation on the real-time extracted pollution feature vector and the model parameter library, determining the pollution type matching degree, and outputting an optimal recognition result.
[0055] The cleaning strategy generation module 24 is used for calculating the required cleaning intensity and cleaning path parameters according to the matching result of the pollution type matching degree, and generating a control instruction; specifically, in this embodiment, the cleaning intensity is a wind pressure of 0.1 MPa; for the probe tip pollution of the VI probe equipment 11, 45° angle jet cleaning is adopted; for the probe sidewall pollution of the VI probe equipment 11, circumferential cleaning is performed; after cleaning, impedance retesting is automatically performed to form a closed-loop quality control;
[0056] The directional cleaning execution system 3 includes a magnetic connector module 31, a three-degree-of-freedom mechanical arm 32, and a wind cleaning head 33 and a laser cleaning head 34 respectively arranged at the front end of the three-degree-of-freedom mechanical arm 32; wherein the magnetic connector module 31 is used for realizing quick disassembly adjustment through magnetic attraction according to different types of to-be-tested board cards of motor trains; the three-degree-of-freedom mechanical arm 32 is used for controlling and adjusting the specific position of the front end thereof according to the control instruction; the wind cleaning head 33 is used for cleaning particulate pollutants (>5 μm); and the laser cleaning head 34 is used for cleaning the oxide layer.
[0057] Embodiment 2
[0058] The specific structure of the self-cleaning VI flying probe detection device for EMU board card maintenance of the embodiment is the same as that of embodiment 1, but the sampling rate of AD conversion is different, the material of the board card to be tested is different, the selected pollution feature vector is different, and the cleaning intensity is different.
[0059] The self-cleaning VI flying probe detection device for EMU board card maintenance of the embodiment comprises a signal acquisition module, a main control unit and a directional cleaning execution system.
[0060] The signal acquisition module comprises a VI probe device and a probe base; the signal acquisition module is used to collect the electrical signals of the test points of the board card to be tested on the probe base in real time through the VI probe device;
[0061] The main control unit is respectively provided with an impedance preprocessing module, a feature extraction module, a pattern matching module and a cleaning strategy generation module; the pattern matching module is connected with the impedance preprocessing module, the feature extraction module and the cleaning strategy generation module in data; the impedance preprocessing module is connected with the feature extraction module in data;
[0062] The impedance preprocessing module is used to process the original electrical signals collected by the signal acquisition module 1, including AD conversion, interference signal filtering and baseline calibration and noise suppression according to impedance characteristic parameters; in the embodiment, the sampling rate of AD conversion is 2MHz.
[0063] The feature extraction module is used to perform convolution operation on the impedance signals and wavelet basis functions with localized characteristics in the time-frequency domain, decompose the impedance signals into components of different frequency bands, and extract pollution feature vectors based on semantic recognition technology; the feature extraction module is provided with a pollution feature recognition model, the pollution feature recognition model is provided with training sets of multiple pollution types such as solder residue, dust deposition and metal oxidation; the feature extraction module can establish an enhanced training subset for different materials and packaging processes of the board card to be tested, perform data enhancement through a generative adversarial network algorithm, and form a high-precision pollution feature recognition model; in the embodiment, the material of the board card to be tested is a ceramic substrate; the pollution feature vector is an organic matter adhesion index.
[0064] The pattern matching module is used to perform Euclidean distance calculation on the real-time extracted pollution feature vectors and model parameter library, determine the pollution type matching degree, and output the optimal recognition result;
[0065] The cleaning strategy generation module is used to calculate the required cleaning intensity and cleaning path parameters according to the matching result of the pollution type matching degree, and generate a control instruction; specifically, in the embodiment, the cleaning intensity is a wind pressure of 0.3MPa; for the probe tip pollution of the VI probe device, 45° angle jet cleaning is adopted; for the probe sidewall pollution of the VI probe device, circumferential cleaning is performed; after cleaning, impedance retesting is automatically performed to form a closed-loop quality control;
[0066] The directional cleaning execution system comprises a magnetic connector module, a three-degree-of-freedom mechanical arm, and a wind power cleaning head and a laser cleaning head arranged at the front end of the three-degree-of-freedom mechanical arm respectively; the magnetic connector module is used to realize quick disassembly and adjustment through magnetic attraction according to different types of measured board cards of the motor train unit; the three-degree-of-freedom mechanical arm is used to control and adjust the specific position at the front end thereof according to a control instruction; the wind power cleaning head is used to clean the particulate pollutants (> 5 mu m); and the laser cleaning head is used to clean the oxide layer.
[0067] Embodiment 3
[0068] The specific structure of the motor train unit board card maintenance self-cleaning VI flying probe detection device in this embodiment is the same as that in embodiment 1, but the sampling rate of AD conversion, the selected pollution feature vector, and the cleaning intensity are different, and specifically:
[0069] The motor train unit board card maintenance self-cleaning VI flying probe detection device of the application comprises a signal acquisition module, a main control unit, and a directional cleaning execution system.
[0070] The signal acquisition module comprises a VI probe device and a probe base; the signal acquisition module is used to acquire the electrical signals of the test points of the board card to be tested on the probe base in real time through the VI probe device;
[0071] The main control unit is respectively provided with an impedance preprocessing module, a feature extraction module, a pattern matching module, and a cleaning strategy generation module; the pattern matching module is connected with the impedance preprocessing module, the feature extraction module, and the cleaning strategy generation module in data; the impedance preprocessing module is connected with the feature extraction module in data;
[0072] The impedance preprocessing module is used to process the original electrical signals collected by the signal acquisition module 1, including AD conversion, interference signal filtering, and baseline calibration and noise suppression according to impedance characteristic parameters; in this embodiment, the sampling rate of AD conversion is 0.5 MHz.
[0073] The feature extraction module is used to perform convolution operation on the impedance signals and wavelet basis functions with localized characteristics in time-frequency domain, decompose the impedance signals into components of different frequency bands, and extract pollution feature vectors based on semantic recognition technology; the feature extraction module is provided with a pollution feature recognition model, which is provided with training sets of multiple pollution types such as solder residue, dust deposition, and metal oxidation; the feature extraction module can establish an enhanced training subset for the material and packaging process of different board cards to be tested, perform data enhancement through a generative adversarial network algorithm, and form a high-precision pollution feature recognition model; in this embodiment, the material of the board card to be tested is FR4; and the pollution feature vector is a particle size distribution spectrum.
[0074] The mode matching module is used to perform Euclidean distance calculation on the real-time extracted pollution feature vector and the model parameter library, determine the pollution type matching degree, and output the optimal recognition result.
[0075] The cleaning strategy generation module is used to calculate the required cleaning intensity and cleaning path parameters according to the matching result of the pollution type matching degree, and generate control instructions. Specifically, in the embodiment, the cleaning intensity is: wind pressure 0.5 MPa. For the probe tip pollution of the VI probe equipment, 45° inclined angle jet cleaning is adopted; for the probe sidewall pollution of the VI probe equipment, circumferential cleaning is performed; after cleaning, impedance re-measurement is automatically performed to form a closed-loop quality control.
[0076] The directional cleaning execution system includes a magnetic connector module, a three-degree-of-freedom mechanical arm, and a wind power cleaning head and a laser cleaning head respectively arranged at the front end of the three-degree-of-freedom mechanical arm. The magnetic connector module is used to realize quick disassembly and adjustment through magnetic attraction according to different types of measured board cards of the motor train unit. The three-degree-of-freedom mechanical arm is used to control and adjust the specific position of the front end thereof according to the control instructions. The wind power cleaning head is used to clean the particulate pollutants (>5 μm). The laser cleaning head is used to clean the oxide layer.
[0077] The self-cleaning type VI flying needle detection device for motor train unit board card maintenance can greatly improve the accuracy of board card VI curve testing, reduce the misjudgment rate, and improve the efficiency of VI flying needle detection of CRH380B platform motor train unit board cards.
[0078] The self-cleaning type VI flying needle detection device for motor train unit board card maintenance analyzes the probe contact impedance obtained through real-time monitoring by using a pollution feature recognition model; determines the pollution type (solder residue / dust adhesion / oxide layer, etc.) according to a historical database, and recognizes key parameters (pollutant particle size, adhesion strength, etc.) according to the pollution features; by integrating a directional cleaning execution system on the VI probe equipment, the corresponding cleaning mode is dynamically triggered, so as to realize high-precision testing and self-cleaning of the VI probe equipment.
[0079] Obviously, the above embodiments are only examples for clear illustration, and not limitation on the embodiments. For ordinary skilled persons in the art, other different forms of changes or variations can be made on the basis of the above description. Here, it is not necessary and impossible to enumerate all the embodiments. The obvious changes or variations derived therefrom are still within the protection scope of the present application.
Claims
1. A self-cleaning VI flying probe testing device for EMU circuit board maintenance, characterized in that, It includes, in sequence: a signal acquisition module (1), a main control unit (2), and a directional cleaning execution system (3); in: The signal acquisition module (1) is used to acquire electrical signals at the test points of the board under test in real time; The main control unit (2) includes: an impedance preprocessing module (21), a feature extraction module (22), a pattern matching module (23), and a cleaning strategy generation module (24); the pattern matching module (23) is connected to the impedance preprocessing module (21), the feature extraction module (22), and the cleaning strategy generation module (24) respectively; the impedance preprocessing module (21) is connected to the feature extraction module (22); the impedance preprocessing module (21) is used to process the original electrical signal acquired by the signal acquisition module (1); the feature extraction module (22) is used to perform convolution operation on the impedance signal with the wavelet basis function with localization characteristics in the time and frequency domain, decompose the impedance signal into components of different frequency bands, and extract the pollution feature vector based on semantic recognition technology; the pattern matching module (23) is used to calculate the Euclidean distance between the real-time extracted pollution feature vector and the model parameter library, determine the pollution type matching degree, and output the optimal recognition result; the cleaning strategy generation module (24) is used to calculate the required cleaning intensity and cleaning path parameters according to the matching result of the pollution type matching degree, and generate control commands; The directional cleaning execution system (3) includes: a magnetic connector module (31), a three-degree-of-freedom robotic arm (32), and a wind-powered cleaning head (33) and a laser cleaning head (34) respectively set at the front end of the three-degree-of-freedom robotic arm (32); wherein, the magnetic connector module (31) is used to achieve quick-release adjustment by magnetic attraction according to the different types of test boards of the EMU; the three-degree-of-freedom robotic arm (32) is used to control and adjust the specific position of its front end according to the control command; the wind-powered cleaning head (33) is used to clean particulate pollutants; and the laser cleaning head (34) is used to clean the oxide layer.
2. The self-cleaning VI flying probe testing device for EMU circuit board maintenance according to claim 1, characterized in that, The signal acquisition module (1) includes: a VI probe device (11) and a probe base (12); the signal acquisition module (1) acquires the electrical signals of the test points of the board under test on the probe base (12) in real time through the VI probe device (11).
3. The self-cleaning VI flying probe testing device for EMU circuit board maintenance according to claim 1, characterized in that, The impedance preprocessing module (21) processes the original electrical signal acquired by the signal acquisition module (1) into an AD conversion.
4. The self-cleaning VI flying probe testing device for EMU circuit board maintenance according to claim 1, characterized in that, The impedance preprocessing module (21) processes the original electrical signal acquired by the signal acquisition module (1) to filter out interference signals.
5. The self-cleaning VI flying probe testing device for EMU circuit board maintenance according to claim 1, characterized in that, The impedance preprocessing module (21) processes the raw electrical signal acquired by the signal acquisition module (1) as baseline calibration and noise suppression.
6. The self-cleaning VI flying probe testing device for EMU circuit board maintenance according to claim 1, characterized in that, The feature extraction module (22) includes a pollution feature recognition model (221), which has a training set for various pollution types.
7. The self-cleaning VI flying probe testing device for EMU circuit board maintenance according to claim 6, characterized in that, The feature extraction module (22) is also used to establish an enhanced training subset for different test boards based on their materials and packaging processes, and to perform data augmentation through adversarial generative network algorithms to form a contamination feature recognition model (221).
8. The self-cleaning VI flying probe testing device for EMU circuit board maintenance according to claim 2, characterized in that, The control instructions generated by the cleaning strategy generation module (24) include: cleaning the probe tip contamination of the VI probe device (11) by spraying at a 45° angle.
9. The self-cleaning VI flying probe testing device for EMU circuit board maintenance according to claim 2, characterized in that, The control instructions generated by the cleaning strategy generation module (24) include: performing a surround cleaning for the probe sidewall contamination of the VI probe device (11).
10. The self-cleaning VI flying probe testing device for EMU circuit board maintenance according to claim 2, characterized in that, The control instructions generated by the cleaning strategy generation module (24) include: automatically performing impedance retesting after cleaning to form a closed-loop quality control.