Processing pretreatment analysis system based on actual conditions of plates

Abstract

The invention relates to the technical field of metal plate processing, in particular to a processing pretreatment analysis system based on the actual condition of a plate, and the system comprises a micro-disturbance trial cutting execution module which is used for executing a micro-disturbance trial cutting action in at least one predetermined local area of the metal plate before formal cutting of the metal plate; the physical signal monitoring module is used for monitoring at least one physical signal in real time in the execution process of the micro-disturbance trial cutting action, and the physical signal represents the response characteristic of the metal plate to the micro-disturbance trial cutting action in the preset local area; the material risk assessment module is used for monitoring physical signals in real time; a cutting path and parameter adjusting module; by performing micro-disturbance trial cutting on the metal plate and monitoring the physical signal before formal cutting, the hidden local risk in the plate can be actively identified and evaluated, so that the method has the advantage that the hidden local risk in the plate can be actively and finely identified and evaluated.

Description

Technical Field

[0001] The present invention relates to the technical field of metal plate processing, and in particular to a processing preprocessing analysis system based on plate real conditions. Background Art

[0002] In the processing of metal sheets, especially in the laser cutting process, the sheets are usually pre-processed and analyzed before formal processing. The existing pre-processing analysis system mainly obtains the macroscopic geometric dimension information of the sheet through sensors, such as thickness, flatness, and some visible surface defect information. The system compares these macroscopic data with the standard model and preset process parameters, and adaptively adjusts the global or segmented parameters of the laser cutting to compensate for the macroscopic differences of the sheet and improve the processing quality. However, in actual production, metal sheets may form some local material anomalies inside due to the influence of their manufacturing or subsequent processing processes, which are difficult to effectively identify by conventional macroscopic detection methods, such as sudden changes in the mechanical properties (such as hardness and toughness) of local areas, or the presence of hidden microcrack initiation points. When the sheet is in a static state, the signal response of these internal anomalies may be very weak, making it difficult to be accurately captured by existing non-destructive testing equipment, or the signal strength does not reach the set alarm threshold.

[0003] Existing machining preprocessing analysis systems often assume that sheet material is uniform at the microscopic level when performing analysis and parameter adjustments. Therefore, when faced with sheet material whose macroscopic geometric parameters appear to meet requirements but harbor hidden localized material anomalies or potential defects, simply adjusting global or segmented machining parameters is difficult to fundamentally resolve machining issues caused by sudden changes in localized material properties or the expansion of potential internal defects under machining loads. This is especially true when machining parts with complex geometric features such as narrow ribs, sharp corners, and densely packed small holes. These localized material issues and potential defects can have a significant impact, easily leading to dimensional deviations, deformation, damage, or even scrap, severely impacting production efficiency and product qualification rates.

[0004] Therefore, the existing pre-processing system lacks the ability to actively and meticulously identify and evaluate these hidden local risks inside the plate. It also lacks a mechanism that can, based on these risk assessment results, instantly and dynamically adjust the laser cutting path or apply special processing parameters to specific local areas before or during formal processing, so as to actively avoid or effectively respond to these potential processing risks.

[0005] In view of the above problems, the existing technology is in urgent need of improvement. Summary of the Invention

[0006] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a processing pretreatment analysis system based on the actual situation of the plate.

[0007] The present invention provides a processing pretreatment analysis system based on the actual situation of plate materials, the system comprising:

[0008] A micro-perturbation trial cutting execution module is used to perform a micro-perturbation trial cutting action on at least one predetermined local area of ​​the metal sheet before formally cutting the metal sheet;

[0009] a physical signal monitoring module, configured to monitor in real time during the execution of the micro-disturbance trial cutting action at least one physical signal, wherein the physical signal represents a response characteristic of the metal sheet to the micro-disturbance trial cutting action in the predetermined local area;

[0010] A material risk assessment module is used to evaluate online whether there is a sudden change in the mechanical properties of the local material or a potential microcrack initiation point in the predetermined local area of ​​the metal sheet based on real-time monitored physical signals, and obtain an assessment result;

[0011] A cutting path and parameter adjustment module is used to adjust the laser cutting process parameters or the laser cutting path when the metal sheet is formally cut and the laser cutting path advances to the predetermined local area if the evaluation result indicates the existence of a sudden change in the mechanical properties of the local material or a potential microcrack initiation point.

[0012] This system effectively controls the quality of metal sheet laser cutting by integrating micro-perturbation test cutting, physical signal monitoring, material risk assessment, and cutting path and parameter adjustment into a single integrated system. Specifically, the micro-perturbation test cutting execution module performs a localized, low-energy test cut in a potential risk area identified based on the geometry of the part being processed, prior to the actual cutting. This process minimally impacts the sheet metal but is sufficient to elicit its true response characteristics in that area. The physical signal monitoring module simultaneously collects the physical signals generated during the test cutting, which directly reflect the sheet metal's mechanical properties or internal structural state at that microscopic scale. The material risk assessment module receives these real-time signals and performs online analysis based on a pre-defined model or algorithm to quickly determine whether the area contains localized material anomalies or potential defects. If the assessment indicates a risk, the cutting path and parameter adjustment module immediately intervenes. When the actual cutting process reaches the risk area, it automatically adjusts laser cutting process parameters (such as power, speed, frequency, focus position, and assist gas pressure) or modifies the cutting path, such as bypassing the risk area or adopting a special cutting strategy. This systematic integration approach enables efficient and stable data transmission and control instructions between functional modules, overcoming potential execution instability issues at the method level. This ensures real-time and accurate risk assessments and enables timely and reliable translation of assessment results into actual machining strategy adjustments, proactively addressing localized material risks in sheet metal before or during machining. In this way, the system effectively identifies and addresses hidden defects that are difficult to detect through conventional testing, significantly improving both the machining success rate for complex geometric features and the overall part qualification rate.

[0013] Compared with the prior art, the present invention has the following beneficial effects:

[0014] By performing micro-disturbance trial cutting on potential risk areas before formal cutting and monitoring the material response in real time, the local material properties and potential defects of the plate can be evaluated online, and based on the evaluation results, the cutting process or path can be dynamically adjusted when the formal cutting reaches the corresponding area. This solves the shortcomings of traditional methods that are difficult to identify hidden material problems and lack targeted local response strategies, and achieves the effect of improving the success rate of complex parts processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a system structure diagram of the present invention.

[0016] In the figure: 101, micro-disturbance trial cutting execution module; 102, physical signal monitoring module; 103, material risk assessment module; 104, cutting path and parameter adjustment module. DETAILED DESCRIPTION

[0017] The embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and are not to be construed as limiting the present invention.

[0018] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the specified features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0019] like Figure 1 A processing pre-processing analysis system based on the actual situation of the plate is shown, and the system includes:

[0020] The micro-perturbation trial cutting execution module 101 is used to perform a micro-perturbation trial cutting action on at least one predetermined local area of ​​the metal sheet before formally cutting the metal sheet;

[0021] The physical signal monitoring module 102 is used to monitor at least one physical signal in real time during the execution of the micro-disturbance test cutting action, where the physical signal represents the response characteristics of the metal sheet to the micro-disturbance test cutting action in a predetermined local area;

[0022] The material risk assessment module 103 is used to evaluate online whether there is a sudden change in the mechanical properties of the metal sheet or a potential microcrack initiation point in a predetermined local area based on the real-time monitored physical signals, and obtain an assessment result;

[0023] The cutting path and parameter adjustment module 104 is used to adjust the laser cutting process parameters or the laser cutting path when the metal sheet is formally cut and the laser cutting path moves to a predetermined local area if the evaluation result indicates that there is a sudden change in the mechanical properties of the local material or a potential microcrack initiation point.

[0024] The micro-perturbation test cutting execution module 101 is an execution unit capable of applying slight, controllable mechanical or energy perturbations to a metal sheet and precisely controlling the location and method of the perturbations. This can be achieved using a low-power laser scribing mechanism, micro-indentation device, or ultrasonic excitation device integrated into a laser cutting head. A predetermined local area refers to a specific small area on the geometric contour of the part to be processed that is predicted based on its shape characteristics and processing experience and is likely to cause quality problems during the laser cutting process due to uneven material properties. This area can include sharp corners, narrow ribs, areas with dense small holes, or complex curved segments. A potential risk area within the geometric features of the part to be processed refers to an area prone to processing defects, determined by analyzing the complexity of the part's geometry, the potential for stress concentration, or the sensitivity to thermal effects based on the part's CAD model or processing drawings. This area can be automatically identified through software algorithms or manually identified through empirical annotation. A physical signal monitoring module 102 is a device capable of real-time acquisition and conversion of physical signals into electrical or digital signals. This can be achieved using an acoustic emission sensor, vibration sensor, temperature sensor, optical sensor, or a combination thereof. Physical signals refer to measurable signals generated by the response of metal sheets to disturbances during micro-disturbance trial cutting, which may include acoustic emission signals, vibration signals, temperature change signals, reflected light or transmitted light signals, etc. The material risk assessment module 103 refers to a processing unit that can receive physical signal data and run a specific algorithm to analyze and judge the local material characteristics of the sheet. It can be implemented using an industrial computer, an embedded processor, or a dedicated signal processing chip. The cutting path and parameter adjustment module 104 refers to a unit that can modify the processing instructions or control parameters of the laser cutting equipment based on the risk assessment results. It can be implemented using a software module that interfaces with the laser cutting machine controller or an independent motion control card.

[0025] This system effectively controls the quality of laser cutting of metal sheets by integrating micro-perturbation test cutting, physical signal monitoring, material risk assessment, and cutting path and parameter adjustment into a coherent whole. Specifically, the micro-perturbation test cutting execution module 101 performs a localized, low-energy test cut in a potential risk area identified based on the geometric characteristics of the part being processed before the actual cutting. This process minimally impacts the sheet metal, but is sufficient to stimulate the sheet metal's true response characteristics in that area. The physical signal monitoring module 102 simultaneously collects the physical signals generated during the test cutting process, which directly reflect the sheet metal's mechanical properties or internal structural state at that microscopic scale. The material risk assessment module 103 receives these real-time signals and performs online analysis based on a pre-set model or algorithm to quickly determine whether the area contains localized material anomalies or potential defects. If the assessment indicates a risk, the cutting path and parameter adjustment module 104 immediately intervenes. When the actual cutting process reaches the risk area, it automatically adjusts laser cutting process parameters (such as power, speed, frequency, focus position, and assist gas pressure) or modifies the cutting path, for example, by bypassing the risk area or adopting a special cutting strategy. This systematic integration approach enables efficient and stable data transmission and control instructions between functional modules, overcoming potential execution instability issues at the method level. This ensures real-time and accurate risk assessments and enables timely and reliable translation of assessment results into actual machining strategy adjustments, proactively addressing localized material risks in sheet metal before or during machining. In this way, the system effectively identifies and addresses hidden defects that are difficult to detect through conventional testing, significantly improving both the machining success rate for complex geometric features and the overall part qualification rate.

[0026] As an embodiment of the present invention, the steps of online evaluating whether there is a sudden change in the local mechanical properties of a metal plate or a potential microcrack initiation point in a predetermined local area and obtaining the evaluation result include:

[0027] Extracting multiple signal features from real-time monitored physical signals;

[0028] Based on the combination of multiple signal features and the values ​​of each signal feature, the comparison is performed with the preset characteristic intervals corresponding to different types or severity of local material mechanical property mutations or potential microcrack initiation points to obtain a comparison result;

[0029] Based on the comparison results, the type or severity level of the local material mechanical property mutation or potential microcrack initiation point in the predetermined local area is determined, and the type or severity level is used as the evaluation result.

[0030] Multiple signal features refer to data or indicators of varying dimensions extracted from real-time monitored physical signals that characterize the material's response to micro-perturbation test cutting in a predetermined local region. These features may include, but are not limited to, signal amplitude, frequency, energy, waveform, and time-domain or frequency-domain statistics. Signal processing techniques can be used to extract these features. The combination of signal features and the numerical values ​​of each feature refer to arranging the extracted signal features in a specific structure or order to form a multidimensional feature vector, taking into account the specific numerical values ​​of each feature in the vector. This combination and numerical information together constitute a comprehensive description of the current material state. Predefined feature intervals corresponding to different types or severities of local material mechanical property mutations or potential microcrack initiation points refer to pre-demarcated regions with specific boundaries or ranges in the multidimensional signal feature space. Each such region is associated with a specific type of local material mechanical property mutation (e.g., abnormal hardness, decreased toughness) or potential microcrack initiation point (e.g., quenching cracks, stress concentration points caused by inclusions), or its severity level (e.g., mild, moderate, severe). These characteristic intervals can be determined based on extensive experimental data, theoretical models, or expert experience. Comparison refers to comparing the characteristic vector formed by the combination of the currently extracted signal features and the numerical values ​​of each signal feature with the aforementioned preset characteristic intervals to determine whether the characteristic vector falls within or is close to a specific characteristic interval. Comparison can be achieved using techniques such as pattern recognition, classification algorithms, or distance metrics. The comparison result refers to the output generated by the comparison process, which indicates which preset characteristic intervals the current characteristic vector is associated with or which characteristic interval it is closest to. Determining the type or severity level refers to identifying, based on the comparison results, the specific type (e.g., hardness mutation, microcracks) or severity level (e.g., mild hardness mutation, severe microcrack risk) of the local material mechanical property mutation or potential microcrack initiation point that most likely corresponds to the current material state. The type or severity level refers to a classification or hierarchical description of the material anomaly or potential defect in a predetermined local area. The type distinguishes different physical manifestations, and the severity level quantifies the extent to which the anomaly or defect may affect subsequent processing.

[0031] This system no longer relies solely on a single physical signal or simple threshold judgment. Instead, it comprehensively considers the combination of multiple signal features and their values, comparing them with pre-set characteristic intervals to more comprehensively reflect the material's true condition. First, multiple signal features are extracted from the real-time monitored physical signal, forming the basis for comprehensive evaluation. Different physical signal features may reflect different aspects of material properties. By extracting multiple signal features, more comprehensive material information can be obtained, avoiding the one-sidedness of a single signal. Second, based on the combination of multiple signal features and their values, they are compared with pre-set characteristic intervals corresponding to different types or severities of local material mechanical property changes or potential microcrack initiation points to obtain a comparison result. This means that this method goes beyond simply determining whether a signal exceeds a threshold. Instead, it combines multiple signal features to form a multidimensional feature vector, which is then compared with pre-set characteristic intervals. Different characteristic intervals correspond to defects of different types and severities. Through this comparison, the type and severity of the defect in the current material can be determined. This method considers the interplay between different signal features, enabling more accurate defect identification. For example, even if the value of a signal feature does not exceed the threshold, if it falls within a high-risk feature interval when combined with other signal features, it can still be judged as a defect. Finally, based on the comparison results, the type or severity level of the local material mechanical property mutation or potential microcrack initiation point in the predetermined local area is determined, and this type or severity level is used as the evaluation result. This means that the evaluation result is not just a simple "yes" or "no" but includes information on the type and severity of the defect. This information can provide more detailed guidance for subsequent cutting process adjustments. For example, if the evaluation results indicate a slight hardness mutation, the laser power can be appropriately increased; if the evaluation results indicate the presence of a serious microcrack initiation point, the cutting path may need to be adjusted to avoid this area. Through the above steps, this method can more accurately and comprehensively assess local material defects in metal sheets, providing a more reliable basis for subsequent cutting process adjustments, thereby improving cutting quality and efficiency. This assessment method, combined with micro-perturbation test cutting and subsequent process adjustments, enables the entire processing pre-processing analysis method to make targeted processing adjustments based on the actual local characteristics of the plate, improving the ability to identify and respond to hidden defects, thereby resolving the problem of conventional systems being unable to handle internal material anomalies that lead to processing failures. This refined assessment, combined with subsequent local process adjustments, is the key to solving the problem. Compared to simple threshold judgment or single signal analysis, the use of multi-feature combination and multi-category / multi-level feature interval comparison can more accurately distinguish local material anomalies of different nature and severity, providing richer and more reliable evaluation information.This refined evaluation result can provide a more guiding basis for subsequent local process parameter or path adjustments, making the adjustments more targeted and effective, thereby improving the processing success rate and quality of plates with hidden local defects.

[0032] As an embodiment of the present invention, the steps of obtaining a comparison result include comparing a combination of multiple signal features and the values ​​of each signal feature with preset characteristic intervals corresponding to local material mechanical property mutations or potential microcrack initiation points of different types or severities, and comprising:

[0033] Obtaining reference signal characteristics of the current batch of metal sheets to be processed at one or more reference positions, and determining reference signal characteristic parameters of the current batch of metal sheets;

[0034] According to the determined reference signal characteristic parameters, the preset characteristic intervals corresponding to local material mechanical property mutations or potential microcrack initiation points of different types or severity are adjusted to obtain the characteristic intervals adjusted by the batch characteristics;

[0035] The combination of multiple signal features and the value of each signal feature are compared with the feature interval adjusted by batch characteristics to obtain a comparison result.

[0036] Reference signal characteristics are obtained for the current batch of metal sheets to be processed at one or more reference locations. Reference locations can be areas of the selected sheet known to be uniform and defect-free, such as the edge or center of the sheet. Reference signal characteristics can be features extracted from the physical signals monitored during micro-perturbation test cutting at these reference locations, such as signal amplitude, frequency, energy, or waveform characteristics. Baseline signal characteristic parameters are determined for the current batch of metal sheets. These parameters can be obtained by statistically analyzing the reference signal characteristics obtained at the reference locations, such as the mean, median, standard deviation, or the intensity of specific frequency components. These parameters characterize the signal characteristics of the current batch of metal sheets within the "normal" region. Preset characteristic intervals are adjusted. Adjustment can refer to modifying the boundaries, center position, or width of the preset characteristic intervals based on the determined baseline signal characteristic parameters. The adjustment can be based on a preset functional relationship, a lookup table, or a model trained using historical data. The batch-characteristic-adjusted characteristic interval refers to the characteristic interval that, after the aforementioned adjustment process, is more suitable for evaluating the current batch of sheets.

[0037] This solution addresses the potential material variations between batches of metal sheets by proposing an evaluation method that dynamically adjusts feature intervals, thereby improving the accuracy and reliability of the evaluation. First, by acquiring reference signal features at one or more reference locations for the current batch of metal sheets to be processed, the overall characteristics of the batch are understood. Then, based on these reference signal features, baseline signal characteristic parameters for the batch, such as the signal mean and variance, are determined. Next, based on the determined baseline signal characteristic parameters, preset feature intervals corresponding to different types or severities of local material mechanical property mutations or potential microcrack initiation points are adjusted to produce batch-adjusted feature intervals. This step is the core of this solution. By adjusting the feature intervals, they are tailored to the characteristics of the current batch of sheets, thereby reducing misjudgments caused by batch variations. For example, if the overall signal of the current batch of sheets is too high, the feature intervals can be shifted upward, and vice versa. Finally, the combination of multiple signal features and the values ​​of each signal feature are compared with the batch-adjusted feature intervals to obtain a comparison result. Because the feature intervals have been adjusted to the characteristics of the current batch of sheets, the comparison results are more accurate and reliable. By introducing batch characteristic adjustment under the basic framework of micro-disturbance trial cutting, signal monitoring and online evaluation, the characteristic interval used for evaluation can dynamically adapt to the actual situation of the current batch, significantly improving the accuracy of evaluating local material anomalies when there are batch differences, thereby better guiding subsequent cutting parameters or path adjustments and reducing the risk of processing failure due to material differences.

[0038] As an embodiment of the present invention, the steps of adjusting preset characteristic intervals corresponding to local material mechanical property mutations or potential microcrack initiation points of different types or severities based on the determined reference signal characteristic parameters to obtain the characteristic intervals adjusted for batch characteristics include:

[0039] For each preset characteristic interval corresponding to different types or severity of local material mechanical property mutations or potential microcrack initiation points, an adjustment rule is determined for each characteristic interval. The adjustment rule defines the change mode of the characteristic parameter of the reference signal determined in response to the characteristic interval, and the adjustment rule corresponding to each characteristic interval is obtained;

[0040] According to the determined reference signal characteristic parameters and the adjustment rules corresponding to each characteristic interval, an adjustment parameter is calculated for each preset characteristic interval, and the adjustment parameter is applied to adjust the corresponding preset characteristic interval to obtain each characteristic interval adjusted for batch characteristics.

[0041] Among them, the characteristic interval refers to a pre-set numerical range, which is used to define whether the characteristic value extracted from the physical signal indicates the presence of a local material anomaly or potential microcrack of a specific type or a specific severity. These intervals can be calculated based on the analysis of known defect samples or theoretical models. The adjustment rule refers to a preset logical relationship or calculation model that describes how a specific characteristic interval should be numerically corrected according to the overall characteristics of the plate batch (represented by the reference signal characteristic parameters). The adjustment rule can be expressed as a function, a lookup table or a series of conditional judgments. The reference signal characteristic parameter refers to one or more values ​​extracted from the physical signal collected from the reference position of the current batch of metal plates, which can represent the overall material characteristics of the batch of plates. The adjustment parameter refers to a specific value or correction factor calculated based on the determined reference signal characteristic parameters and the corresponding adjustment rule, which is used to adjust the upper and lower limits or range of the corresponding preset characteristic interval.

[0042] When performing material risk assessment on a predetermined local area of ​​a metal sheet, this solution first predefines adjustment rules for each predefined characteristic interval, corresponding to different types or severities of localized material mechanical property changes or potential microcrack initiation points. These adjustment rules are not applied uniformly; rather, they are established for each characteristic interval corresponding to a specific defect type or severity level. This allows for a more detailed description of how each characteristic interval should respond to changes in the overall material properties of the sheet batch. For example, a characteristic interval associated with hardness anomalies may primarily respond to changes in a baseline parameter for overall sheet hardness, while a characteristic interval associated with internal stress may primarily respond to changes in a baseline parameter for internal residual stress. These adjustment rules define how a characteristic interval changes in response to the baseline signal characteristic parameters. During the actual assessment process, after obtaining the baseline signal characteristic parameters for the current batch of metal sheets to be processed, the system calculates a specific adjustment parameter for each predefined characteristic interval based on the previously defined adjustment rules. This adjustment parameter quantifies the degree to which the batch characteristics affect that specific characteristic interval. Subsequently, the calculated adjustment parameters are applied to their corresponding preset characteristic intervals, and the numerical range of the interval is corrected to obtain a characteristic interval that has been adjusted for batch characteristics. By comparing the signal features extracted from the predetermined local area with this batch-adjusted characteristic interval, it is possible to more accurately determine whether there are local material anomalies or potential microcracks of a specific type or severity in the area. In this way, this solution incorporates the impact of plate batch differences into the process of determining the characteristic interval, allowing the evaluation criteria to adapt to the actual situation of the current plate and improve the accuracy of the evaluation. This method of determining respective adjustment rules based on different characteristic intervals and calculating and applying adjustment parameters is a further refinement and optimization of the simple adjustment of characteristic intervals in the previous solution, making the evaluation process more targeted and accurate.

[0043] As an embodiment of the present invention, the step of determining the adjustment rules for each characteristic interval includes:

[0044] Obtaining a record of defect assessment for a feature interval determined by applying a specific adjustment rule associated with the feature interval in the metal sheet processing history, and an actual cutting quality result of the metal sheet in the corresponding area corresponding to the defect assessment record;

[0045] Compare and analyze the acquired defect assessment records with the actual cutting quality results to identify any systematic deviations between the currently applied specific adjustment rules in guiding the defect assessment based on the characteristic interval and the actual cutting quality performance;

[0046] If a systematic deviation is identified, the specific adjustment rule that causes the deviation is adjusted or recalibrated based on the characteristics of the systematic deviation to form an optimized adjustment rule for the feature interval for subsequent processing, and the optimized adjustment rule is used as the adjustment rule corresponding to the feature interval.

[0047] Defect assessment records refer to the system's determination of the presence of localized material mechanical property changes or potential microcrack initiation points in a localized area of ​​the sheet metal during sheet metal processing, based on the comparison of micro-perturbation test cutting signal characteristics with characteristic intervals, and their type and severity. These records can be stored in a structured data format, such as database records, log files, or dedicated data objects. Actual cutting quality results refer to the objective evaluation of the cut quality of the sheet metal area corresponding to the defect assessment area after the actual cutting of the sheet metal. These results can be obtained through manual visual inspection, automated optical inspection, or physical property testing, and recorded in a quantitative or categorical manner, such as the flatness grade of the cut section, the degree of slag adhesion, and the presence of cracks or chipping. Comparative analysis refers to the process of comparing and correlating defect assessment records with actual cutting quality results, aiming to identify any consistency or discrepancies between the assessment results and the actual situation. This can be achieved using statistical methods, data mining techniques, or machine learning algorithms, such as calculating assessment accuracy, false positive rate, and false negative rate, or constructing a predictive model to analyze the correlation between the assessment results and quality results. Systematic bias refers to non-random, regularly occurring differences between assessment results and actual quality results, discovered during comparative analysis. This bias may manifest as a consistent underestimation of the severity of a certain defect type or a consistent miscalculation of a particular signal feature combination. This bias may stem from the adjustment rule failing to accurately reflect the complex relationship between signal features and actual material properties, or failing to adequately account for other influencing factors during the manufacturing process. Adjustment or recalibration involves modifying or optimizing the currently applied adjustment rule based on the nature and extent of the identified systematic bias to reduce or eliminate it. This can be achieved manually or semi-automatically using a rule-based expert system, or by automatically optimizing the parameters or structure of the adjustment rule through historical data training using machine learning methods such as regression analysis, classification algorithms, or reinforcement learning. An optimized adjustment rule is one that, after adjustment or recalibration, more accurately reflects the relationship between signal features and actual material properties, thereby improving defect assessment accuracy. This can manifest as changes in the numerical values ​​of parameters within the adjustment rule or modifications to the logical structure of the adjustment rule.

[0048] This technical solution continuously optimizes the adjustment rules used to adjust feature intervals based on baseline signal characteristic parameters by incorporating a feedback mechanism based on historical processing data. Specifically, the system first obtains records of defect assessments performed using the adjustment rule associated with a specific feature interval in the metal sheet processing history, along with the actual cutting quality results for the corresponding sheet area corresponding to these assessment records. This historical data forms the basis for evaluating the effectiveness of the current adjustment rule. The acquired defect assessment records are then compared with the actual cutting quality results for analysis. This analysis aims to systematically compare the assessment accuracy and identify any systematic deviations between the current adjustment rule's guidance on defect assessment and actual cutting quality performance. For example, if a signal feature combination is assessed as a minor defect according to the current rule, but historical data shows that this area frequently exhibits severe quality issues during actual cutting, this indicates a systematic underestimation of risk within the adjustment rule. If this systematic deviation is identified through comparative analysis, the specific adjustment rule causing the deviation is adjusted or recalibrated based on the specific characteristics of the deviation. This adjustment process uses historical data as training samples to refine the parameters or logic of the adjustment rule to more accurately predict actual quality performance. For example, the weight of a particular signal feature in a rule can be increased, or a threshold can be modified. After adjustment or recalibration, an optimized adjustment rule for that feature interval is generated for subsequent processing. This optimized rule is then adopted to guide future evaluations of that feature interval. Through this iterative optimization process based on historical data feedback, the adjustment rule continuously learns and adapts to the complexities and uncertainties of actual processing, ensuring better alignment with actual cutting quality performance. This effectively complements the previously described approach, which only adjusts based on the baseline signal feature parameters of the current batch. This adjustment not only accounts for overall batch-to-batch variations but also calibrates the accuracy of the rule itself through historical experience, thereby improving the accuracy of defect assessment and providing a more reliable basis for subsequent adjustments to process parameters or paths. This continuous learning and optimization of the adjustment rule based on historical processing data enables the system to more accurately identify and quantify hidden risks within the sheet metal, thereby guiding more effective adjustments to processing strategies and ultimately improving sheet metal cutting quality.

[0049] As an embodiment of the present invention, the steps of comparing and analyzing the acquired defect assessment records with the actual cutting quality results to identify a systematic deviation between the defect assessment based on the characteristic interval guided by the currently applied specific adjustment rule and the actual cutting quality performance include:

[0050] Obtain data quality characterization information for each recorded data point in the defect assessment record and each result data point in the actual cutting quality result. The data quality characterization information reflects the timeliness of data collection, the integrity of data records, or the degree of external interference of the data at each data point;

[0051] Based on the data quality characterization information obtained for each data point, determine the usability status of each data point in the subsequent control analysis or the weight value of its contribution to the analysis result;

[0052] According to the determined availability status or contribution weight value of each data point, the record data points in the defect assessment record and the result data points in the actual cutting quality result are screened or weighted to form a quality-optimized defect assessment record and a quality-optimized actual cutting quality result;

[0053] The quality-optimized defect assessment records are compared with the quality-optimized actual cutting quality results to identify any systematic deviations between the currently applied specific adjustment rules in guiding the defect assessment based on the characteristic interval and the actual cutting quality performance.

[0054] Data quality information refers to auxiliary information associated with each data point, describing its reliability or validity. This information can be recorded in the form of a timestamp, a marker, a numerical score, or a text description. Data collection timeliness refers to the closeness between the time the data was recorded or acquired and the time of the event it actually reflects. This information can be characterized by time difference, a timestamp, or a timeliness rating. Data record completeness refers to whether all required information for a data point is recorded comprehensively and accurately. This information can be characterized by field fill rate, missing value ratio, or a completeness score. Data exposure refers to the extent to which data is affected by environmental noise, equipment anomalies, or other unexpected factors during collection, transmission, or storage. This information can be characterized by interference signal strength, noise level, or interference level. Availability status refers to whether a data point is considered sufficiently reliable for use in subsequent comparative analysis. This information can be represented by a Boolean value or an enumeration value. Contribution weight refers to the relative influence of a data point on the analysis results in subsequent comparative analysis. This information can be represented by a numerical value within a predetermined range. Determining the availability status of each data point in the subsequent comparative analysis or its contribution weight to the analysis results refers to evaluating the reliability of each data point based on the acquired data quality characterization information, and judging whether the data point is suitable for analysis or the proportion it should occupy in the analysis. Screening or weighted processing refers to the process of processing the original data set based on the availability status or contribution weight of the data point. Screening can be achieved by removing data points with an availability status of unavailable; weighted processing can be achieved by multiplying the value of each data point by its corresponding contribution weight when calculating the analysis index. The defect assessment records that have undergone quality processing and the actual cutting quality results that have undergone quality processing refer to the data sets used for comparative analysis after screening or weighting processing.

[0055] The solution improves the accuracy of comparative analysis by introducing a data quality assessment mechanism. First, for each data point in the historical defect assessment records and actual cut quality results, data quality information is associated with it. This information provides quantitative or qualitative descriptions of the timeliness of data collection, the completeness of the record, and the degree of external interference. Based on this quality information, the system assesses the reliability of each data point in subsequent analysis and determines its usability status or contribution weight to the analysis results. For example, data points with poor timeliness, incomplete records, or severe interference may have their usability status marked as unusable or their contribution weight assigned to a lower value. The raw data is then processed based on the determined usability status or contribution weight. Data points with an unusable usability status are excluded, while when using contribution weights, the influence of each data point in the calculation is adjusted according to its weight. In this way, the impact of low-quality data on the overall analysis results is effectively reduced or even eliminated. Finally, the quality-processed data set is used for comparative analysis to identify systematic deviations between defect assessments and actual cut quality. Because the analysis relies on more reliable data, the identified deviation patterns are more accurate, providing a solid foundation for subsequent targeted adjustments or recalibration of the adjustment rules. This approach, combined with a scheme for optimizing adjustment rules based on historical data, eliminates the need for the optimization process to rely solely on the surface consistency of the raw data and instead deeply considers the inherent quality of the data. This significantly improves the accuracy and robustness of the adjustment rules, allowing defect assessments based on these rules to more accurately predict actual processing results, ultimately improving the processing quality of sheet metal.

[0056] As an embodiment of the present invention, the step of determining the usability status of each data point in a subsequent control analysis or its contribution weight to the analysis result based on the acquired data quality characterization information of each data point includes:

[0057] Based on the acquired data quality characterization information of each data point, the data quality characterization information is quantified;

[0058] Assigning weights to each piece of data quality representation information after quantification;

[0059] Based on the quantified data quality characterization information and their respective weights, the quality score of the data point is calculated;

[0060] Based on the quality score, the usability status of the data point in the subsequent control analysis or the weight of its contribution to the analysis results is determined.

[0061] Data quality information refers to various types of information reflecting the quality of a data point, such as the time between data collection and the current analysis time, whether key fields in the data record are missing or abnormal, and whether there are known external factors that may affect data accuracy during the data collection process. Quantifying data quality information involves converting this qualitative or semi-qualitative data quality information into numerical form. For example, converting the timeliness of data collection into a time decay factor, converting data record completeness into a percentage or score, and converting the degree of external interference into an impact coefficient. Assigning weights to each type of quantified data quality information involves assigning an importance coefficient to each type of quantified quality information based on its varying impact on the overall quality of the data point. For example, the weight of data collection timeliness may be higher than that of data record completeness. The quality score of a data point is a single numerical value calculated by comprehensively considering each type of quantified data quality information and its respective weights. This value reflects the overall data quality level of the data point. Determining the usability status of a data point in subsequent control analysis or its contribution weight to the analysis results based on the quality score means judging whether the data point is suitable for participating in the subsequent control analysis (usability status) based on the calculated quality score, or determining the degree of influence of the data point on the analysis results when participating in the analysis (contribution weight value). The higher the quality score, the higher the usability of the data point or the greater the contribution weight value.

[0062] The steps of determining the usability status of the data point in the subsequent control analysis or the weight of its contribution to the analysis results based on the quality score include:

[0063] According to a preset mapping function, the quality score is converted into a contribution weight value within a predetermined range, and the contribution weight value increases as the quality score increases.

[0064] The steps of converting the quality score into a contribution weight value within a predetermined range according to a preset mapping function, wherein the contribution weight value increases as the quality score increases, include:

[0065] Get the quality score S of the data point;

[0066] Determine the steepness parameter k and midpoint parameter S_mid of the mapping function;

[0067] According to the quality score S, the steepness parameter k and the midpoint parameter S_mid, the contribution weight value W is calculated by the following functional relationship: W = 1 / (1 + exp(-k*(S-S_{mid})));

[0068] The steepness parameter k is used to control the sensitivity of the quality score change to the contribution weight value, and the midpoint parameter S_mid is used to define the midpoint of the mapping between the quality score and the contribution weight value.

[0069] The quality score S is the numerical value obtained by quantifying the data quality of a data point. It reflects the overall quality level of the data point, including the timeliness of data collection, the completeness of data recording, and the degree of external interference. A mapping function is a mathematical relationship used to convert an input value (quality score S) into an output value (contribution weight W). The steepness parameter k is a coefficient in the Sigmoid function that controls the slope of the function curve, that is, the sensitivity of the contribution weight to changes in the quality score. The midpoint parameter S_mid is another parameter in the Sigmoid function that defines the center point of the function curve. That is, when the quality score S equals S_mid, the contribution weight W is close to 0.5. The functional relationship W = 1 / (1 + exp(-k*(S - S_{mid}))) is a specific form of the Sigmoid function that can map any real-number input to an output between 0 and 1. Its curve is S-shaped, with good smoothness and nonlinear mapping capabilities.

[0070] This solution achieves a nonlinear conversion from quality score to contribution weight by introducing a sigmoid function-based mapping relationship. First, the quality score S of each data point is obtained. This score serves as the basis for measuring the reliability of the data point. Next, key parameters of the mapping function are determined: the steepness parameter k and the midpoint parameter S_mid. The settings of these two parameters are crucial, as they determine how the quality score is converted into the contribution weight. The steepness parameter k controls the slope of the mapping curve. A larger k value results in a significant change in the contribution weight near S_mid, making it suitable for scenarios that require a strict distinction between high-quality and medium-quality data points. A smaller k value results in a flatter curve, minimizing the impact of quality score changes on the weight, making it suitable for scenarios that require a smooth transition of weights. The midpoint parameter S_mid defines the center of the mapping curve; that is, a quality score of S_mid corresponds to a contribution weight of approximately 0.5. By adjusting S_mid, the focus area of ​​the mapping (the area most sensitive to weight changes) can be shifted to different quality score ranges. For example, setting it at a higher quality score ensures that only very high-quality data points receive a higher weight. Finally, using the obtained quality score S and the determined parameters k and S_mid, the final contribution weight W is calculated using a Sigmoid function. The Sigmoid function naturally constrains its output values ​​to be between 0 and 1, which aligns with the physical meaning of contribution weight. This nonlinear mapping approach more flexibly and accurately reflects the complex relationship between the quality score and the contribution of a data point in subsequent analysis. For example, parameters can be set so that the weight of low-quality data points rapidly approaches zero, while the weight of high-quality data points gradually approaches one. Alternatively, a high steepness can be set near a critical quality score threshold to achieve a rapid increase in weight. This adjustable nonlinear mapping allows for more precise control over the impact of data points of varying quality on the analysis results, avoiding the potential biases introduced by simple linear mapping. This refined weight allocation mechanism, combined with the previously described determination of data point availability or contribution weights, and the screening or weighting based on this, results in a quality-optimized data set. This makes subsequent comparative analysis of defect assessment records and actual cutting quality results more accurate and reliable, ultimately helping to identify more precise adjustment rules and improve sheet metal processing quality.

[0071] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. Various changes and improvements are possible without departing from the spirit and scope of the present invention, and such changes and improvements fall within the scope of the invention as claimed.

Claims

1. A processing pretreatment analysis system based on the actual situation of plate, characterized by: The system comprises: A micro-perturbation trial cutting execution module is used to perform a micro-perturbation trial cutting action on at least one predetermined local area of ​​the metal sheet before formally cutting the metal sheet; a physical signal monitoring module, configured to monitor in real time during the execution of the micro-disturbance trial cutting action at least one physical signal, wherein the physical signal represents a response characteristic of the metal sheet to the micro-disturbance trial cutting action in the predetermined local area; A material risk assessment module is used to evaluate online whether there is a sudden change in the mechanical properties of the local material or a potential microcrack initiation point in the predetermined local area of ​​the metal sheet based on real-time monitored physical signals, and obtain an assessment result; A cutting path and parameter adjustment module is used to adjust the laser cutting process parameters or the laser cutting path when the metal sheet is formally cut and the laser cutting path advances to the predetermined local area if the evaluation result indicates the existence of a sudden change in the mechanical properties of the local material or a potential microcrack initiation point.

2. A plate processing pre-processing analysis system based on real-time conditions according to claim 1, characterized in that: The step of online evaluating whether there is a local material mechanical property mutation or a potential microcrack initiation point in the predetermined local area of ​​the metal plate to obtain the evaluation result includes: Extracting multiple signal features from the real-time monitored physical signal; Comparing the combination of the multiple signal features and the values ​​of the respective signal features with preset characteristic intervals corresponding to local material mechanical property mutations or potential microcrack initiation points of different types or severities to obtain a comparison result; According to the comparison result, the type or severity level of the local material mechanical property mutation or potential microcrack initiation point in the predetermined local area is determined, and the type or severity level is used as the evaluation result.

3. The plate processing pretreatment analysis system based on the actual situation of the plate according to claim 2 is characterized in that: The step of comparing the combination of the multiple signal features and the value of each signal feature with preset characteristic intervals corresponding to local material mechanical property mutations or potential microcrack initiation points of different types or severities to obtain a comparison result includes: Acquire reference signal characteristics of a current batch of metal sheets to be processed at one or more reference positions, and determine reference signal characteristic parameters of the current batch of metal sheets; Adjusting the preset characteristic intervals corresponding to local material mechanical property mutations or potential microcrack initiation points of different types or severities according to the determined reference signal characteristic parameters to obtain characteristic intervals adjusted for batch characteristics; The combination of the plurality of signal features and the value of each signal feature are compared with the feature interval adjusted by the batch characteristics to obtain the comparison result.

4. The plate processing pre-processing analysis system based on the actual situation of the plate according to claim 3 is characterized in that: The step of adjusting the preset characteristic intervals corresponding to local material mechanical property mutations or potential microcrack initiation points of different types or severities based on the determined reference signal characteristic parameters to obtain the characteristic intervals adjusted for batch characteristics includes: Determine the adjustment rules for each characteristic interval; According to the determined reference signal characteristic parameters and in accordance with the adjustment rules corresponding to the obtained characteristic intervals, an adjustment parameter is calculated for each of the preset characteristic intervals, and the adjustment parameter is applied to adjust the corresponding preset characteristic interval to obtain each characteristic interval adjusted for batch characteristics.

5. The plate processing pre-processing analysis system based on the actual situation of the plate according to claim 4 is characterized in that: The adjustment rule defines the change mode of the characteristic interval in response to the determined reference signal characteristic parameter, and the adjustment rule corresponding to each characteristic interval is obtained.

6. The plate processing pre-processing analysis system based on the actual situation of the plate according to claim 4 is characterized in that: The steps of determining the adjustment rules for each characteristic interval include: Obtaining a record of defect assessment performed on a characteristic interval determined by applying a specific adjustment rule associated with the characteristic interval in the metal sheet processing history, and an actual cutting quality result of the metal sheet in a corresponding area corresponding to the defect assessment record; Comparing and analyzing the acquired defect assessment record with the actual cutting quality result to identify a systematic deviation between the currently applied specific adjustment rule in guiding the defect assessment based on the characteristic interval and the actual cutting quality performance; If the systematic deviation is identified, the specific adjustment rule that causes the deviation is adjusted or recalibrated based on the characteristics of the systematic deviation to form an optimized adjustment rule for the feature interval for subsequent processing, and the optimized adjustment rule is used as the adjustment rule corresponding to the feature interval.

7. The plate processing pre-processing analysis system based on the actual situation of the plate according to claim 6 is characterized in that: The step of comparing and analyzing the acquired defect assessment record with the actual cutting quality result to identify a systematic deviation between the currently applied specific adjustment rule in guiding the defect assessment based on the characteristic interval and the actual cutting quality performance includes: Acquire data quality characterization information of each record data point in the defect assessment record and each result data point in the actual cutting quality result; Based on the acquired data quality characterization information of each data point, determining the usability status of each data point in a subsequent control analysis or the weight value of its contribution to the analysis result; screening or weighting the record data points in the defect assessment record and the result data points in the actual cutting quality result according to the determined availability status or contribution weight value of each data point, so as to form a quality-optimized defect assessment record and a quality-optimized actual cutting quality result; The quality-optimized defect assessment record is compared with the quality-optimized actual cutting quality result to identify the systematic deviation between the specific adjustment rule currently applied when guiding the defect assessment based on the characteristic interval and the actual cutting quality performance.

8. The plate processing pre-processing analysis system based on the actual situation of the plate according to claim 7 is characterized in that: The data quality characterization information reflects the timeliness of data collection, the integrity of data records, or the degree to which data is affected by external interference at each data point.

9. The plate processing pre-processing analysis system based on the actual situation of the plate according to claim 7, characterized in that: The step of determining the usability status of each data point in the subsequent comparative analysis or the weight value of its contribution to the analysis result based on the acquired data quality characterization information of each data point includes: Based on the acquired data quality characterization information of each data point, quantifying the data quality characterization information; Assigning respective weights to each piece of data quality characterization information after the quantization process; Calculating the quality score of the data point based on the quantified data quality characterization information and their respective weights; The usability status of the data point in the subsequent control analysis or the weight value of its contribution to the analysis result is determined based on the quality score.

10. The plate processing pre-processing analysis system based on the actual situation of the plate according to claim 9, characterized in that: The step of determining the usability status of the data point in the subsequent control analysis or the weight value of its contribution to the analysis result according to the quality score includes: The quality score is converted into a contribution weight value within a predetermined range according to a preset mapping function, and the contribution weight value increases as the quality score increases.

Images