A method for online detection of laser cutting parts replacement

By collecting laser cutting current timing data in real time and dividing cutting events, and making part replacement judgments based on the reference mode, the problems of untimely and inaccurate detection in the prior art are solved, and the precise detection of laser cutting switches is realized.

CN118518155BActive Publication Date: 2024-11-15SUXIN IOT TECH (NANJING) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410981042.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-11-15
Estimated Expiration
2044-07-22

AI Technical Summary

Technical Problem

The existing laser cutting equipment is not timely and accurately detected during the replacement process. Manual monitoring or image recognition methods have problems such as complex installation, difficulty in disassembly, and inconvenient maintenance, which affects the accuracy and timeliness of the detection.

Method used

By collecting the current timing data output from laser cutting in real time, dividing the cutting events, and judging the replacement conditions based on the reference events, selecting the judgment basic mode for comparison and judgment, and obtaining the replacement detection results in real time.

Benefits of technology

It realizes accurate judgment of laser cutting switches, improves the accuracy and timeliness of detection, can quickly respond to changes in the cutting process, and reduces the complexity and difficulty of detection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118518155B_ABST
    Figure CN118518155B_ABST
Patent Text Reader

Abstract

The present invention discloses an online detection method for part replacement in laser cutting. The method collects current time series data in the laser cutting process and divides the cutting events, obtains a number of cutting events and establishes a judgment benchmark, performs part replacement detection on the cutting events based on the judgment benchmark, and performs detection by comparing the differences between two groups of currents in three dimensions: duration, current mean and excess length. Finally, an accurate judgment result is given to the event of part replacement. In addition, for external influencing factors such as abnormal interruption and long-term shutdown, further supplementary part replacement detection is performed by recording historical judgment benchmarks and comparing them with newly generated judgment benchmarks, thereby effectively improving the accuracy of part replacement detection.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent welding and cutting, and in particular relates to an online detection method for laser cutting replacement parts. Background Art

[0002] Laser cutting refers to a technical means of cutting materials using a laser beam. It is often used in the field of industrial manufacturing and processing. The principle of laser cutting is to guide high-power laser output through optical devices, melt the material through the generated laser beam, and remove the irradiated part to complete the cutting of the material.

[0003] Existing laser cutting equipment usually adopts batch cutting to perform specific cutting on a certain workpiece. After completing this batch of workpieces, the workpieces are replaced and the next batch of different workpieces are cut. Under the condition of automatic loading and unloading, the switching of different workpieces represents different output powers and different cutting times, which is an important information to be monitored. The current monitoring method generally adopts manual monitoring or image recognition. Although it is equipped with online detection devices such as image sensors, these devices often have problems such as complex installation, difficult disassembly, and inconvenient maintenance. They cannot quickly respond to changes in the cutting process, affecting the accuracy and timeliness of detection. Summary of the invention

[0004] Purpose of the invention: In view of the problems existing in the above-mentioned background technology, the present invention provides an online detection method for laser cutting part replacement, which collects the cutting current and divides the cutting events, and makes cutting judgments based on the benchmark events. If the part replacement judgment conditions are met, it is judged that a part replacement has occurred.

[0005] Technical solution: A laser cutting replacement online detection method, comprising the following steps:

[0006] Step S1, collecting the current time series data output by the laser cutting in real time, and capturing the cutting event based on the current time series data;

[0007] Step S2: making judgments based on the captured cutting events and selecting a basic pattern for judgment;

[0008] Step S3: on the basis of obtaining the judgment basic pattern, the subsequently captured cutting events are compared with the pattern to obtain the replacement detection result in real time.

[0009] Furthermore, the specific method of capturing the cutting event in step S1 includes:

[0010] Set the current judgment threshold th1, take the first current point greater than th1 in the current time series data as the starting point of the first cutting event, and continuously record the current points until the last current point greater than th1 in the continuous current interval is found as the end point of the first cutting event; the set of all current points between the starting point and the end point is a complete cutting event, which is represented by current_list.

[0011] Furthermore, the specific method of selecting and determining the basic pattern in step S2 includes:

[0012] Capture the cutting event. When the pattern is not set, store the current sequence current_list of the cutting event in the pattern_list in sequence, where pattern_list represents a set of alternative events for setting the basic pattern for judgment; when the number of current_list sequences stored in pattern_list is equal to 3, perform pattern setting judgment: when the difference between the maximum cutting time and the minimum cutting time in the three current_lists is less than or equal to the preset threshold th8, select the current_list corresponding to the middle cutting time as the basic pattern for judgment, otherwise remove the first current_list in pattern_list, continue to capture cutting events, and perform pattern setting judgment; record the new pattern generated each time.

[0013] Furthermore, after obtaining the judgment benchmark pattern, if there is a recorded last judgment benchmark last_pattern at this time, a comparison is performed: when the difference in duration between pattern and last_pattern is less than or equal to th9, and the difference in current mean is less than or equal to th10, it means that no component replacement has occurred, otherwise it means that a component replacement has occurred; th9 and th10 are preset thresholds.

[0014] Furthermore, the specific method of obtaining the replacement detection result in step S3 includes:

[0015] Set a first counter count, a second counter current_mean_count and a third counter over_length_count;

[0016] Compare the captured current_list sequence with the judgment benchmark pattern in turn. The specific method is as follows:

[0017] (1) When the absolute value of the difference between the length of current_list and the length of pattern is greater than the preset threshold th2, the first counter count is incremented by 1, otherwise the counter count is reset to zero;

[0018] (2) When the difference between the current point mean in current_list and the current point mean in pattern is greater than the preset threshold th3, the second counter current_mean_count is incremented by 1, otherwise the current_mean_count counter is reset to zero;

[0019] (3) When the length of current_list exceeds the length of pattern and the excess length is greater than the preset threshold th4, the third counter over_length_count is incremented by 1, otherwise the over_length_count counter is cleared;

[0020] Continuously capture several cutting events and perform the above (1)-(3) judgments. When the conditions are met,

[0021] 1) count>th5

[0022] 2) current_mean_count>th6

[0023] 3) over_length_count>th7

[0024] When any of the conditions is met, it means that a replacement has occurred; at this time, pattern is set to None, and the first counter count, the second counter current_mean_count and the third counter over_length_count are all set to 0; the above th5, th6 and th7 are all preset thresholds.

[0025] Furthermore, a counter interval_time is set for recording the interval between adjacent cutting events. When interval_time is greater than a preset threshold idleTimes, pattern is set to None and pattern_list is cleared, and the first counter count, the second counter current_mean_count and the third counter over_length_count are all set to 0.

[0026] Compared with the prior art solutions, the technical solution adopted by the present invention has the following beneficial effects:

[0027] The present invention first establishes a judgment reference pattern for laser cutting events within a certain continuous time period, and performs part replacement detection on the newly captured cutting events based on the pattern. The detection is performed by comparing the differences between the two sets of current sets in three dimensions: duration, current mean, and excess length, and finally gives an accurate judgment result for the part replacement event. In addition, the present invention also performs corresponding supplementary processing for external influencing factors such as abnormal interruption and long-term stop. On the basis of obtaining the new pattern, it is compared with the recorded last_pattern to further determine whether there is a part replacement after a long-term interruption, so that the part replacement detection result is more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a flow chart of the online detection method for laser cutting replacement parts provided by the present invention. DETAILED DESCRIPTION

[0029] The present invention will be further explained below in conjunction with the accompanying drawings.

[0030] like Figure 1 As shown, the present invention provides an online detection method for component replacement based on laser cutting output current time series data, which divides several cutting events based on the laser cutting output current, selects the optimal cutting event current sequence as a judgment reference pattern, and judges subsequent cutting events based on this pattern. When the component replacement detection condition is met, it is judged that a component replacement has occurred. Specifically,

[0031] Step S1, dividing a number of cutting events according to the laser cutting output current timing data.

[0032] Set the current judgment threshold th1, find the first current point greater than th1 in the output current time series data, and use it as the starting point of the first cutting event. Record them continuously until the last current point greater than th1 in the continuous current interval is found. The continuous current interval is a complete cutting event, represented by current_list. The setting of th1 depends on the basic current of other devices besides the laser cutter.

[0033] Step S2: Capture the cutting event and set the basic pattern for judgment.

[0034] Cutting starts, and the cutting event is captured. When the pattern is not set, the current sequence current_list of the cutting event is stored in the pattern_list in sequence, where pattern_list represents a set of alternative events for setting the basic pattern for judgment. When the number of current_list sequences stored in pattern_list is equal to 3, a pattern setting judgment is performed: when the difference between the maximum cutting time and the minimum cutting time among the three current_lists is less than or equal to the preset threshold th8, the current_list corresponding to the middle cutting time is selected as the basic pattern for judgment, otherwise the first current_list in the pattern_list is removed, and the cutting event continues to be captured to perform pattern setting judgment. Among them, the difference between the maximum cutting time and the minimum cutting time is less than or equal to the preset threshold th8, which means that the length of the three current_lists has not changed significantly, so the current_list with the middle length is taken as the judgment reference pattern. The above-mentioned cutting time can be judged by the corresponding current_list set length. The longer the current_list is, the longer the cutting time is. Record the new pattern generated each time.

[0035] In the actual cutting process, due to various possible abnormal interruptions, work start and end and other external factors, when a new pattern is generated, it is necessary to compare it with the last recorded pattern, that is, last_pattern, to determine whether a part change has occurred before and after. The specific judgment method includes: when the time difference between pattern and last_pattern is less than or equal to th9, and the difference in current mean is less than or equal to th10, it means that the pattern is not much different from last_pattern, that is, the current cutting event is consistent with the workpiece in the previous continuous cutting event, and no part change has occurred, otherwise it means that a part change has occurred. Among them, th9 and th10 are preset thresholds.

[0036] Step S3: Perform a replacement detection on the newly captured cutting event current_list sequence according to the judgment reference pattern obtained in step S2. Specifically, a first counter count, a second counter current_mean_count and a third counter over_length_count are set.

[0037] Compare the captured current_list sequence with the judgment benchmark pattern in turn. The specific method is as follows:

[0038] (1) When the absolute value of the difference between the length of current_list and the length of pattern is greater than the preset threshold th2, it means that there is a significant difference between current_list and pattern in the dimension of duration. At this time, the first counter count representing the duration dimension is increased by 1, otherwise the counter count is reset to zero. In this embodiment, th2=5.

[0039] (2) When the difference between the current point mean in current_list and the current point mean in pattern is greater than the preset threshold th3, it means that there is a significant difference between current_list and pattern in the dimension of current mean. At this time, the counter current_mean_count representing the current mean dimension is increased by 1, otherwise the current_mean_count counter is reset to zero. In this embodiment, th3=0.5A.

[0040] (3) When the length of current_list exceeds the length of pattern, and the excess length is greater than the preset threshold th4, it means that the duration of current_list at this time obviously exceeds the length of pattern. At this time, the counter over_length_count indicating the excess length is increased by 1, otherwise the counter over_length_count is reset to zero. In this embodiment, th4=20.

[0041] Continuously capture cutting events and perform the above (1)-(3) judgments. When the conditions are met,

[0042] 1) count>th5

[0043] 2) current_mean_count>th6

[0044] 3) over_length_count>th7

[0045] When any of the above conditions is met, it means that a replacement has occurred. At this time, pattern is set to None, and the first counter count, the second counter current_mean_count and the third counter over_length_count are all set to 0. The above th5, th6 and th7 are all preset thresholds. In this embodiment, th5=5, th6=2, th7=2.

[0046] In this embodiment, a counter interval_time is also provided for recording the interval between adjacent cutting events. When interval_time is greater than the preset threshold idleTimes, it means that the interval between adjacent cutting events is too long, and there may be an abnormal interruption. At this time, the replacement detection of continuous cutting events will no longer be performed. The pattern should be set to None and the pattern_list should be cleared, and the first counter count, the second counter current_mean_count and the third counter over_length_count should all be set to 0.

[0047] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for online detection of laser cutting replacement parts. It is characterized in that The following steps are involved: Step S1, collecting the current time series data output by the laser cutting in real time, and capturing the cutting event based on the current time series data; Step S2: making judgments based on the captured cutting events and selecting a basic judgment template; The specific method of selecting and judging the basic template includes: Capture cutting events. When no basic judgment template is currently set, store the current sequence current_list of the cutting event in pattern_list in sequence, where pattern_list represents a set of candidate events for setting the basic judgment template; when the number of current_list sequences stored in pattern_list is equal to 3, perform judgment on the setting of the basic judgment template: when the difference between the maximum cutting time and the minimum cutting time in the three current_lists is less than or equal to the preset threshold th8, select the current_list corresponding to the middle cutting time as the basic judgment template, otherwise remove the first current_list in pattern_list, continue to capture cutting events, and perform template setting judgment; record the new basic judgment template generated each time; Step S3: on the basis of obtaining the basic judgment template, the subsequently captured cutting events are compared with the basic judgment template to obtain the replacement detection result in real time.

2. According to claim 1, a laser cutting replacement online detection method, It is characterized in that The specific method of capturing the cutting event in step S1 includes: Set the current judgment threshold th1, take the first current point in the current time series data that is greater than the current judgment threshold th1 as the starting point of the first cutting event, and continuously record the current points until the last current point in the continuous current interval that is greater than the current judgment threshold th1 is found as the end point of the first cutting event; the set of all current points between the starting point and the end point is a complete cutting event, which is represented by current_list.

3. According to claim 1, a laser cutting replacement online detection method, It is characterized in that After obtaining the judgment basic template, if there is a recorded judgment basic template last_pattern at this time, a comparison is performed: when the difference in time length between the judgment basic template and last_pattern is less than or equal to the preset threshold th9, and the difference in current mean is less than or equal to the preset threshold th10, it means that no component replacement has occurred, otherwise it means that a component replacement has occurred.

4. According to claim 1, a laser cutting replacement online detection method, It is characterized in that The specific method for obtaining the replacement detection result in step S3 includes: Set a first counter count, a second counter current_mean_count and a third counter over_length_count; Compare the captured current_list sequence with the judgment basic template in turn. The specific method is as follows: (1) When the absolute value of the difference between the length of current_list and the length of the basic template is greater than the preset threshold th2, the first counter count is incremented by 1, otherwise the first counter count is reset to zero; (2) When the difference between the current point mean in current_list and the current point mean in the judgment base template is greater than the preset threshold th3, the second counter current_mean_count is incremented by 1, otherwise the second counter current_mean_count is reset to zero; (3) When the length of current_list exceeds the length of the basic template, and the excess length is greater than the preset threshold th4, the third counter over_length_count is incremented by 1, otherwise the third counter over_length_count is reset to zero; Continuously capture several cutting events and perform the above (1)-(3) judgments. When the conditions are met, 1) count>preset threshold th5 2) current_mean_count>preset threshold th6 3) over_length_count>preset threshold th7 When any of the following conditions is met, it means that a replacement has occurred; at this time, the judgment base template is set to None, and the first counter count, the second counter current_mean_count and the third counter over_length_count are all set to 0.

5. The method for online detection of laser cutting replacement parts according to claim 1, It is characterized in that Set a counter interval_time for recording the intervals between adjacent cutting events. When interval_time is greater than a preset threshold idleTimes, set the basic template to None and clear pattern_list, and set the first counter count, the second counter current_mean_count and the third counter over_length_count to 0.