A numerical control plate cutting optimization method and system based on path planning

Abstract

This invention relates to the field of CNC cutting technology, and proposes a method and system for optimizing CNC plate cutting based on path planning. The method includes: performing global property detection on the plate to be cut by the CNC plate cutter to obtain a material response feature sequence; performing distribution characteristic inversion on the local stiffness distribution and local damping distribution to obtain frequency self-adaptive parameters; performing cooperative frequency alignment and confidence feature encoding on the frequency self-adaptive parameters and the current operating frequency of the CNC plate cutter to construct a contact state confidence map; performing anti-interference topology reconstruction on the cutting motion trajectory to obtain an anti-interference path plan; performing mutual information entropy evaluation and iterative optimization correction on the anti-interference path plan and the detected data to obtain a steady-state cutting configuration; and performing residual stress compensation and hierarchical compilation on the steady-state cutting configuration to obtain stress adaptive control instructions. This invention can improve the efficiency of optimizing CNC plate cutting based on path planning.

Description

Technical Field

[0001] This invention relates to the field of CNC cutting technology, and in particular to an optimization method and system for CNC plate cutting based on path planning. Background Technology

[0002] Existing CNC plate cutting technology lacks a comprehensive physical property testing system for the plate to be cut. It cannot acquire multi-channel time-domain response waveforms of the plate through wide-frequency sweep excitation, making it difficult to complete the inversion of substrate physical property parameters and the fusion of multi-dimensional features. It cannot generate accurate material response feature sequences, and consequently cannot achieve accurate inversion of the local stiffness distribution and local damping distribution of the plate. As a result, frequency self-adaptive parameters cannot be generated, the operating frequency of the CNC plate cutting machine cannot be aligned with the physical properties of the plate, and a confidence spectrum of the contact state between the equipment and the plate cannot be constructed. The coupling state of the cutting condition lacks quantitative evaluation basis, and the stability of equipment operation is difficult to guarantee.

[0003] Traditional cutting path planning does not rely on contact state confidence maps for anti-disturbance topology reconstruction, making it impossible to locate the distribution of disturbance sources on the board surface and complete trajectory obstacle avoidance bias and smoothing. This results in weak anti-interference capability of the cutting trajectory. Furthermore, it does not use mutual information entropy to assess the correlation between the anti-disturbance path planning and the board detection data. The cutting layout scheme cannot be iteratively optimized to form a steady-state cutting configuration, and trajectory distortion compensation and hierarchical compilation are not performed for residual cutting stress. This easily leads to problems such as cutting trajectory distortion, layout conflicts, deterioration of cutting accuracy, and reduced board yield, resulting in overall cutting efficiency and quality failing to reach ideal levels. Therefore, improving the efficiency, accuracy, and stability of CNC board cutting machines has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides a method and system for optimizing the cutting of sheet metal in a CNC slitting machine based on path planning, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, this invention provides an optimization method for CNC plate cutting based on path planning, comprising: P1. Perform full-domain physical property testing on the material to be cut by the CNC board cutting machine to obtain the material response characteristic sequence of the material to be cut. P2. Based on the material response characteristic sequence, the distribution characteristics of the local stiffness distribution and local damping distribution of the plate to be cut are inverted to obtain the frequency self-adaptive parameters of the plate to be cut. P3. Align the frequency self-adaptation parameters with the current operating frequency of the CNC board cutting machine, and encode the confidence features of the aligned collaborative control parameters to construct a confidence map of the contact state between the CNC board cutting machine and the board to be cut. P4. Based on the contact state confidence map, the cutting motion trajectory of the CNC plate opening machine is reconstructed to obtain the anti-interference path planning of the cutting motion trajectory. P5. Evaluate the mutual information entropy between the anti-interference path planning and the detected data of the plate to be cut, and iteratively optimize and correct the cutting layout scheme of the plate to be cut based on the evaluation results to obtain the steady-state cutting configuration of the cutting layout scheme. P6. Perform residual stress compensation on the steady-state cutting configuration, and compile the compensated trajectory control strategy in a hierarchical manner to obtain the stress adaptive control command for the plate to be cut.

[0006] In a preferred embodiment, the step of performing full-domain physical property testing on the material to be cut by the CNC slitting machine to obtain the material response characteristic sequence of the material to be cut includes: A wide-frequency sweep excitation was performed on the material to be cut by the CNC plate cutting machine, and the multi-channel time-domain response waveform of the material to be cut was acquired at different excitation frequencies. Based on the multi-channel time-domain response waveform, the physical properties of the material to be cut are analyzed by inversion to obtain the substrate physical property parameters of the material to be cut. Multidimensional feature fusion of the substrate physical property parameters yields the global physical property tensor field of the plate to be cut; By performing ordered feature mapping on the global property tensor field, the material response feature sequence of the plate to be cut is obtained.

[0007] In a preferred embodiment, the step of performing distribution characteristic inversion on the local stiffness distribution and local damping distribution of the plate to be cut based on the material response characteristic sequence to obtain the frequency self-adaptive parameters of the plate to be cut includes: Local feature decoupling is performed on the material response feature sequence to obtain a subset of local response features of the plate to be cut; Based on the local response feature subset, mechanical parameters are inverted in local blocks of the plate to be cut to obtain the local stiffness distribution parameters and local damping distribution parameters of the plate to be cut. The frequency response amplitude spectrum of the plate to be cut is obtained by superimposing the local stiffness distribution parameters and the local damping distribution parameters with their frequency response characteristics. Frequency domain feature matching is performed on the frequency response amplitude spectrum to obtain the frequency self-adaptive parameters of the plate to be cut.

[0008] In a preferred embodiment, the step of coordinating the frequency self-adaptation parameters with the current operating frequency of the CNC slitting machine, and encoding the aligned coordinating control parameters with confidence features to construct a confidence map of the contact state between the CNC slitting machine and the material to be cut, includes: Frequency offset coherent analysis is performed on the frequency self-adaptation parameters and the current operating frequency of the CNC board opening machine to obtain the frequency offset characteristic quantity between the frequency self-adaptation parameters and the current operating frequency. Based on the frequency offset characteristic, the working frequency of the CNC plate cutting machine is normalized in real time to obtain the working condition coupling characteristics between the CNC plate cutting machine and the plate to be cut. The coupling gain of the working condition coupling characteristics is optimized, and the optimized cooperative control parameters are embedded and encoded to obtain the contact state feature spectrum between the CNC plate cutter and the plate to be cut. Confidence topological reconstruction of the contact state feature spectrum yields a confidence spectrum of the contact state between the CNC board cutting machine and the board to be cut.

[0009] In a preferred embodiment, the step of optimizing the coupling gain of the working condition coupling characteristics and performing feature embedding encoding on the optimized cooperative control parameters to obtain the contact state feature spectrum between the CNC slitting machine and the plate to be cut includes: The operating condition coupling features are de-aggregated and separated to obtain the main modal components and residual perturbation components of the operating condition coupling features; Anomaly fluctuation screening was performed on the residual perturbation components to obtain outlier feature fragments in the residual perturbation components; Outlier feature segments are removed from the residual perturbation components, and the purified residual components obtained by removal are coherently superimposed and fused with the main mode components to obtain the collaborative control parameters of the operating condition coupling features. Locally sensitive encoding is performed on the collaborative control parameters, and the encoding results are reduced and aggregated to obtain the contact state feature spectrum between the CNC slitting machine and the plate to be cut.

[0010] In a preferred embodiment, the step of performing anti-interference topology reconstruction on the cutting motion trajectory of the CNC plate-opening machine based on the contact state confidence map to obtain anti-interference path planning for the cutting motion trajectory includes: Disturbance spatial localization is performed on the contact state confidence spectrum to obtain the distribution of disturbance sources on the surface of the plate to be cut; Based on the distribution of disturbance sources, the original cutting trajectory of the CNC board opening machine is reconstructed by obstacle avoidance bias to obtain the preliminary obstacle avoidance trajectory of the CNC board opening machine. The initial obstacle avoidance trajectory is smoothed to obtain a smooth obstacle avoidance trajectory for the CNC plate opening machine; The anti-interference accuracy of the smooth obstacle avoidance trajectory is tuned to obtain the anti-interference path planning for the cutting motion trajectory.

[0011] In a preferred embodiment, the step of evaluating the mutual information entropy between the anti-interference path planning and the detected data of the material to be cut, and iteratively optimizing and correcting the cutting layout scheme of the material to be cut based on the evaluation results to obtain the steady-state cutting configuration of the cutting layout scheme, includes: The mutual information entropy of the anti-interference path planning and the detected data of the plate to be cut is extracted to obtain the correlation parameter between the anti-interference path planning and the detected data. The correlation parameter is embedded into the current cutting layout scheme of the material to be cut in order to identify the weak correlation areas of the current cutting layout scheme; Targeted conflict resolution is performed on the weak areas, and the resolved layout scheme is optimized, reconstructed and adjusted to obtain the candidate cutting layout scheme for the plate to be cut. The candidate cutting layout schemes are iteratively converged to obtain the steady-state cutting configuration of the cutting layout scheme.

[0012] In a preferred embodiment, the step of extracting mutual information entropy between the anti-interference path planning and the detected data of the material to be cut, to obtain the correlation parameter between the anti-interference path planning and the detected data, includes: Obtain the coordinates of trajectory points for anti-interference path planning; Based on the coordinates of the trajectory points, spatial interpolation sampling is performed on the detected data of the plate to be cut to obtain the detection response amplitude of the detected data; By performing discrete interval statistics on trajectory points and detection response amplitudes, the joint frequency, edge frequency, and total number of discrete intervals of trajectory point coordinates and detection response amplitudes are obtained; By mapping the joint frequency and edge frequency using probability distributions, the mutual information entropy between the anti-interference path planning and the detected data is obtained. The formula for calculating the mutual information entropy is as follows: ; in, Represents mutual information entropy. This represents the total number of discrete intervals representing the coordinates of the trajectory points. This represents the total number of discrete intervals of the detected response amplitude. This indicates that the coordinates of the trajectory point fall into the first... The interval and the detection response amplitude falls within the first interval. Joint frequency of each interval, This indicates that the coordinates of the trajectory point fall into the first... Edge frequencies of each interval This indicates that the detection response amplitude falls within the first... Edge frequencies of each interval The total number of samples representing the coordinates of trajectory points; The mutual information entropy is normalized and compressed to obtain the correlation parameter between the anti-interference path planning and the detected data.

[0013] In a preferred embodiment, the residual stress compensation for the steady-state cutting configuration and the hierarchical compilation of the compensated trajectory control strategy to obtain the stress adaptive control command for the plate to be cut include: The residual stress field is reconstructed for the steady-state cutting configuration to obtain the spatial distribution map of the residual stress of the plate to be cut; Based on the spatial distribution map of residual stress, trajectory distortion compensation is performed on the key cutting segments in the steady-state cutting configuration to obtain the cutting trajectory offset of the key cutting segments. Using the cutting trajectory offset as the guiding vector, stress-guided reshaping of the steady-state cutting configuration is performed, and the reshaping cutting control trajectory is analyzed in a hierarchical manner to obtain the hierarchical control parameters of the steady-state cutting configuration. The layered control parameters are adapted and compiled layer by layer to obtain the stress adaptive control command for the plate to be cut.

[0014] To address the aforementioned problems, this invention also provides a path planning-based optimization system for CNC plate cutting, the system comprising: The full-domain physical property detection module is used to perform full-domain physical property detection on the material to be cut by the CNC plate cutting machine in order to obtain the material response characteristic sequence of the material to be cut. The distribution characteristic inversion module is used to perform distribution characteristic inversion on the local stiffness distribution and local damping distribution of the plate to be cut based on the material response characteristic sequence, so as to obtain the frequency self-adaptive parameters of the plate to be cut. The collaborative alignment and confidence coding module is used to perform collaborative frequency alignment between the frequency self-adaptation parameters and the current operating frequency of the CNC board cutting machine, and to perform confidence feature coding on the aligned collaborative control parameters in order to construct a confidence map of the contact state between the CNC board cutting machine and the board to be cut. The anti-disturbance topology reconstruction module is used to perform anti-disturbance topology reconstruction on the cutting motion trajectory of the CNC plate-opening machine based on the contact state confidence map, so as to obtain the anti-disturbance path planning of the cutting motion trajectory. The mutual information evaluation and iterative optimization module is used to evaluate the mutual information entropy between the anti-interference path planning and the detected data of the plate to be cut, and to iteratively optimize and correct the cutting layout scheme of the plate to be cut based on the evaluation results, so as to obtain the steady-state cutting configuration of the cutting layout scheme. The stress compensation and hierarchical compilation module is used to perform residual stress compensation on the steady-state cutting configuration and to perform hierarchical compilation on the compensated trajectory control strategy to obtain the stress adaptive control command for the plate to be cut.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention, through full-domain physical property detection and wide-frequency sweep excitation acquisition of the material to be cut, can completely acquire the multi-channel time-domain response waveform of the material, accurately complete the inversion of the substrate physical property parameters and the fusion of multi-dimensional features, and generate a standardized material response feature sequence. Based on this sequence, the spatial distribution characteristics of the local stiffness and local damping of the material can be efficiently inverted, and frequency self-adaptive parameters can be stably generated to achieve precise coordination and alignment between the working frequency of the CNC cutting machine and the physical properties of the material. Through confidence feature encoding, the contact state confidence map is constructed, realizing the quantitative characterization and stable control of the coupling condition between the equipment and the material, and providing reliable state support and parameter basis for the cutting operation.

[0016] 2. This invention achieves disturbance-resistant topology reconstruction of the cutting trajectory based on contact state confidence maps, accurately locating disturbance sources and realizing trajectory obstacle avoidance bias and smoothing, significantly improving the anti-interference capability and operational stability of the cutting trajectory. Through mutual information entropy evaluation, it quantifies the correlation between path planning and detection data, completing iterative optimization of the cutting layout scheme and generating a steady-state configuration. Combined with residual stress field reconstruction and trajectory distortion compensation, it effectively avoids cutting deformation and layout conflicts. The stress-adaptive control commands generated through hierarchical compilation comprehensively improve the accuracy of the cutting trajectory, operational stability, and sheet metal yield, achieving efficient, precise, and stable operation of the CNC slitting machine. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating an optimization method for CNC plate cutting based on path planning, provided in an embodiment of the present invention. Figure 2 A functional module diagram of a CNC board cutting optimization system based on path planning is provided in an embodiment of the present invention. Figure 3 A line graph comparing the normalized mutual information entropy of three paths provided in an embodiment of the present invention; Figure 4 A bar chart comparing the cumulative offset of three trajectories under different numbers of disturbance sources, provided in an embodiment of the present invention; Figure 5 This is a line graph comparing the trajectory offset before and after residual stress compensation according to an embodiment of the present invention. Figure 6 A line graph showing the comparison of the optimization error of the present invention and the traditional solution as a function of iteration steps, provided in an embodiment of the present invention.

[0018] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0019] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0020] This application provides a path planning-based optimization method for CNC plate cutting. The execution entity of this path planning-based optimization method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the path planning-based optimization method for CNC plate cutting can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0021] Reference Figure 1 The diagram shown is a flowchart illustrating a path planning-based optimization method for CNC plate cutting in an embodiment of the present invention. In this embodiment, the path planning-based optimization method for CNC plate cutting in an embodiment includes: P1. Perform full-domain physical property testing on the material to be cut by the CNC board cutting machine to obtain the material response characteristic sequence of the material to be cut. In this embodiment of the invention, the step of performing full-domain physical property testing on the material to be cut by the CNC board cutting machine to obtain the material response characteristic sequence of the material to be cut includes: A wide-frequency sweep excitation was performed on the material to be cut by the CNC plate cutting machine, and the multi-channel time-domain response waveform of the material to be cut was acquired at different excitation frequencies. Based on the multi-channel time-domain response waveform, the physical properties of the material to be cut are analyzed by inversion to obtain the substrate physical property parameters of the material to be cut. Multidimensional feature fusion of the substrate physical property parameters yields the global physical property tensor field of the plate to be cut; By performing ordered feature mapping on the global property tensor field, the material response feature sequence of the plate to be cut is obtained.

[0022] A continuous and uniformly varied wideband sweep excitation is applied to the entire area of ​​the plate to be cut carried by the CNC plate cutting machine. The excitation signal smoothly transitions from the low frequency band to the high frequency band and completely covers the entire frequency range of the plate's response. The excitation points are evenly distributed in the edge and center areas of the plate to ensure that the excitation coverage is complete. During the entire excitation process, multiple sets of sensor acquisition units are arranged along the length and width of the plate. Each sensor acquisition unit synchronously captures the physical vibration signal generated by the plate after being excited in real time. The captured continuous vibration signal is converted into an electrical signal waveform that varies with time. The response data generated by the plate at each excitation frequency point is completely recorded and integrated to form a multi-channel time-domain response waveform.

[0023] The acquired multi-channel time-domain response waveform is decomposed segment by segment and point by point to extract core features such as response intensity change, time sequence duration, and waveform fluctuation pattern. The decomposed waveform features are matched one by one with the corresponding rules of the physical properties of the board. Based on the matching results, the basic physical properties of the board are determined. The inversion analysis of the board properties is completed by layer-by-layer analysis and feature correspondence. Finally, the substrate physical property parameters that can comprehensively characterize the basic physical properties of the board are determined.

[0024] The obtained substrate physical property parameters are systematically integrated according to the spatial location dimension and physical property dimension of the board. Different types of substrate physical property parameters in the same spatial location are associated and bound, and the same type of substrate physical property parameters in different spatial locations are spatially arranged. Through cross-dimensional feature association and data integration, the scattered physical property parameters are integrated into a unified overall data structure, forming a global physical property tensor field that can completely reflect the distribution state of the physical properties of the board.

[0025] Data within the global property tensor field is extracted in a directional manner according to the spatial distribution order and physical property classification order of the board material. The spatially distributed property features in the tensor field are linearly transformed according to a fixed arrangement rule, transforming the spatially distributed features into continuous and ordered linear feature data, completing the ordered feature mapping of the global property tensor field, and finally forming a material response feature sequence that can be directly used for subsequent process processing.

[0026] The beneficial effects are as follows: by performing full-domain broadband frequency sweep excitation and multi-channel synchronous acquisition on the material to be cut, the physical response information of the material in the entire frequency range can be obtained completely and without omission, thus completely avoiding the impact of local data loss or acquisition deviation on subsequent processing. Through refined feature decomposition and directional matching analysis of time-domain response waveforms, the substrate physical property parameters of the material can be accurately determined, ensuring the authenticity and accuracy of the basic physical property data. Through multi-dimensional feature fusion, the scattered and independent physical property parameters are integrated into a unified data field covering the entire domain, realizing a complete and global characterization of the physical properties of the material. Through ordered feature mapping, the spatial tensor field is transformed into a standardized linear sequence, providing stable and accurate core basic data for subsequent operations such as material mechanical property inversion, equipment frequency coordination alignment, cutting trajectory anti-disturbance reconstruction, and cutting layout iterative optimization. This significantly improves the execution accuracy, operational stability, and process continuity of the entire cutting optimization system.

[0027] P2. Based on the material response characteristic sequence, the distribution characteristics of the local stiffness distribution and local damping distribution of the plate to be cut are inverted to obtain the frequency self-adaptive parameters of the plate to be cut. In this embodiment of the invention, the step of performing distribution characteristic inversion on the local stiffness distribution and local damping distribution of the plate to be cut based on the material response feature sequence to obtain the frequency self-adaptive parameters of the plate to be cut includes: Local feature decoupling is performed on the material response feature sequence to obtain a subset of local response features of the plate to be cut; Based on the local response feature subset, mechanical parameters are inverted in local blocks of the plate to be cut to obtain the local stiffness distribution parameters and local damping distribution parameters of the plate to be cut. The frequency response amplitude spectrum of the plate to be cut is obtained by superimposing the local stiffness distribution parameters and the local damping distribution parameters with their frequency response characteristics. Frequency domain feature matching is performed on the frequency response amplitude spectrum to obtain the frequency self-adaptive parameters of the plate to be cut.

[0028] The generated material response feature sequence is directionally split according to the actual spatial region division of the board to be cut. The common feature data of the overall region of the board in the sequence is completely separated from the individual feature data of the independent local region of the board. The exclusive feature data corresponding to each independent local region is collected and aggregated. Redundant feature data that cannot characterize the local region characteristics in the sequence is removed. Feature data that can accurately reflect the physical response state of each local region is fully retained. The local feature decoupling of the material response feature sequence is completed, and finally the local response feature subset of the board to be cut is obtained.

[0029] The local response feature subset is precisely mapped to each independent local block of the plate to be cut. Based on the physical response change trend and characteristic manifestation of the local response feature subset, the mechanical load-bearing characteristics and energy dissipation characteristics of each local block are analyzed one by one. Through the direct correspondence between the characteristic manifestation and the mechanical properties, the stiffness distribution state and damping distribution state of each local block are determined, and the mechanical parameter inversion of the local blocks of the plate to be cut is completed. Finally, the local stiffness distribution parameters and local damping distribution parameters of the plate to be cut are obtained.

[0030] The frequency response variation characteristics corresponding to the local stiffness distribution parameters and the frequency response variation characteristics corresponding to the local damping distribution parameters are synchronously merged according to the same frequency node. The response characteristics of the two distribution parameters at the same frequency are integrated and merged, so that the frequency response characteristics of the two distribution parameters form a unified and coherent frequency response data, which fully presents the comprehensive response state of the plate in the entire frequency range. The frequency response characteristics of the local stiffness distribution parameters and the local damping distribution parameters are superimposed, and finally the frequency response amplitude spectrum of the plate to be cut is obtained.

[0031] The amplitude variation characteristics and frequency response characteristics corresponding to different frequency positions are extracted from the frequency response amplitude spectrum. The extracted frequency domain characteristics are compared one by one with the frequency characteristics that can keep the board material in a stable physical response. The frequency characteristic data that perfectly matches the physical characteristics of the board material are selected, and the frequency parameters that are suitable for the overall and local characteristics of the board material are determined. The frequency domain characteristics of the frequency response amplitude spectrum are matched, and finally the frequency self-adaptive parameters of the board material to be cut are obtained.

[0032] The beneficial effects are as follows: by performing refined local feature decoupling on the material response feature sequence, the exclusive response features of each local area of ​​the board can be accurately separated, avoiding interference from global features in the extraction of local characteristics. Based on the subset of local response features, the mechanical parameter inversion can accurately obtain the complete spatial distribution of local stiffness and local damping of the board, providing accurate mechanical data support for the board characteristic analysis. By superimposing the frequency response characteristics of local stiffness distribution parameters and local damping distribution parameters, a frequency response amplitude spectrum that comprehensively characterizes the full frequency response law of the board can be formed, fully presenting the physical response state of the board at different frequencies. The frequency self-adaptive parameters generated by frequency domain feature matching of the frequency response amplitude spectrum can perfectly match the physical characteristics of the board itself, providing accurate core parameters for the subsequent coordinated frequency alignment between the CNC board cutting machine and the board, ensuring the accuracy and stability of frequency matching during the cutting control process, and improving the overall adaptation effect of the cutting operation.

[0033] P3. Align the frequency self-adaptation parameters with the current operating frequency of the CNC board cutting machine, and encode the confidence features of the aligned collaborative control parameters to construct a confidence map of the contact state between the CNC board cutting machine and the board to be cut. In this embodiment of the invention, the step of coordinating the frequency self-adaptation parameters with the current operating frequency of the CNC board cutting machine, and encoding the confidence features of the aligned coordinating control parameters to construct a confidence map of the contact state between the CNC board cutting machine and the board to be cut, includes: Frequency offset coherent analysis is performed on the frequency self-adaptation parameters and the current operating frequency of the CNC board opening machine to obtain the frequency offset characteristic quantity between the frequency self-adaptation parameters and the current operating frequency. Based on the frequency offset characteristic, the working frequency of the CNC plate cutting machine is normalized in real time to obtain the working condition coupling characteristics between the CNC plate cutting machine and the plate to be cut. The coupling gain of the working condition coupling characteristics is optimized, and the optimized cooperative control parameters are embedded and encoded to obtain the contact state feature spectrum between the CNC plate cutter and the plate to be cut. Confidence topological reconstruction of the contact state feature spectrum yields a confidence spectrum of the contact state between the CNC board cutting machine and the board to be cut.

[0034] The step of coordinating the frequency self-adaptation parameters with the current operating frequency of the CNC board cutting machine, and encoding the confidence features of the aligned coordinating control parameters to construct a confidence map of the contact state between the CNC board cutting machine and the board to be cut, includes: Frequency offset coherent analysis is performed on the frequency self-adaptation parameters and the current operating frequency of the CNC board opening machine to obtain the frequency offset characteristic quantity between the frequency self-adaptation parameters and the current operating frequency. Based on the frequency offset characteristic, the working frequency of the CNC plate cutting machine is normalized in real time to obtain the working condition coupling characteristics between the CNC plate cutting machine and the plate to be cut. The coupling gain of the working condition coupling characteristics is optimized, and the optimized cooperative control parameters are embedded and encoded to obtain the contact state feature spectrum between the CNC plate cutter and the plate to be cut. Confidence topological reconstruction of the contact state feature spectrum yields a confidence spectrum of the contact state between the CNC board cutting machine and the board to be cut.

[0035] The frequency trajectory corresponding to the frequency self-adaptation parameters is synchronously compared with the trajectory of the CNC board cutting machine's current actual operating frequency throughout the entire process without interruption. The frequency change trend, interrelation status, and frequency offset of the two frequencies are tracked point by point during the operation. All offset data and coherent correlation data of the two frequencies from start-up to continuous operation are recorded. All recorded data are centrally integrated to form a complete feature data set. Through full trajectory comparison and data integration, the frequency offset coherent analysis of the frequency self-adaptation parameters and the current operating frequency of the CNC board cutting machine is completed, and finally the frequency offset characteristic quantity between the frequency self-adaptation parameters and the current operating frequency is obtained.

[0036] Based on the frequency offset direction, frequency offset degree, and frequency coherence correlation state determined by the frequency offset characteristic quantity, the operating frequency of the CNC board cutting machine is continuously and uninterruptedly corrected and adjusted in real time. The operating frequency of the equipment is gradually brought closer to and fully matched with the standard frequency corresponding to the frequency self-adaptation parameter. This allows the operating frequency of the equipment to form a synchronous and coordinated operating mode with the adapted frequency of the board to be cut, so that the equipment and the board can achieve a stable adapted combination state in the frequency dimension. Through continuous real-time correction and coordinated adjustment, the real-time frequency normalization of the operating frequency of the CNC board cutting machine is completed, and finally the working condition coupling characteristics between the CNC board cutting machine and the board to be cut are obtained.

[0037] The operating condition coupling features are split in a targeted manner according to the degree of influence of the features on the coupling state. The core stable features that can dominate the coupling state between the equipment and the plate are separated from the overall coupling features. The secondary fluctuation features that accompany the coupling state are divided separately. The core feature set and the secondary fluctuation feature set are completely separated. The feature de-aggregation and separation of the operating condition coupling features are completed through targeted splitting and classification. Finally, the main modal components and residual disturbance components of the operating condition coupling features are obtained.

[0038] The residual disturbance component is monitored and its characteristics are verified segment by segment throughout the entire process. The characteristic data segments that deviate from the normal coupling fluctuation law in the residual disturbance component are identified one by one. These abnormal characteristic data segments that do not conform to the stable coupling state are completely extracted and collected separately. Through segment-by-segment monitoring and accurate identification, the abnormal fluctuation screening of the residual disturbance component is completed, and finally the outlier characteristic segments in the residual disturbance component are obtained.

[0039] Outlier feature fragments are completely removed from the residual disturbance components, leaving only pure residual feature data without abnormal interference to form purified residual components. These purified residual components and the principal modal components are then merged segment by segment according to their corresponding spatial locations and runtime sequences, allowing the two types of feature data to fuse together to form unified and stable coupled feature data. The coherent superposition and fusion of the purified residual components and the principal modal components is achieved through outlier removal and segment-by-segment fusion, ultimately yielding the collaborative control parameters of the working condition coupled features. These collaborative control parameters refer to a set of specific control quantities obtained after optimizing the coupling gain of the working condition coupled features, used to characterize the optimal collaborative working state between the CNC board cutter and the board to be cut. The generation process of these collaborative control parameters is as follows: first, the working condition coupled features are de-aggregated and separated to obtain the principal modal components and residual disturbance components; then, abnormal fluctuations in the residual disturbance components are screened and outlier feature fragments are removed; finally, the purified residual components and the principal modal components are coherently superimposed and fused to obtain the collaborative control parameters.

[0040] The collaborative control parameters are oriented and encoded according to the actual contact area between the CNC board cutter and the board to be cut. Each set of feature data in the collaborative control parameters is mapped to the specific spatial position of the contact between the equipment and the board. Each set of feature data is assigned a unique position pointing mark, so that the feature data and the contact position form a precise correspondence. The local sensitive coding of the collaborative control parameters is completed through position correspondence and marking.

[0041] All feature data that have completed local sensitive coding are directionally integrated and compacted to eliminate invalid gaps and redundant data between scattered coded features. The scattered coded feature data are merged into a coherent, unified and compact overall feature set. The dimension reduction and aggregation of the coding results are completed through integration, compaction and redundancy elimination, and finally the contact state feature spectrum between the CNC board cutting machine and the board to be cut is obtained.

[0042] All feature data in the contact state feature spectrum are arranged topologically according to the actual contact space layout between the CNC slitting machine and the material to be cut. The feature data corresponding to each contact position is precisely bound to the contact confidence state at that position. A spatial confidence structure that can completely cover all contact areas between the equipment and the material is constructed. The confidence topology reconstruction of the contact state feature spectrum is completed through topological arrangement and confidence binding, and finally the contact state confidence map of the CNC slitting machine and the material to be cut is obtained.

[0043] The beneficial effects are as follows: by using frequency self-adaptive parameters and full-process frequency offset coherent analysis of the current working frequency of the CNC cutting machine, the offset and correlation information of the two frequencies can be completely captured, providing a precise and comprehensive basis for the frequency adjustment of the equipment. Real-time frequency normalization based on frequency offset characteristics can keep the working frequency of the equipment and the adaptive frequency of the board synchronized and coordinated, forming a stable and adapted working condition coupling state. Feature de-aggregation and separation can clearly distinguish the core coupling features and disturbance features. Abnormal fluctuation screening can accurately remove invalid outlier data in residual disturbances. Coherent superposition and fusion can generate pure and stable collaborative control parameters. Locally sensitive coding can make feature data correspond precisely to the contact position. Dimension reduction and aggregation can make feature data more compact and coherent. The contact state confidence map constructed by confidence topology reconstruction can comprehensively quantify the confidence level of the entire contact area between the equipment and the board, and fully present the stable state of the interaction between the equipment and the board. This provides comprehensive and reliable state support for the anti-disturbance topology reconstruction of the subsequent cutting motion trajectory, and significantly improves the stability, accuracy and controllability of the equipment operation and the interaction between the equipment and the board during the cutting operation.

[0044] P4. Based on the contact state confidence map, the cutting motion trajectory of the CNC plate opening machine is reconstructed to obtain the anti-interference path planning of the cutting motion trajectory. In this embodiment of the invention, the step of performing anti-interference topology reconstruction on the cutting motion trajectory of the CNC plate-opening machine based on the contact state confidence map to obtain anti-interference path planning for the cutting motion trajectory includes: Disturbance spatial localization is performed on the contact state confidence spectrum to obtain the distribution of disturbance sources on the surface of the plate to be cut; Based on the distribution of disturbance sources, the original cutting trajectory of the CNC board opening machine is reconstructed by obstacle avoidance bias to obtain the preliminary obstacle avoidance trajectory of the CNC board opening machine. The initial obstacle avoidance trajectory is smoothed to obtain a smooth obstacle avoidance trajectory for the CNC plate opening machine; The anti-interference accuracy of the smooth obstacle avoidance trajectory is tuned to obtain the anti-interference path planning for the cutting motion trajectory.

[0045] The contact confidence map is traversed to determine the contact confidence state of each spatial location of the material to be cut. Areas where the confidence state does not meet the stability standard are marked as interference areas. Based on the specific location of the marked interference areas on the surface of the material, spatial locking is completed. All locked interference areas are collected according to their actual spatial locations to form a complete spatial distribution result of interference. The spatial localization of interference in the contact confidence map is completed, and finally the distribution of disturbance sources on the surface of the material to be cut is obtained.

[0046] By comparing all spatial locations of the disturbance source with the original cutting trajectory of the CNC board opening machine segment by segment, the trajectory segments in the original cutting trajectory that coincide with or are adjacent to the location of the disturbance source are identified. These trajectory segments are then directionally offset and adjusted so that the trajectory segments are completely away from the spatial range covered by the disturbance source. The adjusted trajectory segments are then connected and integrated, and the normal segments in the original cutting trajectory that are not affected by interference are retained. This completes the obstacle avoidance offset reconstruction of the original cutting trajectory, and finally the preliminary obstacle avoidance trajectory of the CNC board opening machine is obtained.

[0047] The initial obstacle avoidance trajectory is smoothed point by point at all trajectory turning points and abrupt changes caused by obstacle avoidance offset. Abrupt trajectory turning points are transformed into coherent transition forms, ensuring that the trajectory runs continuously without interruption. All uneven connecting parts in the trajectory are eliminated, making the entire trajectory smooth and uniform. This completes the smoothing of the initial obstacle avoidance trajectory, ultimately resulting in a smooth obstacle avoidance trajectory for the CNC plate opening machine.

[0048] The anti-interference capability of each running point of the smooth obstacle avoidance trajectory is verified. For points that do not meet the anti-interference standard during verification, a small directional adjustment is made to ensure that all points on the trajectory can resist the running influence caused by the disturbance of the plate surface. The adjusted trajectory is then shaped as a whole to lock the final running form and spatial position of the trajectory, and the anti-interference accuracy of the smooth obstacle avoidance trajectory is tuned. Finally, the anti-interference path planning of the cutting motion trajectory is obtained.

[0049] The beneficial effects are as follows: by using the contact state confidence map to locate the spatial distribution of all disturbance sources on the board surface, the spatial distribution of all disturbance sources can be accurately locked, providing a clear positional basis for the anti-disturbance adjustment of the cutting trajectory. Based on the distribution of disturbance sources, the obstacle avoidance offset reconstruction of the original cutting trajectory can make the trajectory completely avoid the interference area, eliminating the impact of disturbance on the cutting operation from the source. Smoothing the initial obstacle avoidance trajectory can eliminate the running stutter caused by trajectory turning and sudden change, ensuring the smoothness of the cutting motion. Adjusting the anti-disturbance accuracy of the smooth obstacle avoidance trajectory can enhance the anti-disturbance capability of the entire trajectory, ensuring that the trajectory operation is stable and without deviation throughout the entire process. The final generated anti-disturbance path planning can adapt to the actual state of the board surface, greatly improving the stability and reliability of the CNC board cutting motion, and ensuring the continuous and smooth execution of the cutting operation.

[0050] P5. Evaluate the mutual information entropy between the anti-interference path planning and the detected data of the plate to be cut, and iteratively optimize and correct the cutting layout scheme of the plate to be cut based on the evaluation results to obtain the steady-state cutting configuration of the cutting layout scheme. In this embodiment of the invention, the step of evaluating the mutual information entropy between the anti-interference path planning and the detected data of the material to be cut, and iteratively optimizing and correcting the cutting layout scheme of the material to be cut based on the evaluation results to obtain the steady-state cutting configuration of the cutting layout scheme, includes: The mutual information entropy of the anti-interference path planning and the detected data of the plate to be cut is extracted to obtain the correlation parameter between the anti-interference path planning and the detected data. The correlation parameter is embedded into the current cutting layout scheme of the material to be cut in order to identify the weak correlation areas of the current cutting layout scheme; Targeted conflict resolution is performed on the weak areas, and the resolved layout scheme is optimized, reconstructed and adjusted to obtain the candidate cutting layout scheme for the plate to be cut. The candidate cutting layout schemes are iteratively converged to obtain the steady-state cutting configuration of the cutting layout scheme.

[0051] The process of extracting mutual information entropy between the anti-interference path planning and the detected data of the material to be cut yields a correlation parameter between the anti-interference path planning and the detected data, including: Obtain the coordinates of trajectory points for anti-interference path planning; Based on the coordinates of the trajectory points, spatial interpolation sampling is performed on the detected data of the plate to be cut to obtain the detection response amplitude of the detected data; By performing discrete interval statistics on trajectory points and detection response amplitudes, the joint frequency, edge frequency, and total number of discrete intervals of trajectory point coordinates and detection response amplitudes are obtained; By mapping the joint frequency and edge frequency using probability distributions, the mutual information entropy between the anti-interference path planning and the detected data is obtained. The formula for calculating the mutual information entropy is as follows: ; in, Represents mutual information entropy. This represents the total number of discrete intervals representing the coordinates of the trajectory points. This represents the total number of discrete intervals of the detected response amplitude. This indicates that the coordinates of the trajectory point fall into the first... The interval and the detection response amplitude falls within the first interval. Joint frequency of each interval, This indicates that the coordinates of the trajectory point fall into the first... Edge frequencies of each interval This indicates that the detection response amplitude falls within the first... Edge frequencies of each interval The total number of samples representing the coordinates of trajectory points; The mutual information entropy is normalized and compressed to obtain the correlation parameter between the anti-interference path planning and the detected data.

[0052] From the starting position to the ending position of the anti-interference path planning, the spatial location information corresponding to each running trajectory point is collected point by point. All location information is organized in an orderly manner according to the order of trajectory operation to ensure that all nodes of the trajectory are covered and no location data is missed. The organized location information is formed into a complete and coherent trajectory location data set, and the comprehensive collection and organization of trajectory point coordinates of anti-interference path planning is completed, finally obtaining the trajectory point coordinates of anti-interference path planning.

[0053] Using the collected anti-interference path planning trajectory point coordinates as the core spatial reference benchmark, the detection data corresponding to each trajectory point coordinate is accurately retrieved from the detected data of the board to be cut. For the blank areas of the detection data that are not directly covered between the trajectory point coordinates, continuous interpolation and completion processing is performed. The completed detection data is bound one-to-one with the trajectory point coordinates to obtain the physical response data of the board at each trajectory point position. The spatial interpolation sampling of the detected data of the board to be cut is completed, and finally the detection response amplitude of the detected data is obtained.

[0054] The trajectory point coordinates are divided into continuous and non-overlapping independent intervals according to the actual spatial distribution range of the board. The detection response amplitude is also divided into continuous and non-overlapping independent intervals according to the actual variation range of the data. The number of data points corresponding to each trajectory point coordinate interval and each detection response amplitude interval is counted one by one. The number of data points corresponding to each trajectory point coordinate interval alone is counted. The number of data points corresponding to each detection response amplitude interval alone is counted. At the same time, the total number of all divided trajectory point coordinate discrete intervals and detection response amplitude discrete intervals is counted. The full-dimensional discrete interval statistics of trajectory points and detection response amplitudes are completed. Finally, the joint frequency, edge frequency and total number of discrete intervals of trajectory point coordinates and detection response amplitudes are obtained.

[0055] The joint frequency and edge frequency obtained from the statistics are precisely matched one by one according to the actual spatial correspondence between the trajectory point coordinates and the detection response amplitude. The frequency data is transformed into a probability distribution state that can intuitively reflect the close relationship between the anti-interference path planning and the detected data. The correlation characteristics between the two are directly determined by the overall performance of the probability distribution and transformed into a unified data form. The probability distribution mapping of the joint frequency and edge frequency is completed, and finally the mutual information entropy between the anti-interference path planning and the detected data is obtained.

[0056] All data used to calculate mutual information entropy comes from the statistical results of discrete intervals of trajectory point coordinates and detection response amplitudes. The total number of discrete intervals for trajectory point coordinates is the total number of continuous, non-overlapping independent intervals obtained by dividing the trajectory point coordinates according to their spatial distribution range. The total number of discrete intervals for detection response amplitudes is the total number of continuous, non-overlapping independent intervals obtained by dividing the detection response amplitudes according to their numerical variation range. The joint frequency of trajectory point coordinates and detection response amplitudes falling into their corresponding discrete intervals, the marginal frequency of trajectory point coordinates falling into their corresponding discrete intervals alone, and the marginal frequency of detection response amplitudes falling into their corresponding discrete intervals alone are all obtained by statistically analyzing the distribution of trajectory points and detection response amplitudes interval by interval. The total number of samples for trajectory point coordinates is the total number of all trajectory point coordinates in the anti-interference path planning. This calculation quantifies the information correlation between the anti-interference path planning and the detected data of the plate to be cut by statistically analyzing the frequency distribution relationship between the trajectory point coordinates and the detection response amplitude in the discrete interval. It reflects the degree of correlation between the two. The higher the correlation between the trajectory point coordinates and the detection response amplitude, the larger the calculated mutual information entropy value. When the two distributions are completely independent and uncorrelated, the mutual information entropy value is zero. As the correlation increases, the mutual information entropy value shows a positive trend.

[0057] The acquired mutual information entropy is standardized and integrated according to a unified and fixed standard range to eliminate numerical differences and data deviations caused by different data intervals. The mutual information entropy is transformed into correlation feature data that conforms to a unified standard, so that the data can be directly applied to the analysis and judgment of the cutting layout scheme. The normalization and compression of mutual information entropy are completed, and finally the correlation parameter between the anti-interference path planning and the detected data is obtained.

[0058] The correlation parameter is precisely bound to each independent layout area of ​​the current cutting layout scheme of the board to be cut. The actual performance of the correlation parameter directly determines the degree of fit between each layout area and the anti-interference path planning. Areas whose fit does not meet the cutting operation standard are uniformly marked. The embedding and analysis of the correlation parameter into the current cutting layout scheme are completed, and finally the weak correlation areas of the current cutting layout scheme are identified.

[0059] Targeted adjustments are made to the identified weak links to completely eliminate layout conflicts and adaptation deviations between the area and the anti-interference path planning. This ensures that the layout of the weak links is stably adapted to the trajectory planning. The overall cutting layout scheme with conflict resolution is then fine-tuned and structurally optimized across the entire area to ensure smooth connection and overall unity between the layout areas. This process completes the targeted conflict resolution and layout scheme optimization and reconstruction of the weak links, ultimately yielding candidate cutting layout schemes for the material to be cut.

[0060] The candidate cutting layout schemes are subjected to multiple cyclical comprehensive verifications, and the rationality of the layout, trajectory adaptability, operational stability and execution feasibility of the schemes are checked one by one. Based on the verification results, the layout details and structural form are continuously fine-tuned until the candidate cutting layout schemes no longer change and fully meet all cutting operation requirements. Finally, the final form and structure of the layout are locked, the iterative convergence judgment of the candidate cutting layout schemes is completed, and the steady-state cutting configuration of the cutting layout scheme is finally obtained.

[0061] The beneficial effects are as follows: comprehensive acquisition of trajectory point coordinates through anti-interference path planning can fully obtain the spatial location information of the trajectory; spatial interpolation sampling of the detected data based on the trajectory point coordinates can accurately fill data gaps and ensure that the detected data fully covers the entire trajectory; the correlation parameter generated by discrete interval statistical probability distribution mapping and normalization compression can objectively and accurately reflect the intrinsic correlation between trajectory planning and plate detection data; embedding the correlation parameter into the cutting layout scheme can quickly and accurately identify weak correlation areas in the layout that are not well adapted to the trajectory; targeted conflict resolution and optimization reconstruction can completely eliminate layout conflicts and improve the adaptability of the layout and the trajectory; iterative convergence judgment can continuously optimize the layout scheme until a stable optimal state is reached; the final generated steady-state cutting configuration can be highly consistent with the anti-interference path planning, comprehensively improving the scientific rationality and adaptability of the cutting layout, ensuring the smooth and efficient execution of the cutting operation process, and improving the yield and quality of plate cutting.

[0062] P6. Perform residual stress compensation on the steady-state cutting configuration, and compile the compensated trajectory control strategy in a hierarchical manner to obtain the stress adaptive control command for the plate to be cut.

[0063] In this embodiment of the invention, the residual stress compensation for the steady-state cutting configuration and the hierarchical compilation of the compensated trajectory control strategy to obtain the stress adaptive control command for the plate to be cut include: The residual stress field is reconstructed for the steady-state cutting configuration to obtain the spatial distribution map of the residual stress of the plate to be cut; Based on the spatial distribution map of residual stress, trajectory distortion compensation is performed on the key cutting segments in the steady-state cutting configuration to obtain the cutting trajectory offset of the key cutting segments. Using the cutting trajectory offset as the guiding vector, stress-guided reshaping of the steady-state cutting configuration is performed, and the reshaping cutting control trajectory is analyzed in a hierarchical manner to obtain the hierarchical control parameters of the steady-state cutting configuration. The layered control parameters are adapted and compiled layer by layer to obtain the stress adaptive control command for the plate to be cut.

[0064] By traversing the entire cutting area of ​​the plate to be cut covered by the steady-state cutting configuration, and combining the material properties of the plate with the heat conduction and plastic deformation state during the cutting process, the direction, range and intensity of stress distribution in each region are analyzed. The stress characteristics of each region are collected and arranged according to the actual spatial position of the plate to construct a continuous stress distribution state that completely covers all cutting areas, and the residual stress field of the steady-state cutting configuration is reconstructed, finally obtaining the spatial distribution map of residual stress of the plate to be cut.

[0065] Based on the stress state of each cutting segment in the residual stress spatial distribution map, the key cutting segment most significantly affected by residual stress in the steady-state cutting configuration is located. The trajectory offset trend and deformation direction of the key cutting segment under the continuous action of residual stress are analyzed. The specific direction and adjustment range of trajectory correction are determined, the trajectory distortion compensation of the key cutting segment is completed, and finally the cutting trajectory offset of the key cutting segment is obtained.

[0066] Using the cutting trajectory offset of the key cutting segment as the guiding vector, the overall cutting trajectory of the steady-state cutting configuration is oriented and adjusted to eliminate the interference of residual stress on the trajectory operation, forming a reshaped cutting control trajectory that is fully adapted to the stress distribution of the plate. Then, the reshaped trajectory is decomposed segment by segment according to the control level of the cutting operation, and the trajectory operation characteristics and execution requirements corresponding to different control levels are separated to complete the hierarchical analysis of the reshaped trajectory, and finally the hierarchical control parameters of the steady-state cutting configuration are obtained.

[0067] The layered control parameters of the steady-state cutting configuration are adapted layer by layer according to the control instruction specifications of the CNC plate cutting machine. The control characteristics of each layer are transformed into instruction forms that the equipment can recognize. The instruction characteristics of each layer are integrated and spliced ​​according to the running sequence of the cutting operation to form a complete and coherent control instruction sequence. The layered adaptation and compilation of the layered control parameters are completed, and finally the stress adaptive control instruction of the plate to be cut is obtained.

[0068] The beneficial effects are as follows: by reconstructing the residual stress field of the steady-state cutting configuration, the residual stress distribution of the entire plate to be cut can be accurately obtained, providing a comprehensive and clear stress state basis for trajectory distortion compensation. The trajectory distortion compensation of the key cutting segment based on the residual stress spatial distribution map can effectively eliminate the interference of residual stress on the cutting trajectory. The stress-guided reshaping with the cutting trajectory offset as the guiding vector can make the cutting trajectory fully adapt to the stress distribution of the plate. The hierarchical analysis of the reshaping trajectory can accurately extract the operating characteristics and execution requirements of different control levels. The stress adaptive control instructions generated by hierarchical adaptation compilation can directly adapt to the operating requirements of the CNC plate cutting machine, ensuring that the cutting trajectory can still maintain a stable and distortion-free operating state under the action of residual stress, and significantly improving the trajectory accuracy of the cutting operation, the forming quality of the plate, and the overall stability of the cutting operation.

[0069] like Figure 2 The diagram shown is a functional block diagram of a CNC board cutting optimization system based on path planning provided in an embodiment of the present invention.

[0070] The path planning-based CNC plate cutting optimization system 100 described in this invention can be installed in an electronic device. Depending on the functions implemented, the path planning-based CNC plate cutting optimization system 100 may include a global property detection module 101, a distributed characteristic inversion module 102, a cooperative alignment and confidence coding module 103, an anti-disturbance topology reconstruction module 104, a mutual information evaluation and iterative optimization module 105, and a stress compensation and hierarchical compilation module 106. The modules described in this invention can also be called units, referring to a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, stored in the memory of the electronic device.

[0071] In this embodiment, the functions of each module / unit are as follows: The full-domain physical property detection module 101 is used to perform full-domain physical property detection on the material to be cut by the CNC board cutting machine in order to obtain the material response characteristic sequence of the material to be cut. The distribution characteristic inversion module 102 is used to perform distribution characteristic inversion on the local stiffness distribution and local damping distribution of the plate to be cut based on the material response characteristic sequence, so as to obtain the frequency self-adaptive parameters of the plate to be cut. The cooperative alignment and confidence coding module 103 is used to perform cooperative frequency alignment between the frequency self-adaptation parameters and the current working frequency of the CNC board cutting machine, and to perform confidence feature coding on the aligned cooperative control parameters in order to construct a confidence map of the contact state between the CNC board cutting machine and the board to be cut. The anti-interference topology reconstruction module 104 is used to perform anti-interference topology reconstruction on the cutting motion trajectory of the CNC plate-opening machine based on the contact state confidence map, so as to obtain the anti-interference path planning of the cutting motion trajectory. The mutual information evaluation and iterative optimization module 105 is used to evaluate the mutual information entropy between the anti-interference path planning and the detected data of the plate to be cut, and to iteratively optimize and correct the cutting layout scheme of the plate to be cut based on the evaluation results, so as to obtain the steady-state cutting configuration of the cutting layout scheme. The stress compensation and hierarchical compilation module 106 is used to perform residual stress compensation on the steady-state cutting configuration and hierarchically compile the compensated trajectory control strategy to obtain the stress adaptive control command of the plate to be cut.

[0072] Figure 3To verify the experimental results of the correlation between different path planning strategies and the material properties data of the board, the path point interval numbers (1-20) are plotted on the horizontal axis, and the normalized mutual information entropy (range 0-1, the higher the value, the better the matching degree between the path and the material properties distribution, and the higher the information utilization efficiency) is plotted on the vertical axis, presenting the changing trends of mutual information entropy for three types of paths: traditional path, disturbance-resistant path, and optimized steady-state path. Data shows that the mutual information entropy of the traditional path generally exhibits a significant fluctuating downward trend. The initial value is approximately 0.32, reaching a peak of 0.54 in interval 4-5 before continuously declining, reaching a minimum of approximately 0.09 in interval 14. Although there is a slight rebound afterward, it remains below 0.4, reflecting that the traditional path planning does not fully utilize the material properties information of the board. As the path progresses, the matching degree with the actual distribution of the board continuously decreases, resulting in weak disturbance resistance. The initial value of the disturbance-resistant path is approximately 0.62, showing a slow downward trend overall, reaching 0.4 in interval 10, and subsequently fluctuating between 0.4 and 0.6. Although better than the traditional path, the mutual information entropy of the traditional path decreases as the path progresses. The continuous decay of mutual information entropy during path advancement indicates that conventional anti-interference strategies can only partially alleviate the problem of declining information matching degree and cannot achieve steady-state maintenance. The optimized steady-state path remains in the high range of 0.7-0.78 throughout the entire process, with minimal fluctuations. It only shows a slight decline in the ranges of 5, 11, and 19, and remains stable overall. This indicates that the optimized steady-state path planning strategy of this invention can continuously maintain a high degree of matching between the path and the material property distribution, and the information utilization efficiency remains at a high level. It solves the problem of information entropy decay during the advancement of traditional paths and conventional anti-interference paths, and provides core support for the stability of subsequent cutting.

[0073] Figure 4To verify the trajectory control capability under multiple disturbance conditions, an experimental comparison chart is used. The number of disturbance sources (2, 3, 4, representing increasing complexity) is plotted on the horizontal axis, and the cumulative trajectory offset (in mm, a higher value indicates greater disturbance impact and poorer stability) is plotted on the vertical axis. The offset performance of the original trajectory, the disturbance-resistant trajectory, and the steady-state iterative trajectory are compared. When the number of disturbance sources is 2 (low complexity condition), the cumulative offset of the original trajectory is approximately 18.2 mm, the disturbance-resistant trajectory is approximately 11.1 mm, and the steady-state iterative trajectory is approximately 8.3 mm. At this point, the difference in offset among the three types of trajectories is not significant, but the steady-state iterative trajectory already shows better control performance. When the number of disturbance sources is 3 (medium complexity condition), the offset of the original trajectory surges to 30.1 mm, an increase of over 65% compared to the case with 2 disturbance sources. The disturbance-resistant trajectory increases to 15.8 mm, an increase of approximately 42%, while the steady-state iterative trajectory only increases to 9.6 mm. With an increase of less than 16%, the offsets of traditional and conventional anti-disturbance trajectories have already increased significantly, while the steady-state iterative trajectory of this invention shows a very small increase in offset, highlighting its stability advantage. When the number of disturbance sources is 4 (high complexity condition), the offset of the original trajectory further climbs to 43.5 mm, an increase of over 139% compared to 2 disturbance sources. The anti-disturbance trajectory increases to 21.7 mm, an increase of approximately 95%, while the steady-state iterative trajectory only increases to 12.3 mm, an increase of approximately 48%, and the values ​​are only 28% of the original trajectory and 57% of the anti-disturbance trajectory. Overall, with the increase in the number of disturbance sources and the increase in the complexity of the condition, the offset of the original trajectory grows exponentially. Although the conventional anti-disturbance trajectory shows improvement, it still cannot suppress the significant increase in offset. However, the steady-state iterative trajectory of this invention always maintains a low offset range, indicating that the strategy can effectively suppress the trajectory offset caused by multiple disturbance sources and maintain excellent trajectory control stability under complex conditions, verifying the effectiveness of the anti-disturbance iterative optimization strategy of this invention in high-interference environments.

[0074] Figure 5To verify the experimental results of the residual stress compensation strategy on trajectory distortion suppression, the residual stress value (unit: MPa, range: 10-80 MPa, representing increasing residual stress level of the plate) is used as the horizontal axis, and the trajectory offset Δx (unit: mm, higher values ​​indicate more severe distortion caused by residual stress) is used as the vertical axis, comparing the trends of trajectory offset before and after compensation. Before compensation, the trajectory offset continuously increases with the increase of residual stress level, and the fluctuation range is large: when the residual stress is 10 MPa, the offset is approximately -0.1 mm, close to zero; when the residual stress rises to 18 MPa, the first peak of 1.8 mm appears; then it drops to 0.6 mm at 25 MPa, and then continues to rise, reaching 2.2 mm at 42 MPa, 3.3 mm at 58 MPa, and 4.7 mm at 75 MPa, showing a significant upward trend overall. This indicates that the higher the residual stress level, the more severe the impact on trajectory distortion, and the distortion has no obvious pattern, making it difficult to control with conventional strategies; after compensation... The trajectory offset remained within a low range of 0-1.3mm throughout the entire process, with minimal fluctuations. As the residual stress increased from 10MPa to 80MPa, the offset consistently fluctuated slightly between 0-1.3mm, with the highest value being only 1.3mm at 70MPa, far lower than the peak value before compensation. Furthermore, there was no obvious upward trend, indicating that the residual stress compensation strategy of this invention can effectively counteract the distortion effect of residual stress on the trajectory. Regardless of the level of residual stress, the trajectory offset can be controlled at an extremely low level, significantly improving the stability and accuracy of the cutting trajectory and solving the problem of trajectory distortion when cutting high residual stress plates.

[0075] Figure 6To verify the experimental comparison of the convergence efficiency and accuracy of the optimization algorithm, the horizontal axis is plotted with the number of optimization iterations (steps 1-10) and the vertical axis is plotted with the optimization error (in %), where a lower value indicates higher optimization accuracy and better convergence. The error change process of the traditional scheme and the scheme of this invention is compared. The initial optimization error of the traditional scheme is 19.2%, showing a slow downward trend overall, with extremely slow convergence speed: from step 1 to step 4, the error decreases from 19.2% to 12.8%, a reduction of only about 33%; from step 5 to step 8, the error decreases from 10.7% to 5.5%, a reduction of about 49%; in subsequent steps 9-10, the error remains around 8.2%, no longer decreasing significantly, and the final error is still above 8%, indicating that the traditional scheme has slow convergence speed and low optimization accuracy. Even after multiple iterations, it cannot achieve effective error convergence, and the optimization effect is limited. The present invention… The initial optimization error of the proposed solution is 5.9%, which is much lower than that of the traditional solution, and the convergence speed is extremely fast: from step 1 to step 3, the error drops rapidly from 5.9% to 0.5%, a reduction of over 91%; from step 4 to step 10, the error remains within the range of 0-0.5% with slight fluctuations, and steadily approaches zero. This shows that the proposed solution can achieve rapid error convergence in very few iterations, and the final optimization accuracy is much higher than that of the traditional solution. It solves the problems of slow convergence and low accuracy of traditional optimization algorithms, and greatly improves the efficiency and final effect of path optimization.

[0076] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0077] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0078] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0079] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0080] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0081] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A path planning based numerical control plate cutting optimization method for a plate cutting machine, characterized in that, The method includes: The material properties of the sheet metal to be cut by the CNC plate cutting machine are tested over the entire range to obtain the material response characteristic sequence of the sheet metal to be cut. Based on the material response feature sequence, the distribution characteristics of the local stiffness distribution and local damping distribution of the plate to be cut are inverted to obtain the frequency self-adaptive parameters of the plate to be cut. The frequency self-adaptation parameters are coordinated with the current working frequency of the CNC plate cutting machine, and the coordinated control parameters obtained after alignment are encoded with confidence features to construct a confidence map of the contact state between the CNC plate cutting machine and the plate to be cut. The coordinated control parameters are used to characterize the control parameters that enable the CNC plate cutting machine and the plate to be cut to achieve the optimal coordinated working state. Based on the contact state confidence map, the cutting motion trajectory of the CNC plate opening machine is reconstructed to obtain the anti-interference path planning of the cutting motion trajectory. The mutual information entropy of the anti-interference path planning and the detected data of the plate to be cut is evaluated, and the cutting layout scheme of the plate to be cut is iteratively optimized and corrected according to the evaluation results to obtain the steady-state cutting configuration of the cutting layout scheme. Residual stress compensation is performed on the steady-state cutting configuration, and the compensated trajectory control strategy is compiled in a hierarchical manner to obtain the stress adaptive control command for the plate to be cut.

2. The path planning based CNC plate cutting optimization method of claim 1, wherein, The process of performing full-domain physical property testing on the material to be cut by the CNC board cutting machine to obtain the material response characteristic sequence of the material to be cut includes: A wide-frequency sweep excitation was performed on the material to be cut by the CNC plate cutting machine, and the multi-channel time-domain response waveform of the material to be cut was acquired at different excitation frequencies. Based on the multi-channel time-domain response waveform, the physical properties of the material to be cut are analyzed by inversion to obtain the substrate physical property parameters of the material to be cut. Multidimensional feature fusion of the substrate physical property parameters yields the global physical property tensor field of the plate to be cut; By performing ordered feature mapping on the global property tensor field, the material response feature sequence of the plate to be cut is obtained.

3. The path planning based CNC plate cutting optimization method of claim 1, wherein, The method involves inverting the distribution characteristics of the local stiffness and damping distributions of the plate to be cut based on the material response feature sequence to obtain the frequency self-adaptive parameters of the plate, including: Local feature decoupling is performed on the material response feature sequence to obtain a subset of local response features of the plate to be cut; Based on the local response feature subset, mechanical parameters are inverted in local blocks of the plate to be cut to obtain the local stiffness distribution parameters and local damping distribution parameters of the plate to be cut. The frequency response amplitude spectrum of the plate to be cut is obtained by superimposing the local stiffness distribution parameters and the local damping distribution parameters with their frequency response characteristics. Frequency domain feature matching is performed on the frequency response amplitude spectrum to obtain the frequency self-adaptive parameters of the plate to be cut.

4. The path planning based CNC plate cutting optimization method of claim 1, wherein, The step of coordinating the frequency self-adaptation parameters with the current operating frequency of the CNC board cutting machine, and encoding the confidence features of the aligned coordinating control parameters to construct a confidence map of the contact state between the CNC board cutting machine and the board to be cut, includes: A frequency offset coherent analysis is performed on the frequency self-adaptation parameters and the current operating frequency of the CNC board opening machine to obtain the frequency offset characteristic quantity between the frequency self-adaptation parameters and the current operating frequency. Based on the frequency offset characteristic, the working frequency of the CNC plate cutting machine is normalized in real time to obtain the working condition coupling characteristics between the CNC plate cutting machine and the plate to be cut. The coupling gain of the working condition coupling characteristics is optimized, and the optimized cooperative control parameters are embedded and encoded to obtain the contact state feature spectrum between the CNC plate cutter and the plate to be cut. Confidence topological reconstruction of the contact state feature spectrum yields a confidence spectrum of the contact state between the CNC board cutting machine and the board to be cut.

5. The path planning based CNC panel saw slab cutting optimization method of claim 4, wherein, The process of optimizing the coupling gain of the working condition coupling characteristics and embedding and encoding the optimized cooperative control parameters to obtain the contact state feature spectrum between the CNC slitting machine and the plate to be cut includes: The operating condition coupling features are de-aggregated and separated to obtain the main modal components and residual perturbation components of the operating condition coupling features; Anomaly fluctuation screening was performed on the residual perturbation components to obtain outlier feature fragments in the residual perturbation components; Outlier feature segments are removed from the residual perturbation components, and the purified residual components obtained by removal are coherently superimposed and fused with the main mode components to obtain the collaborative control parameters of the operating condition coupling features. Locally sensitive encoding is performed on the collaborative control parameters, and the encoding results are reduced and aggregated to obtain the contact state feature spectrum between the CNC slitting machine and the plate to be cut.

6. The method for optimizing sheet metal cutting in a CNC slitting machine based on path planning as described in claim 1, characterized in that, The method of performing anti-interference topology reconstruction on the cutting motion trajectory of the CNC plate-opening machine based on the contact state confidence map to obtain anti-interference path planning for the cutting motion trajectory includes: Disturbance spatial localization is performed on the contact state confidence spectrum to obtain the distribution of disturbance sources on the surface of the plate to be cut; Based on the distribution of disturbance sources, the original cutting trajectory of the CNC board opening machine is reconstructed by obstacle avoidance bias to obtain the preliminary obstacle avoidance trajectory of the CNC board opening machine. The initial obstacle avoidance trajectory is smoothed to obtain a smooth obstacle avoidance trajectory for the CNC plate opening machine; The anti-interference accuracy of the smooth obstacle avoidance trajectory is tuned to obtain the anti-interference path planning for the cutting motion trajectory.

7. The path planning based CNC plate cutting optimization method of claim 1, wherein, The process of evaluating the mutual information entropy between the anti-interference path planning and the detected data of the material to be cut, and iteratively optimizing and correcting the cutting layout scheme of the material to be cut based on the evaluation results to obtain the steady-state cutting configuration of the cutting layout scheme, includes: The mutual information entropy of the anti-interference path planning and the detected data of the plate to be cut is extracted to obtain the correlation parameter between the anti-interference path planning and the detected data. The correlation parameter is embedded into the current cutting layout scheme of the material to be cut in order to identify the weak correlation areas of the current cutting layout scheme; Targeted conflict resolution is performed on the weak areas, and the resolved layout scheme is optimized, reconstructed and adjusted to obtain the candidate cutting layout scheme for the plate to be cut. The candidate cutting layout schemes are iteratively converged to obtain the steady-state cutting configuration of the cutting layout scheme.

8. The path planning based CNC panel saw slab cutting optimization method of claim 7, wherein, The process of extracting mutual information entropy between the anti-interference path planning and the detected data of the material to be cut yields a correlation parameter between the anti-interference path planning and the detected data, including: Obtain the coordinates of trajectory points for anti-interference path planning; Based on the coordinates of the trajectory points, spatial interpolation sampling is performed on the detected data of the plate to be cut to obtain the detection response amplitude of the detected data; By performing discrete interval statistics on trajectory points and detection response amplitudes, the joint frequency, edge frequency, and total number of discrete intervals of trajectory point coordinates and detection response amplitudes are obtained; By mapping the joint frequency and edge frequency using probability distributions, the mutual information entropy between the anti-interference path planning and the detected data is obtained. The formula for calculating the mutual information entropy is as follows: ; in, Represents mutual information entropy. This represents the total number of discrete intervals representing the coordinates of the trajectory points. This represents the total number of discrete intervals of the detected response amplitude. This indicates that the coordinates of the trajectory point fall into the first... The interval and the detection response amplitude falls within the first interval. Joint frequency of each interval, This indicates that the coordinates of the trajectory point fall into the first... Edge frequencies of each interval This indicates that the detection response amplitude falls within the first... Edge frequencies of each interval The total number of samples representing the coordinates of trajectory points; The mutual information entropy is normalized and compressed to obtain the correlation parameter between the anti-interference path planning and the detected data.

9. The path planning based CNC panel cutting optimization method of claim 1, wherein, The process of compensating for residual stress in the steady-state cutting configuration and performing hierarchical compilation of the compensated trajectory control strategy to obtain stress adaptive control instructions for the plate to be cut includes: The residual stress field is reconstructed for the steady-state cutting configuration to obtain the spatial distribution map of the residual stress of the plate to be cut; Based on the spatial distribution map of residual stress, trajectory distortion compensation is performed on the key cutting segments in the steady-state cutting configuration to obtain the cutting trajectory offset of the key cutting segments. Using the cutting trajectory offset as the guiding vector, stress-guided reshaping of the steady-state cutting configuration is performed, and the reshaping cutting control trajectory is analyzed in a hierarchical manner to obtain the hierarchical control parameters of the steady-state cutting configuration. The layered control parameters are adapted and compiled layer by layer to obtain the stress adaptive control command for the plate to be cut.

10. A path planning-based optimization system for CNC plate cutting machine, characterized in that, The system for implementing the path planning-based optimization method for cutting sheet metal in a CNC board cutting machine as described in claim 1 includes: The full-domain physical property detection module is used to perform full-domain physical property detection on the material to be cut by the CNC plate cutting machine in order to obtain the material response characteristic sequence of the material to be cut. The distribution characteristic inversion module is used to perform distribution characteristic inversion on the local stiffness distribution and local damping distribution of the plate to be cut based on the material response characteristic sequence, so as to obtain the frequency self-adaptive parameters of the plate to be cut. The collaborative alignment and confidence coding module is used to perform collaborative frequency alignment between the frequency self-adaptation parameters and the current operating frequency of the CNC board cutting machine, and to perform confidence feature coding on the aligned collaborative control parameters in order to construct a confidence map of the contact state between the CNC board cutting machine and the board to be cut. The anti-interference topology reconstruction module is used to perform anti-interference topology reconstruction on the cutting motion trajectory of the CNC plate-opening machine based on the contact state confidence map, so as to obtain the anti-interference path planning of the cutting motion trajectory. The mutual information evaluation and iterative optimization module is used to evaluate the mutual information entropy between the anti-interference path planning and the detected data of the plate to be cut, and to iteratively optimize and correct the cutting layout scheme of the plate to be cut based on the evaluation results, so as to obtain the steady-state cutting configuration of the cutting layout scheme. The stress compensation and hierarchical compilation module is used to perform residual stress compensation on the steady-state cutting configuration and to perform hierarchical compilation on the compensated trajectory control strategy to obtain the stress adaptive control command for the plate to be cut.

Images