Online monitoring and early warning device and method for tool wear in cutting machining of zinc steel protective fence pipe

Abstract

The invention discloses an online monitoring and early warning device and method for tool wear in zinc steel protective fence pipe cutting, and the device comprises a visual module, a matrix-type sonic sensor, an alarm module, an equipment controller, and an image and vibration analysis model constructed at a cloud end. A noise monitoring scheme is added on the basis of a visual monitoring scheme, periodic detection is carried out through a visual module, meanwhile, uninterrupted detection is carried out during cutting each time through a sonic sensor, and when the sonic sensor monitors abnormity, the visual module is immediately started to carry out emergency detection, so that the noise monitoring effect is improved. On one hand, compared with a pure noise monitoring scheme, a visual scheme is more accurate and more reliable, and compared with a pure visual monitoring scheme, the cost is lower, and the workload of an analysis model is lower, so that early warning can be quickly made when the cutting blade is abnormal, and dangerous cases caused by excessive abrasion of the cutting blade are avoided.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of zinc steel pipe processing, in particular to a device and method for monitoring and early warning of tool wear in zinc steel guard rail pipe cutting processing. BACKGROUND

[0002] Zinc steel guard rail pipes are usually square steel pipes, which need to be cut into target lengths according to use requirements during processing, and subsequent bending is also required. During length cutting, a saw blade is usually used for cutting processing.

[0003] In traditional cutting processing, the tool will gradually wear out as the number of cuts increases. In addition, during cutting, the saw blade may have abnormal wear due to saw blade defects, uneven pipe density, different cutting speeds, and cutting operation technical problems, including: 1. Excessive wear causes the actual service life of the saw blade to be inconsistent with the expected service life; 2. Saw blade breakage and edge collapse; 3. Saw blade deformation and skewing; The above situations will affect the cutting end face quality, motor life, and personnel safety. In existing processing operations, the sound during cutting, abnormal torque of the motor, and additional protective cover are usually used to make judgments and provide protection based on the experience of the operator. However, this method is not accurate and timely, and in many cases, the saw blade is severely worn out when abnormal torque and clear noise are detected. At this time, the saw blade may collapse at any time, which poses a danger. Therefore, a device and method for monitoring and early warning of tool wear in zinc steel guard rail pipe cutting processing are provided. SUMMARY

[0004] The purpose of the present application is to provide a device and method for monitoring and early warning of tool wear in zinc steel guard rail pipe cutting processing to solve the problems raised in the background.

[0005] To achieve the above purpose, the present application provides the following technical solution: a device for monitoring and early warning of tool wear in zinc steel guard rail pipe cutting processing, comprising: An image detection unit includes at least two vision modules for appearance detection from the front and side, respectively; A noise detection unit includes a plurality of sound wave sensors for detecting abnormal noise during cutting; An analysis module includes a device controller arranged at a local end and a cloud server, and an analysis model is built in the cloud server, the monitoring data of the visual module and the acoustic sensor are transmitted to the analysis model for feature analysis after being collected by the device controller, and the analysis result is fed back to the device controller; An alarm module includes an alarm lamp and / or a buzzer for early warning according to the analysis result fed back by the analysis model.

[0006] Preferably, a partition is further included, a cutting blade movably arranged on the top of the partition, a through slot is formed on the partition for the cutting blade to pass through, the visual module is installed on the top of the partition, and the acoustic sensor is arranged in the partition in a matrix form, so that the acoustic sensor collects acoustic signals when the cutting blade moves to the position below the through slot for cutting, and the visual module collects image signals when the cutting blade moves to the position above the through slot.

[0007] Preferably, a protective cover is installed on the outer wall of the cutting blade, positioning columns are installed at the four corners of the partition, springs are movably sleeved on the outer surfaces of the lower ends of the positioning columns, so that the protective cover abuts against the partition and drives the partition to move downward when the cutting blade moves downward, so as to avoid metal chips generated by cutting from entering the top of the partition.

[0008] Preferably, a pneumatic dust removal module is connected to the protective cover, the pneumatic dust removal module blows away the metal chips after cutting is completed, so that the surface of the cutting blade is free of metal chips during image signal collection.

[0009] A method for monitoring and early warning of tool wear in cutting processing of zinc steel protective fence pipe, comprising the following steps: S1, a visual module, a matrix acoustic sensor, an alarm module and a device controller are arranged at a local end, and an image and vibration analysis model is built in a cloud end; S2, the visual module periodically collects images of the cutting blade through the device controller; S3, the analysis model analyzes the degree of tool wear according to the collected images, and the alarm module is started through the device controller when a certain feature reaches a threshold value by setting multiple feature thresholds; S4, the matrix acoustic sensor continuously monitors the cutting vibration signals of the cutting blade, and the device controller controls the visual module to detect immediately when the analysis model detects abnormal signals; S5, an alarm database is established according to abnormal vibration signals and abnormal images, which is used for training the analysis model.

[0010] Preferably, in step S, one set of vision modules acquires the blade image from the front end of the cutting blade; the other set of vision modules acquires the sidewall image from the side of the cutting blade, so as to extract multiple abnormal features in different directions.

[0011] Preferably, the abnormal features include blade breakage, blade deformation along the cutting direction, blade deformation perpendicular to the cutting direction, blade tip breakage, blade thickness wear, and blade length wear.

[0012] Preferably, in step S, the frame rate of the vision module is matched with the blade rotation speed, and the matching formula is as follows;

[0013] Where FPS is the frame rate of the vision module, rpm is the rotation speed of the cutting blade, S is the number of teeth of the cutting blade, and K is the redundancy coefficient, and K∈[1.2,1.5]. During the blade image acquisition process, the image acquisition range includes the image of the cutting blade exposed at the bottom of the protective cover. The number of shots in a single detection cycle is S, so as to realize the detection of blade deformation in the vertical cutting direction, blade tip breakage, blade thickness wear and blade length wear for each tooth. During the side wall image acquisition process, the image acquisition range includes the cutting blade exposed at the bottom of the protective cover. By tilting the vision module, the area of ​​the image acquired in a single acquisition is one-quarter of the cutting blade. Four images are taken within a single monitoring cycle, and then the images are stitched together to form a complete side image of the cutting blade.

[0014] Preferably, in step S4, the matrix acoustic sensor acquires cutting noise from multiple directions. In the analysis model, normal noise from cutting with different sized blades and different types of cutting materials is recorded in the external environment and in an environment that shields other mechanical noise as basic training samples. At the same time, noise from different wear types of cutting blades is recorded as comparison samples. After the matrix acoustic sensor acquires the actual samples, the target noise signal is extracted after filtering and noise reduction. Then, the analysis model is used for analysis and comparison to determine whether the cutting blade is worn and the type of wear.

[0015] Preferably, in step S5, the alarm database incorporates real image data of cutting blade wear and cutting noise data to improve the accuracy of subsequent analysis by the analysis model. At the same time, the alarm database manually adjusts the feature threshold range according to the actual cutting effect of the cut object under different wear levels to improve the utilization rate of the cutting blade.

[0016] The technical effects and advantages of this invention are as follows: 1. This online monitoring and early warning method for tool wear during the cutting and processing of zinc-steel guardrail pipes adds a noise monitoring scheme to the visual monitoring scheme. The visual module performs periodic checks to reduce its operating frequency, while an acoustic sensor performs continuous checks during each cut to ensure safe monitoring of abnormalities between visual inspection cycles. When the acoustic sensor detects an anomaly, the visual module is immediately activated for "emergency detection." This combined approach is more accurate and reliable than a pure noise monitoring scheme, and also lower in cost and workload for the analysis model. This allows for rapid early warning when the cutting blade malfunctions, preventing dangerous situations caused by excessive blade wear.

[0017] 2. The online monitoring and early warning method for tool wear during the cutting and processing of zinc-steel guardrail pipes sets noise abnormality thresholds and image abnormality thresholds respectively. The noise abnormality threshold is set higher so that it can be detected when the cutting blade shows a small degree of wear; the image abnormality threshold is set lower so that the cutting blade can continue to be used within a reasonable wear range, thereby improving the utilization rate of the cutting blade.

[0018] 3. The online monitoring and early warning device for tool wear during the cutting and processing of zinc steel guardrail pipes, by setting up a partition and a pneumatic dust removal module, allows the cylinder to drive the cutting blade downward during cutting, so that the protective cover comes into contact with the partition. At this time, the cutting blade is inserted into the through groove, preventing cutting metal chips from entering the top of the partition and causing damage to the vision module. At the same time, after cutting, the pneumatic dust removal module blows the cutting blade chips away, thereby preventing metal chips from interfering with image detection. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the overall outer surface of the online monitoring and early warning device of the present invention; Figure 2 This is a schematic diagram of the bottom structure of the online monitoring and early warning device of the present invention; Figure 3 This is a side view of the cutting blade of the present invention; Figure 4 This is a schematic diagram of the end face of the cutting blade of the present invention; Figure 5 This is a flowchart of the online monitoring and early warning method of the present invention.

[0020] In the diagram: 1. Vision module; 1-1. Blade breakage; 1-2. Blade deformation along the cutting direction; 1-3. Blade deformation perpendicular to the cutting direction; 1-4. Blade tip breakage; 1-5. Blade thickness wear; 1-6. Blade length wear; 2. Acoustic sensor; 3. Alarm module; 4. Partition; 42. Through slot; 5. Pneumatic dust removal module; 6. Cutting blade; 7. Protective cover; 8. Spring; 9. Positioning post. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Example 1: The present invention provides, as follows Figures 1 to 4 The device shown is an online monitoring and early warning device for tool wear during the cutting and machining of zinc-steel guardrail pipes, comprising: An image detection unit includes at least two vision modules 1, which perform appearance detection from the front and side respectively through the two vision modules 1. The noise detection unit includes multiple acoustic wave sensors 2, which are used to detect abnormal noise during the cutting process. The analysis module includes a device controller located locally and a cloud server. The cloud server contains an analysis model. After the device controller collects the monitoring data from the vision module 1 and the acoustic sensor 2, it transmits the data to the analysis model for feature analysis and then feeds the analysis results back to the device controller. Alarm module 3, which includes an alarm light and / or a buzzer, is used to issue early warnings based on the analysis results fed back by the analysis model.

[0023] It also includes a partition 4, on the top of which is a cutting blade 6 that can move up and down. The partition 4 has a through groove 42 for the blade to pass through. The vision module 1 is installed on the top of the partition 4. The acoustic sensors 2 are embedded in the partition 4 in a matrix so that when the blade moves to the bottom of the through groove 42 for cutting, the acoustic sensors 2 collect acoustic signals; when the blade moves to the top of the through groove 42, the vision module 1 collects image signals.

[0024] A protective cover 7 is installed on the outer wall of the cutting blade 6, and positioning posts 9 are installed at the four corners of the partition 4. A spring 8 is movably sleeved on the lower outer surface of the positioning post 9 so that when the cutting blade 6 moves down, the protective cover 7 abuts against the partition 4 and drives the partition 4 to move down, so as to prevent the metal chips generated by cutting from entering the top of the partition 4.

[0025] The protective cover 7 is connected to a pneumatic dust removal module 5. After cutting, the pneumatic dust removal module 5 blows away metal chips so that there is no metal chip interference on the surface of the cutting blade 6 when the image signal is acquired.

[0026] Working principle: When in use, the cutting blade 6 is installed inside the machine body and is driven to rise and fall by a cylinder. At the same time, the cutting blade 6 is connected to the servo motor via a belt. The protective cover 7 is installed outside the cutting blade 6. During cutting, the cylinder drives the cutting blade 6 to move downward, so that the protective cover 7 contacts the partition 4. At this time, the cutting blade 6 is inserted into the through groove 42. Then the cylinder continues to move downward, driving the partition 4 to move downward. At this time, the spring 8 is compressed. As the cutting blade 6 continues to move downward, it contacts the pipe and cuts it. At this time, the acoustic sensor 2 is activated to continuously monitor the cutting noise during the cutting process. Then the noise is preprocessed by the equipment controller and the signal is transmitted to the cloud analysis model. The analysis model performs online analysis to determine whether the cutting blade 6 is worn. After cutting is completed, the pneumatic dust removal module 5 is activated, blowing air downwards to clean the surface of the cutting blade 6. Then, the cylinder drives the cutting blade 6 to move upwards and reset. At this time, the servo motor speed drops to idle. The user can set the monitoring cycle of the vision module 1 according to the usage requirements. Optionally, the cycle can be set to the number of pipes cut (e.g., one detection every 10 pipes cut); optionally, the cycle can be set to time (e.g., one detection every 30 minutes); optionally, the cycle can be set to the number of rotations of the cutting blade 6 (e.g., one detection every 200,000 rotations of the blade, calculated based on the motor speed and working time). During detection, the two vision modules 1 acquire images of the end face and side face of the cutting blade 6 respectively. Then, the acquired images are preprocessed by the equipment controller and transmitted to the cloud analysis model. The analysis model performs online analysis to determine whether there is wear on the cutting blade 6. When wear is detected, the alarm module 3 is activated by the equipment controller to trigger an alarm. Furthermore, during use, when the acoustic signal from the acoustic sensor 2 is deemed abnormal, the vision module 1 will be activated immediately after the cutting blade 6 moves upward following the completion of this round of cutting to perform detection.

[0027] Example 2: The present invention provides, as follows Figures 1 to 5 The method for online monitoring and early warning of tool wear during the cutting and machining of zinc-steel guardrail pipes, as shown, includes the following steps: S1, Locally deployed visual module 1, matrix-style acoustic sensor 2, alarm module 3, and device controller, to build image and vibration analysis models in the cloud; S2, Vision module 1 periodically acquires images of the cutting blade 6 by setting the monitoring cycle through the device controller; S3. The analysis model analyzes the degree of tool damage based on the acquired images. By setting multiple feature thresholds, when a certain feature reaches the threshold, the alarm module 3 is activated by the device controller. S4. The matrix-type acoustic wave sensor 2 continuously monitors the cutting vibration signal of the cutting blade 6. When the analysis model detects an abnormal signal, the equipment controller immediately controls the vision module 1 to perform detection. In the initial state, the noise abnormality threshold is set higher and the image abnormality threshold is set lower, so that the cutting blade 6 can be judged and detected when it has a small degree of wear. The vision module 1 performs the detection. When the vision module 1 determines that the wear level is not suitable for the cutting blade to be replaced, it automatically increases the noise abnormality threshold (including the noise amplitude threshold and the noise frequency threshold) to prevent the acoustic wave sensor 2 from continuously triggering "emergency detection". This ensures that the cutting blade 6 can continue to be used within a reasonable wear level, improves the utilization rate of the cutting blade, and avoids the acoustic wave sensor 2 from missing monitoring and causing danger. S5. Establish an alarm database based on abnormal vibration signals and abnormal images for training and analysis models.

[0028] In step S1, one set of vision modules 1 acquires the blade image of the cutting blade 6 from the front end; another set of vision modules 1 acquires the sidewall image of the cutting blade 6 from the side, so as to extract multiple abnormal features in different directions.

[0029] Abnormal characteristics include blade breakage 1-1, blade deformation along the cutting direction 1-2, blade deformation perpendicular to the cutting direction 1-3, blade tip breakage 1-4, blade thickness wear 1-5, and blade length wear 1-6.

[0030] In step S1, the frame rate of vision module 1 is matched with the blade rotation speed, and the matching formula is as follows;

[0031] Where FPS is the frame rate of vision module 1, rpm is the rotation speed of cutting blade 6, S is the number of teeth of cutting blade 6, and K is the redundancy coefficient, and K∈[1.2,1.5]. During the blade image acquisition process, the image acquisition range includes the image of the cutting blade 6 exposed at the bottom of the protective cover 7. The number of shots in a single detection cycle is S, so as to realize the detection of blade vertical cutting direction deformation 1-3, blade end breakage 1-4, blade thickness wear 1-5 and blade length wear 1-6 for each tooth; During the side wall image acquisition process, the image acquisition range includes the cutting blade 6 exposed at the bottom of the protective cover 7. By tilting the vision module 1, the area of ​​the image acquired in a single acquisition is one-quarter of that of the cutting blade 6. Four images are captured in a single monitoring cycle, and then the images are stitched together to form a complete side image of the cutting blade 6.

[0032] In step S4, the matrix acoustic sensor 2 acquires cutting noise from multiple directions. In the analysis model, normal noise during cutting of different sizes of cutting blades 6 and different types of cutting objects is recorded in the external environment and in an environment that shields other mechanical noise as basic training samples. At the same time, noise of different wear types of cutting blades 6 is recorded as comparison samples. After the matrix acoustic sensor 2 acquires the actual samples, the target noise signal is extracted after filtering and noise reduction. Then, the analysis model is used for analysis and comparison to determine whether the cutting blade 6 has wear and the type of wear.

[0033] In step S5, the alarm database incorporates real image data of the wear of the cutting blade 6 and cutting noise data to improve the accuracy of subsequent analysis of the analysis model. At the same time, the alarm database manually adjusts the feature threshold range according to the actual cutting effect of the cutting object under different wear levels to improve the utilization rate of the cutting blade 6.

[0034] Working Principle: In this method, a noise monitoring scheme is added to the visual monitoring scheme. The visual module 1 performs periodic detection to reduce the operating frequency of the visual module 1. At the same time, the acoustic sensor 2 performs uninterrupted detection during each cut to ensure the safety of abnormal detection within the periodic interval of the visual detection. When the acoustic sensor 2 detects an anomaly, the visual module 1 is immediately activated for "emergency detection". By setting the initial noise anomaly threshold higher and the image anomaly threshold lower, the detection is performed when the cutting blade 6 shows a small degree of wear. When the visual module 1 determines that the wear level is not suitable for replacing the cutting blade, it automatically raises the noise anomaly threshold to prevent the acoustic sensor 2 from continuously triggering "emergency detection". This ensures that the cutting blade 6 can continue to be used within a reasonable wear level, improves the utilization rate of the cutting blade 6, and avoids the acoustic sensor 2 from missing detection and causing danger. Compared with pure noise monitoring, the visual monitoring solution is more accurate and reliable. Compared with pure visual monitoring, it is also less expensive and requires less work on the analysis model. This allows for rapid early warning when the cutting blade 6 malfunctions, preventing dangerous situations caused by excessive wear of the cutting blade 6, while avoiding the high cost burden of continuous visual monitoring.

[0035] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An online monitoring and early warning device for tool wear during the cutting and machining of zinc-coated steel guardrail pipes, characterized in that, include: An image detection unit comprising at least two vision modules (1) for performing appearance detection from the front and side, respectively, through the two vision modules (1); The noise detection unit includes multiple acoustic sensors (2) for detecting abnormal noise during the cutting process. The analysis module includes a device controller and a cloud server located locally. The cloud server contains an analysis model. The device controller collects the monitoring data of the vision module (1) and the acoustic sensor (2) and transmits it to the analysis model for feature analysis. The analysis results are then fed back to the device controller. An alarm module (3) includes an alarm light and / or a buzzer, used to issue an early warning based on the analysis results fed back by the analysis model.

2. The online monitoring and early warning device for tool wear during the cutting and machining of zinc-steel guardrail pipes according to claim 1, characterized in that, It also includes a partition (4), on the top of which is a cutting blade (6) that can move up and down. The partition (4) has a through groove (42) for the blade to pass through up and down. The vision module (1) is installed on the top of the partition (4). The acoustic sensors (2) are embedded in the partition (4) in a matrix, so that when the blade moves to the bottom of the through groove (42) to cut, the acoustic sensors (2) collect acoustic signals; when the blade moves to the top of the through groove (42), the vision module (1) collects image signals.

3. The online monitoring and early warning device for tool wear during the cutting and processing of zinc-steel guardrail pipes according to claim 2, characterized in that, The outer wall of the cutting blade (6) is fitted with a protective cover (7), and the four corners of the partition (4) are fitted with positioning posts (9). The lower outer surface of the positioning post (9) is movably fitted with a spring (8) so that when the cutting blade (6) moves down, the protective cover (7) abuts against the partition (4) and drives the partition (4) to move down, so as to prevent the metal chips generated by cutting from entering the top of the partition (4).

4. The online monitoring and early warning device for tool wear during the cutting and machining of zinc-steel guardrail pipes according to claim 3, characterized in that, The protective cover (7) is connected to a pneumatic dust removal module (5). After the cutting is completed, the pneumatic dust removal module (5) blows away metal chips pneumatically so that there is no metal chip interference on the surface of the cutting blade (6) when the image signal is acquired.

5. A method for online monitoring and early warning of tool wear during the cutting and machining of zinc-steel guardrail pipes, based on the online monitoring and early warning device for tool wear during the cutting and machining of zinc-steel guardrail pipes as described in any one of claims 1-4, characterized in that... Includes the following steps: S1. Locally arranged vision module (1), matrix acoustic wave sensor (2), alarm module (3) and device controller, and image and vibration analysis model are constructed in the cloud; S2, the vision module (1) periodically acquires images of the cutting blade (6) by setting the monitoring cycle through the equipment controller; S3. The analysis model analyzes the degree of tool damage based on the collected images. By setting multiple feature thresholds, when a certain feature reaches the threshold, the alarm module (3) is activated by the device controller. S4. The matrix-type acoustic sensor (2) continuously monitors the cutting vibration signal of the cutting blade (6). When the analysis model detects an abnormal signal, the equipment controller immediately controls the vision module (1) to perform detection. S5. Establish an alarm database based on abnormal vibration signals and abnormal images for training and analysis models.

6. The method for online monitoring and early warning of tool wear during the cutting and machining of zinc-steel guardrail pipes according to claim 5, characterized in that, In step S1, one set of vision modules (1) acquires the blade image of the cutting blade (6) from the front end; another set of vision modules (1) acquires the sidewall image of the cutting blade (6) from the side, so as to extract multiple abnormal features in different directions.

7. The method for online monitoring and early warning of tool wear during the cutting and machining of zinc-steel guardrail pipes according to claim 6, characterized in that, The abnormal features include blade breakage (1-1), blade deformation along the cutting direction (1-2), blade deformation perpendicular to the cutting direction (1-3), blade tip breakage (1-4), blade thickness wear (1-5), and blade length wear (1-6).

8. The method for online monitoring and early warning of tool wear during the cutting and machining of zinc-steel guardrail pipes according to claim 6, characterized in that, In step S1, the frame rate of the vision module (1) is matched with the blade rotation speed, and the matching formula is:

9. Where FPS is the frame rate of the vision module (1), rpm is the rotation speed of the cutting blade (6), S is the number of teeth of the cutting blade (6), K is the redundancy coefficient, and K∈[1.2,1.5]; During the blade image acquisition process, the image acquisition range includes the image of the cutting blade (6) exposed at the bottom of the protective cover (7). The number of shots in a single detection cycle is S, so as to realize the detection of blade deformation (1-3), blade end breakage (1-4), blade thickness wear (1-5), and blade length wear (1-6) for each tooth. During the side wall image acquisition process, the image acquisition range includes the cutting blade (6) exposed at the bottom of the protective cover (7). By tilting the vision module (1), the area of ​​the image acquired in a single acquisition is one-quarter of the cutting blade (6). Four shots are taken in a single monitoring cycle, and then the images are stitched together to form a complete side image of the cutting blade (6).

10. The method for online monitoring and early warning of tool wear during the cutting and machining of zinc-steel guardrail pipes according to claim 5, characterized in that, In step S4, the matrix acoustic sensor (2) acquires cutting noise from multiple directions. In the analysis model, normal noise from cutting blades (6) of different sizes and different types of cutting objects is recorded in the external environment and in an environment that shields other mechanical noise as basic training samples. At the same time, noise from different wear types of cutting blades (6) is recorded as comparison samples. After the matrix acoustic sensor (2) acquires the actual samples, the target noise signal is extracted after filtering and noise reduction. Then, the analysis model is used for analysis and comparison to determine whether the cutting blade (6) has wear and the type of wear.

11. The method for online monitoring and early warning of tool wear during the cutting and machining of zinc-steel guardrail pipes according to claim 5, characterized in that, In step S5, the alarm database improves the accuracy of subsequent analysis by adding real image data of the wear of the cutting blade (6) and cutting noise data. At the same time, the alarm database manually adjusts the feature threshold range according to the actual cutting effect of the cut object under different wear levels, thereby improving the utilization rate of the cutting blade (6).

Images