Cutting precision control and movement cooperation system of rotating disc type laser plate splitting machine

Abstract

The invention relates to the field of cutting precision control, in particular to a cutting precision control and movement coordination system of a rotating disc type laser plate splitting machine, which comprises a parameter database module used for storing historical data of the whole life cycle of the rotating disc type laser plate splitting machine; the precision prediction model module ensures that the model can accurately predict the cutting track deviation and the machining precision change; the motion cooperative control module is used for dynamically planning the optimal motion track of the laser cutting head and calibrating the motion synchronism of the cutting head and the turntable in real time; the intelligent monitoring and diagnosis module is used for identifying laser power abnormity and synchronization deviation problems and sending out early warning; according to the method, while precise prediction of the cutting precision and intelligent optimization of motion cooperative dynamic control are achieved, the equipment running state and the machining effect are presented through a graphical interface, the cutting precision control of the laser plate splitting machine is promoted to develop towards the intelligent and flexible direction, and the machining precision stability and the production efficiency are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of cutting precision control, and more particularly to a cutting precision control and motion coordination system for a turntable laser panel splitter. Background Art

[0002] In the production scenario of turntable laser panel separators, with the rapid development of the electronics manufacturing industry and the continuous upgrading of the demand for precision plate processing, laser panel separators, as core production equipment, face higher challenges in processing accuracy and operation stability. The traditional production management model can no longer meet actual needs, and there are many problems that need to be solved urgently. Currently, the operation monitoring of laser panel separators involves a variety of equipment such as position sensors, power meters, and speed encoders. The data formats, sampling frequencies, and accuracy collected by different types of sensors vary significantly, and historical operation data, process parameters, and quality inspection results are scattered and stored in different systems. There is a lack of systematic integration, resulting in low data utilization. Staff need to manually summarize and analyze a large amount of raw data, which is not only time-consuming and labor-intensive, but also prone to parameter matching errors, data traceability difficulties, and other problems, seriously affecting production scheduling efficiency and quality control level.

[0003] At the same time, multi-source heterogeneous data cannot be directly used for process optimization and precision improvement due to the lack of unified processing and analysis standards, which greatly limits the intelligent processing level of the equipment; in terms of processing precision control, existing methods mostly rely on manual experience to adjust process parameters, and it is difficult to accurately predict the precision loss caused by cutting trajectory deviation and material deformation in real time, resulting in an increase in scrap rate; in the face of processing needs for plates of different batches and materials, the formulation of an adaptive process plan requires manual review of historical records and repeated debugging. The process is cumbersome and the stability and efficiency of the plan are difficult to guarantee, making flexible production impossible; in addition, the equipment operating status and processing data are mainly displayed through paper records or a single interface, lacking intuitive and dynamic visualization methods. It is difficult for managers to fully grasp the health status of the equipment, the adaptability of process parameters and the production progress. Decisions rely on experience rather than data support, which seriously affects the stability and economy of production.

[0004] Therefore, there is an urgent need for a system that can integrate the operating data of the laser panel separator, unify data processing standards, accurately predict processing accuracy, generate process plans, and visualize production and equipment information to achieve intelligent management of the entire process from data collection to decision support. Summary of the Invention

[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides the following technical solutions: The present invention provides a cutting precision control and motion coordination system for a turntable laser panel separator, comprising: a parameter database module for storing historical data of the turntable laser panel separator throughout its entire life cycle;

[0006] Real-time sensing acquisition module, used to collect real-time data of the laser depaneling machine;

[0007] The precision prediction model module is used to train the precision prediction model using historical data and real-time data to achieve early prediction of cutting trajectory deviation and processing accuracy;

[0008] The motion coordination control module includes: a trajectory planning unit, which is used to dynamically plan the optimal motion trajectory of the laser cutting head according to the plate processing path and the motion state of the turntable;

[0009] Synchronous adjustment unit, used to calibrate the synchronization between the movement of the laser cutting head and the rotation of the turntable in real time, ensuring that the deviation between the cutting position and the preset path is controlled within the threshold range;

[0010] The error compensation module is used to dynamically adjust the laser focal length and cutting speed based on the results of the accuracy prediction model and real-time operation data to compensate for the accuracy loss caused by mechanical errors and material deformation;

[0011] The visual monitoring and diagnosis module includes: a status monitoring unit, which is used to display the operating parameters and cutting trajectory simulation of the laser depaneling machine through a graphical interface;

[0012] Fault diagnosis unit, used to identify potential mechanical failures and issue early warning prompts;

[0013] The process optimization decision module includes: an effect analysis unit, which is used to perform statistical analysis on cutting accuracy data and processing efficiency indicators and generate optimization suggestions;

[0014] The solution generation unit is used to automatically generate personalized cutting parameter configuration solutions based on the processing requirements of different batches of plates and combined with historical optimal process solutions to assist operators in making decisions.

[0015] Technical effects and advantages of the present invention:

[0016] 1. This invention uses a CNN-LSTM hybrid network architecture to identify abnormal conditions such as cutting trajectory deviation, power anomaly, and synchronization lag. This process effectively avoids the accuracy problems caused by traditional manual monitoring due to experience differences and misses and misjudgments, significantly reducing the risk of batch scrap and equipment damage caused by the lack of timely intervention in processing anomalies.

[0017] 2. The present invention not only achieves static optimization of process parameters based on historical data in the parameter database, but also achieves real-time optimization of the precision prediction model through a dynamic adaptive training mechanism. This continuous intelligent optimization mechanism enables the present invention to quickly adapt to complex processing scenarios such as different plate materials and thickness tolerances, significantly improving the system's adaptability and processing precision stability for flexible production scenarios involving multiple varieties and small batches.

[0018] 3. While achieving precise accuracy prediction and coordinated motion control, the present invention intuitively presents the equipment operating status, cutting trajectory simulation, and precision deviation data through a visual monitoring platform. In addition, operators can adjust production parameters in real time and predict the processing effects of different process solutions. This effectively promotes the cutting precision control of laser panel splitters towards intelligent and digital directions, significantly improving production efficiency and quality control levels. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a schematic diagram of the overall structure of the present invention.

[0020] Figure 2 This is a step diagram for planning the optimal motion trajectory of the laser cutting head of the present invention. DETAILED DESCRIPTION

[0021] The technical solutions of the present invention will be described clearly and completely below in conjunction with the drawings in the present invention. In addition, the forms of the various structures described in the following embodiments are merely examples. The cutting precision control and motion coordination system of a turntable laser panel splitter involved in the present invention is not limited to the various structures described in the following embodiments. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0022] Reference Figure 1 The present invention provides a cutting precision control and motion coordination system for a turntable laser panel splitter, comprising:

[0023] Parameter database module, used to store historical data of the entire life cycle of the turntable laser panel separator;

[0024] Real-time sensing acquisition module, used to collect real-time data of the laser depaneling machine;

[0025] The precision prediction model module is used to train the precision prediction model using historical data and real-time data to achieve early prediction of cutting trajectory deviation and processing accuracy;

[0026] The motion coordination control module includes: a trajectory planning unit, which is used to dynamically plan the optimal motion trajectory of the laser cutting head according to the plate processing path and the motion state of the turntable;

[0027] Synchronous adjustment unit, used to calibrate the synchronization between the movement of the laser cutting head and the rotation of the turntable in real time, ensuring that the deviation between the cutting position and the preset path is controlled within the threshold range;

[0028] The error compensation module is used to dynamically adjust the laser focal length and cutting speed based on the results of the accuracy prediction model and real-time operation data to compensate for the accuracy loss caused by mechanical errors and material deformation;

[0029] The visual monitoring and diagnosis module includes: a status monitoring unit, which is used to display the operating parameters and cutting trajectory simulation of the laser depaneling machine through a graphical interface;

[0030] Fault diagnosis unit, used to identify potential mechanical failures and issue early warning prompts;

[0031] The process optimization decision module includes: an effect analysis unit, which is used to perform statistical analysis on cutting accuracy data and processing efficiency indicators and generate optimization suggestions;

[0032] The solution generation unit is used to automatically generate personalized cutting parameter configuration solutions based on the processing requirements of different batches of plates and combined with historical optimal process solutions to assist operators in making decisions.

[0033] The specific implementation of the present invention includes the following steps:

[0034] Parameter database module, used to store historical data of the entire life cycle of the turntable laser panel separator;

[0035] Furthermore, it should be specifically explained in this embodiment that the historical data covers the initial parameters of the equipment when it leaves the factory, the material property data of different types of plates, including physical property parameters such as plate thickness, hardness, thermal conductivity, and the process parameters of the corresponding plates during the cutting process, operating parameters such as laser power, cutting speed, pulse frequency, and focal length adjustment value, and also includes the operating status data of each core component of the equipment, including the turntable motor speed fluctuation record, laser cutting head positioning deviation data, sensor feedback vibration and temperature change data, and quality inspection data after the cutting is completed. In addition, it also includes equipment maintenance records, such as component replacement time, calibration cycle and parameter comparison data before and after calibration, as well as various command parameters generated during user operation and corresponding cutting effect data;

[0036] It needs to be explained that these historical data form a complete data chain through structured storage, which can not only provide parameter reference basis for new cutting tasks, but also provide training samples for the precision prediction model module. By extracting the characteristics of normal working conditions and abnormal working conditions in historical data, the system's ability to predict potential faults is improved. At the same time, through the long-term accumulated equipment operation data, the performance degradation trend of the equipment can be analyzed to provide data support for the formulation of maintenance plans, realize the refined management of the entire life cycle of the equipment, and ensure the basic supporting role of historical data in the process of cutting precision control and motion coordinated optimization.

[0037] Real-time sensing acquisition module, used to collect real-time data of the laser depaneling machine;

[0038] What needs to be specifically explained in this embodiment is that the real-time data includes the laser cutting head position information, the turntable speed, the plate positioning deviation and the laser power feedback data during the cutting process; the laser head cutting position information data is collected by the grating ruler installed on the moving axis of the laser cutting head, the speed data is collected by the encoder built into the turntable drive motor, the positioning deviation data is collected by the visual sensor at the edge of the plate positioning platform, and the real-time power feedback data is collected by the power sensor at the output end of the laser generator. Various sensors collect data at a sampling frequency of not less than 1kHz, and transmit it to the accuracy prediction module in real time via industrial Ethernet.

[0039] It should be explained that the prediction model is preprocessed before training, including noise filtering of the collected raw data, using the Kalman filter algorithm to eliminate high-frequency noise generated by sensor jitter, and filling the missing values ​​in the data with interpolation to ensure the continuity of the data sequence; time synchronization processing of heterogeneous data from different sensors, aligning multi-source data to the same time dimension based on the system's unified timestamp to avoid data deviation caused by sampling delay; standardization of parameters of different magnitudes such as laser power and position coordinates, mapping the data to the 0-1 range to eliminate the impact of dimensional differences on model training; extraction of characteristic parameters in the data, such as the rate of change of cutting head motion acceleration, the fluctuation coefficient of turntable speed, the standard deviation of power feedback, etc., to enhance the characterization ability of the data; at the same time, outlier detection and correction of the data, by setting a reasonable threshold range to identify abnormal data points that exceed normal working conditions, and combining the valid data at adjacent moments for smoothing correction to ensure the accuracy and consistency of the data input to the prediction model, provide a high-quality data source for subsequent model training, and ensure the model's prediction accuracy for cutting accuracy and motion coordination status.

[0040] The precision prediction model module is used to train the precision prediction model using historical data and real-time data to achieve early prediction of cutting trajectory deviation and processing accuracy;

[0041] Furthermore, it is necessary to specifically explain in this embodiment that the process of training the accuracy prediction model is as follows:

[0042] A1. Data preparation: Extract historical data from the parameter database module and combine it with real-time data to build a multidimensional dataset and standardize the data.

[0043] A2. Feature Engineering: While extracting key features that affect cutting accuracy from the raw data, we construct time series features by sliding the time window to capture the dynamic changes of the data.

[0044] A3. Model construction phase: A hybrid architecture is used to build a prediction model. Long-short-term memory networks are used to process time series data to capture dynamic trends in the cutting process. Convolutional neural networks are used to extract spatial features, spatially model the cutting trajectory, and automatically assign weights to different features through an attention mechanism.

[0045] A4. Model training and optimization phase: The dataset is divided into training, validation, and test sets in a ratio of 7:2:1. The mean squared error is used as the loss function, and the Adam optimization algorithm is combined for model training. An early stopping strategy is used to prevent overfitting, and the Bayesian optimization algorithm is used to automatically search for the optimal hyperparameter combination to improve the model's generalization ability and prediction accuracy.

[0046] It should be explained that the specific process of predicting the cutting trajectory deviation and processing accuracy is as follows: During the cutting process, the current equipment status data vector St = [st1, st2, ..., stn] and the process parameter vector Pt = [pt1, pt2, ..., ptm] are collected in real time, and the cutting trajectory prediction value T for the next N time steps is generated through the trained prediction model Fθ t+1:t+N =Fθ(St, Pt), θ represents the model parameters; the predicted trajectory T t+1:t+N and theoretical trajectory Perform point-by-point comparison and calculate the trajectory deviation vector The deviation can be expressed as ||Ei||2(i=t+1,...,t+N), and the direction is represented by the unit vector Determine; at the same time, based on the mapping relationship between features and accuracy For the mapping function, through the function Predict key quality indicators Q = [q1, q2, ..., qk], where Q represents cutting width, heat-affected zone size, etc. The prediction results are compared with the accuracy threshold vector Q required by the process. * =q1 * ,q2 * ,...,qk * By comparing the discriminant function Determine whether there is a risk of precision exceeding the tolerance, δi represents the allowable deviation threshold of the i-th quality indicator.

[0047] What needs further explanation is that the model adopts an online learning mechanism. As new data continues to accumulate, the incremental learning algorithm is used to update the model regularly to adapt to changes brought about by equipment performance degradation and process parameter adjustments. At the same time, uncertainty quantification technology is introduced to provide confidence intervals in the prediction results to evaluate the reliability of the prediction. When the prediction uncertainty exceeds the threshold, the system is triggered to recalibrate. In addition, through feature importance analysis, the parameters that have the greatest impact on accuracy are identified to provide data support for process optimization and ensure that the cutting process is always in a high-precision state.

[0048] The motion coordination control module includes: a trajectory planning unit, which is used to dynamically plan the optimal motion trajectory of the laser cutting head according to the plate processing path and the motion state of the turntable;

[0049] Furthermore, as a feasible preferred embodiment, it is necessary to explain the process of planning the optimal motion trajectory of the laser cutting head as shown in the attached figure. Figure 2 As shown:

[0050] C1. Discrete the sheet metal processing path into a set of spatial points and extract geometric feature parameters. Combined with the turntable motion parameters, the model is used to predict its future state.

[0051] It should be explained that based on the geometric characteristics of the plate processing path, it is discretized into a series of ordered spatial point sets B = {b1, b2, ..., bn}, bi = xi, yi, zi) as three-dimensional space coordinates, and the curvature of each point is calculated by numerical differentiation and rate of change of direction At the same time, the current position θ(t) and speed of the turntable are obtained in real time. and acceleration Information, through the kinematic model Predict the motion state of the turntable in the future processing time period;

[0052] C2. Construct a multi-objective optimization model with the goals of minimizing tracking error, maximizing velocity, and minimizing acceleration rate, and clarify the dynamic and process constraints;

[0053] It should be explained that the specific process of building a multi-objective optimization model is to minimize the cutting path tracking error. bi is the theoretical trajectory point, ci is the actual trajectory point, and the cutting speed is maximized T is the total processing time, and the acceleration rate of change is minimized As the optimization target, a multi-objective optimization problem is formed, minJ = (J1, -J2, J3); the dynamic constraints of the laser cutting head are comprehensively considered: maximum speed constraint: vi≤vmax, maximum acceleration constraint: ai≤amax, maximum jerk constraint: ji≤jmax; and process constraints: minimum laser dwell time constraint: tstay,i≥tstay,min, power variation range: Pmin≤Pi≤Pmax, vi, ai, ji are the speed, acceleration and jerk of the cutting head in the i-th segment of the trajectory, tstay,i is the laser dwell time at that point, and Pi is the laser power at the corresponding position;

[0054] C3, using the improved particle swarm optimization algorithm to solve the model, and iterating through the adaptive weight and speed update formula to obtain the optimal solution;

[0055] It should be explained that when the improved particle swarm optimization algorithm is used to solve the optimization model, the particle position is represented by X k =(v1, a1, ...vn, an), and update the velocity formula V k+1 =dV k +c1r1(Pk, best-X k )+c2r2(Gbest-X k ), iterative update, d is the adaptive weight c1 and c2 are learning factors, r1 and r2 are random numbers, Pk,best is the individual optimal position, and Gbest is the global optimal position;

[0056] C4. Use the B-spline curve to fit the initial trajectory and achieve trajectory smoothing while ensuring the accuracy threshold.

[0057] It should be explained that when smoothing the generated initial trajectory, an m-order B-spline curve is used to fit the trajectory points: m(u), Ni, m(u) are B-spline basis functions, u∈[0, 1] is a parameter variable. By adjusting the control vertex Pi and the node vector U={u0,u1,…un+m+1}, under the premise of ensuring the trajectory accuracy max||S(u)-Pi||2≤∈, ∈ is the accuracy threshold, reducing the mutation points in the trajectory and reducing the difficulty of motion control.

[0058] It should be noted that in the trajectory planning process, a time synchronization strategy is adopted to ensure the precise coordination of the laser cutting head and the turntable movement. By establishing a unified time reference, the spatial points on the plate processing path are mapped to the time domain, and a time-parameterized trajectory is generated, so that the position of the laser cutting head at each time point is strictly matched with the rotation angle of the turntable; at the same time, a dynamic safety distance constraint is introduced. According to the current speed and acceleration of the laser cutting head, the minimum safety distance between it and the edge of the plate is adjusted in real time to prevent the risk of collision due to motion inertia; in addition, for complex curve processing scenarios, a curvature adaptive speed planning method is adopted to automatically reduce the cutting speed in areas with larger curvature to ensure the consistency of cutting quality.

[0059] Synchronous adjustment unit, used to calibrate the synchronization between the movement of the laser cutting head and the rotation of the turntable in real time, ensuring that the deviation between the cutting position and the preset path is controlled within the threshold range;

[0060] Furthermore, it should be specifically explained in this embodiment that the synchronization adjustment unit realizes synchronization calibration through the following mechanism: first, a multi-sensor fusion positioning system based on Kalman filtering is constructed, and the grating position data z1(t) of the laser cutting head and the angle data z2(t) of the turntable encoder are fused with the observation data z3(t) of the visual positioning system, and the optimal estimation is realized through the state equation x(t)=Ax(t-1)+w(t-1) and the observation equation z(t)=Hx(t)+v(x), where x(t)=[x(t), y(t), θ(t), x1(t), y1(t), θ1(t)] T , is the system state vector, x(t), y(t), θ(t) are velocity components, including the cutting head position, turntable angle and corresponding speed, A is the state transfer matrix, H is the observation matrix, w(t) and v(t) are process noise and observation noise respectively. The Kalman gain is used to eliminate the measurement noise and delay of a single sensor to obtain an accurate relative position relationship;

[0061] Then, a predictive control algorithm is used to predict the motion trajectory of the next Np time steps based on the current motion state x^(t) and historical data through the kinematic model x^(t+k|t)=f(x^(t+k-1|t),u(t+k-1)), where k=1, 2, ..., Np, and u(t) is the control input vector; the phase difference is defined as θdisk is the actual angle of the turntable, θlaster is the corresponding angle of the cutting head, when it is detected hour, To allow the phase difference threshold, the synchronous regulation unit immediately generates a compensation control instruction Kc is the proportional control gain matrix, which fine-tunes the movement speed and acceleration of the laser cutting head and makes small corrections to the rotation speed of the turntable, achieving dynamic synchronization between the two through dual closed-loop control.

[0062] It should be explained that the synchronization adjustment unit adopts a hybrid control architecture that combines time triggering and event triggering: in the stable cutting stage, the system performs synchronization calibration according to a fixed sampling period to ensure the continuity of control; and when events such as turntable acceleration and deceleration, and sudden changes in the cutting path are detected, the system immediately triggers the emergency synchronization program, quickly adjusts the control parameters, and reduces the synchronization error during the transition process; in addition, the unit also introduces an adaptive gain scheduling mechanism to dynamically adjust the control gain according to working parameters such as cutting speed and path curvature, thereby enhancing control sensitivity during high-speed cutting and improving control stability in fine processing areas.

[0063] The error compensation module is used to dynamically adjust the laser focal length and cutting speed based on the results of the accuracy prediction model and real-time operation data to compensate for the accuracy loss caused by mechanical errors and material deformation;

[0064] Furthermore, it should be specifically explained that the error compensation module implements error compensation through the following steps: First, based on the laser cutting head position p(t) = (x(t), y(t), z(t)), turntable speed w(t), and plate positioning deviation σ(t) obtained by the real-time sensing acquisition module, combined with the trajectory deviation etraj(t) and processing accuracy prediction result q(t) output by the accuracy prediction model, a multivariate error prediction model is established. The model uses the Gaussian process regression algorithm and defines the error prediction function as: ∈(t) = X(t) = [p(t), w(t), σ(t), q(t)] is the device state parameter vector, K(.) is the Gaussian kernel function, and the model also outputs the confidence interval of the prediction to quantify the prediction uncertainty.

[0065] Next, the dynamic compensation unit adopts a hierarchical compensation strategy for different types of error sources: For systematic errors caused by thermal deformation of the mechanical structure, a temperature-error mapping model ∈(t)=f(T(t),trun) is established, where T(t) is the ambient temperature and trun is the equipment running time. The least squares method is used to fit the obtained ∈thermal(t)=a1T(t)+a2trun+a3, where a1, a2, and a3 are fitting coefficients. The thermal deformation is predicted based on the ambient temperature and equipment running time, and the laser focal length and cutting path are adjusted in advance.

[0066] For the local deformation error caused by the release of internal stress in the plate, the deformation is calculated using the thin plate bending theory. D is the bending stiffness, EI is the bending stiffness, The surface topography of the plate is compensated by adjusting the laser head posture angle; during the cutting process, the system continuously monitors random factors such as laser power fluctuations and airflow disturbances, estimates random errors in real time through the Kalman filter, and dynamically adjusts the cutting speed and pulse frequency, and suppresses the error accumulation through the error accumulation suppression formula. To ensure that the error is always controlled within the allowable range.

[0067] The visual monitoring and diagnosis module includes: a status monitoring unit, which is used to display the operating parameters and cutting trajectory simulation of the laser depaneling machine through a graphical interface;

[0068] What needs to be specifically explained in this embodiment is that the graphical interface adopts a multi-window partition layout design, and a real-time data dashboard area is set on the left side of the main interface to display core operating parameters in the form of dynamic dashboards, bar charts, line charts, etc., including a comparison curve between the real-time value and the set value of the laser power, the current speed of the turntable and the speed fluctuation curve, a dynamic display of the X / Y / Z axis position coordinates of the laser cutting head, and real-time monitoring values ​​of the ambient temperature and humidity. All parameters are marked with normal threshold ranges, and when the parameters exceed the threshold, they will automatically switch to a red warning display.

[0069] The central area of ​​the main interface is the cutting trajectory simulation window, which uses three-dimensional visualization technology to render the motion trajectory of the laser cutting head and the rotation status of the turntable in real time. The theoretical cutting path is displayed as a green line, and the actual cutting trajectory is superimposed with a blue line. The trajectory deviation is intuitively presented through color difference, and mouse dragging and zooming operations are supported to view local details. The accuracy deviation value of the current position is marked in real time next to the trajectory, including linear deviation and angular deviation in the X-axis and Y-axis directions. The deviation value is dynamically updated as the cutting process progresses. A precision analysis panel is set on the right side of the interface, which uses a scatter plot to show the distribution of historical trajectory deviations, and a heat map to show the accuracy distribution characteristics of different areas, helping operators quickly identify areas with weak accuracy.

[0070] Fault diagnosis unit, used to identify potential mechanical failures and issue early warning prompts;

[0071] Furthermore, the present embodiment needs to specifically explain that, first, a multi-source feature library is constructed to collect equipment operation data in real time, and mechanical fault features, sensor abnormality features, and control system features are extracted through time domain analysis, frequency domain analysis, and signal modal decomposition;

[0072] Secondly, a hybrid diagnostic model is used for fault classification: for known fault types, a classifier is trained based on the random forest algorithm, and the real-time feature vector is matched with the fault template in the feature library to output the fault type and confidence level. For unknown faults, the isolation forest algorithm is used to identify abnormal features that deviate from the normal pattern, mark them as unidentified anomalies, and trigger the manual confirmation process. At the same time, combined with fault tree analysis, the fault propagation path of each component is sorted out to locate the root cause of the fault.

[0073] Finally, a three-level early warning mechanism is established: the first-level warning targets minor abnormalities, prompting attention to the stability of a certain sensor; the second-level warning targets potential faults, pushing a schematic diagram of the fault location and preliminary troubleshooting steps; the third-level warning targets serious faults, immediately triggering an audible and visual alarm, and generating a detailed fault report to assist in rapid repair.

[0074] The process optimization decision module includes: an effect analysis unit, which is used to perform statistical analysis on cutting accuracy data and processing efficiency indicators and generate optimization suggestions;

[0075] Furthermore, it should be specifically explained in this embodiment that the specific process of statistical analysis is as follows: first, a multi-dimensional data indicator system is constructed, covering cutting accuracy data and processing efficiency indicators, historical data and real-time data of the real-time sensing acquisition module are extracted from the parameter database to form an analysis data set, and the time series decomposition method is used to separate the trend items, periodic items and random items in the accuracy data, and identify the attenuation law of accuracy with processing batches or equipment operation time; at the same time, variance analysis and correlation analysis are applied to calculate the correlation coefficients between process parameters such as laser power, cutting speed, and pulse frequency and accuracy / efficiency indicators, and the significance of the parameter influence is judged by the F test and P value, and the key parameters that play a dominant role in quality and efficiency are located; in addition, the fluctuation range of the accuracy indicators is monitored by the control chart, abnormal data points that exceed the statistical control limit are identified, and the causes of the abnormalities are traced in combination with the records of process parameter changes.

[0076] It should be explained that the optimization suggestions are specifically as follows: for scenarios with high precision requirements, it is recommended to adjust the laser power and defocus to reduce the trajectory deviation to within the process threshold; for efficiency improvement needs, it is recommended to optimize the idle travel and pulse frequency of the cutting path to shorten the processing time per unit area; for quality stability issues, it is recommended to dynamically adjust the auxiliary gas pressure and pulse interval to reduce edge burrs and heat-affected zone fluctuations; all suggestions are accompanied by parameter adjustment range and expected effects, and the adjustment constraints are marked to ensure the feasibility and safety of the suggestions.

[0077] The solution generation unit is used to automatically generate personalized cutting parameter configuration solutions based on the processing requirements of different batches of plates and combined with historical optimal process solutions to assist operators in making decisions.

[0078] As a preferred feasible embodiment, it needs to be explained that the specific process of generating a personalized cutting parameter configuration plan is: first, the processing requirements of the current batch of plates are analyzed through the feature extraction module, and the core feature parameters are extracted, including plate material type, thickness tolerance, surface treatment level, batch size, target cutting accuracy level and special processing requirements. At the same time, the pre-processed data of the real-time sensing acquisition module is called to construct a batch feature vector containing 10 dimensions.

[0079] A three-level retrieval mechanism for the historical solution library is initiated based on feature vectors: the first level uses material and thickness as search keywords to screen out historical solution sets for similar plates, retaining solutions with a processing pass rate ≥95%; the second level calculates the matching degree between the current feature vector and the historical solution characteristics through the cosine similarity algorithm, and selects solutions with a similarity ≥80% as the candidate set; the third level scores the adaptability of the candidate solutions based on the current equipment status parameters, using the formula score = w1*pass rate + w2*efficiency value + w3 equipment adaptability. The top three solutions are selected to enter the parameter fusion link. After the parameters are generated, the cutting process is simulated through the digital twin system to output the predicted cutting accuracy, processing time and energy consumption indicators. If the simulation results meet the process requirements, a solution report containing a parameter configuration table, applicable scenario description and expected results is generated; if not, the parameters are reversely fine-tuned based on the simulation deviation, and the simulation is repeated until the standards are met; after the final solution is confirmed by the operator, it is automatically converted into a parameter code that can be recognized by the control system.

[0080] Secondly: The drawings of the embodiments disclosed in the present invention only involve structures related to the embodiments disclosed in the present invention. Other structures may refer to conventional designs. The same embodiment and different embodiments of the present invention may be combined with each other without conflict.

[0081] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A rotary laser panel cutting machine cutting precision control and motion coordination system, characterized in that: include: Parameter database module, used to store historical data of the entire life cycle of the turntable laser panel separator; Real-time sensing acquisition module, used to collect real-time data of the laser depaneling machine; The precision prediction model module is used to train the precision prediction model using historical data and real-time data to achieve early prediction of cutting trajectory deviation and processing accuracy; The motion coordination control module includes: a trajectory planning unit, which is used to dynamically plan the optimal motion trajectory of the laser cutting head according to the plate processing path and the motion state of the turntable; Synchronous adjustment unit, used to calibrate the synchronization between the movement of the laser cutting head and the rotation of the turntable in real time, ensuring that the deviation between the cutting position and the preset path is controlled within the threshold range; The error compensation module is used to dynamically adjust the laser focal length and cutting speed based on the results of the accuracy prediction model and real-time operation data to compensate for the accuracy loss caused by mechanical errors and material deformation; The visual monitoring and diagnosis module includes: a status monitoring unit, which is used to display the operating parameters and cutting trajectory simulation of the laser depaneling machine through a graphical interface; Fault diagnosis unit, used to identify potential mechanical failures and issue early warning prompts; The process optimization decision module includes: an effect analysis unit, which is used to perform statistical analysis on cutting accuracy data and processing efficiency indicators and generate optimization suggestions; The solution generation unit is used to automatically generate personalized cutting parameter configuration solutions based on the processing requirements of different batches of plates and combined with historical optimal process solutions to assist operators in making decisions.

2. The cutting precision control and motion coordination system for a rotary laser panel splitter according to claim 1, characterized in that: The real-time data includes laser cutting head position information, turntable speed, plate positioning deviation and laser power feedback data during the cutting process.

3. The cutting precision control and motion coordination system for a rotary laser panel splitter according to claim 1, characterized in that: The process of predicting the cutting trajectory deviation is as follows: During the cutting process, the current equipment status data vector St = [st1, st2, ..., stn] and the process parameter vector Pt = [pt1, pt2, ..., ptm] are collected in real time, and the cutting trajectory prediction value T for the next N time steps is generated through the trained prediction model Fθ t+1:t+N =Fθ(St, Pt), θ represents the model parameters; the predicted trajectory T t+1:t+N and theoretical trajectory Perform point-by-point comparison and calculate the trajectory deviation vector The deviation can be expressed as ||Ei||2(i=t+1,...,t+N), and the direction is represented by the unit vector Sure.

4. The cutting precision control and motion coordination system for a rotary laser panel splitter according to claim 1, characterized in that: The process of predicting machining accuracy is as follows: Based on the mapping relationship between features and accuracy For the mapping function, through the function Predict key quality indicators Q = [q1, q2, ..., qk], where Q represents cutting width, heat-affected zone size, etc. The prediction results are compared with the accuracy threshold vector Q required by the process. * =q1 * ,q2 * ,...,qk * By comparing the discriminant function Determine whether there is a risk of precision exceeding the tolerance, δi represents the allowable deviation threshold of the i-th quality indicator.

5. The cutting precision control and motion coordination system for a rotary laser panel splitter according to claim 1, characterized in that: The process of planning the optimal motion trajectory of the laser cutting head is as follows: C1. Discrete the sheet metal processing path into a set of spatial points and extract geometric feature parameters. Combined with the turntable motion parameters, the model is used to predict its future state. C2. Construct a multi-objective optimization model with the goals of minimizing tracking error, maximizing velocity, and minimizing acceleration rate, and clarify the dynamic and process constraints; C3, using the improved particle swarm optimization algorithm to solve the model, and iterating through the adaptive weight and speed update formula to obtain the optimal solution; C4. Use the B-spline curve to fit the initial trajectory and achieve trajectory smoothing while ensuring the accuracy threshold.

6. The cutting precision control and motion coordination system for a rotary laser panel splitter according to claim 5, characterized in that: The specific process of constructing the multi-objective optimization model is: minimizing the cutting path tracking error ci is the actual trajectory point, bi is the theoretical path point; cutting speed is maximized T is the total processing time; the acceleration rate of change is minimized As the optimization target, a multi-objective optimization problem is formed, minJ = (J1, -J2, J3).

7. The cutting precision control and motion coordination system for a rotary laser panel splitter according to claim 5, characterized in that: When the improved particle swarm optimization algorithm is used to solve the optimization model, the particle position is represented by X k =(v1, a1, ...vn, an), and update the velocity formula V k+1 =dV k +c1r1(Pk, best-X k )+c2r2(Gbest-X k ), iterative update, d is the adaptive weight c1 and c2 are learning factors, r1 and r2 are random numbers, Pk,best is the individual optimal position, and Gbest is the global optimal position.

8. The cutting precision control and motion coordination system for a rotary laser panel splitter according to claim 5, characterized in that: The B-spline curve fitting initial trajectory formula is: Ni,m(u) is the B-spline basis function, u∈[0,1] is the parameter variable, and Pi is the control vertex.

9. The cutting precision control and motion coordination system for a rotary laser panel splitter according to claim 1, characterized in that: The error compensation module implements error compensation by establishing a multivariable error prediction model. The specific formula is: X(t) = [p(t), w(t), σ(t), q(t)] is the device state parameter vector, K(.) is the Gaussian kernel function, and the model also outputs the confidence interval of the prediction to quantify the prediction uncertainty.

Images