CN-121982650-A - Monitoring method and system for screw looseness-preventing glue coating process

CN121982650ACN 121982650 ACN121982650 ACN 121982650ACN-121982650-A

Abstract

The invention discloses a monitoring method and a system for a screw anti-loosening glue coating process, and relates to the technical field of coating monitoring. The method comprises the steps of obtaining a plurality of window screw sets coated by anti-loosening glue, obtaining a plurality of window coating sampling sequences, obtaining a plurality of window coating thread image frame group sequences, determining trend characteristic vectors of the anti-loosening glue coating process, adjusting coating process parameters and adjusting preset monitoring window adaptability, and monitoring the screw anti-loosening glue coating process based on the adjusted updated monitoring window. The technical problems that in the prior art, the coating quality of the screw glue is based on single detection, the overall trend analysis of a continuous coating process is lacking, so that the coating process abnormality is difficult to identify in time are solved, and the technical effects of continuous monitoring and early warning of the process abnormality of the screw glue coating process are realized by carrying out time sequence correlation and trend analysis on a plurality of monitoring window coating image sequences.

Inventors

QING LINFENG

Assignees

深圳市耐极塑胶五金有限公司

Dates

Publication Date: 20260505
Application Date: 20260131

Claims (10)

1. The method for monitoring the screw looseness-preventing glue coating process is characterized by comprising the following steps of: Acquiring a plurality of window screw sets coated by anti-loosening glue in a plurality of continuous preset monitoring windows, wherein each window screw set is provided with a coating time stamp; Clamping the window screw sets through V-shaped grooves respectively, enabling each window screw to rotate around the axis of each window screw at a preset rotating speed according to the sequence of the coating time stamps, enabling the threaded area coated with the anti-loosening glue to sequentially pass through a monitoring view field in the rotating process, and obtaining a plurality of window coating sampling sequences; in the screw rotating process, high-precision imaging is carried out on the thread local areas of the window coating sampling sequences, so that a plurality of window coating thread image frame group sequences are obtained; executing trend analysis of the anti-loosening glue coating process on the plurality of window coating thread image frame group sequences based on the sequence of the coating time stamps, and determining trend feature vectors of the anti-loosening glue coating process; And carrying out coating process parameter adjustment and preset monitoring window adaptability adjustment on the anti-loosening glue coating process according to the trend characteristic vector of the anti-loosening glue coating process, and monitoring the screw anti-loosening glue coating process based on the adjusted updated monitoring window.
2. The method for monitoring a screw anti-loosening glue coating process according to claim 1, wherein during screw rotation, high-precision imaging is performed on a thread local area of the plurality of window coating sampling sequences to obtain a plurality of window coating thread image frame group sequences, comprising: setting imaging trigger frequency based on the thread pitch and a preset rotating speed, wherein the imaging trigger frequency is used for enabling the thread rotation angle corresponding to two adjacent imaging to meet a preset angle threshold; and performing high-precision imaging on the threaded local areas of the window coating sampling sequences according to a preset imaging trigger frequency by adopting a high-magnification optical imaging device to obtain a plurality of window coating threaded image frame group sequences, wherein each window coating threaded image frame group corresponds to one window screw.
3. A method of monitoring a screw anti-loosening glue coating process as claimed in claim 2, wherein at least one of a coaxial light source, a diffuse light source or a polarized light source is applied to a localized area of the screw thread for illumination during high precision imaging.
4. The method for monitoring a screw anti-loosening glue coating process according to claim 1, wherein the step of performing anti-loosening glue coating process trend analysis on the plurality of window coating thread image frame group sequences based on the sequence of coating time stamps, determining an anti-loosening glue coating process trend feature vector comprises: Extracting a first window coating thread image frame group from the plurality of window coating thread image frame group sequences, and executing glue coating state analysis to obtain a first window coating thread glue coating state feature vector; adding the first window coating thread glue coating state feature vector into a plurality of window coating thread glue coating state features; And executing trend analysis of the anti-loosening glue coating process on the characteristic vectors of the coating states of the window coating thread glue based on the sequence of the coating time stamps, and determining the trend characteristic vectors of the anti-loosening glue coating process.
5. The method of claim 4, wherein extracting a first window coating thread image frame set from the sequence of window coating thread image frame sets, performing glue coating state analysis, and obtaining a first window coating thread glue coating state feature vector, comprises: Performing glue area identification segmentation on the first window coating thread image frame group to obtain a glue area pixel set group; analyzing the relative position relation of the glue area pixel set group from two dimensions of the glue in the axial direction and the circumferential direction of the thread to obtain a glue covering position feature group; Calculating the area occupation ratio of glue characteristic pixels based on the glue area pixel set group to obtain an area group for the glue characteristic pixels, wherein the area of the glue characteristic pixels is used for representing the glue coverage degree in the current thread image; based on the glue area pixel set groups, analyzing the continuous distribution of the detection glue pixels of each glue area pixel set in the axial direction or the circumferential direction to obtain a glue single-frame distribution continuity feature group; and integrating the glue covering position feature group, the glue feature pixel area group and the glue single-frame distribution continuity feature group to obtain a first window coating thread glue coating state feature vector.
6. The method of claim 4, wherein integrating the glue covering position feature set, the glue feature pixel area set, and the glue single frame distribution continuity feature set to obtain the first window coating thread glue coating state feature vector comprises: Carrying out average value calculation on the glue covering position characteristic group and the glue characteristic pixel area group to obtain a glue covering position average value characteristic and a glue characteristic pixel area average value; according to the image frame acquisition time of the first window coating thread image frame group, carrying out continuity adjacent consistency identification on the glue single-frame distribution continuity feature group to obtain glue distribution continuity features; And splicing the glue covering position average value characteristic, the glue characteristic pixel area average value and the glue distribution continuity characteristic to obtain a first window coating thread glue coating state characteristic vector.
7. The method of claim 4, wherein the step of performing a trend analysis of the anti-loosening glue coating process on the plurality of window coating thread glue coating status feature vectors based on a sequence of coating time stamps, determining the anti-loosening glue coating process trend feature vector, comprises: serializing the plurality of window coating thread glue coating state feature vectors based on the sequence of the coating time stamps to obtain a window coating thread glue coating state feature vector sequence; Extracting a first window coating thread glue coating state characteristic vector and a second window coating thread glue coating state characteristic vector from the window coating thread glue coating state characteristic vector sequence; Performing trend feature vector interaction on the first window coating thread glue coating state feature vector and the second window coating thread glue coating state feature vector to obtain a first interaction coating thread glue coating state feature vector; And carrying out trend feature vector interaction on a second window coating thread glue coating state feature vector in the window coating thread glue coating state feature vector sequence by using the first interaction coating thread glue coating state feature vector, and the like until the end of the sequence is reached, so as to obtain a trend feature vector of the anti-loosening glue coating process.
8. The method of claim 7, wherein the step of performing trend feature vector interaction on the first window coating thread glue coating state feature vector and the second window coating thread glue coating state feature vector to obtain a first interaction coating thread glue coating state feature vector comprises: Performing adjacent interaction deviation recognition on the first window coating thread glue coating state feature vector and the second window coating thread glue coating state feature vector from three dimensions of a glue coating position average value feature, a glue feature pixel area average value and a glue distribution continuity feature to obtain a glue coating position average value feature deviation, a glue feature pixel area average value deviation and a glue distribution continuity feature deviation; normalizing the glue coverage position mean value characteristic deviation, the glue characteristic pixel area mean value deviation and the glue distribution continuity characteristic deviation to obtain a first trend interaction matrix; And performing convolution interaction on the second window coating thread glue coating state characteristic vector by using the first trend interaction matrix to obtain the first interaction coating thread glue coating state characteristic vector.
9. The method for monitoring a screw anti-loosening glue coating process according to claim 1, wherein the step of performing coating process parameter adjustment and preset monitoring window adaptation adjustment on the anti-loosening glue coating process according to the trend feature vector of the anti-loosening glue coating process, and performing monitoring of the screw anti-loosening glue coating process based on the adjusted updated monitoring window comprises the steps of: acquiring current coating process parameters, carrying out parameter adjustment by combining the trend feature vector of the anti-loosening glue coating process, acquiring adjustment coating process parameters, and transmitting the adjustment coating process parameters to a coating control module for adjustment; performing coating stability identification based on the trend feature vector of the anti-loosening glue coating process to obtain a coating stability coefficient; And adaptively adjusting the preset monitoring window according to the size of the coating stability coefficient to obtain an updated monitoring window.
10. A monitoring system for a screw anti-loosening glue coating process, for implementing a method for monitoring a screw anti-loosening glue coating process according to any one of claims 1-9, the system comprising: The data acquisition component is used for acquiring a plurality of window screw sets coated by anti-loosening glue in a plurality of continuous preset monitoring windows, wherein each window screw set is provided with a coating time stamp; The monitoring assembly is used for clamping the window screw sets through the V-shaped grooves respectively, enabling each window screw to rotate around the axis of each window screw at a preset rotating speed according to the sequence of the coating time stamps, enabling the threaded area coated with the anti-loosening glue to sequentially pass through a monitoring view field in the rotating process, and obtaining a plurality of window coating sampling sequences; The image acquisition component is used for carrying out high-precision imaging on the thread local areas of the window coating sampling sequences in the screw rotating process to obtain a plurality of window coating thread image frame group sequences; The characteristic analysis component is used for executing trend analysis of the anti-loosening glue coating process on the plurality of window coating thread image frame group sequences based on the sequence of the coating time stamps and determining trend characteristic vectors of the anti-loosening glue coating process; And the parameter adjusting component is used for adjusting coating process parameters and adjusting adaptability of a preset monitoring window of the anti-loosening glue coating process according to the trend characteristic vector of the anti-loosening glue coating process, and monitoring the screw anti-loosening glue coating process based on the adjusted updated monitoring window.

Description

Monitoring method and system for screw looseness-preventing glue coating process Technical Field The invention relates to the technical field of coating monitoring, in particular to a monitoring method and a system for a screw anti-loosening glue coating process. Background Screws are widely used as fasteners in mechanical assembly and industrial manufacturing, and their anti-loosening properties directly affect the structural reliability and safety of the product. In the actual production process, anti-loosening glue is generally coated on the surface of the screw thread to form a bonding or locking effect after assembly, so that the anti-vibration and anti-loosening capabilities of the threaded connection are improved. The coating quality of the anti-loosening glue, such as the coating position, coating continuity and glue quantity distribution, is a key factor influencing the anti-loosening effect. The existing screw anti-loosening glue coating quality detection mode mostly adopts sampling detection or a static detection method based on a single image to judge the coating state of a single screw at a certain moment. The detection mode is difficult to reflect the overall change condition of the anti-loosening glue in the continuous production process, and when parameter drift, unstable glue spraying or fluctuation of the glue state occur to coating equipment, the anti-loosening glue can be discovered only after a large number of unqualified screws are generated, so that hysteresis and sporadic risks exist. In addition, as the screw is in a continuous circulation state in the coating process, the coating quality of the anti-loosening glue has obvious time correlation characteristics, the single detection result is difficult to distinguish between sporadic coating abnormality and continuous process abnormality, and the stability and evolution trend of the coating process cannot be effectively evaluated. Disclosure of Invention The application provides a monitoring method and a system for a screw anti-loosening glue coating process, which solve the technical problems that in the prior art, the screw glue coating quality is based on single detection, and the coating process is abnormal and difficult to identify in time due to lack of integral trend analysis of a continuous coating process. In a first aspect of the application, a method for monitoring a screw loosening prevention glue coating process is provided, the method comprising: The method comprises the steps of obtaining a plurality of window screw sets coated by anti-loosening glue in a continuous mode, obtaining a plurality of window screw sets coated by the anti-loosening glue, clamping the window screw sets through V-shaped grooves respectively, enabling each window screw to rotate around the axis of each window screw according to the sequence of the coating time stamps, enabling a threaded area coated with the anti-loosening glue to sequentially pass through a monitoring view field in the rotating process to obtain a plurality of window coating sampling sequences, carrying out high-precision imaging on the threaded local areas of the window coating sampling sequences in the screw rotating process to obtain a plurality of window coating thread image frame group sequences, carrying out trend analysis of the anti-loosening glue coating process on the window coating thread image frame group sequences based on the sequence of the coating time stamps, determining trend characteristic vectors of the anti-loosening glue coating process, carrying out coating process parameter adjustment and preset monitoring window adaptability adjustment on the anti-loosening glue coating process according to the trend characteristic vectors of the anti-loosening glue coating process, and carrying out monitoring glue anti-loosening coating process on the monitoring window based on the updated monitoring window after adjustment. In a second aspect of the present application, there is provided a monitoring system for a screw anti-loosening glue coating process, the system comprising: The anti-loosening glue coating system comprises a data acquisition component, a monitoring component, a parameter adjustment component, an image acquisition component and a characteristic analysis component, wherein the data acquisition component acquires a plurality of window screw sets coated by anti-loosening glue in a plurality of continuous preset monitoring windows, each window screw set is provided with a coating time stamp, the plurality of window screw sets are respectively clamped through V-shaped grooves according to the sequence of the coating time stamps, each window screw rotates around the axis of each window screw at a preset rotating speed, a threaded area coated with the anti-loosening glue sequentially passes through a monitoring view field in the rotating process to acquire a plurality of window coating sampling sequences, the image acquisition component performs high-preci