A cutting quality real-time monitoring method, electronic equipment and storage medium

Abstract

The application discloses a cutting quality real-time monitoring method, electronic equipment and storage medium, relates to the technical field of laser cutting, and comprises the following steps: target laser perforation cutting is monitored through a high-precision sensor, a feedback signal when target laser and material initial contact is acquired, and the feedback signal is analyzed according to a pre-cutting quality evaluation strategy, and a first cutting quality coefficient is obtained; a cutting signal time sequence is dynamically monitored through an intelligent monitor, cutting characteristic information is extracted, and the cutting characteristic information is input into a cutting quality evaluation model, and a second cutting quality coefficient is obtained; finally, the first and second cutting quality coefficients are averaged, a target cutting quality index is calculated, and the cutting quality index is used for representing a cutting quality condition. The technical problems that the laser cutting quality monitoring means is single and cannot be dynamically evaluated in real time in the prior art are solved, and the technical effects that the laser cutting quality can be dynamically monitored and accurately quantitatively evaluated in real time are achieved.

Description

Technical Field

[0001] This application relates to the field of laser cutting technology, and in particular to a method for real-time monitoring of cutting quality, an electronic device, and a storage medium. Background Technology

[0002] With the widespread application of laser processing technology, laser cutting, as a highly efficient and precise processing method, has been widely used in industries such as automotive, aerospace, and electronics. However, existing laser cutting quality monitoring technologies typically rely on single static monitoring methods, such as monitoring single parameters like laser power and gas pressure. This makes it difficult to comprehensively characterize the material adaptability before cutting and the dynamic quality characteristics during the cutting process. Consequently, defects that occur during cutting (such as rough kerfs, incomplete perforations, or uneven cutting) are difficult to detect and address in a timely manner. Especially in complex workpiece materials or high-speed cutting tasks, existing technologies cannot effectively monitor and evaluate the entire process before, during, and after cutting, and lack the ability to dynamically quantify and predict cutting quality. Therefore, how to achieve real-time dynamic monitoring and accurate evaluation of laser cutting quality has become a pressing technical challenge in the field of laser processing.

[0003] Currently, the relevant technologies suffer from the problem that laser cutting quality monitoring methods are limited and cannot be dynamically evaluated in real time. Summary of the Invention

[0004] This application provides a method, electronic device, and storage medium for real-time monitoring of cutting quality, thereby solving the technical problem that existing laser cutting quality monitoring methods are limited and cannot provide real-time dynamic evaluation.

[0005] This application provides a method for real-time monitoring of cutting quality, including:

[0006] High-precision sensors are used to monitor the laser perforation cutting of the target material, obtaining contact point feedback signals. These feedback signals refer to the initial contact signals between the target laser and the target material during the laser perforation cutting process. A pre-cutting quality assessment strategy is used to evaluate and analyze these contact point feedback signals, yielding a first cutting quality coefficient. A smart monitor dynamically monitors the timing of the laser perforation cutting signals, acquiring cutting characteristic information of this timing. This cutting characteristic information is used as input data for a quality assessment model during cutting, and output data, including a second cutting quality coefficient, is obtained. The average of the first and second cutting quality coefficients is taken as the target cutting quality index, which characterizes the cutting quality of the laser perforation cutting of the target material.

[0007] This application also provides an electronic device, including:

[0008] A memory is used to store executable instructions; a processor is used to implement a method for real-time monitoring of cutting quality when executing the executable instructions stored in the memory.

[0009] This application also provides a computer-readable storage medium, comprising:

[0010] It stores a computer program that, when executed by a processor, implements a method for real-time monitoring of cutting quality.

[0011] The proposed method, electronic device, and storage medium for real-time monitoring of cutting quality first utilize a high-precision sensor to monitor the target laser perforation cutting, acquiring the feedback signal at the initial contact between the target laser and the material. This feedback signal is then analyzed according to a pre-cutting quality assessment strategy to derive a first cutting quality coefficient. Next, an intelligent monitor dynamically monitors the timing of the cutting signal, extracting cutting feature information and inputting it into a quality assessment model during cutting to obtain a second cutting quality coefficient. Finally, the average of the first and second cutting quality coefficients is used to calculate the target cutting quality index, which characterizes the cutting quality. This achieves the technical effect of real-time dynamic monitoring and precise quantitative assessment of laser cutting quality. Attached Figure Description

[0012] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0013] Figure 1 A flowchart illustrating a real-time cutting quality monitoring method provided in this application embodiment;

[0014] Figure 2 This application provides a schematic diagram of the cutting signal timing construction process for a real-time cutting quality monitoring method according to an embodiment of the present application.

[0015] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0016] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below.

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application will be provided in conjunction with the accompanying drawings. The described embodiments should not be considered as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] In the following description, references to "some embodiments" describe a subset of all possible embodiments. However, it is understood that "some embodiments" can be the same or different subsets of all possible embodiments and can be combined with each other without conflict. The terms "first" and "second" are used merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only.

[0019] This application provides a method for real-time monitoring of cutting quality, such as... Figure 1 As shown, the method includes:

[0020] Step S100 involves monitoring the target laser perforation cutting using a high-precision sensor to obtain a contact point feedback signal. This contact point feedback signal refers to the feedback signal at the initial contact between the target laser and the target material during the laser perforation cutting process. Specifically, the high-precision sensor monitors the target laser perforation cutting process in real time, acquiring the feedback signal at the initial contact between the laser and the material. This signal reflects the state of laser energy transfer and material surface interaction. The feedback signal collected by the sensor includes laser reflection intensity, material heating temperature change, vibration response, etc. After filtering, noise reduction, and feature extraction, key parameters such as focus deviation and temperature change rate are generated. These parameters are used to evaluate whether the laser power, focus position, and auxiliary gas pressure meet preset requirements. Simultaneously, by comparing with the cutting conditions, real-time adjustments to the laser power, focus position, and auxiliary gas pressure are made to optimize the initial cutting conditions and ensure cutting stability and accuracy. Finally, this feedback signal generates a first cutting quality coefficient, providing a crucial reference for pre-cutting quality assessment and subsequent process monitoring.

[0021] Step S200: The pre-cutting quality assessment strategy is retrieved to evaluate and analyze the contact point feedback signal, resulting in a first cutting quality coefficient. Specifically, the pre-cutting quality assessment strategy is retrieved to analyze the contact point feedback signal, identifying predetermined pre-cutting factors such as laser power, focal point position deviation, gas pressure, gas purity, and laser cutting distance. The feedback signal is mapped to each factor, and relevant parameters are extracted, such as laser power stability, focal point offset, and instantaneous auxiliary gas pressure. These parameters are then subjected to variational weighting, assigning different weights based on material properties and actual working conditions to highlight the influence of key factors. Through comprehensive calculation, a first cutting quality coefficient is generated to characterize the quality level of the initial cutting conditions. The closer the quality coefficient is to the ideal state, the more thorough the pre-cutting preparation; a low value indicates potential problems such as focal point offset, power fluctuation, or abnormal gas parameters. Ultimately, this coefficient is used in the cutting control system to achieve real-time optimization of cutting parameters, providing stable and superior initial conditions for the subsequent cutting process.

[0022] In one possible implementation, the pre-cutting quality assessment strategy is retrieved to evaluate and analyze the feedback signal at the contact point, obtaining a first cutting quality coefficient. Step S200 further includes step S210, reading the predetermined pre-cutting factors in the pre-cutting quality assessment strategy. Specifically, the pre-cutting quality assessment strategy is first parsed, and the predetermined pre-cutting factors are extracted, including key parameters such as laser power, laser beam focal position deviation, gas pressure, gas purity, and laser cutting distance. These factors define the target value range for the cutting operation; for example, the laser power should be between 85% and 95%, the focal position deviation should not exceed ±0.2 mm, and the gas purity should be higher than 98.5%. Subsequently, the system matches these target values ​​with the actual parameters of the cutting equipment. By calling the equipment module to calibrate the laser power, adjust the focal position, detect the gas pressure and purity, and confirm the laser cutting distance, the system ensures that the equipment status meets the requirements of the assessment strategy. At the same time, the extracted factor target values ​​are structured and stored as a standardized factor list, providing a basis for subsequent data analysis and real-time monitoring. The target range is dynamically adjusted in conjunction with the cutting task to enhance the adaptability of the strategy. Ultimately, these factor target values ​​serve as input data for subsequent evaluation and analysis, supporting the generation of the first cutting quality coefficient and laying the foundation for cutting quality monitoring.

[0023] Step S220 involves iterating through the contact point feedback signal to obtain pre-cutting factor parameters. Specifically, the system uses a high-precision sensor to analyze the contact point feedback signal and sequentially extracts the pre-cutting factor parameters, including laser power, laser beam focal point position deviation, gas pressure, gas purity, and laser cutting distance. First, the system extracts laser energy data from the feedback signal, calculates the deviation between the actual power value and the predetermined power value, and marks abnormal intervals. Next, it analyzes the laser beam focal point position, quantifies the degree of focal point deviation using a positioning algorithm, and generates corresponding parameters. Subsequently, the system analyzes the gas pressure and purity, extracts actual parameter values ​​by combining gas flow rate, density, and spectral characteristics, and compares them with predetermined thresholds to screen out substandard cases. Finally, the system measures the laser cutting distance to confirm whether the straight-line distance between the laser and the material meets the requirements. Through iterative analysis, the system converts each factor in the contact point feedback signal into specific pre-cutting factor parameters, generating structured data containing actual values, deviation values, and abnormal markers, providing a reliable basis for subsequent cutting quality assessment and real-time control.

[0024] Step S230: Perform variation weighting processing on the pre-cutting factor parameters to obtain the first cutting quality coefficient. Specifically, after obtaining the pre-cutting factor parameters, the system calculates the first cutting quality coefficient through variation weighting processing. First, according to a predetermined pre-cutting quality assessment strategy, an initial weight is assigned to each factor (such as laser power, focal position deviation, gas pressure, gas purity, and laser cutting distance). Variation analysis is performed based on the deviation between the actual values ​​and the theoretical standard values, and the weights of each factor are dynamically adjusted. Factors with large variations have increased weights to highlight their impact on the assessment results, while factors with small variations have correspondingly decreased weights. Subsequently, the adjusted factor weights are weighted and calculated with the corresponding parameter values, and the total weighted value is accumulated. The result is then normalized to a value within the range of 0 to 1 to generate the first cutting quality coefficient. This coefficient directly reflects the overall quality level of the pre-cutting parameters, providing a scientific basis for the quality assessment of the subsequent cutting process.

[0025] Step S240, wherein the predetermined pre-cutting factors include laser power, laser beam focal point position deviation, gas pressure, gas purity, and laser cutting distance, and the laser cutting distance refers to the straight-line distance between the target laser and the target material. Specifically, in the pre-cutting quality assessment, the predetermined pre-cutting factors include laser power, laser beam focal point position deviation, gas pressure, gas purity, and laser cutting distance, which are crucial to the cutting quality. Laser power determines the efficiency of material melting by heat; insufficient power will lead to incomplete cutting, while excessive power may cause overheating. Laser beam focal point position deviation affects cutting accuracy; focal point deviation will lead to increased kerf width or unevenness. Gas pressure affects cutting speed and slag removal effect; insufficient pressure may lead to slag buildup, while excessive pressure will damage the cutting texture. Gas purity has a significant impact on chemical reactions; low purity may generate impurities or reduce the protective effect. Laser cutting distance directly affects laser energy distribution and cutting accuracy; excessive distance leads to energy attenuation, while insufficient distance may damage the equipment. The system collects these factor data through high-precision sensors and compares them with preset standard values ​​to comprehensively evaluate the state before cutting, providing a precise basis for subsequent quality optimization and the calculation of the first cutting quality coefficient.

[0026] Step S300 involves dynamically monitoring the cutting signal timing of the target laser perforation cutting using an intelligent monitor, and acquiring cutting characteristic information from the cutting signal timing. Specifically, during the laser perforation cutting process, the intelligent monitor dynamically monitors the cutting signal timing, collecting and recording various signal data in real time, such as laser power changes, plasma light radiation intensity, and arc signal intensity, fully reflecting the dynamic changes in the cutting process from start to finish. The intelligent monitor incorporates high-precision sensors and a data processing module to analyze and process the collected signal timing data, extracting key cutting characteristic information, such as the cutting speed change trend, laser power output fluctuations, plasma intensity peak values, and the stability of the cutting width signal. These characteristic information are optimized through noise removal using filtering algorithms and signal pattern capture using time series analysis to ensure their accuracy and representativeness, and multi-dimensional data are simultaneously analyzed to form a comprehensive cutting characteristic description. Finally, the extracted characteristic information is output in a standardized form as a key input to the subsequent cutting quality assessment model, providing a basis for real-time cutting quality assessment and process parameter optimization, effectively improving the quality and efficiency of laser perforation cutting.

[0027] In one possible implementation, such as Figure 2The process involves dynamically monitoring the cutting signal timing of the target laser perforation cutting using an intelligent monitor, and acquiring cutting characteristic information from the cutting signal timing. Step S300 further includes step S310, where the target laser perforation cutting is monitored using a plasma sensor in the intelligent monitor to obtain the first plasma intensity at the first moment. Specifically, during the laser perforation cutting process, the intelligent monitor uses a plasma sensor to monitor the plasma radiation signal generated when the laser first contacts the target material, obtaining the plasma intensity at the first moment. The spectral signal captured by the sensor is converted into numerical data through photoelectric conversion, representing the thermal effect and energy transfer efficiency of the laser-material interaction. This signal undergoes anti-interference transmission and filtering to ensure data integrity and accuracy, and is recorded in the cutting signal database. The intelligent monitor further analyzes the plasma intensity value to determine whether it is within a reasonable range. If it deviates from the normal value, it indicates a problem with the laser power, gas assistance, or the material surface. This value serves as the initial data point for the cutting signal timing, providing a crucial basis for subsequent cutting status monitoring and quality assessment.

[0028] Step S320: Extract the cutting control device group from the intelligent monitor, and monitor the predetermined cutting control indicators of the target laser perforation cutting at the first moment through the cutting control device group to obtain the first cutting control parameter group. Specifically, extracting the cutting control device group from the intelligent monitor and performing real-time monitoring is crucial during the target laser perforation cutting process. The cutting control device group includes multiple components such as a laser power controller, a motion platform controller, an auxiliary gas pressure regulator, and an optical focus adjustment system. These devices work in coordination to collect key control indicators of the cutting. At the first moment, each component records data such as laser power, cutting path movement speed, auxiliary gas pressure and flow rate, and laser focus position deviation, and transmits the collected monitoring data to the intelligent monitor through a communication network. After processing the data, the monitor generates the first cutting control parameter group, which includes real-time values, time stamps, and information related to the cutting status, providing an accurate data basis for subsequent cutting speed calculation and dynamic adjustment, ensuring that the cutting process is efficient and stable.

[0029] Step S330: Read the predetermined relative weight allocation of the predetermined cutting control indicators. Specifically, reading the predetermined relative weight allocation of the predetermined cutting control indicators is a crucial step in optimizing cutting quality during the target laser perforation cutting process. First, based on the different influences of indicators such as laser power, material type, kerf width, defocusing amount, and auxiliary gas pressure, the relative weights of each indicator are determined through experimental data, industry standards, and process requirements. For example, high-precision cutting may place greater emphasis on laser power and focal position, while thick plate cutting has a higher weight for auxiliary gas pressure. Next, the system extracts relevant records from the historical cutting database, analyzes the cutting quality under different conditions, and uses algorithms such as multi-factor regression analysis or principal component analysis to generate preliminary weight values. These values ​​are then adjusted in conjunction with the optimization objective to form the final weight allocation scheme. Finally, these weights are embedded in the subsequent calculation of cutting control parameters. By weighted calculation of key parameters such as cutting speed, the system comprehensively reflects the influence of different indicators on cutting quality, thereby ensuring that the evaluation and optimization are more in line with actual working conditions and improving cutting efficiency and quality consistency.

[0030] Step S340: The first cutting control parameter group is weighted according to the predetermined relative weight allocation to obtain the first cutting speed. Specifically, in real-time monitoring of target laser perforation cutting, determining the first cutting speed by weighting the first cutting control parameter group according to the predetermined relative weight allocation is a crucial step. The system first acquires the cutting control parameter group at the first moment in real time through the cutting control equipment group, including parameter values ​​such as laser power, material type, kerf width, defocusing amount, and auxiliary gas pressure. Subsequently, a pre-set weight allocation is extracted according to the cutting process requirements. This weight value is set according to the degree of influence of each parameter on the cutting quality. For example, thin plate cutting may focus more on laser power and focal position, while thick plate cutting focuses more on auxiliary gas pressure and material type. By multiplying each parameter value by its corresponding weight and summing them, the comprehensive first cutting speed is calculated. After the calculation is completed, the system verifies the result to ensure that the speed is within a reasonable range. If an anomaly is detected, an alarm will be triggered and the weights or parameter values ​​will be recalibrated. Ultimately, the first cutting speed integrates the influence of various cutting control indicators, providing accurate and reliable data support for subsequent cutting quality evaluation and optimization.

[0031] Step S350: The first cutting index is obtained by weighting the first plasma intensity and the first cutting speed. Specifically, calculating the first cutting index by weighting the first plasma intensity and the first cutting speed is a key step in real-time monitoring of cutting quality. The system first obtains the plasma intensity value at the first moment from the plasma sensor. This value reflects the energy release characteristics of the laser-material interaction and is closely related to the melting and penetration efficiency of the material. Simultaneously, the system calculates the first cutting speed based on the cutting control parameter set and preset weight allocation. This speed integrates multiple influencing factors such as laser power, kerf width, and material type. Subsequently, based on the cutting process requirements and material characteristics, the system sets weight coefficients for the plasma intensity and cutting speed, and calculates the first cutting index by combining the two using a weighted formula. This index reflects both the current cutting efficiency and quality and can be used for subsequent time-series analysis and process optimization. Finally, the system verifies and records the cutting index to ensure it is within a reasonable range. If an anomaly is detected, an alarm is triggered to adjust the parameters, and the data is stored in the database to provide a basis for subsequent cutting process optimization.

[0032] Step S360: Establish the cutting signal timing sequence based on the first correspondence between the first cutting index and the first time point. Specifically, establishing the cutting signal timing sequence based on the correspondence between the first cutting index and the first time point is a crucial step in real-time monitoring of cutting quality. The system first calculates the first cutting index and binds it to the corresponding first time point timestamp, forming data points reflecting the cutting state at a specific time. Then, the system uses time as the main axis, concatenating multiple data points to construct a complete cutting signal timing curve, dynamically displaying the changing trend of cutting quality with time as the horizontal axis and the cutting index as the vertical axis. During this process, the system verifies the completeness and accuracy of the data in real time, detecting and correcting abnormal fluctuations or discontinuities to ensure the reliability of the timing curve. The final generated cutting signal timing sequence is stored in the monitoring system, providing not only a basis for real-time monitoring but also serving as a data foundation for subsequent analysis and process optimization, used to identify the causes of quality fluctuations, adjust cutting parameters, and improve cutting quality and stability.

[0033] In one possible implementation, the cutting control device group in the intelligent monitor is extracted, and the predetermined cutting control indicators of the target laser perforation cutting at the first moment are monitored through the cutting control device group to obtain a first cutting control parameter group. Step S320 further includes step S321, whereby the predetermined cutting control indicators include laser power, material type, kerf width, defocusing amount, and auxiliary gas pressure. Specifically, the predetermined cutting control indicators include laser power, material type, kerf width, defocusing amount, and auxiliary gas pressure, which constitute a multi-dimensional dynamic control system for the cutting process. Laser power directly affects the laser energy transfer efficiency and material melting effect. The system adjusts the power in real time to ensure it is within the target range, avoiding overheating or incomplete cutting. Material type determines the matching of cutting parameters. The system adjusts the laser wavelength, power, and cutting speed according to the material characteristics to ensure process adaptability. The kerf width, as a precision indicator, is dynamically adjusted by comparing it with the target value in real time to ensure dimensional accuracy. The defocusing amount affects the laser energy concentration, which in turn affects the cutting depth and surface quality. The system adjusts the defocusing amount by monitoring the laser focal point position to maintain optimal cutting conditions. Assist gas pressure, by regulating the flow rate and pressure of the injected gas, removes slag, cools the cutting zone, and prevents oxidation, thereby improving cutting efficiency and cut quality. Real-time monitoring and optimization of these indicators provide a solid foundation for establishing cutting signal timing and evaluating cutting quality.

[0034] In one possible implementation, the predetermined relative weight allocation of the predetermined cutting control index is read. Step S330 further includes step S331, extracting the first historical cutting record from the historical laser cutting database. Specifically, to extract the first historical cutting record from the historical laser cutting database, a connection is first established with the database through the data extraction module to access complete historical data containing cutting process parameters, equipment settings, process monitoring data, and cutting quality results. Next, extraction rules are formulated based on filtering conditions (such as time period, material type, or equipment model), and cutting records that meet the requirements are filtered out through query statements or data interfaces. Subsequently, data cleaning is performed to remove missing values, outliers, or entries with inconsistent formats to ensure the integrity and accuracy of the data. The cleaned data is classified and organized, and the cutting control parameters (such as laser power, cutting speed, kerf width, etc.) are stored in a structured manner and timestamps or task identifiers are attached to improve traceability. Finally, the organized records are output in a standardized format (such as CSV or JSON files) or stored in a temporary database to provide high-quality data support for subsequent cutting parameter analysis and model optimization.

[0035] Step S332: Based on the predetermined cutting control indicators, the first historical cutting control parameter set is obtained by traversing the first historical cutting records. Specifically, to obtain the first historical cutting control parameter set by traversing the first historical cutting records based on the predetermined cutting control indicators, indicators such as laser power, material type, kerf width, defocusing amount, and auxiliary gas pressure are first identified, and their corresponding fields in the historical records are matched. The historical records are scanned one by one using a traversal algorithm or query program to extract parameter values ​​related to the predetermined indicators, and these values ​​are stored according to indicator categories. Simultaneously, the extracted data is preprocessed to remove outliers and fill in missing values, ensuring data integrity and accuracy. Timestamps, material batches, and other information are appended during the traversal process for further analysis. Finally, the extracted control parameters are organized into a structured first historical cutting control parameter set, laying the data foundation for subsequent correlation analysis and cutting quality optimization.

[0036] Step S333 involves performing a many-to-one correlation analysis between the first historical cutting control parameter group and the first historical cutting speed in the first historical cutting record to obtain the first correlation analysis result. Specifically, to perform the many-to-one correlation analysis between the first historical cutting control parameter group and the cutting speed in the first historical cutting record, the control parameter data (such as laser power, material type, kerf width, defocusing amount, and auxiliary gas pressure) and the corresponding cutting speed in the historical database are first cleaned to remove outliers and missing items to ensure data integrity. Subsequently, a correlation analysis method (such as Pearson correlation coefficient, multiple linear regression, etc.) is selected, with the control parameter group as the independent variable and the cutting speed as the dependent variable. A mathematical model is constructed to calculate the correlation coefficient or regression coefficient between each control parameter and the cutting speed, quantifying the influence of each parameter on the cutting speed. In the analysis, the accuracy and stability of the model are verified through cross-validation and residual evaluation, and the influence weight and contribution of each parameter are compiled to generate the first correlation analysis result, providing a reference for cutting control optimization.

[0037] Step S334: Analyze the results of the first correlation analysis and determine the predetermined relative weight allocation of the predetermined cutting control indicators. Specifically, after analyzing the results of the first correlation analysis, the correlation coefficients between each cutting control parameter (such as laser power, material type, kerf width, defocusing amount, and auxiliary gas pressure) and the cutting speed are interpreted to quantify the influence of each parameter on the cutting speed. These data are then sorted to determine the primary and secondary influencing parameters. Based on the actual controllability of the parameters, equipment limitations, and process requirements, corresponding weights are assigned. For example, high-correlation parameters such as laser power are given higher weights, while low-correlation parameters are given lower weights. Subsequently, the weights are standardized through weighted calculation or normalization to a sum of 1 (or 100%). To verify the rationality of the weights, they are applied to historical cutting cases for simulation evaluation, and the consistency between the simulation results and the actual effects is compared. Weights are adjusted as necessary. Finally, through multiple rounds of optimization, a scientific and reasonable predetermined cutting control indicator weight allocation scheme is generated, providing an accurate basis for cutting quality evaluation.

[0038] In one possible implementation, the cutting signal time series is established based on a first correspondence between the first cutting index and the first moment. Step S360 further includes step S361, establishing a plasma intensity time series based on a second correspondence between the first moment and the first plasma intensity. Specifically, the plasma sensor in the intelligent monitor collects the first plasma intensity at the first moment during the laser perforation cutting process and records the correspondence between this time point and the intensity value, serving as the basic data for constructing the plasma intensity time series. Subsequently, plasma intensity values ​​at multiple time points are continuously collected at different stages of the cutting process, and these data are arranged in chronological order to form a complete plasma intensity time series. During this process, the data is cleaned, including removing outliers, filling in missing data, and using a smoothing algorithm to optimize intensity fluctuations, ensuring the authenticity and completeness of the time series data. By establishing a second correspondence model between plasma intensity and time, a mathematical fitting method is used to describe the intensity change trend at different stages of the cutting process, ultimately generating time series data that reflects the dynamic characteristics of the plasma signal, providing a reliable basis for quality assessment and parameter optimization of the laser cutting process.

[0039] Step S362: Establish a cutting speed time series based on the third correspondence between the first time point and the first cutting speed. Specifically, during the laser cutting process, the cutting control equipment group collects and records the first cutting speed value at the first time point, establishes a third correspondence between this time point and the corresponding cutting speed value, and uses it as the initial data for the cutting speed time series. Subsequently, cutting speed data at multiple time points are continuously collected in real time, and a preliminary cutting speed time series is generated in chronological order. The data is then cleaned and preprocessed to remove outliers, fill in missing data, and use a smoothing algorithm to eliminate high-frequency noise. The data is then combined with cutting process parameters (such as laser power, material type, kerf width, etc.) to perform trend fitting, constructing a complete cutting speed time series that accurately reflects the dynamic changes in speed during the cutting process. This provides a reliable basis for quality assessment and parameter optimization, and also supports real-time monitoring and analysis of the cutting process.

[0040] Step S363: Obtain the target prediction time, and sequentially analyze the plasma intensity time series to obtain the predicted plasma intensity at the target prediction time, and analyze the cutting speed time series to obtain the predicted cutting speed at the target prediction time. Specifically, in real-time monitoring of laser cutting quality, the target prediction time is first determined as a key cutting parameter for predicting the target prediction time. This time can be dynamically generated based on the current cutting process or real-time condition changes. Subsequently, by analyzing the established plasma intensity time series, a time series model is used to fit and analyze the historical plasma intensity data, and the predicted plasma intensity at the target prediction time is extrapolated and calculated. Simultaneously, based on the cutting speed time series, methods such as linear regression or multinomial fitting are used to analyze historical cutting speed data, capture its dynamic change patterns, and predict the cutting speed value at the target time. Finally, the predicted plasma intensity and cutting speed values ​​are combined to provide a basis for the prediction and optimization of cutting quality, while supporting intelligent control and anomaly warning, improving the stability and accuracy of the cutting process.

[0041] Step S364: The predicted cutting index is obtained by weighting the predicted plasma intensity and the predicted cutting speed. Specifically, in the laser cutting quality prediction process, the predicted cutting index is obtained by weighting the predicted plasma intensity and the predicted cutting speed. First, the predicted plasma intensity and predicted cutting speed are extracted, reflecting the energy release characteristics of the laser and the material, and the dynamic performance of the cutting head movement speed, respectively. Next, weights are assigned to these two parameters according to the cutting process objectives. The weight allocation needs to be combined with actual needs; for example, cutting speed has a higher weight when efficiency is emphasized, while plasma intensity has a higher weight when precision is emphasized. Subsequently, these two parameters are normalized to eliminate dimensional differences, and they are weighted and summed using weighting factors to finally calculate the predicted cutting index. This index, as a comprehensive evaluation index, can characterize the overall performance of the cutting process in real time. A higher index indicates better cutting quality and efficiency; if the index is low, it suggests the need to optimize parameters such as laser power, cutting speed, or auxiliary gas pressure, providing a reliable basis for optimizing the cutting process and improving quality.

[0042] Step S365: Analyze the predicted cutting feature information of the predicted cutting signal timing using the cutting quality assessment model to obtain the predicted cutting quality coefficient. Specifically, in real-time monitoring of laser cutting quality, to obtain the predicted cutting index, the predicted plasma intensity and predicted cutting speed at the target prediction time need to be weighted. First, set weight values ​​according to the importance of their impact on cutting quality, and standardize the predicted plasma intensity and predicted cutting speed to ensure consistent data dimensions and avoid affecting calculation accuracy due to differences. Then, multiply the standardized predicted plasma intensity and predicted cutting speed by their corresponding weight values, and calculate using the formula: Predicted Cutting Index = Weight 1 × Predicted Plasma Intensity + Weight 2 × Predicted Cutting Speed; Predicted Cutting Index = Weight 1 × Predicted Plasma Intensity + Weight 2 × Predicted Cutting Speed; Predicted Cutting Index = Weight 1 × Predicted Plasma Intensity + Weight 2 × Predicted Cutting Speed, where the sum of the weight values ​​is 1. Ultimately, the predicted cutting index serves as a comprehensive evaluation indicator, reflecting the combined impact of plasma characteristics and cutting speed at the target prediction time on cutting performance. It provides a basis for cutting quality assessment and parameter optimization, and supports the early identification of quality problems and adjustment of cutting strategies, thereby improving cutting efficiency and quality stability.

[0043] Step S366: The average of the first cutting quality coefficient and the predicted cutting quality coefficient is taken as the predicted cutting quality index. The predicted cutting quality index characterizes the predicted cutting quality of the target laser perforation cutting at the predicted time. Specifically, to characterize the overall cutting quality of the target laser perforation cutting at the predicted time, the predicted cutting quality index needs to be calculated. First, the first cutting quality coefficient and the predicted cutting quality coefficient are obtained. The first cutting quality coefficient reflects the influence of pre-cutting factors (such as laser power, focal position deviation, gas pressure, etc.) on the initial quality state. The predicted cutting quality coefficient, based on plasma intensity time series, cutting speed time series, and other characteristics, is output by the quality assessment model during cutting, reflecting the dynamic quality state at the predicted time. Next, the average of the two is taken as the predicted cutting quality index, with the formula: Predicted cutting quality index = (First cutting quality coefficient + Predicted cutting quality coefficient) / 2. This calculation integrates the dual influences on quality before and during cutting, ensuring coverage of the dynamic characteristics of the entire cutting process. Ultimately, the predicted cutting quality index serves as a comprehensive indicator to intuitively reflect the cutting quality level, determine whether the expected quality requirements are met, and provide data support for parameter adjustment and process optimization, thereby improving the stability and efficiency of cutting quality.

[0044] Step S400: The cutting feature information is used as input data for the cutting quality assessment model, and output data is obtained through the cutting quality assessment model. The output data includes a second cutting quality coefficient. Specifically, the cutting feature information consists of time-domain and frequency-domain features extracted from the cutting signal time-series data, such as parameters like plasma intensity, cutting speed, and heat input fluctuations, which comprehensively reflect the dynamic changes and real-time status during the cutting process. The processed feature information is input into the cutting quality assessment model. This model combines machine learning technology and historical cutting data to perform deep learning and comprehensive analysis on the relationship between input features and cutting quality, such as identifying whether plasma intensity fluctuations are too large and whether the cutting speed matches the material type. The model calculates these influencing factors using internal algorithms to generate output data, including the second cutting quality coefficient. This coefficient quantitatively describes the real-time quality during the cutting process, such as edge quality, cutting depth consistency, and material melting degree. The second cutting quality coefficient, as a key indicator of dynamic quality in the cutting process, provides data support for subsequent quality judgment, parameter optimization, and process adjustment, ensuring a more accurate and efficient cutting process.

[0045] In one possible implementation, the cutting feature information is used as input data for a cutting quality assessment model, and output data is obtained through the cutting quality assessment model. The output data includes a second cutting quality coefficient. Step S400 further includes step S410, acquiring a cutting sample, which has an identifier for a sample cutting data group. Specifically, acquiring a cutting sample is a crucial step in the cutting quality assessment process, used to extract and analyze actual data from the cutting process. The cutting sample is a target material sample selected during laser cutting and assigned an independent sample cutting data group identifier for tracking and analysis. The data group identifier records the physical properties of the sample, cutting parameters, equipment information, and process conditions, ensuring that the data throughout the cutting process is bound to the specific sample. The system selects samples and generates identifiers based on preset rules or real-time conditions, used to store the time point, environmental conditions, and cutting parameters of the cutting process, while also supporting comparison with other samples or standard cutting results. By acquiring samples and data identifiers, basic data support is provided for extracting signal features, assessing cutting quality, and optimizing the process, and a reliable basis is laid for tracing anomalies.

[0046] Step S420: Sequentially collect the time-domain and frequency-domain features of the sample cutting signal sequence from the sample cutting data set, and assemble them into sample cutting feature information. Specifically, during the cutting quality assessment process, the signal timing of the sample cutting data set is sequentially collected and its time-domain and frequency-domain features are extracted to form complete sample cutting feature information. In the time-domain analysis, parameters such as signal amplitude, peak value, mean, and standard deviation are calculated to capture the time variation trend during the cutting process, such as the stability of laser output and vibration during the cutting process. Subsequently, the signal is converted to the frequency domain through Fourier transform, and features such as frequency components, main frequency peak, and frequency band energy distribution are extracted to identify high-frequency interference or low-frequency anomalies that may exist during the cutting process. After unified formatting, the extracted time-domain and frequency-domain features are integrated into structured cutting feature information, providing comprehensive and accurate input data for the subsequent cutting quality assessment model, and providing important support for optimizing the cutting process and diagnosing problems.

[0047] Step S430: Based on machine learning principles, the cutting quality coefficients of the specimen based on the specimen cutting feature information and the specimen cutting data set are trained to obtain the quality assessment model during cutting. Specifically, the quality assessment model during cutting is constructed based on machine learning principles. First, the specimen cutting feature information and the cutting quality coefficients in the specimen cutting data set are used as training data. The time-domain features (such as amplitude and mean) and frequency-domain features (such as dominant frequency and spectral energy distribution) of the specimen signal are used as model input variables, and the cutting quality coefficients are used as output targets. The input dataset is optimized through data preprocessing, denoising, and standardization, and a suitable data-driven algorithm (such as neural network or random forest) is selected for modeling. During training, the model parameters are adjusted through iterative optimization algorithms to minimize prediction errors, and cross-validation is used to improve the model's generalization ability. After training is completed, independent validation data is used to evaluate the model performance, and the model parameters are optimized based on error analysis and the coefficient of determination to ensure that the model can accurately predict the cutting quality coefficients. The final model can process cutting feature information in real time and output high-precision cutting quality coefficients, providing data support for dynamic monitoring and process optimization of the cutting process, and effectively improving cutting efficiency and quality stability.

[0048] Step S500: The average of the first cutting quality coefficient and the second cutting quality coefficient is taken as the target cutting quality index, wherein the target cutting quality index is used to characterize the cutting quality of the target laser perforation cutting. Specifically, the target cutting quality index is calculated by integrating the average of the first and second cutting quality coefficients, and is used to comprehensively characterize the overall quality of the target laser perforation cutting. The first cutting quality coefficient is based on the pre-cutting evaluation strategy and is obtained by analyzing the feedback signal of the contact point, reflecting the influence of factors such as laser power, focus deviation, and gas pressure on the cutting quality before cutting; the second cutting quality coefficient is calculated by dynamically monitoring the cutting signal timing through an intelligent monitor, extracting cutting feature information, and using a quality evaluation model during cutting, reflecting the real-time influence of actual operating parameters on quality during cutting. By calculating the average and integrating the quality information before and after cutting, the calculation results are ensured to be accurate and reliable, while providing a scientific basis for cutting process optimization. The target cutting quality index not only clearly reflects the overall cutting quality performance, but also indicates potential anomalies in the cutting process, providing an important reference for subsequent adjustment of process parameters and improvement of equipment, thereby improving cutting efficiency and the stability of finished product quality.

[0049] In one possible implementation, the average of the first cutting quality coefficient and the second cutting quality coefficient is taken as the target cutting quality index, wherein the target cutting quality index is used to characterize the cutting quality of the target laser perforation cutting. Step S500 further includes step S510, which monitors the target laser perforation cutting using a backlight reflection sensor to obtain the backlight reflection light signal timing sequence. Specifically, the target laser perforation cutting is monitored using a backlight reflection sensor to capture the light signal reflected from the surface of the target material in real time during the cutting process and form a complete backlight reflection light signal timing sequence. The sensor is installed in the cutting equipment and can highly sensitively sense the changes in the intensity of reflected light generated when the laser interacts with the material. These changes are affected by laser power, cutting speed, cutting depth, and material properties. The sensor continuously records the intensity of reflected light at each moment with a high sampling rate and transmits the data to the processing unit in real time. The data processing unit performs noise removal and anomaly correction on the signal to ensure the authenticity and accuracy of the data, and organizes and generates the backlight reflection light signal timing sequence in chronological order. This time series records the dynamic changes in reflected light intensity, providing basic data support for analyzing cutting quality, identifying cutting anomalies, and optimizing process parameters. This enables real-time monitoring and dynamic adjustment of the cutting process, improving cutting accuracy and efficiency.

[0050] Step S520: Extract the target backlight reflected light signal corresponding to the target time in the backlight reflected light signal time sequence, and obtain the target backlight reflected light signal intensity corresponding to the target backlight reflected light signal. Specifically, when analyzing the backlight reflected light signal time sequence, the target time is first extracted from the recorded data. This target time is determined based on the cutting process requirements or specific events (such as the initial contact between the laser and the material or a critical time point). By analyzing the light signal time sequence, the backlight reflected light signal corresponding to the target time is accurately located, and complete signal characteristics are extracted. During the extraction process, the system automatically filters noise interference or abnormal data to ensure signal accuracy. Subsequently, the intensity value of the target backlight reflected light signal is further analyzed and obtained to reflect the interaction state between the laser and the material at the target time, such as changes in the reflectivity of the material surface or the instantaneous influence of laser power. The extracted signal intensity not only provides an important basis for real-time evaluation of cutting quality but also provides data support for optimization and adjustment decisions in the cutting process, thereby improving cutting accuracy and efficiency.

[0051] Step S530: If the target backlight reflected light signal intensity is less than a predetermined perforation detection threshold, the total target cutting time is determined in conjunction with the target time. Specifically, during the cutting process, the perforation status is accurately determined by comparing the target backlight reflected light signal intensity with the predetermined perforation detection threshold. When the backlight reflected light signal intensity is detected to be lower than the threshold, the system determines that the laser has completed or is close to completing the perforation operation, and calculates the duration from the start of cutting to the signal intensity falling below the threshold in conjunction with the target time (the initial time point when the laser begins to act on the material), thereby determining the total target cutting time. The entire process relies on real-time monitoring and continuous sampling to ensure the integrity and accuracy of the signal data. At the same time, the signal fluctuations are smoothed through a dynamic data analysis module to eliminate environmental interference and noise. The determination of the total target cutting time not only provides a key time reference for cutting quality assessment but also lays an important data foundation for process optimization and efficiency analysis.

[0052] Step S540: Obtain the target cutting fitness based on the target total cutting time and the target cutting quality index. Specifically, during the cutting quality evaluation process, the target cutting fitness is calculated by comprehensively analyzing the target total cutting time and the target cutting quality index. The target total cutting time reflects material penetration efficiency and process stability; a shorter time indicates high cutting efficiency, while a longer time may indicate equipment performance issues or complex material properties. The target cutting quality index comprehensively quantifies multi-dimensional cutting effects such as cutting accuracy and edge smoothness. After inputting the two parameters into the fitness evaluation model, a weighted calculation is performed according to preset weights. Cutting time, which is correlated with efficiency, is assigned a negative weight, while the cutting quality index, which is correlated with accuracy, is assigned a positive weight, ultimately generating the target cutting fitness value. A high fitness score indicates good cutting efficiency and quality, and a reasonable process setting. If the score is low, adjustments to laser power, cutting speed, or auxiliary gas parameters are needed, along with optimization of equipment and material flow to improve the fitness level. The fitness results not only evaluate the current process but also provide optimization references for subsequent tasks.

[0053] This application embodiment employs a high-precision sensor to monitor the target laser perforation cutting, acquiring the feedback signal at the initial contact between the target laser and the material. This feedback signal is then analyzed according to a pre-cutting quality assessment strategy to derive a first cutting quality coefficient. A smart monitor dynamically monitors the timing of the cutting signal, extracts cutting feature information, and inputs it into a cutting quality assessment model to obtain a second cutting quality coefficient. Finally, the average of the first and second cutting quality coefficients is used to calculate the target cutting quality index, which characterizes the cutting quality. This achieves the technical effect of real-time dynamic monitoring and precise quantitative assessment of laser cutting quality.

[0054] Based on the foregoing embodiments, this application also provides an electronic device and a computer-readable storage medium storing a computer program. When the computer program is executed by the processor of the electronic device, it can implement the methods described in any of the preceding embodiments.

[0055] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention, showing a block diagram of an exemplary electronic device suitable for implementing the embodiments of the present invention. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of the present invention. This electronic device is in the form of a general-purpose computing device, and its components may include, but are not limited to, an input device 301, a processor 302, a memory 303, and an output device 304. The processor 302 may be one or more; the memory 303 may include a computer-readable medium and at least one program product having a set (at least one) of program modules configured to perform the functions of the embodiments of this application.

[0056] The memory 303 shown in this embodiment of the invention can be any combination of one or more computer-readable media. The computer-readable storage medium can be, but is not limited to, infrared, semiconductor systems, devices or components, or any combination thereof, used to store software programs, computer-executable programs and modules, such as the program instructions / modules corresponding to a real-time cutting quality monitoring method in this embodiment of the invention. The processor 302 executes various functional applications and data processing of the computer device by running the software programs, instructions and modules stored in the memory 303, thereby realizing the above-mentioned real-time cutting quality monitoring method.

[0057] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for real-time monitoring of cutting quality, characterized in that, include: The laser perforation cutting of the target is monitored by a high-precision sensor to obtain the contact point feedback signal. The contact point feedback signal refers to the feedback signal when the target laser and the target material make initial contact during the laser perforation cutting of the target. The pre-cutting quality assessment strategy is retrieved to evaluate and analyze the contact point feedback signal to obtain a first cutting quality coefficient. This includes: reading a predetermined pre-cutting factor from the pre-cutting quality assessment strategy; traversing the predetermined pre-cutting factor in the contact point feedback signal to obtain pre-cutting factor parameters; and performing a variation weighting process on the pre-cutting factor parameters to obtain the first cutting quality coefficient. The predetermined pre-cutting factor includes laser power, laser beam focal point position deviation, gas pressure, gas purity, and laser cutting distance, where the laser cutting distance refers to the straight-line distance between the target laser and the target material. The timing sequence of the laser perforation cutting of the target is dynamically monitored by an intelligent monitor, and the cutting characteristic information of the timing sequence is obtained, including: monitoring the laser perforation cutting of the target through a plasma sensor in the intelligent monitor to obtain a first plasma intensity at a first moment; extracting the cutting control device group in the intelligent monitor, and monitoring the predetermined cutting control index of the laser perforation cutting of the target at the first moment through the cutting control device group to obtain a first cutting control parameter group; reading the predetermined relative weight allocation of the predetermined cutting control index; performing a weighted calculation on the first cutting control parameter group according to the predetermined relative weight allocation to obtain a first cutting speed; weighting the first plasma intensity and the first cutting speed to obtain a first cutting index; and establishing the timing sequence of the cutting signal according to the first correspondence between the first cutting index and the first moment. The cutting feature information is used as input data for a cutting quality assessment model, and output data is obtained through the cutting quality assessment model. The output data includes a second cutting quality coefficient. The process includes: acquiring a cutting sample, the cutting sample having an identifier for a sample cutting data group; sequentially collecting the sample time-domain features and sample frequency-domain features of the sample cutting signal time sequence in the sample cutting data group, and forming sample cutting feature information; training the cutting quality coefficient based on the sample cutting feature information and the sample cutting data group based on machine learning principles to obtain the cutting quality assessment model. The average of the first cutting quality coefficient and the second cutting quality coefficient is taken as the target cutting quality index, wherein the target cutting quality index is used to characterize the cutting quality of the target laser perforation cutting.

2. The method as described in claim 1, characterized in that, The predetermined cutting control parameters include laser power, material type, kerf width, defocusing amount, and auxiliary gas pressure.

3. The method as described in claim 1, characterized in that, Reading the predetermined relative weight allocation of the predetermined cutting control index includes: Extract the first historical cutting record from the historical laser cutting database; The first historical cutting control parameter group is obtained by traversing the first historical cutting record based on the predetermined cutting control index. A many-to-one correlation analysis is performed between the first historical cutting control parameter group and the first historical cutting speed in the first historical cutting record to obtain the first correlation analysis result; Analyze the results of the first correlation analysis and determine the predetermined relative weight allocation of the predetermined cutting control index.

4. The method as described in claim 1, characterized in that, Also includes: Based on the second correspondence between the first time point and the first plasma intensity, a plasma intensity time series is established; Based on the third correspondence between the first moment and the first cutting speed, a cutting speed timing sequence is established; The target prediction time is obtained, and the predicted plasma intensity at the target prediction time is obtained by analyzing the plasma intensity time series in sequence, and the predicted cutting speed at the target prediction time is obtained by analyzing the cutting speed time series in sequence. The predicted cutting index is obtained by weighting the predicted plasma intensity and the predicted cutting speed. The predicted cutting quality coefficient is obtained by analyzing the predicted cutting feature information of the predicted cutting signal time sequence through the cutting quality assessment model. The average of the first cutting quality coefficient and the predicted cutting quality coefficient is taken as the predicted cutting quality index, wherein the predicted cutting quality index is used to characterize the predicted cutting quality of the target laser perforation cutting at the target prediction time.

5. The method as described in claim 1, characterized in that, Also includes: The laser perforation cutting of the target is monitored by a backlight reflection sensor to obtain the timing sequence of the backlight reflection light signal; Extract the target backlight reflected light signal corresponding to the target time in the backlight reflected light signal time sequence, and obtain the target backlight reflected light signal intensity corresponding to the target backlight reflected light signal; If the intensity of the target backlight reflected light signal is less than the predetermined perforation detection threshold, the total cutting time of the target is determined in conjunction with the target time. The target cutting adaptability is obtained based on the total target cutting time and the target cutting quality index.

6. An electronic device, characterized in that, The electronic device includes: Memory, used to store executable instructions; The processor, when executing executable instructions stored in the memory, implements the real-time cutting quality monitoring method according to any one of claims 1 to 5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements a real-time cutting quality monitoring method as described in any one of claims 1-5.

Images