An automated monitoring method and system for laser cutting operations of metal sheets

Abstract

The present application discloses an automated monitoring method and system for metal sheet laser cutting operations, which relates to the technical field of laser cutting monitoring methods. The present application provides an automated detection method for metal sheet laser cutting operations, including the following operating steps: obtaining the cutting length error rate and laser volatility of the metal sheet within a preset time period; evaluating an anomaly score based on the cutting length error rate and laser volatility; if the anomaly score is greater than a preset threshold, obtaining the image information of the metal sheet; evaluating a cutting quality score based on the image information; if the cutting quality score is greater than a target score, outputting a misjudgment result of the anomaly score and tracing the cause of the misjudgment; if the cutting quality score is less than the target score, obtaining the type of cutting defect and adjusting the cutting strategy.

Description

Technical Field

[0001] The present application relates to the technical field of laser cutting monitoring methods, and particularly to an automatic monitoring method and system for laser cutting operations of metal sheets. Background Art

[0002] In the metal processing industry, laser cutting technology has been widely used due to its high precision, high efficiency, and good cutting quality. However, in the actual production process, due to various factors such as equipment aging, material property changes, and laser power fluctuations, some unforeseen errors often occur in laser cutting operations. These errors not only affect the dimensional accuracy of the cut parts but may also cause quality defects on the cutting surface. Traditionally, the detection of these cutting qualities mainly relies on manual visual inspection or post - sampling inspection, with low inspection efficiency and being greatly affected by human factors, making it difficult to achieve real - time and comprehensive quality control. Summary of the Invention

[0003] The main purpose of the present application is to provide an automatic monitoring method and system for laser cutting operations of metal sheets, aiming to solve the technical problems in the prior art that the quality detection of laser cutting of metal sheets mainly relies on manual visual inspection or post - sampling inspection, with low inspection efficiency and being greatly affected by human factors.

[0004] To achieve the above purpose, in the first aspect, the present application provides an automatic detection method for laser cutting operations of metal sheets, including the following operation steps:

[0005] Obtain the cutting length error rate and laser volatility within a preset time period of the metal sheet;

[0006] Evaluate the anomaly score according to the cutting length error rate and laser volatility;

[0007] If the anomaly score is greater than the preset threshold, obtain the image information of the metal sheet;

[0008] Evaluate the cutting quality score according to the image information;

[0009] If the cutting quality score is greater than the target score, output the misjudgment result of the anomaly score and trace the cause of the misjudgment;

[0010] If the cutting quality score is less than the target score, obtain the cutting defect type and adjust the cutting strategy.

[0011] Optionally, the step of obtaining the cutting length error rate and laser volatility within a preset time period of the metal sheet includes:

[0012] According to the actual spatial coordinates of the laser cutting head and the reference spatial coordinates at the corresponding moment, obtain the actual movement path and reference movement path of the laser cutting head within a preset time period;

[0013] Obtain the cutting length error rate according to the actual movement path and the reference movement path;

[0014] The calculation formula of the cutting length error rate is as follows:

[0015]

[0016] where d i represents the Euclidean distance between the actual spatial coordinates and the reference coordinates at the i-th moment, n is the total number of sampling points within the preset time period, L is the total length of the reference path within the preset time period, and w i is the weight factor.

[0017] Optionally, the step of obtaining the cutting length error rate and the laser volatility within the preset time period of the metal sheet further includes:

[0018] Obtain the average power and standard deviation of the laser beam within this time period according to the preset number of power output data collected within the preset time period;

[0019] Obtain the laser fluctuation power according to the average power and the standard deviation.

[0020] Optionally, the step of evaluating the anomaly score according to the cutting length error rate and the laser volatility includes:

[0021] Configure weight coefficients for the cutting length error rate and the laser volatility respectively, and obtain the anomaly score through weighted calculation;

[0022] If the anomaly score is less than the preset threshold, output that the cutting length error rate and the laser volatility are within the error range;

[0023] If the anomaly score is greater than the target threshold and less than the preset threshold, mark the preset time as the time period that needs to be observed with key attention.

[0024] Optionally, the step of evaluating the cutting quality score according to the image information includes:

[0025] Preprocess the front image information and the back image information according to the front image information and the back image information of the metal sheet located in the cutting area to obtain a grayscale image;

[0026] Extract the features of the cutting area according to the grayscale image, and obtain the cutting defect type and the defect degree;

[0027] Configure weight coefficients for the defect degree according to the cutting defect type, and perform weighted calculation on all obtained cutting defect types to obtain the cutting quality score.

[0028] Optionally, the step of extracting the features of the cutting area based on the grayscale image and obtaining the cutting defect type and defect degree includes:

[0029] The formula for obtaining the defect degree is as follows:

[0030]

[0031] where w j represents the weight of the j-th feature, and f j (·) represents the quantization function corresponding to the j-th feature, and the feature value ij is the j-th feature value of the defect type i.

[0032] Optionally, the step of configuring the weight coefficient of the defect degree according to the cutting defect type, performing weighted calculation on all obtained cutting defect types, and obtaining the cutting quality score includes:

[0033] The formula for the cutting quality score is as follows:

[0034]

[0035] where α i is the weight coefficient of the i-th defect type, and D i is the defect degree of the i-th defect type.

[0036] Optionally, the step of outputting the misjudgment result of the abnormal score and tracing the cause of misjudgment if the cutting quality score is greater than the target score includes:

[0037] If the cutting quality score is greater than the target score, output the misjudgment result of the abnormal score;

[0038] According to the misjudgment result of the abnormal score, obtain whether there is a corresponding historical case record;

[0039] If so, obtain the historical cutting information according to the historical case record and trace the cause of misjudgment;

[0040] If not, sequentially check the data acquisition device, cutting parameters, and process to trace the cause of misjudgment.

[0041] Optionally, the step of obtaining the cutting defect type and adjusting the cutting strategy if the cutting quality score is less than the target score includes:

[0042] If the cutting quality score is less than the target score, obtain the cutting defect type;

[0043] According to the cutting defect type, obtain the cause of the defect;

[0044] According to the cause of the defect, adjust the cutting parameters and / or optimize the cutting path.

[0045] In a second aspect, the present application provides an automated detection system for laser cutting operation of metal sheets, including:

[0046] An error rate and volatility acquisition module, which is configured to acquire the cutting length error rate and laser volatility of the metal sheet within a preset time period;

[0047] An abnormal score evaluation module, which is configured to evaluate an abnormal score according to the cutting length error rate and laser volatility;

[0048] An image information acquisition module, which is configured to acquire the image information of the metal sheet if the abnormal score is greater than a preset threshold;

[0049] A cutting quality score evaluation module, which is configured to evaluate a cutting quality score according to the image information;

[0050] A traceability module, which is configured to output a misjudgment result of the abnormal score and trace the reason for the misjudgment if the cutting quality score is greater than the target score;

[0051] A cutting strategy adjustment module, which is configured to acquire the cutting defect type and adjust the cutting strategy if the cutting quality score is less than the target score.

[0052] Beneficial effects that the present application can achieve:

[0053] An automated detection method and system for laser cutting operation of metal sheets proposed in the embodiments of the present application can more accurately identify potential problems in the cutting process by acquiring the cutting length error rate and laser volatility and evaluating the abnormal score. When the abnormal score exceeds the preset threshold, the image information of the metal sheet is further acquired, and image processing technology is used to evaluate the cutting quality score. This step not only improves the accuracy of detection but also realizes an intuitive evaluation of the cutting quality. If the cutting quality score is greater than the target score, a misjudgment result of the abnormal score is output and the reason for the misjudgment is traced, which helps to optimize the evaluation model of the abnormal score, reduce the misjudgment rate, and improve the reliability of the detection system. If the cutting quality score is less than the target score, the cutting strategy is adjusted by identifying the cutting defect type. This instant feedback and adjustment mechanism can significantly reduce the incidence of cutting defects, improve the product qualification rate and production efficiency. Description of the Drawings

[0054] Figure 1 It is a schematic flowchart of the automated monitoring method for laser cutting operation of metal sheets of the present application.

[0055] The realization, functional features, and advantages of the objectives of the present application will be further described with reference to the embodiments and the accompanying drawings. Detailed Embodiments

[0056] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0057] It should be noted that all directional indications (such as up, down, left, right, front, back...) in the embodiments of the present invention are only used to explain the relative position relationship and movement conditions between components in a specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indications will also change accordingly.

[0058] In the present invention, unless otherwise clearly defined and limited, terms such as "connection" and "fixation" should be understood in a broad sense. For example, "fixation" can be a fixed connection, a detachable connection, or integrated; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two components or the interaction relationship between two components, unless otherwise clearly limited. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0059] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of such features. In addition, the meaning of "and / or" appearing throughout the text includes three parallel solutions. Taking "A and / or B" as an example, it includes solution A, solution B, or a solution that satisfies both A and B at the same time. In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the fact that those of ordinary skill in the art can implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0060] Embodiment 1

[0061] Referring to Figure 1 , the first embodiment of the present application provides an automatic detection method for laser cutting operation of metal sheets, including the following operation steps:

[0062] S10. Obtain the cutting length error rate and laser volatility of the metal sheet within a preset time period.

[0063] Optionally, the spatial coordinate position of the metal sheet during the cutting process is monitored in real time through a high-precision positioning system (such as an optical locator or a mechanical locator), the movement path within a preset time period is obtained, and compared with the preset movement path, and the cutting length error rate is calculated. At the same time, through a laser power monitor or a laser stability analyzer, the output power fluctuation of the laser source is monitored in real time, and the laser volatility is calculated. The preset time period can be set according to actual needs, such as every minute, every five minutes, or after cutting a specific length.

[0064] S20. Evaluate the anomaly score according to the cutting length error rate and the laser volatility.

[0065] Optionally, the cutting length error rate refers to the deviation ratio between the actual cutting length and the preset cutting length, which reflects the accuracy and stability of the cutting operation. The cutting length error rate may be affected by various factors such as the accuracy of the cutting equipment, material properties (such as thickness, hardness), cutting speed, and laser power. A higher cutting length error rate means that the processed metal sheet may not meet the processing requirements, resulting in unqualified products or the need for additional processing to adjust the size. The laser volatility refers to the degree of fluctuation of the output power of the laser source during the cutting process, which reflects the stability and reliability of the laser cutting system. The laser volatility may be affected by factors such as laser source aging, unstable power supply, and poor cooling system performance. A higher laser volatility will lead to unstable cutting quality, and problems such as incomplete cutting, increased burrs, and enlarged heat affected zone may occur, affecting the appearance and performance of the cut parts. By evaluating the anomaly score, it is judged whether there is an abnormal condition in the cutting process.

[0066] S30. If the anomaly score is greater than the preset threshold, obtain the image information of the metal sheet.

[0067] Optionally, when the anomaly score is greater than the preset threshold, it indicates that there may be some faults, resulting in the quality of the processed metal sheet not meeting expectations. At this time, a high-resolution industrial camera or an infrared thermal imager and other image acquisition devices are used to take pictures or videos of the cut metal sheet to further judge whether the cutting quality meets expectations. Only when the anomaly score is greater than the preset threshold, the image information is obtained and the image information is analyzed, making rational use of resources. Through the image information, cutting defects can be more accurately identified, improving the accuracy of detection.

[0068] S40. Evaluate the cutting quality score according to the image information.

[0069] Optionally, use an image processing algorithm to process and analyze the acquired image information, and extract features such as the contour of the cutting edge and the texture of the cutting surface. According to the preset cutting quality evaluation criteria, quantify and score the extracted features to obtain a cutting quality score. The objective evaluation of cutting quality is achieved through image processing technology, reducing the influence of human factors. The cutting quality score provides an accurate basis for subsequent judgments and decisions.

[0070] S50. If the cutting quality score is greater than the target score, output the misjudgment result of the abnormal score and trace the cause of the misjudgment.

[0071] Optionally, when the cutting quality score is higher than the target score, it indicates that the cutting quality of the metal sheet meets the expectations and is judged as a misjudgment of the abnormal score. Review and analyze the measurement data of the cutting length error rate and laser volatility and the image processing results to find out the cause of the misjudgment. By tracing the cause of the misjudgment, the evaluation model and parameter settings can be optimized to improve the accuracy of detection. Reduce the occurrence of misjudgment situations and improve production efficiency.

[0072] S60. If the cutting quality score is less than the target score, obtain the cutting defect type and adjust the cutting strategy.

[0073] According to the image information and cutting quality evaluation criteria, identify the specific cutting defect types (such as burrs, cracks, incomplete cuts, etc.), and adjust the cutting parameters (such as laser power, cutting speed, focal length, etc.) or cutting path planning according to the defect types. By identifying the cutting defect types and adjusting the cutting strategy, the cutting quality can be significantly improved. Improve the product qualification rate and production efficiency, and reduce production costs.

[0074] Embodiment 2

[0075] Based on Embodiment 1, this embodiment provides an automatic detection method for laser cutting operations of metal sheets, including the following operating steps:

[0076] S10. Obtain the cutting length error rate and laser volatility of the metal sheet within a preset time period.

[0077] Optionally, the step of obtaining the cutting length error rate and laser volatility of the metal sheet within a preset time period includes:

[0078] S101. According to the actual spatial coordinates of the laser cutting head and the reference spatial coordinates at the corresponding moment, obtain the actual movement path and reference movement path of the laser cutting head within a preset time period.

[0079] Specifically, a high-precision positioning system (such as an optical positioner or a mechanical positioner) is used to record the actual spatial coordinates of the laser cutting head in a preset time period in real time. The preset time period can be set according to actual needs, such as every minute, every five minutes, or after cutting a specific length. According to the cutting operation plan or the preset cutting path, the reference spatial coordinates at the corresponding moment are obtained. The actual spatial coordinates and the reference spatial coordinates are time-synchronized to ensure that they correspond at the same time point. Based on the time-synchronized coordinate data, the actual movement path and the reference movement path of the laser cutting head are plotted. By recording the actual spatial coordinates of the laser cutting head in real time, the actual movement during the cutting process can be accurately reflected, and the introduction of the reference spatial coordinates provides a basis for calculating the cutting length error rate.

[0080] S102. Obtain the cutting length error rate according to the actual movement path and the reference movement path.

[0081] The calculation formula for the cutting length error rate is as follows:

[0082]

[0083] where d i represents the Euclidean distance between the actual spatial coordinates and the reference coordinates at the i-th moment, n is the total number of sampling points within the preset time period, L is the total length of the reference path within the preset time period, and w i is the weight factor. According to the importance of the cutting operation or specific requirements, a weight factor w i is assigned to each sampling point.

[0084] Specifically, the greater the deviation between the spatial coordinates d i at adjacent moments, the greater its contribution to the error rate. The greater the deviation between the spatial coordinates at adjacent moments indicates a greater degree of jumping of the laser cutting head and a larger error rate. It should be noted that this weighting method may make the error rate very sensitive to individual large deviations. Therefore, in practical applications, an appropriate weight factor w i needs to be selected according to the specific situation. The 1 / L in the formula is to normalize the error rate to the length of the entire reference path. represents the sum of the weighted Euclidean distances of all sampling points, reflecting the overall deviation between the actual movement path and the reference movement path. represents the sum of the weight factors of all sampling points, which is used to average the weighted Euclidean distance. By introducing the weight factor, different parts of the cutting process can be focused on to different degrees according to actual needs, improving the flexibility of the evaluation. By obtaining the cutting length error rate, a reliable data basis is provided for subsequent anomaly detection, cutting quality evaluation, and cutting strategy adjustment.

[0085] Calculate the Euclidean distance between the actual spatial coordinates and the reference coordinates at each sampling point within a preset time period. The formula for the Euclidean distance is as follows:

[0086]

[0087] Where, (x i , y i , z i ) represents the actual spatial coordinates of the laser cutting head at the i-th moment, and (x i ′ , y i ′ , z i ′ ) represents the reference spatial coordinates of the laser cutting head at the i-th moment.

[0088] Optionally, the step of obtaining the cutting length error rate and the laser fluctuation rate within a preset time period of the metal sheet further includes:

[0089] S103. Obtain the average power and standard deviation of the laser beam within this time period according to a preset number of power output data collected within the preset time period.

[0090] Specifically, within the preset time period, use a power monitor to collect the power output data of the laser source at a certain sampling frequency. Ensure that the number of collected power output data is large enough to represent the laser power change situation within this time period. Calculate the average value of the collected power output data to obtain the average power of the laser beam. Calculate the deviation between the collected power output data and the average power, and obtain the average value of the sum of the squares of these deviations, and then take the square root to obtain the standard deviation.

[0091] S104. Obtain the laser fluctuation power according to the average power and the standard deviation.

[0092] Specifically, the ratio of the standard deviation to the average power is used to represent the relative fluctuation degree of the laser power output. The laser fluctuation rate reflects the stability of the laser power output. The larger the fluctuation rate, the greater the fluctuation of the laser power output and the worse the stability. The laser fluctuation power provides an intuitive quantitative index for evaluating the stability of the laser beam during the laser cutting process. By monitoring the change of the laser fluctuation power, the abnormal fluctuation of the laser source can be detected in time, providing a basis for subsequent abnormal detection and processing.

[0093] S20. Evaluate the abnormal score according to the cutting length error rate and the laser fluctuation rate.

[0094] Optionally, the step of evaluating the abnormal score according to the cutting length error rate and the laser fluctuation rate includes:

[0095] S201. Configure weight coefficients for the cutting length error rate and the laser volatility respectively, and obtain the anomaly score through weighted calculation.

[0096] Specifically, according to the specific requirements and importance of the cutting operation, reasonable weight coefficients are configured for the cutting length error rate and the laser volatility respectively. The magnitude of the weight coefficient reflects the relative importance of the corresponding indicator in the calculation of the anomaly score. Use the weighted calculation formula, such as anomaly score = cutting length error rate × first weight coefficient + laser volatility × second weight coefficient, to calculate the anomaly score. The sum of the first weight coefficient and the second weight coefficient is equal to one. Ensure that the configuration of the weight coefficients and the weighted calculation process comply with the actual requirements and industry standards to ensure the accuracy and fairness of the evaluation. By configuring weight coefficients for the cutting length error rate and the laser volatility and calculating the anomaly score through weighting, the impact of the two indicators on the quality of the cutting operation can be comprehensively considered, improving the comprehensiveness and accuracy of the evaluation. The adjustability of the weight coefficients makes the evaluation method more flexible and can adapt to the changes in different cutting operation scenarios and requirements.

[0097] S202. If the anomaly score is less than the preset threshold, output that the cutting length error rate and the laser volatility are within the error range.

[0098] Specifically, set a reasonable preset threshold as the standard for judging whether the cutting length error rate and the laser volatility are within the acceptable range. When the anomaly score is less than the preset threshold, the system automatically outputs the information that the cutting length error rate and the laser volatility are within the error range, indicating that the current cutting operation quality meets the predetermined requirements, and the cutting can continue without intervention.

[0099] S203. If the anomaly score is greater than the target threshold and less than the preset threshold, mark the preset time as the period that needs to be observed with key emphasis.

[0100] Specifically, set a range between the preset threshold and the target threshold. This range is within the acceptable range but can be used as the period that needs to be observed with key emphasis. The target threshold is usually lower than the preset threshold and is used to give an early warning of possible abnormal situations. When the anomaly score falls within this range, the system automatically marks the current period as the period that needs to be observed with key emphasis and issues a corresponding prompt or alarm. The operator can, according to the prompt or alarm, or when the cutting quality decreases subsequently, conduct a more detailed review and analysis of the marked period to discover and solve potential problems in a timely manner. It can discover and give an early warning of possible abnormal situations in the cutting operation in advance, improving the predictability and initiative of detection.

[0101] S30. If the anomaly score is greater than the preset threshold, obtain the image information of the metal sheet.

[0102] S40. Evaluate the cutting quality score according to the image information.

[0103] Optionally, the step of evaluating the cutting quality score according to the image information includes:

[0104] S401. Preprocess the front image information and the back image information of the metal sheet in the cutting area according to the front image information and the back image information of the metal sheet, and obtain a grayscale image.

[0105] Specifically, read the front and back images of the metal sheet from the image acquisition device or the storage medium, convert the color image into a grayscale image. Usually, the weighted average method is used, such as the grayscale formula of the RGB image: Gray = 0.299*R + 0.587*G + 0.114*B. Apply a filter (such as a Gaussian filter) to remove the noise in the image, improve the accuracy of subsequent feature extraction, and normalize the grayscale value to the range of [0, 1] or [0, 255] for subsequent processing.

[0106] S402. Extract the features of the cutting area according to the grayscale image, and obtain the cutting defect type and the defect degree.

[0107] Specifically, use an edge detection algorithm (such as Canny edge detection) to identify the contour of the cutting area. The cutting area can be segmented from the background, and methods such as threshold segmentation and region growing can be used. Extract the geometric features (such as length, width, shape, etc.) and texture features (such as gray level co-occurrence matrix, wavelet transform, etc.) of the cutting area. Based on the extracted features, use a machine learning or deep learning model (such as SVM, CNN, etc.) to identify the type and degree of the cutting defect (such as the size and depth of the defect). Cutting defect types include: incomplete cutting defect, the cutting does not completely penetrate the material, and the gray value in the back image is significantly higher than that in the front image; burr defect, there are small and sharp protrusions on the cutting edge; smoothness defect, the cutting surface is uneven, with obvious texture or edge mutation, etc.

[0108] Partial defect detection, through a simplified Python code example as follows:

[0109]

[0110]

[0111]

[0112] Optionally, the step of extracting the features of the cutting area according to the grayscale image and obtaining the cutting defect type and the defect degree includes:

[0113] The formula for obtaining the defect degree is as follows:

[0114]

[0115] Among them, w j represents the weight of the j-th feature, reflecting the importance of this feature in evaluating the defect degree; f j (·) represents the quantization function corresponding to the j-th feature, and the feature value ij is the j-th feature value of the defect type i, and m represents the number of features.

[0116] The specific form of the quantization function corresponding to the j-th feature depends on the nature of this feature and its application in evaluating the cutting quality. Specific feature examples are as follows:

[0117] Gray value feature:

[0118]

[0119] Among them, the above formula is used to describe the overall brightness of the region, N represents the number of pixel points in the region, and the gray value i,j,k represents the gray value of the k-th pixel point under the j-th feature (i.e., the gray value feature) of the i-th defect type.

[0120] Gray value statistical feature:

[0121]

[0122] Among them, the above formula is used to describe the dispersion degree of the gray values in the region, and μ i,j represents the average value of the gray values in the region.

[0123] Texture feature:

[0124]

[0125] Among them, the above formula is used to describe the severity of the local gray value change in the image. L is the number of gray levels, and P(l,k) represents the element value in the l-th row and k-th column of the gray level co-occurrence matrix, representing the probability of the adjacent occurrence of the gray value l and the gray value k.

[0126] S403. According to the cutting defect type, configure the weight coefficient of the defect degree, and perform weighted calculation on all obtained cutting defect types to obtain the cutting quality score.

[0127] Optionally, the step of configuring the weight coefficient of the defect degree according to the cutting defect type, performing weighted calculation on all obtained cutting defect types, and obtaining the cutting quality score includes:

[0128] The formula for the cutting quality score is as follows:

[0129]

[0130] Among them, α i is the weight coefficient of the i-th defect type, D i is the degree of defect of the i-th defect type, and n is the number of defect types.

[0131] Configure reasonable weight coefficients for each cutting defect type to reflect its impact on the overall cutting quality. Analyze historical data or expert experience to understand the importance of different defect types to cutting quality. According to the severity, occurrence frequency, and impact on the performance of the final product of the defect, assign a weight coefficient α i to each defect type, ensuring that the sum of the weight coefficients is 1 to maintain consistency in subsequent calculations. According to the cutting quality score, the quality of the cutting operation can be objectively evaluated. Different score intervals can be set to correspond to different quality levels (such as excellent, good, qualified, unqualified, etc.).

[0132] S50. If the cutting quality score is greater than the target score, output the misjudgment result of the abnormal score and trace the cause of the misjudgment.

[0133] Optionally, the step of if the cutting quality score is greater than the target score, output the misjudgment result of the abnormal score and trace the cause of the misjudgment includes:

[0134] S501. If the cutting quality score is greater than the target score, output the misjudgment result of the abnormal score.

[0135] Specifically, if the cutting quality score is greater than the target score, that is, the quality is within the expected range, indicating that there is an error in the calculation of the abnormal score, then mark it as a misjudgment of the abnormal score and immediately output the misjudgment result.

[0136] S502. According to the misjudgment result of the abnormal score, obtain whether there is a corresponding historical case record.

[0137] Specifically, the system queries the historical case record library corresponding to the misjudgment result of the abnormal score to retrieve whether there are similar or identical misjudgment situations, including misjudgment types, reasons, and solutions. Use historical experience to quickly locate problems and reduce the troubleshooting time. It helps to identify repetitive errors and promote the formulation of continuous improvement and preventive measures.

[0138] S503. If there is a corresponding historical case record, obtain the historical cutting information according to the historical case record and trace the cause of the misjudgment.

[0139] Specifically, if there are corresponding historical cases, extract the historical cutting information in the cases, such as cutting materials, process parameters, equipment status, etc. Compare the similarities and differences between the current cutting operation and the historical cases, and analyze the reasons for misjudgment. Record and analyze the reasons for misjudgment to provide a basis for correcting the model or improving the process. Through learning from historical cases, improve the accuracy and efficiency of problem-solving. Promote knowledge accumulation and sharing, and reduce the risk of similar misjudgments in the future.

[0140] S504. If there is no corresponding historical case record, sequentially check the data acquisition equipment, cutting parameters, and processes to trace the reasons for misjudgment.

[0141] Specifically, if there is no corresponding historical case, start checking from the data acquisition equipment to check whether sensors, cameras, etc. are working properly and whether the data is accurate. Then check the cutting parameters, such as cutting speed, power, focal length, etc., to see if they are set within a reasonable range. Finally, check the cutting process, including cutting path planning, material handling, cooling method, etc., to see if it meets the specifications. Comprehensively and systematically check the possible reasons to ensure that the problem is fundamentally solved.

[0142] S60. If the cutting quality score is less than the target score, obtain the cutting defect type and adjust the cutting strategy.

[0143] The step of, if the cutting quality score is less than the target score, obtaining the cutting defect type and adjusting the cutting strategy includes:

[0144] Optionally, if the cutting quality score is less than the target score, obtain the cutting defect type;

[0145] S601. According to the cutting defect type, obtain the cause of the defect;

[0146] S602. According to the cause of the defect, adjust the cutting parameters and / or optimize the cutting path.

[0147] Specifically, according to the cutting defect types pointed out in the cutting quality assessment report (such as cracks, burrs, incomplete cutting, etc.), further analyze the causes of these defects. The cause analysis may involve multiple aspects such as cutting material properties, cutting equipment status, cutting parameter settings, cutting path planning, and operator skills. According to the analysis results of the cause of the defect, formulate corresponding improvement measures. If the defect is related to cutting parameters (such as cutting speed, power, focal length, etc.), adjust these parameters to a more appropriate range. If the defect is related to cutting path planning, optimize the cutting path, reduce unnecessary turns and overlaps, and improve cutting efficiency and quality. By precisely identifying the defect type and analyzing its cause, and then targeted adjusting of cutting parameters and optimizing the cutting path, the cutting quality can be significantly improved to meet the higher standards of customer requirements.

[0148] Embodiment 3

[0149] Based on Embodiment 1, this embodiment provides an automated detection system for laser cutting operations of metal sheets, including:

[0150] An error rate and volatility acquisition module, which is configured to acquire the cutting length error rate and laser volatility of the metal sheet within a preset time period;

[0151] An abnormal score evaluation module, which is configured to evaluate an abnormal score according to the cutting length error rate and laser volatility;

[0152] An image information acquisition module, which is configured to acquire the image information of the metal sheet if the abnormal score is greater than a preset threshold;

[0153] A cutting quality score evaluation module, which is configured to evaluate a cutting quality score according to the image information;

[0154] A traceability module, which is configured to output a misjudgment result of the abnormal score and trace the cause of the misjudgment if the cutting quality score is greater than the target score;

[0155] A cutting strategy adjustment module, which is configured to acquire the type of cutting defect and adjust the cutting strategy if the cutting quality score is less than the target score.

[0156] The above are only the preferred embodiments of the present application, and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made by using the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present application.

Claims

1. An automated detection method for laser cutting operation of metal sheets, characterized in that, Including: Obtain the cutting length error rate and laser volatility within a preset time period of the metal sheet; Evaluate the anomaly score according to the cutting length error rate and laser volatility; If the anomaly score is greater than the preset threshold, obtain the image information of the metal sheet; Evaluate the cutting quality score according to the image information; If the cutting quality score is greater than the target score, output the misjudgment result of the anomaly score and trace the cause of misjudgment; If the cutting quality score is less than the target score, obtain the cutting defect type and adjust the cutting strategy; The step of evaluating the cutting quality score according to the image information includes: According to the front image information and back image information of the metal sheet in the cutting area, preprocess the front image information and back image information to obtain a grayscale image; Extract the features of the cutting area according to the grayscale image, and obtain the cutting defect type and defect degree; Configure the weight coefficient of the defect degree according to the cutting defect type, and perform weighted calculation on all obtained cutting defect types to obtain the cutting quality score; The step of extracting the features of the cutting area according to the grayscale image and obtaining the cutting defect type and defect degree includes: The formula for obtaining the defect degree is as follows: Among them, represents the weight of the j-th feature, represents the quantization function corresponding to the j-th feature, is the j-th feature value of the defect type i.

2. The automated inspection method for laser cutting operation of metal sheets according to claim 1, wherein, The step of obtaining the cutting length error rate and laser volatility within a preset time period of the metal sheet includes: According to the actual spatial coordinates of the laser cutting head and the reference spatial coordinates at the corresponding moment, obtain the actual movement path and reference movement path of the laser cutting head within a preset time period; Obtain the cutting length error rate according to the actual movement path and reference movement path; The calculation formula of the cutting length error rate is as follows: Among them, represents the Euclidean distance between the actual spatial coordinates and the reference coordinates at the i-th moment, n is the total number of sampling points within the preset time period, and L is the total length of the reference path within the preset time period. is the weight factor.

3. The automated detection method for laser cutting operation of metal sheets according to claim 1, characterized in that, The step of obtaining the cutting length error rate and laser volatility within a preset time period of the metal sheet further includes: Obtain the average power and standard deviation of the laser beam within this time period according to the preset number of power output data collected within a preset time period; Obtain the laser fluctuation power according to the average power and standard deviation; 4. The automated detection method for metal sheet laser cutting operations according to claim 1, wherein The step of evaluating the anomaly score according to the cutting length error rate and laser volatility includes: Configure weight coefficients for the cutting length error rate and laser volatility respectively, and obtain the anomaly score through weighted calculation; If the anomaly score is less than the preset threshold, output that the cutting length error rate and laser volatility are within the error range; If the anomaly score is greater than the target threshold and less than the preset threshold, mark the preset time as the time period that needs to be observed key; 5. The automated detection method for laser cutting operation of metal sheets according to claim 1, characterized in that, The step of configuring the weight coefficient of the defect degree according to the cutting defect type and performing weighted calculation on all obtained cutting defect types to obtain the cutting quality score includes: The formula for the cutting quality score is as follows: Among them, is the weight coefficient of the i-th type of defect, is the degree of defect of the i-th type of defect.

6. The automated detection method for laser cutting operation of metal sheets according to claim 1, wherein The step of outputting the misjudgment result of the anomaly score and tracing the cause of misjudgment if the cutting quality score is greater than the target score includes: If the cutting quality score is greater than the target score, output the misjudgment result of the anomaly score; According to the misjudgment result of the anomaly score, obtain whether there is a corresponding historical case record; If so, obtain the historical cutting information according to the historical case record and trace the cause of misjudgment; If not, sequentially check the data acquisition device, cutting parameters and processes to trace the cause of misjudgment.

7. The automated detection method for laser cutting operation of metal sheets according to claim 1, wherein, The step of obtaining the cutting defect type and adjusting the cutting strategy if the cutting quality score is less than the target score includes: If the cutting quality score is less than the target score, obtain the cutting defect type; According to the cutting defect type, obtain the cause of the defect; According to the cause of the defect, adjust the cutting parameters and / or optimize the cutting path.

8. An automated detection system for laser cutting operations of metal sheets, characterized in that, It includes: An error rate and volatility acquisition module configured to acquire the cutting length error rate and laser volatility of the metal sheet within a preset time period; An abnormal score evaluation module configured to evaluate an abnormal score according to the cutting length error rate and laser volatility; An image information acquisition module configured to acquire the image information of the metal sheet if the abnormal score is greater than a preset threshold; A cutting quality score evaluation module configured to evaluate the cutting quality score according to the image information; A traceability module configured to output a misjudgment result of the abnormal score and trace the cause of the misjudgment if the cutting quality score is greater than the target score; A cutting strategy adjustment module configured to obtain the cutting defect type and adjust the cutting strategy if the cutting quality score is less than the target score; The step of evaluating the cutting quality score according to the image information includes: Preprocess the front image information and the back image information of the metal sheet in the cutting area according to the front image information and the back image information of the metal sheet in the cutting area to obtain a grayscale image; Extract the features of the cutting area according to the grayscale image to obtain the cutting defect type and the defect degree; Configure the weight coefficient of the defect degree according to the cutting defect type, and perform weighted calculation on all obtained cutting defect types to obtain the cutting quality score; The step of extracting the features of the cutting area according to the grayscale image to obtain the cutting defect type and the defect degree includes: The formula for obtaining the defect degree is as follows: Among them, represents the weight of the j-th feature, represents the quantization function corresponding to the j-th feature, is the j-th feature value of the defect type i.

Images