CN214311786U

CN214311786U - Engine piston assembly recognition device that makes mistakes

Info

Publication number: CN214311786U
Application number: CN202023302695.9U
Authority: CN
Inventors: 杨洋; 毛扬; 张�林
Original assignee: SAIC GM Wuling Automobile Co Ltd
Current assignee: SAIC GM Wuling Automobile Co Ltd
Priority date: 2020-12-31
Filing date: 2020-12-31
Publication date: 2021-09-28
Anticipated expiration: 2030-12-31

Abstract

The utility model discloses an engine piston assembly recognition device that makes mistakes. The device comprises a video processing and identifying unit, a video acquisition module, a controller, an RFID reader-writer, an electromagnetic induction sensor and a shaping module, wherein the video acquisition module is connected with the video processing and identifying unit, the controller is respectively connected with the video processing and identifying unit, the RFID reader-writer and the shaping module, and the electromagnetic induction sensor is connected with the shaping module. The utility model discloses whether can the automatic identification piston dress mistake, the artifical discernment that has relatively improved the recognition accuracy and recognition rate, can obviously improve the assembly speed of engine piston. The utility model is used for the method that the discernment piston makeed mistakes changes the back a little and can be used for the discernment that other parts made mistakes.

Description

Engine piston assembly recognition device that makes mistakes

Technical Field

The utility model belongs to the technical field of automobile engine automatic assembly, concretely relates to recognition device makes mistakes in engine piston assembly.

Background

At present, the assembly of an automobile engine is carried out in a manufacturing environment with multiple parts collinear, and potential failure risks such as missing installation, misinstallation and the like exist due to different products. For example, the piston is installed, and the installation can be rotated by 180 degrees due to the cylindrical shape, so that misinstallation is caused; when the pistons are replaced in batches, the same engine piston uses pistons produced by different manufacturers to cause misassembly; when the pistons of different models are co-linear when the model is changed, the pistons of different models are mixed and loaded to cause misloading. Piston misloading can cause serious quality problems; the resulting rework and scrap increases the manufacturing cost of the engine.

In order to reduce the error rate of the assembly of the engine piston, various methods for preventing the error of the piston are adopted in each base, for example, an operator detects through visual files in the assembly process, personnel in a lower station rechecks, and a special camera (such as Kangnai vision IS840 series, the unit price IS about 6 thousands, and IS used for detecting the machine type characteristic identifier on the surface of the piston) IS used. The method has the problems of low visual inspection efficiency, incapability of finding wrong installation in time, high cost, long time consumption and the like of the special camera.

SUMMERY OF THE UTILITY MODEL

In order to solve the above-mentioned problem that exists among the prior art, the utility model provides an engine piston assembly recognition device that makes mistakes.

In order to achieve the above purpose, the utility model adopts the following technical scheme:

the device for identifying the assembly errors of the engine piston comprises a video processing identification unit, a video acquisition module, a controller, an RFID reader-writer, an electromagnetic induction sensor and a shaping module, wherein the video acquisition module is connected with the video processing identification unit, the controller is respectively connected with the video processing identification unit, the RFID reader-writer and the shaping module, and the electromagnetic induction sensor is connected with the shaping module.

Compared with the prior art, the utility model discloses following beneficial effect has:

the utility model discloses a set up video processing identification element, video acquisition module, controller, RFID read write line, electromagnetic induction sensor and plastic module, through the video image of the engine piston on the automatic acquisition station and carry out image processing, whether can the automatic identification piston misloading to write in the identification result in the RFID label on the station engine tray, so that reprocess the error correction and handle. Compared with the existing manual identification, the method improves the identification accuracy and the identification speed, and can obviously improve the assembly speed of the engine piston. In addition, the method for identifying the piston error (after slight modification) can also be used for identifying the error of other parts.

Drawings

Fig. 1 is the embodiment of the present invention discloses a schematic diagram of an engine piston assembly error recognition device.

In the figure: the system comprises a video processing and identifying unit, a 2-video collecting module, a 3-controller, a 4-RFID reader-writer, a 5-shaping module and a 6-electromagnetic induction sensor.

Detailed Description

The present invention will be described in further detail with reference to the accompanying drawings.

The embodiment of the utility model provides an engine piston assembly recognition device that makes mistakes, as shown in figure 1, the device includes video processing identification unit 1, video acquisition module 2, controller 3, RFID read write line 4, electromagnetic induction sensor 6 and plastic module 5, and video acquisition module 2 links to each other with video processing identification unit 1, and controller 3 links to each other with video processing identification unit 1, RFID read write line 4 and plastic module 5 respectively, and electromagnetic induction sensor 6 links to each other with plastic module 5.

In this embodiment, the device mainly comprises a video processing and identifying unit 1, a video collecting module 2, a controller 3, an RFID reader-writer 4, an electromagnetic induction sensor 6 and a shaping module 5. The connection relationship of the modules is shown in fig. 1, and the function and principle of each module are described below.

And the video acquisition module 2 is used for acquiring a video image of the engine piston on the current station and sending the acquired video image to the video processing and identifying unit 1. The video acquisition module 2 generally comprises cameras, and the selection of the number and the installation position of the cameras is based on the principle that the video images of the piston can be comprehensively, clearly and effectively shot. Since the number of pistons is generally more than one, it is difficult to effectively capture a video image of each piston using only one camera. For this reason, the prior art generally employs a servo system to take pictures from different angles by rotating the camera. This treatment is not only costly but also time consuming. This embodiment has solved the not enough that prior art set up servo through the quantity that increases the camera.

The video processing identification unit 1 is mainly used for identifying whether the piston is in error. The video signal input by the video acquisition module 2 is subjected to image processing to extract feature points, and then the extracted feature points are input into a trained neural network error recognition model to judge whether an error occurs or not, and a recognition result is sent to the controller 3. The video processing and identifying unit 1 can adopt a raspberry development board. The ras pberry development board is a linux system development board, supports java, python and other languages, and has the characteristics of small volume, low cost, simplicity in use and maintenance and the like. The image processing and error recognition algorithm can be implemented by opencv software. opencv is a cross-platform computer vision and machine learning software library issued based on BSD license (open source), and contains various algorithms such as image processing, pattern recognition, three-dimensional reconstruction, object tracking and machine learning.

And the controller 3 is mainly used for acquiring information related to an engine at an assembly station and outputting control signals to coordinate the work of other modules. Communicating with the video processing and identifying unit 1, receiving the error identification result, and sending a control instruction to the RFID reader-writer 4 to write the identification result into an RFID label on the engine tray at the current station; when receiving a high or low level signal which is output by the shaping module 5 and indicates that the engine is in place, outputting a control instruction to the video processing and identifying unit 1, controlling the video acquisition module 2 to start shooting a video image of the piston by the video processing and identifying unit 1, and outputting the control instruction to the RFID reader-writer 4 to read information in the RFID label. The information in the RFID tag comprises whether the current assembly state of the engine is qualified, the engine model, the unique identification code of the engine and the like.

The RFID reader/writer 4 is mainly used for reading information from the RFID tag or writing information to the RFID tag (if the information is qualified). The RFID label is arranged on the tray of the engine, when the engine is in place, the RFID label is close to the RFID reader-writer 4, and the RFID reader-writer 4 carries out information reading and writing operation on the RFID label under the control of the controller 3.

The electromagnetic induction sensor 6 is equivalent to an electromagnetic coil and is mainly used for generating an engine in-place signal based on the principle of electromagnetic induction. An induction block (such as an iron block) is fixed on the tray of the engine, when the engine is in place, the induction block is close to the electromagnetic induction sensor 6, and the electromagnetic induction sensor 6 outputs an induction voltage signal. The induced voltage signal is an irregular pulse signal and is sent to the shaping module 5 for shaping. Many sensors such as hall elements are commonly used, but this embodiment is not limited to this.

And the shaping module 5 is used for changing the irregular voltage signal output by the electromagnetic induction sensor 6 into a (high or low) level signal and sending the signal to the controller 3. The controller 3 performs engine-in-place detection based on a change in the level signal (e.g., from low to high). The shaping module 5 may adopt a comparison circuit composed of an operational amplifier, or may adopt a schmitt shaping circuit.

As an optional embodiment, the video capture module 2 includes a first camera and a second camera, the first camera is used for capturing video images of the first piston and the second piston, the second camera is used for capturing video images of the third piston and the fourth piston, and the first camera and the second camera are respectively connected to the video processing and recognition unit 1 through a USB cable.

The embodiment provides a technical scheme of the video acquisition module 2. In this embodiment, the video capture module 2 is composed of two cameras, i.e., a first camera and a second camera, and each camera is responsible for capturing video images of two pistons. Thus, the present embodiment is designed for a typical four cylinder engine. Because the two cameras are adopted to shoot the video images of the four pistons at the same time, a servo system can be omitted, and the image acquisition speed is improved.

As an alternative embodiment, the controller 3 is a GE-PLC controller 3, and is connected to the video processing and identifying unit 1 through a network cable.

In the present embodiment, the controller 3 is a GE-PLC controller 3. Siemens and GE are two major brands of PLC. Siemens is German corporation, and the processor of Siemens PLC adopts 8096 product sequence chip; GE is a company of America, and the processor of GE-PLC uses the Pentium series of i5, i6 chips. The GE-PLC controller 3 is connected with the video processing identification unit 1 through a network cable, and is in data communication with the video processing identification unit 1 by adopting a TCP protocol.

The above description is only for the description of several embodiments of the present invention, but the scope of the present invention should not be considered as the protection scope of the present invention, in which all the equivalent changes or modifications or the equal-scale enlargement or reduction etc. made according to the design spirit of the present invention should be considered as falling into the protection scope of the present invention.

Claims

1. The device for identifying the assembly errors of the engine piston is characterized by comprising a video processing identification unit, a video acquisition module, a controller, an RFID reader-writer, an electromagnetic induction sensor and a shaping module, wherein the video acquisition module is connected with the video processing identification unit, the controller is respectively connected with the video processing identification unit, the RFID reader-writer and the shaping module, and the electromagnetic induction sensor is connected with the shaping module.

2. The engine piston assembly error recognition device of claim 1, wherein the video capture module comprises a first camera and a second camera, the first camera is used for capturing video images of the first piston and the second piston, the second camera is used for capturing video images of the third piston and the fourth piston, and the first camera and the second camera are respectively connected with the video processing recognition unit through a USB cable.

3. The engine piston assembly error recognition device of claim 1, wherein the controller is a GE-PLC controller connected to the video processing recognition unit via a network cable.