Method and apparatus for detecting typical fault tolerance margin of 3d visual defect diagnosis equipment

Abstract

The application discloses a 3D visual defect diagnosis equipment typical fault fault tolerance margin detection method and device, with different object distance with external defect battery sample, circular platform is driven by motor at different rotating speed, motor is fixed on vibration platform with different frequency and different amplitude, with the increase of object distance, the difference between the object distance of the previous battery sample which cannot identify the known defect information and the minimum object distance is taken as the object distance fault tolerance margin; with the increase of rotating speed, the difference between the rotating speed corresponding to the previous defect information which cannot identify the known defect information and the rated rotating speed is taken as the rotating speed fault tolerance margin; with the increase of amplitude, the absolute value of the difference between the position corresponding to the previous defect information which cannot identify the known defect information and the rated position is taken as the vibration amplitude fault tolerance margin; with the increase of vibration frequency, the difference between the vibration frequency corresponding to the previous defect information which cannot identify the known defect information and the rated frequency is taken as the vibration frequency fault tolerance margin.

Description

Technical Field

[0001] This invention belongs to the field of dynamic detection technology of machine vision, specifically a device and method for detecting the tolerance margin of several typical faults in 3D vision defect diagnosis equipment. Background Technology

[0002] Machine vision has been widely used in various industries. 3D vision defect diagnosis equipment is an important component of machine vision, and its performance directly affects the application of machine vision technology. In 3D vision defect diagnosis equipment, dynamic vision detection has certain technical difficulties. Since the object being tested is in a non-static state, it is easy to cause misjudgment due to excessive detection error. For example, 3D vision defect diagnosis equipment for automotive power battery surface defects is generally used on automotive power battery production lines. The automotive power battery is in a moving state on the conveyor mechanism. When the performance of the 3D vision defect diagnosis equipment is not very good, misdiagnosis of automotive power battery surface defects often occurs. The reasons for this misdiagnosis fault are as follows: (1) Vibration of the transmission mechanism causes the detection error of the component, which is called vibration error fault; (2) Change in the distance between the photosensitive element and the object being tested causes the detection error of the component, which is called object distance error fault; (3) The excessive speed of the object being tested causes the detection error of the component, which is called speed error fault. Currently, methods to address these misdiagnosis faults include reducing vibration of the transmission mechanism, fixing the distance between the photosensitive element and the object under test, and controlling the movement speed of the object. However, these methods cannot completely solve the problems, and misdiagnosis faults in these three types of diagnostic equipment remain typical and unavoidable faults in dynamic visual inspection processes. Although the tolerance for misdiagnosis errors in these three types of inspection equipment is constantly improving with the continuous development of intelligent software for dynamic visual defect diagnosis equipment, the specific margin of error tolerance for misdiagnosis faults remains unknown, and automatic assessment of this margin is still not possible. Summary of the Invention

[0003] The purpose of this invention is to solve the above-mentioned problems existing in the prior art, and to propose a method and device for detecting the fault tolerance margin of typical faults in 3D visual defect diagnosis equipment. The invention uses dynamic visual defect diagnosis equipment to detect the fault tolerance margin of three typical faults: vibration error fault, object distance error fault, and speed error fault, to measure the fault tolerance performance of the diagnostic equipment for misdiagnosed faults, and to realize automatic evaluation of fault tolerance margin.

[0004] To achieve the above objectives, the technical solution adopted by the typical fault tolerance margin detection device of the 3D visual defect diagnosis equipment of the present invention is as follows: An automatic fault tolerance margin assessment device is installed inside the greenhouse. This device includes a horizontal circular tray and a vertical circular tray drive motor. The housing of the circular tray drive motor is fixedly connected to a cylindrical motor fixing component made of ferrous material, and the output axis is fixedly connected to the circular tray upwards. The cylindrical motor fixing component is located inside a cylindrical cylinder liner. The outer wall of a cylindrical electromagnet fixing component directly below the cylindrical motor fixing component is fixedly connected to the inner wall of the cylindrical cylinder liner. The cylindrical electromagnet is fitted in the central through hole of the cylindrical electromagnet fixing component and can slide up and down but cannot rotate. The upper section of the cylindrical electromagnet extends above the cylindrical electromagnet fixing component, and an electromagnet coil is fitted around the upper section. A spring is placed on the outside of the electromagnet coil. All ends are supported on the bottom surface of the cylindrical motor fixing component, and the lower end is fixedly connected to the cylindrical electromagnet fixing component; the output axis of the cylindrical electromagnet telescopic drive stepper motor is connected to the cylindrical electromagnet through a nut pair, and the cylindrical electromagnet can slide up and down; several test battery samples are mounted on the upper surface of the circular tray along the circumferential direction, with the distance from the edge of the circular tray increasing by equal value. Each test battery sample has the same battery sample defect and a unique battery sample information QR code. The content of the QR code is the edge distance of the corresponding battery sample. The camera lens of the visual defect diagnosis equipment is aimed at the defect and QR code of the test battery sample; the MCU evaluation component is connected to the visual defect diagnosis equipment through the input interface, and is connected to the circular tray drive motor, the cylindrical electromagnet telescopic drive stepper motor, and the electromagnet coil through the output interface.

[0005] The detection method of the typical fault tolerance margin detection device for the 3D vision defect diagnosis equipment adopts the following technical solution:

[0006] Step 1): The battery sample defect information and QR code information set on the battery sample to be tested are used as known defect information. When there is no fault, the circular tray drive motor rotates at the rated speed V0, the cylindrical electromagnet extension drive stepper motor is at the rated position S0, and the electromagnet coil oscillates on and off at the rated frequency F0.

[0007] Step 2): The visual defect diagnosis equipment collects the defect information and QR code information of the tested battery sample. As the object distance increases, the difference between the object distance of the previous battery sample that cannot identify defect information and the object distance of the first battery sample with the smallest object distance is used as the object distance tolerance margin.

[0008] Step 3): Only change the rotational speed V0 of the circular tray drive motor, rotating it at speeds V0+v, V0+2v, ..., V0+nv respectively, where v is the speed increment each time and n is the number of times the speed increases. Then the visual defect diagnosis device collects n kinds of defect information Fv(V0+iv), i = 1, 2, ..., n. As the speed increases, the difference between the speed corresponding to the previous defect information Fv(V0+(i-1)v) that cannot identify the known defect information Fv(V0+iv) and the rated speed V0 is taken as the speed tolerance margin.

[0009] Step 4): Make the circular tray drive motor return to the rated speed V0 and rotate. Control the cylindrical electromagnet to extend and retract, drive the stepper motor to move down to different positions S0-s, S0-2s, ... S0-ks respectively. s is the position reduction each time, and k is the number of times the position moves down. The amplitude of the cylindrical motor fixed part continuously increases, and k kinds of defect information Fs(S0-is) are collected. As the amplitude increases, the absolute value of the difference between the position corresponding to the previous Fs(S0-(i-1)s) of the Fs(S0-is) that cannot identify known defect information and the rated position S0 is taken as the vibration amplitude tolerance margin.

[0010] Step 5): Control the cylindrical electromagnet to extend and retract, drive the stepper motor to return to the rated position S0. The electromagnet coil oscillates on and off at different frequencies of F0+f, F0+2f, ..., F0+jf, where f is the increment of the vibration frequency, j = 1, 2, 3, ..., to obtain j kinds of defect information Ff(F0+if), i = 1, 2, ..., j. As the vibration frequency increases, the difference between the vibration frequency corresponding to the previous defect information Ff(F0+(i-1)f) that cannot identify the known defect information Ff(F0+if) and the rated frequency F0 is taken as the vibration frequency tolerance margin.

[0011] The present invention has the following advantages after adopting the above technical solution:

[0012] (1) The present invention uses the error fault tolerance margin to measure the error fault tolerance performance of dynamic visual defect diagnosis equipment, providing a new means to evaluate the performance of dynamic visual defect diagnosis equipment.

[0013] (2) The present invention constructs a fault tolerance margin assessment mechanical device that integrates transmission mechanism vibration change, object distance change and test object movement speed change. A circular platform is constructed in a closed box. Test battery samples with surface defects with different object distances are installed around the circular platform. The circular platform can rotate at different speeds under the drive of a motor. The motor is fixed on a vibration platform with different frequencies and amplitudes. It can effectively realize the assessment of the fault tolerance margin of vibration error fault, object distance error fault and speed error fault.

[0014] (3) The present invention constructs an electromechanical device with programmable vibration amplitude and frequency, which can effectively simulate various vibration states and lay the foundation for automatic evaluation of the fault tolerance margin of vibration error faults. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the overall structure of the typical fault tolerance margin detection device of the 3D visual defect diagnosis equipment of the present invention.

[0016] Figure 2 for Figure 1 Enlarged schematic diagram of the internal structure of the automatic fault tolerance margin assessment device;

[0017] Figure 3 for Figure 2 The layout of the tested battery samples in the circular tray;

[0018] Figure 4 for Figure 3 A schematic diagram of the structure of a battery sample under test facing the visual defect diagnosis equipment;

[0019] Figure 5 for Figure 1 and Figure 2 Control circuit logic block diagram;

[0020] Figure 6 This is a flowchart of the fault tolerance margin detection method described in this invention.

[0021] The components in the attached diagram are numbered and named as follows: 1. Greenhouse, 2. Automatic fault tolerance margin assessment device, 3. First wireless transceiver, 4. Second wireless transceiver, 5. Host computer, 6. Tested battery sample, 7. Circular tray, 8. Motor and tray connector, 9. Tray drive motor shaft, 10. Circular tray drive motor, 11. Cylindrical motor fixing component, 12. Visual defect diagnosis equipment, 13. Support frame, 14. Fault tolerance margin MCU assessment component, 15. Cylindrical cylinder liner, 16. Cylindrical electromagnet fixing component, 17. Cylindrical electromagnet telescopic drive stepper motor, 18. Electromagnet telescopic drive stepper motor shaft, 19. Cylindrical electromagnet, 20. Electromagnet upper and lower sliding keyway, 21. Electromagnet coil, 22. Spring, 23. Fixed base, 29. Battery sample defect, 30. Battery sample information QR code. Detailed Implementation

[0022] See Figure 1This invention discloses a typical fault tolerance margin detection device for 3D visual defect diagnosis equipment, comprising a sealed cubic chamber 1. Inside the cubic chamber 1 is an automatic fault tolerance margin assessment device 2, and a first wireless transceiver 3 is placed inside the automatic fault tolerance margin assessment device 2. To prevent dust in the air from affecting the visual inspection probe, the automatic fault tolerance margin assessment device 2 is placed inside the sealed cubic chamber 1. Outside the cubic chamber 1 is a host computer 5 and a second wireless transceiver 4. The second wireless transceiver 4 is connected to the host computer 5 via a communication line and communicates wirelessly with the first wireless transceiver 3. The host computer 5 facilitates human-computer interaction, and communication between the host computer 5 and the automatic fault tolerance margin assessment device 2 is achieved through the first wireless transceiver 3 and the second wireless transceiver 4.

[0023] See Figure 2 The automatic fault tolerance margin assessment device 2 shown mainly consists of three parts: a distance and velocity error fault generator, a vibration error fault generator, and a fault tolerance margin assessment component. The distance and velocity error fault generator includes a circular tray 7 and a circular tray drive motor 10. The circular tray drive motor 10 is vertically arranged, with its lower motor housing fixedly connected to a cylindrical motor fixture 11. A blind hole is formed in the center of the upper section of the cylindrical motor fixture 11, and the motor housing extends into and is fixed within the blind hole. The output shaft of the circular tray drive motor 10 is a tray drive motor shaft 9, which is vertically upward and connected to the circular tray 7 via a motor-tray connector 8. The circular tray 7 is horizontally arranged, and the battery sample 6 under test is mounted on its upper surface. The operation of the circular tray drive motor 10 drives the circular tray 7 and the battery sample 6 under test to rotate synchronously.

[0024] Directly below the distance and velocity error generator is the vibration error generator, which includes a cylindrical cylinder liner 15, a cylindrical electromagnet fixing component 16, a cylindrical electromagnet telescopic drive stepper motor 17, a cylindrical electromagnet 19, an electromagnet sliding keyway 20, an electromagnet coil 21, and a spring 22. The outermost part of the vibration error generator is the cylindrical cylinder liner 15, within which are housed the cylindrical motor fixing component 11, the cylindrical electromagnet fixing component 16, the cylindrical electromagnet telescopic drive stepper motor 17, the cylindrical electromagnet 19, the electromagnet coil 21, and the spring 22.

[0025] The bottom of the cylindrical cylinder liner 15 and the housing of the cylindrical electromagnet telescopic drive stepper motor 17 are both fixedly connected to the fixed base 23, which is horizontally arranged. The outer wall of the cylindrical electromagnet fixing member 16 is fixedly connected to the inner wall of the cylindrical cylinder liner 15. The outer wall of the cylindrical motor fixing member 11 does not contact the inner wall of the cylindrical cylinder liner 15, leaving a gap. The cylindrical electromagnet fixing member 16 is directly below the cylindrical motor fixing member 11, with a space between them. A through hole is opened in the center of the cylindrical electromagnet fixing member 16, and an electromagnet sliding keyway 20 is provided on the inner wall of the through hole.

[0026] A cylindrical electromagnet telescopic drive stepper motor 17 is arranged vertically, and its output shaft is the electromagnet telescopic drive stepper motor shaft 18. The electromagnet telescopic drive stepper motor shaft 18 is coaxially fixedly connected to a cylindrical electromagnet 19. The cylindrical electromagnet 19 is fitted into the through hole in the center of the cylindrical electromagnet fixing member 16. A sliding key is provided on the outer wall of the cylindrical electromagnet 19. The sliding key cooperates with the upper and lower sliding keyway 20 of the electromagnet. The upper and lower sliding keyway 20 of the electromagnet is connected to the cylindrical electromagnet fixing member 16, so that the cylindrical electromagnet 19 can slide up and down along the cylindrical electromagnet fixing member 16 but cannot rotate.

[0027] The output shaft of the electromagnet telescopic drive stepper motor shaft 18 is inserted into the threaded hole of the cylindrical electromagnet 19, and is connected to the cylindrical electromagnet 19 through an internal and external thread structure to form a nut pair. When the electromagnet telescopic drive stepper motor shaft 18 rotates forward and backward, it pushes the cylindrical electromagnet 19 to slide up and down.

[0028] The upper section of the cylindrical electromagnet 19 extends above the cylindrical electromagnet fixing member 16. An electromagnet coil 21 is fitted around the upper section of the cylindrical electromagnet 19. The distance between the top surface of the electromagnet coil 21 and the bottom surface of the cylindrical motor fixing member 11 is always greater than the farthest distance between the top surface of the cylindrical electromagnet 19 and the bottom surface of the cylindrical motor fixing member 11.

[0029] Several springs 22 are evenly arranged on the outside of the electromagnet coil 21; this invention uses six springs 22. The upper ends of all springs 22 are supported on the bottom surface of the cylindrical motor fixing member 11, and the lower ends are fixedly connected to the cylindrical electromagnet fixing member 16. The springs 22 and the electromagnet coil 21 are located in the space between the cylindrical electromagnet fixing member 16 and the cylindrical motor fixing member 11. The springs 22 support the cylindrical motor fixing member 11, allowing it to vibrate up and down within the cylindrical cylinder liner 15.

[0030] The cylindrical motor holder 11 is made of ferrous material. When current flows through the electromagnet coil 21, the cylindrical electromagnet 19 generates a magnetic field, attracting the cylindrical motor holder 11 directly above it, causing the cylindrical motor holder 11 to move downwards and compressing the spring 22 downwards. When the electromagnet coil 21 is de-energized, the cylindrical electromagnet 19 loses its magnetic field, releasing the cylindrical motor holder 11. Under the action of the spring 22, the cylindrical motor holder 11 returns to its original position, thus generating one vibration. The amplitude of the vibration is determined by the distance between the cylindrical electromagnet 19 and the cylindrical motor holder 11, which can be adjusted by the number of steps the cylindrical electromagnet drives the stepper motor 17 to rotate. That is, by rotating the cylindrical electromagnet and the stepper motor 17 by different numbers of steps, the cylindrical electromagnet 19 slides up and down different distances in the circular hole. Therefore, the amplitude of the vibration can be determined by the number of steps the cylindrical electromagnet drives the stepper motor 17 to rotate.

[0031] The fault tolerance margin assessment component is installed next to the object distance and speed error fault generator. It includes a visual defect diagnostic device 12 and an MCU evaluation component 14. The visual defect diagnostic device 12 has a built-in camera lens, which is mounted on the same horizontal line as the battery sample 6 under test, and the camera lens is directly aimed at the battery sample 6 under test. Both the visual defect diagnostic device 12 and the MCU evaluation component 14 are fixed on the support frame 13.

[0032] See Figure 3 Several battery samples 6 to be tested are distributed along the circumference on a circular tray 7. This invention takes five battery samples 6 as an example, namely battery samples A, B, C, D and E. The radial distances of the samples from the edge of the circular tray 7 along the diameter direction are different. This radial distance is called the edge distance. The edge distances are ordered from smallest to largest as follows: the edge distance of the first battery sample A is L1, the edge distance of the second battery sample B is L2, the edge distance of the third battery sample C is L3, the edge distance of the fourth battery sample D is L4, and the edge distance of the fifth battery sample E is L5. The edge distance L1 is the smallest, and the values ​​of L1, L2, L3, L4 and L5 increase sequentially. The difference between the edge distances of two adjacent battery samples is t. L1 is added to t, 2t, 3t and 4t respectively to get L2 = L1 + t, L3 = L1 + 2t, L4 = L1 + 3t and L5 = L1 + 4t. In this way, the object distances of the five battery samples 6 to the lens of the visual defect diagnosis device 12 are different.

[0033] Combination Figure 4The same battery sample defect 29 was set on all five tested battery samples 6. For the five different tested battery samples 6, the defect information was defined as: SS1, SS2, SS3, SS4, and SS5. Each tested battery sample 6 had a unique battery sample information QR code 30. The content of the sample information QR code 30 was the margin of the corresponding battery sample; that is, the content of the sample information QR code 30 for tested battery samples A, B, C, D, and E was the margin L1, L2, L3, L4, and L5, respectively. Both the battery sample defect 29 and the QR code 30 were set on the side facing the visual defect diagnosis device 12, with the camera lens pointed at the tested battery sample 6, the battery sample defect 29, and the battery sample information QR code 30.

[0034] See Figure 5 The MCU evaluation component 14 has its own MCU controller, which consists of an ARM main control device and input / output interface circuits. It communicates with the host computer 5 outside the sealed cubic greenhouse 1 through the first wireless transceiver 3, connects to the visual defect diagnosis device 12 through the input interface, and receives information from the visual defect diagnosis device 12. It connects to the circular tray drive motor 10, the cylindrical electromagnet telescopic drive stepper motor 17, and the electromagnet coil 21 through the output interface, respectively, and issues control commands to the circular tray drive motor 10, the cylindrical electromagnet telescopic drive stepper motor 17, and the electromagnet coil 21 to control the speed and start / stop of the circular tray drive motor 10, the number of steps and start / stop of the cylindrical electromagnet telescopic drive stepper motor 17, and the on / off of the current in the electromagnet coil 21.

[0035] See Figure 6 , and then combine Figure 1-5 The fault tolerance margin detection method of the present invention is as follows:

[0036] Step 1: The pre-set known defect information SS1, SS2, SS3, SS4, SS5 and the margins L1, L2, L3, L4, L5 of the sample information QR code 30 for each tested battery sample 6 are pre-saved in the host computer 5.

[0037] The host computer 5 sends commands to the MCU evaluation component 14 via the second wireless transceiver 4 and the first wireless transceiver 3. The MCU controller in the MCU evaluation component 14 controls the circular tray drive motor 10 to rotate at the rated speed V0, and simultaneously controls the cylindrical electromagnet extension drive stepper motor 17 to be at the rated position S0 (height), and controls the electromagnet coil 21 to oscillate on and off at the rated frequency F0, completing the initialization settings. These three initialization rated values ​​are the safe values ​​when the visual defect diagnosis device 12 diagnoses it as fault-free. At this time, the edge distance L1 is the safe value when there is no fault. The circular tray 7 drives the tested battery sample 6 to rotate at the speed V0, the cylindrical electromagnet 19 is at the rated position S0, and the entire object distance and speed error fault generator above the spring 22 (including the tested battery sample 6) vibrates up and down at the rated frequency F0.

[0038] Step 2: The visual defect diagnosis device 12 continuously collects the defect and QR code information of the tested battery sample 6 on the circular tray 7, and obtains five different defect and QR code information. Each information contains two parts: battery sample defect information 29 and sample information QR code 30. Although the original battery sample defects 29 on the tested battery sample 6 are exactly the same, the information of the battery sample defects 29 collected will differ because the object distances of the five tested battery samples 6 from the lens of the visual defect diagnosis device 12 are different. That is, the difference between the defect information SS1, SS2, SS3, SS4, and SS5 of the five battery sample defects 29 is caused by the different lens object distances. The battery sample information QR code 30 records the corresponding different lens object distances as L1, L2, L3, L4, and L5. Since the software of the visual defect diagnosis device 12 has a fault-tolerant function, although the defect information SS1, SS2, SS3, SS4, and SS5 of the five battery sample defects 29 are different, after processing by the software of the visual defect diagnosis device 12, if the increase in object distance is not very large, the visual defect diagnosis device 12 will still identify the correct defect information SS1, SS2, SS3, SS4, and SS5. If, as the object distance increases, the visual defect diagnosis device 12 fails to identify the correct defect information SS1, SS2, SS3, SS4, and SS5, the object distance tolerance margin F(t) for the object distance error fault is determined based on the battery sample for which the visual defect diagnosis device 12 cannot identify the defect information. The difference between the object distance of the preceding battery sample and the smallest object distance L1 of the first battery sample is used as the object distance tolerance margin F(t). For example, if the visual defect diagnosis device 12 cannot identify the defect information SS4 of the fourth battery sample D, and the preceding battery sample of the fourth battery sample D is the third battery sample C, the difference between the object distance L3 of the third battery sample C and the object distance L1 of the first battery sample is 2t (L3 = L1 + 2t). Therefore, the object distance tolerance margin F(t) = 2t, which is the tolerance margin for detecting an object distance error fault. The same logic applies to the others. The visual defect diagnosis device 12 transmits the processed object distance tolerance margin F(t) = 2t to the host computer 5.

[0039] To obtain a more accurate object distance tolerance margin, the number of tested battery samples 6 with different edge distances can be increased, or the difference t between the edge distances of two adjacent battery samples can be decreased, i.e., the difference between the edge distances of two adjacent battery samples can be reduced. The visual defect diagnosis device 12 transmits the processed object distance error fault tolerance margin to the host computer 5.

[0040] Step 3: The host computer 5 sends a command to change only the rotational speed V0 of the circular tray drive motor 10. At this time, the cylindrical electromagnet extension drive stepper motor 17 remains at its rated position S0, and the electromagnet coil 21 remains on and off at its rated frequency F0. Using the object distance error fault tolerance margin F(t) obtained in Step 2 as a reference (this is the object distance error fault tolerance margin obtained at the rated speed V0), the object distance corresponding to the object distance error fault tolerance margin F(t) is kept constant. The circular tray drive motor 10 is rotated at different speeds V0+v, V0+2v, ..., V0+nv, where v is the speed increment for each increment, and n is the number of speed increases, n = 1, 2, 3, ... The visual defect diagnosis device 12 continuously collects information from the tested battery sample 6. With each increase in rotational speed, the defect information of the battery sample defect 29 collected by the visual defect diagnosis device 12 will differ, resulting in n different defect information sets for the battery sample defect 29. These n defect information sets are defined as Fv(V0+iv), i = 1, 2, ..., n. At this point, the differences in defect information between battery samples with the same object distance are caused by different rotational speeds. Because the software of the visual defect diagnosis device 12 has fault tolerance, although the defect information Fv(V0+iv) of the n battery sample defects 29 are different, after processing by the software of the visual defect diagnosis device 12, if the increase in rotational speed is not significant, the visual defect diagnosis device 12 can still identify the preset known defect information from Fv(V0+iv). If, as the rotational speed increases, the visual defect diagnosis device 12 cannot identify known defect information from Fv(V0+iv), the difference between the rotational speed corresponding to the pre-defined known defect information Fv(V0+(i-1)v) and the rated rotational speed V0 is used as the rotational speed tolerance margin F(v). For example, if the visual defect diagnosis device 12 cannot identify the third defect information Fv(V0+3v), then the preceding defect information of the third defect information Fv(V0+3v) is the second defect information Fv(V0+2v), and the rotational speed corresponding to Fv(V0+2v) is V0+2v. Subtracting the rotational speed V0+2v from the rated rotational speed V0 gives 2v, so the rotational speed tolerance margin F(v) = 2v, which is the tolerance margin for detecting a rotational speed error fault. The rest can be deduced similarly. The visual defect diagnosis device 12 transmits the processed rotational speed tolerance margin F(v) to the host computer 5.

[0041] Step 4: The host computer 5 sends a command to the MCU evaluation component 14 to return the circular tray drive motor 10 to its rated speed, i.e., rotating at speed V0. At this time, the electromagnet coil 21 remains on and off at its rated frequency F0. The fault tolerance margin F(t) for the object distance error obtained in Step 2 is used as the standard; this is the fault tolerance margin for the object distance error obtained at the rated position S0. The object distance corresponding to the fault tolerance margin F(t) is kept constant. The MCU controller controls the cylindrical electromagnet extension drive stepper motor 17 to be at different positions S0-s, S0-2s, ..., S0-ks, i.e., the cylindrical electromagnet 19 continuously moves downwards, where s is the position decrease each time, and k is the number of position decreases, k = 1, 2, 3, ... . Since the vibration amplitude is determined by the distance between the cylindrical electromagnet 19 and the cylindrical motor fixing component 11, the vibration amplitude increases by s each time the position of the cylindrical electromagnet 19 decreases. The visual defect diagnosis device 12 continuously collects information from the tested battery sample 6. As the cylindrical electromagnet 19 moves downward, the distance between the cylindrical motor fixing part 11 and the cylindrical electromagnet 19 increases, and the vibration amplitude of the cylindrical motor fixing part 11 also increases, making it more difficult for the visual defect diagnosis device 12 to identify defects in the battery sample. Each time the cylindrical electromagnet 19 moves downward, the information on the battery sample defect 29 collected by the visual defect diagnosis device 12 will differ, resulting in k types of defect information for the battery sample 29. This k types of defect information for the battery sample 29 are defined as: Fs(S0-is), i = 1, 2, ..., k. At this point, the differences in defect information between battery sample defects 29 at the same object distance are caused by different vibration amplitudes. Because the software of the visual defect diagnosis device 12 has a fault-tolerant function, although the defect information Fs(S0-is) of the k types of battery samples 29 are different, after processing by the software of the visual defect diagnosis device 12, if the increase in vibration amplitude is not very large, the visual defect diagnosis device 12 can still identify the preset known defect information from the defect information Fs(S0-is). If the visual defect diagnosis device 12 cannot identify the preset known defect information as the vibration amplitude increases, based on the Fs(S0-is) that the visual defect diagnosis device 12 cannot identify, the absolute value of the difference between the position of the previous Fs(S0-(i-1)s) of the unidentified preset known defect information Fs(S0-is) and the rated position S0 is taken as the vibration amplitude fault margin F(s). For example, if the visual defect diagnosis device 12 cannot identify the fourth defect information Fs(S0-4s), then the defect information preceding the fourth defect information Fs(S0-4s) is the third defect information Fs(S0-3s). The absolute value of the difference between the position S0-3s of Fs(S0-3s) and the rated position S0 is 3s. Therefore, the vibration amplitude tolerance margin F(s) = 3s, which is the tolerance margin for detecting vibration amplitude error faults.The rest follow the same logic. The visual defect diagnosis device 12 transmits the processed vibration amplitude error fault tolerance margin F(s) to the host computer 5.

[0042] Step 5: The host computer 5 sends a command to the MCU evaluation component 14 to control the extension and retraction of the cylindrical electromagnet, driving the stepper motor 17 to return to the rated position S0. At this time, the circular tray drive motor 10 continues to rotate at a speed of V0, based on the object distance error fault tolerance margin F(t) obtained in Step 2. This is the object distance error fault tolerance margin obtained at the rated frequency F0. The object distance corresponding to the object distance error fault tolerance margin F(t) remains unchanged. The MCU controller controls the electromagnet coil 21 to oscillate on and off at different frequencies of F0+f, F0+2f, ..., F0+jf, where f is the increment of the vibration frequency, and j = 1, 2, 3, ... The visual defect diagnosis device 12 continuously collects information from the tested battery sample 6. As the vibration frequency of the cylindrical motor fixture 11 increases, the visual defect diagnosis device 12 will find it more difficult to identify defects in the battery sample. As a result, the collected information on battery sample defects 29 will differ, resulting in defect information for j types of battery sample defects 29. This defect information for these j types of battery sample defects 29 is defined as: Ff(F0+if), i = 1, 2, ..., j. The difference in defect information between battery samples with the same object distance and defect 29 is caused by different vibration frequencies. Since the software of the visual defect diagnosis device 12 has a fault tolerance function, although the defect information Ff(F0+if) of the j types of battery samples 29 are different, after processing by the software of the visual defect diagnosis device 12, if the vibration amplitude does not increase significantly, the visual defect diagnosis device 12 can still identify the preset known defect information from Ff(F0+if). If the visual defect diagnosis device 12 cannot identify the preset known defect information from Ff(F0+if) as the vibration frequency increases, the difference between the vibration frequency corresponding to the previous defect information Ff(F0+(i-1)f) that cannot be identified by the visual defect diagnosis device 12 and the rated frequency F0 is taken as the vibration frequency tolerance margin F(f). For example, if the visual defect diagnosis device 12 cannot identify the fifth defect information Ff(F0+5f), then the preceding defect information of the fifth defect information Ff(F0+5f) is the fourth defect information Ff(F0+4f). The difference between the vibration frequency F0+4f of the fourth defect information Ff(F0+4f) and the rated frequency F0 is 4f. Therefore, the vibration frequency tolerance margin F(f) = 4f, which is the tolerance margin for detecting a vibration frequency error fault. The rest can be deduced similarly. The visual defect diagnosis device 12 transmits the processed vibration frequency tolerance margin F(f) to the host computer 5.

[0043] Using the above method, under the coordinated control of the host computer and MCU controller, the fault tolerance margins of three typical faults—vibration error fault, object distance error fault, and speed error fault—are automatically detected.

Claims

1. A typical fault tolerance margin detection device for 3D visual defect diagnosis equipment, wherein an automatic fault tolerance margin assessment device is installed inside the greenhouse, characterized in that: The fault tolerance margin automatic assessment device includes a horizontal circular tray (7) and a vertical circular tray drive motor (10). The housing of the circular tray drive motor (10) is fixedly connected to a cylindrical motor fixture (11) made of ferrous material, and the output axis is fixedly connected to the circular tray (7). The cylindrical motor fixing part (11) is inside the cylindrical cylinder liner (15). The outer wall of the cylindrical electromagnet fixing part (16) directly below the cylindrical motor fixing part (11) is fixedly connected to the inner wall of the cylindrical cylinder liner (15). The cylindrical electromagnet (19) is fitted in the through hole in the middle of the cylindrical electromagnet fixing part (16) and can slide up and down but cannot rotate. The upper section of the cylindrical electromagnet (19) is higher than the cylindrical electromagnet fixing part (16) and the upper section is fitted with an electromagnet coil (21). Several springs (22) are evenly arranged outside the electromagnet coil (21). The upper ends of all springs (22) are supported on the bottom surface of the cylindrical motor fixing part (11), and the lower ends are fixedly connected to the cylindrical electromagnet fixing part (16). The output axis of the cylindrical electromagnet extension drive stepper motor (17) is connected to the cylindrical electromagnet (19) via a nut. The inner side wall of the through hole in the center of the cylindrical electromagnet fixing part (16) is provided with an electromagnet up and down sliding keyway (20). The outer side wall of the cylindrical electromagnet (19) is provided with a sliding key. The sliding key cooperates with the electromagnet up and down sliding keyway (20) so that the cylindrical electromagnet (19) slides up and down along the cylindrical electromagnet fixing part (16) but cannot rotate. On the surface of the circular tray (7), several battery samples to be tested are arranged in a circumferential direction along the circumference, with the distance from the edge of the circular tray (7) increasing in equal order. Each battery sample to be tested has the same battery sample defect and a QR code with different edge distance information. The camera lens of the visual defect diagnosis device (12) is aimed at the defect and QR code of the battery sample to be tested. The MCU evaluation component (14) is connected to the visual defect diagnosis device (12) through the input interface, and is connected to the circular tray drive motor (10), the cylindrical electromagnet telescopic drive stepper motor (17), and the electromagnet coil (21) through the output interface.

2. The typical fault tolerance margin detection device for 3D vision defect diagnosis equipment according to claim 1, characterized in that: The distance between the top surface of the electromagnet coil (21) and the bottom surface of the cylindrical motor fixture (11) is always greater than the farthest distance between the top surface of the cylindrical electromagnet (19) and the bottom surface of the cylindrical motor fixture (11).

3. A detection method for a typical fault tolerance margin detection device for 3D vision defect diagnosis equipment as described in any one of claims 1-2, characterized in that... Includes the following steps: Step 1): The battery sample defect information and QR code information already set on the battery sample to be tested are used as known defect information. When there is no fault, the circular tray drive motor (10) rotates at the rated speed V0, and the cylindrical electromagnet extends and retracts to drive the step. The motor (17) is in the rated position S0, and the electromagnet coil (21) oscillates on and off at the rated frequency F0; Step 2): The visual defect diagnosis device (12) collects the defect information and QR code information of the tested battery sample. As the object distance increases, the visual defect diagnosis device (12) will be unable to identify the correct defect information. Based on the battery sample that the visual defect diagnosis device (12) cannot identify the defect information, the difference between the object distance of the battery sample before the battery sample that cannot identify the defect information and the object distance of the smallest first battery sample is used as the object distance tolerance margin for the object distance error fault. Step 3): Only change the rotation speed V0 of the circular tray drive motor (10), rotate at speeds V0+v, V0+2v, ..., V0+nv respectively, where v is the speed increment each time and n is the number of times the speed increases. Then the visual defect diagnosis device (12) collects n kinds of defect information Fv(V0+iv), i=1, 2, ..., n. As the speed increases, the difference between the speed corresponding to the previous defect information Fv(V0+(i-1)v) that cannot identify the known defect information Fv(V0+iv) and the rated speed V0 is taken as the speed tolerance margin. Step 4): Make the circular tray drive motor (10) return to the rated speed V0 and rotate. Control the cylindrical electromagnet to extend and retract to drive the stepper motor (17) to move down to different positions S0-s, S0-2s, ... S0-ks respectively. s is the position reduction each time, and k is the number of times the position moves down. The amplitude of the cylindrical motor fixing part (11) continuously increases and collects k kinds of defect information Fs(S0-is). As the amplitude increases, the absolute value of the difference between the position corresponding to the previous Fs(S0-(i-1)s) of the Fs(S0-is) that cannot identify known defect information and the rated position S0 is taken as the vibration amplitude tolerance margin. Step 5): Control the cylindrical electromagnet to extend and retract, drive the stepper motor (17) to return to the rated position S0, and the electromagnet coil (21) oscillates on and off at different frequencies of F0+f, F0+2f, ..., F0+jf respectively, where f is the increment of the vibration frequency, j=1, 2, 3, ..., to obtain j kinds of defect information Ff(F0+if), i=1, 2, ..., j. As the vibration frequency increases, the difference between the vibration frequency corresponding to the previous defect information Ff(F0+(i-1)f) that cannot identify the known defect information is taken as the vibration frequency tolerance margin.

4. The detection method according to claim 3, characterized in that: Increasing the number of battery samples with different edge distances can yield a more accurate object distance tolerance margin.

5. The detection method according to claim 3, characterized in that: Reducing the difference in edge distance between two adjacent battery samples under test can yield a more accurate object distance tolerance margin.

6. The detection method according to claim 3, characterized in that: The MCU evaluation component (14) controls the electromagnet coil (21) to be energized, and the cylindrical electromagnet (19) generates a magnetic field, which attracts the cylindrical motor fixing part (11) directly above it, causing the cylindrical motor fixing part (11) to move downward, driving the spring (22) to compress downward; when the electromagnet coil (21) is de-energized, the cylindrical electromagnet (19) loses its magnetic field, releases the cylindrical motor fixing part (11), and under the action of the spring (22), the cylindrical motor fixing part (11) returns to its original position, thus generating one vibration.

7. The detection method according to claim 6, characterized in that: The amplitude of the vibration is determined by the distance between the cylindrical electromagnet 19 and the cylindrical motor fixture (11), and this distance is adjusted by the number of steps the cylindrical electromagnet drives the stepper motor (17) to rotate.

8. The detection method according to claim 3, characterized in that: The MCU evaluation component (14) communicates with the host computer (5) located outside the greenhouse via a wireless transceiver.

9. The detection method according to claim 8, characterized in that: The known defect information and edge distance information of each tested battery sample are pre-set in the host computer (5).

Images