Dynamic control method and system for stamping machine tools

Abstract

The present invention discloses a dynamic control method and system for a stamping machine tool, which relates to the field of intelligent manufacturing. The method includes: obtaining product design information of stamping parts to be produced, performing modeling and restoration, and obtaining a target stamping part model; configuring a camera view to perform image acquisition and export, and obtaining a two-dimensional model image set; synchronizing to a stamping feature recognition network to perform stamping feature recognition; performing process step sequence sorting, obtaining a target process sequence, performing reverse engineering analysis, and obtaining multiple process stamping models; mapping and performing stamping process parameter analysis, and obtaining multiple groups of stamping process parameter sequences; generating a stamping dynamic control sequence, and executing batch production of stamping parts to be produced. The method solves the technical problem that the existing stamping machine tool control relies on manual experience, resulting in insufficient stamping precision and low production efficiency, and achieves the technical effect of reducing the reliance on manual experience, thereby improving stamping precision and production efficiency.

Description

Technical Field

[0001] The present application relates to fields related to intelligent manufacturing, and in particular to a dynamic control method and system for a stamping machine tool. Background Art

[0002] In industrial production, stamping is a key processing step for manufacturing various metal parts. At present, stamping machine tool control often relies on traditional control strategies, which are highly dependent on the experience of engineers. This results in the need for readjustment and testing every time a new product is produced, wasting a lot of time and resources. In addition, due to the lack of accurate model and algorithm support, traditional stamping control methods are difficult to achieve high-precision production requirements. At the same time, the need to frequently adjust and optimize control strategies will seriously affect production efficiency.

[0003] In the current related technologies, the control of stamping machine tools relies on manual experience, which leads to technical problems such as insufficient stamping accuracy and low production efficiency. Summary of the invention

[0004] The present application provides a dynamic control method and system for stamping machine tools, and adopts technical means such as modeling and restoration that introduces product design information, image acquisition, stamping feature recognition, process step sequence sorting, and stamping process parameter analysis to achieve dynamic control of stamping machine tools, thereby achieving the technical effect of reducing dependence on manual experience and thus improving stamping accuracy and production efficiency.

[0005] The present application provides a dynamic control method for a stamping machine tool, comprising:

[0006] Interactively obtain product design information of the stamping part to be produced, and perform modeling and restoration based on the product design information to obtain a target stamping part model;

[0007] Preset a standard product acquisition angle, and configure the camera viewing angle based on the standard product acquisition angle to perform image acquisition and export on the target stamping part model to obtain a two-dimensional model image set;

[0008] Pre-constructing a stamping feature recognition network, and synchronizing the two-dimensional model image set to the stamping feature recognition network for stamping feature recognition, and outputting K types of stamping features, where K is a positive integer;

[0009] Pre-constructing a stamping feature process mapping, and sorting the process steps of the K stamping features according to the stamping feature process mapping to obtain a target process sequence, and performing reverse engineering analysis on the target stamping part model according to the target process sequence to obtain multiple process stamping models;

[0010] Using the K stamping features to map the multiple process stamping models to perform stamping process parameter analysis, and obtain multiple sets of stamping process parameter sequences;

[0011] The multiple groups of stamping process parameter sequences are fitted to the target process sequence to generate a stamping dynamic control sequence, and the stamping dynamic control sequence is used to execute batch production of the stamping parts to be produced.

[0012] In a possible implementation, a stamping feature recognition network is pre-built, and the two-dimensional model image set is synchronized to the stamping feature recognition network for stamping feature recognition, K stamping features are output, and the following processing is performed:

[0013] Pre-constructing a random operator library, and randomly calling M preset operators from the random operator library according to the M product acquisition angles of the standard product acquisition angle, where M is a positive integer;

[0014] Interactively obtain a standard stamping feature set, and perform stamping feature identification on multiple real-collected image sets of multiple sample stamping parts obtained interactively based on the standard stamping feature set to obtain multiple stamping identification image sets;

[0015] Constructing and generating M stamping feature recognition sub-channels based on the M preset operators, and using the multiple stamping mark image sets to construct the M stamping feature recognition sub-channels;

[0016] The M stamping feature recognition sub-channels are arranged in parallel, and a feature aggregation channel is configured at the output ends of the M stamping feature recognition sub-channels to complete the construction of the stamping feature recognition network;

[0017] Synchronize the mapping of the M model images of the two-dimensional model image set to the M stamping feature recognition sub-channels of the stamping feature recognition network to perform stamping feature recognition, and obtain M stamping feature subsets;

[0018] The M stamping feature subsets are synchronized to the feature aggregation channel for repeated feature aggregation, and the K stamping features are output.

[0019] In a possible implementation, a stamping feature process mapping is pre-constructed, and the process steps of the K stamping features are sorted in order according to the stamping feature process mapping to obtain a target process sequence, and the target stamping part model is reverse-engineered according to the target process sequence to obtain multiple process stamping models, and the following processing is performed:

[0020] interactively obtaining a standard stamping process sequence, wherein a plurality of stamping processes constituting the standard stamping process sequence are mapped to a plurality of shape feature sets;

[0021] Associatively storing the plurality of stamping processes and the plurality of shape feature sets based on the knowledge graph, and generating the stamping feature process mapping;

[0022] The K types of stamping features are matched with the stamping feature process mapping to perform stamping process calling to obtain multiple target calling processes;

[0023] Rearrange the processes of the plurality of target calling processes based on the standard stamping process sequence to generate the target process sequence;

[0024] Reverse engineering analysis is performed on the target stamping part model according to the target process sequence to obtain the multiple process stamping models.

[0025] In a possible implementation, reverse engineering analysis is performed on the target stamping part model according to the target process sequence to obtain the multiple process stamping models, and the following processing is performed:

[0026] The K types of stamping features are divided by using the multiple target calling processes to obtain multiple groups of stamping features;

[0027] Reversely sorting the plurality of groups of stamping features according to the target process sequence to obtain a stamping feature group sequence;

[0028] Execute sequential call on the stamping feature group sequence to obtain a first group of sequential stamping features, and perform feature elimination on the target stamping part model with the first group of sequential stamping features as constraints to obtain a first process stamping model;

[0029] Execute sequential call on the stamping feature group sequence to obtain a second group of sequential stamping features, and perform feature elimination on the first process stamping model with the second group of sequential stamping features as constraints to obtain a second process stamping model;

[0030] By analogy, the stamping feature group sequence is used as a feature elimination constraint, and multiple rounds of feature elimination and model retention are performed on the target stamping part model to obtain the multiple process stamping models.

[0031] In a possible implementation, the K stamping features are used to perform stamping process parameter analysis on the multiple process stamping model mappings to obtain multiple sets of stamping process parameter sequences, and the following processing is performed:

[0032] Reversely call the stamping feature group sequence to obtain a first reverse stamping feature group, and reversely call the multiple process stamping models to obtain a first reverse process stamping model;

[0033] The first set of reverse stamping features is used to traverse the first reverse process stamping model to perform parameter calling to obtain a first stamping deformation parameter set;

[0034] Similarly, the stamping feature group sequence and the multiple process stamping models are reversely called, and parameter calls are performed based on the reverse call result mapping to obtain multiple stamping deformation parameter sets;

[0035] Pre-constructing a stamping process matching network, wherein the stamping process matching network includes a plurality of stamping process matching branches mapped to the plurality of stamping processes, and the plurality of stamping process matching branches are connected in parallel;

[0036] According to the target process sequence, the stamping process matching network is activated by stamping process matching branch mapping to obtain a plurality of target matching branches;

[0037] The plurality of stamping deformation parameter sets are mapped and synchronized to the plurality of target matching branches according to the target process sequence to obtain the plurality of sets of stamping process parameter sequences.

[0038] In a possible implementation, a stamping process matching network is pre-constructed, wherein the stamping process matching network includes a plurality of stamping process matching branches mapped to the plurality of stamping processes, and the plurality of stamping process matching branches are connected in parallel, and the following processing is performed:

[0039] Interactively obtaining a plurality of sample stamping deformation dimensions, a plurality of sample stamping control parameters, and a plurality of sample stamping tool calls of a first stamping process among the plurality of stamping processes;

[0040] Based on the knowledge graph, the plurality of sample stamping deformation dimensions, the plurality of sample stamping control parameters and the plurality of sample stamping tool calls are associated and stored to construct a first stamping process matching layer;

[0041] Interactively obtaining a sample stamping sequence of the plurality of sample stamping deformation sizes, and generating a first stamping process sorting layer based on the sample stamping sequence;

[0042] Connecting the output end of the first stamping process matching layer to the input end of the first stamping process sorting layer to construct and generate a first stamping process matching branch;

[0043] By analogy, the multiple stamping process matching branches mapped to the multiple stamping procedures are constructed and obtained, and the multiple stamping process matching branches are connected in parallel to construct and generate the stamping process matching network.

[0044] In a possible implementation, reverse engineering analysis is performed on the target stamping part model according to the target process sequence to obtain the multiple process stamping models, and then the following processing is performed:

[0045] Grouping the multiple process stamping models adjacently according to the target process sequence to obtain multiple groups of process stamping models;

[0046] Obtaining a first group of process stamping models from the multiple groups of process stamping models;

[0047] Performing a stamping intersection analysis on the first group of process stamping models, and if a stamping intersection area exists, performing a stamping deviation analysis on the stamping intersection area, and optimizing the target process sequence based on the stamping deviation analysis result;

[0048] Similarly, by performing stamping intersection analysis on the multiple groups of process stamping models, and optimizing the target process sequence based on the analysis results, an optimized process sequence is generated;

[0049] The plurality of process stamping models are sequenced and optimized based on the optimized process sequence.

[0050] The present application also provides a dynamic control system for a stamping machine tool, comprising:

[0051] A target stamping part model acquisition module, which is used to interactively obtain product design information of the stamping part to be produced, and perform modeling and restoration according to the product design information to obtain a target stamping part model;

[0052] A two-dimensional model image set acquisition module, which is used to preset a standard product acquisition angle, and configure a camera viewing angle based on the standard product acquisition angle to perform image acquisition and export on the target stamping part model to obtain a two-dimensional model image set;

[0053] A stamping feature recognition module, which is used to pre-build a stamping feature recognition network, synchronize the two-dimensional model image set to the stamping feature recognition network for stamping feature recognition, and output K types of stamping features, where K is a positive integer;

[0054] Multiple process stamping model acquisition modules, the multiple process stamping model acquisition modules are used to pre-build a stamping feature process mapping, and sort the process steps of the K types of stamping features according to the stamping feature process mapping to obtain a target process sequence, and perform reverse engineering analysis on the target stamping part model according to the target process sequence to obtain multiple process stamping models;

[0055] A plurality of stamping process parameter sequence acquisition modules, wherein the plurality of stamping process parameter sequence acquisition modules are used to perform stamping process parameter analysis on the plurality of process stamping model mappings using the K stamping features to obtain a plurality of stamping process parameter sequences;

[0056] A stamping dynamic control sequence generation module is used to fit the multiple sets of stamping process parameter sequences to the target process sequence, generate a stamping dynamic control sequence, and use the stamping dynamic control sequence to execute batch production of the stamping parts to be produced.

[0057] The dynamic control method and system for stamping machine tools proposed in this application first interactively obtain the product design information of the stamping parts to be produced, perform modeling and restoration based on the product design information, obtain the target stamping part model, and then preset the standard product acquisition angle, configure the camera angle based on the standard product acquisition angle to perform image acquisition and export on the target stamping part model, obtain a two-dimensional model image set, then pre-build a stamping feature recognition network, synchronize the two-dimensional model image set to the stamping feature recognition network for stamping feature recognition, and output K types of stamping features, where K is a positive integer, and then pre-build a stamping feature process mapping, according to the stamping feature. The feature process mapping arranges the process steps of K kinds of stamping features in order to obtain the target process sequence, and performs reverse engineering analysis on the target stamping part model according to the target process sequence to obtain multiple process stamping models. Then, K kinds of stamping features are used to map multiple process stamping models to perform stamping process parameter analysis to obtain multiple groups of stamping process parameter sequences. Finally, the multiple groups of stamping process parameter sequences are fitted to the target process sequence to generate a stamping dynamic control sequence. The stamping dynamic control sequence is used to execute batch production of the stamping parts to be produced, thereby achieving the technical effect of reducing dependence on manual experience and thus improving stamping accuracy and production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0059] Figure 1 A schematic flow chart of a dynamic control method for a stamping machine tool provided in an embodiment of the present application;

[0060] Figure 2 A schematic structural diagram of a dynamic control system for a stamping machine tool provided in an embodiment of the present application. DETAILED DESCRIPTION

[0061] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0062] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0063] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0064] The present application embodiment provides a dynamic control method for a stamping machine tool, such as Figure 1 As shown, the method includes:

[0065] Step S100, interactively obtain the product design information of the stamping parts to be produced, and perform modeling and restoration based on the product design information to obtain the target stamping part model. Specifically, communicate with the product design department or relevant personnel to collect the product design information of the stamping parts to be produced, including CAD drawings, 3D models, product specifications, material types, etc. Start the modeling software, perform modeling based on the collected product design information, and gradually build various parts of the stamping parts to be produced according to the product design requirements, including the main structure, detailed features, assembly holes, etc., verify the built model, and obtain the target stamping part model.

[0066] Step S200, preset a standard product acquisition angle, and configure the camera viewing angle based on the standard product acquisition angle to perform image acquisition and export on the target stamping part model to obtain a two-dimensional model image set. Specifically, according to the product structure characteristics and stamping process requirements of the target stamping part model, determine the key parts and angles that need to be collected, analyze and preset multiple standard product acquisition angles based on product design drawings, 3D models or actual samples, adjust the position and angle of the camera according to the preset standard product acquisition angle, use the camera to perform multi-angle image acquisition on the target stamping part model, organize the collected images, and form a two-dimensional model image set.

[0067] Step S300, pre-construct a stamping feature recognition network, and synchronize the two-dimensional model image set to the stamping feature recognition network for stamping feature recognition, and output K types of stamping features, where K is a positive integer. Specifically, based on deep learning technology, a stamping feature recognition network is pre-trained, and a large number of collected image data sets containing different stamping features are used to train the stamping feature recognition network, so that the stamping feature recognition network can accurately identify various features of stamping parts. The two-dimensional model image set is pre-processed and loaded into the stamping feature recognition network. The stamping feature recognition network processes the input image layer by layer and identifies K different stamping features in the image, including shape features, size features, texture features, etc.

[0068] In a possible implementation, step S300 further includes step S310, pre-building a random operator library, and randomly calling to obtain M preset operators from the random operator library according to the M product acquisition angles of the standard product acquisition angle, where M is a positive integer. Specifically, the random operator library contains a variety of different types of neural network models or modules for performing different feature extraction and conversion tasks, and randomly selecting M preset operators from the random operator library according to the M product acquisition angles of the standard product acquisition angle, and each product acquisition angle corresponds to a specific preset operator for extracting stamping features related to the product acquisition angle. Step S320, interactively obtain a standard stamping feature set, and perform stamping feature identification on multiple real-collected image sets of multiple sample stamping parts obtained interactively based on the standard stamping feature set to obtain multiple stamping identification image sets. Specifically, the standard stamping feature set contains descriptions and definitions of various standard features of stamping parts, which are used for feature identification and recognition. The standard stamping feature set is used to locate and mark the stamping features in multiple real-sampled image sets of multiple sample stamping parts, and generate multiple stamping identification image sets containing feature location and type information. Step S330, based on the M preset operators, M stamping feature recognition sub-channels are constructed and generated, and the M stamping feature recognition sub-channels are constructed using the multiple stamping identification image sets. Specifically, M preset operators selected from a random operator library are used to construct M independent stamping feature recognition sub-channels, each of which is responsible for extracting stamping features from different angles and levels, and multiple stamping identification image sets are used to train and optimize the M stamping feature recognition sub-channels, so that the stamping feature recognition sub-channels can accurately identify the stamping features in the image. Step S340, the M stamping feature recognition sub-channels are arranged in parallel, and a feature aggregation channel is configured at the output end of the M stamping feature recognition sub-channels to complete the construction of the stamping feature recognition network. Specifically, the M stamping feature recognition sub-channels are connected in parallel to form a multi-channel feature extraction network, and each stamping feature recognition sub-channel independently extracts stamping features and outputs the results. A feature aggregation channel is configured at the output end of the M stamping feature recognition sub-channels, and the feature aggregation channel is used to fuse and aggregate the features extracted by each stamping feature recognition sub-channel to generate a final feature representation. Step S350, the M model images of the two-dimensional model image set are mapped and synchronized to the M stamping feature recognition sub-channels of the stamping feature recognition network for stamping feature recognition to obtain M stamping feature subsets. Specifically, the M model images in the two-dimensional model image set are respectively mapped to M stamping feature recognition sub-channels, and the M model images correspond to M product acquisition angles respectively. Each stamping feature recognition sub-channel extracts and recognizes features of the corresponding model image to generate M stamping feature subsets.Step S360, synchronize the M stamping feature subsets to the feature aggregation channel for repeated feature aggregation, and output the K stamping features. Specifically, the M stamping feature subsets are input into the feature aggregation channel, which fuses and aggregates these stamping feature subsets, eliminates redundant and repeated feature information, and generates the final K stamping feature outputs. This implementation method enables the stamping feature recognition network to extract and recognize stamping features more comprehensively by introducing a random operator library and constructing multiple parallel stamping feature recognition sub-channels, thereby achieving the technical effect of improving the accuracy and robustness of feature recognition.

[0069] Step S400, pre-constructing a stamping feature process mapping, and sorting the process steps of the K stamping features according to the stamping feature process mapping to obtain a target process sequence, and reverse engineering analysis of the target stamping part model according to the target process sequence to obtain multiple process stamping models. Specifically, the production data of historical stamping parts is collected and analyzed, including the processing sequence of various stamping features, the molds used and the process parameters, etc., and based on the collected data, a mapping relationship between stamping features and processes is constructed. The stamping feature process mapping defines the sequence of different stamping features in the processing process and the required process steps. The identified K stamping features are input into the stamping feature process mapping, and the position and sequence of each stamping feature in the process steps are determined according to the mapping relationship, and a stamping feature sequence arranged in the process step sequence is output, that is, the target process sequence. Based on the target process sequence, starting from the final target stamping part model, the stamping features and shape changes required for each process step are reversely analyzed, and the intermediate stamping part model in each process step is gradually restored using CAD or other reverse engineering tools. Through reverse engineering analysis, a series of intermediate stamping part models are obtained, which represent the product status under different process steps. The obtained intermediate stamping part models are verified to obtain the multiple process stamping models.

[0070] In a possible implementation, step S400 further includes step S410, interactively obtaining a standard stamping process sequence, wherein a plurality of stamping processes constituting the standard stamping process sequence are mapped to a plurality of shape feature sets. Specifically, the standard stamping process sequence is obtained from a process expert or a database through a user interface or other interactive methods, and the standard stamping process sequence is a reasonable arrangement of a series of stamping processes based on experience and knowledge. The fact that a plurality of stamping processes constituting the standard stamping process sequence should be mapped to a plurality of shape feature sets indicates that each stamping process is associated with a specific shape feature set, and the shape feature set describes the shape and features of the stamped parts that need to be achieved in the stamping process. Step S420, the plurality of stamping processes and the plurality of shape feature sets are associated and stored based on the knowledge graph to generate the stamping feature process mapping. Specifically, a knowledge graph is constructed. The knowledge graph is a structured knowledge representation method used to show the association and dependency between different entities. The knowledge graph contains stamping processes, shape feature sets and the association relationship between them. In the knowledge graph, multiple stamping processes and multiple shape feature sets are stored in association, that is, each stamping process is connected to the corresponding shape feature set in the knowledge graph to form a clear mapping relationship. Based on the information stored in association, a stamping feature process mapping is generated. The stamping feature process mapping is a structured representation that clarifies the correspondence between different stamping features and stamping processes. Step S430, the K types of stamping features are matched with the stamping feature process mapping to perform stamping process calls, and multiple target calling processes are obtained. Specifically, the identified K types of stamping features are matched with the stamping feature process mapping, and the stamping process corresponding to the K types of stamping features is found by comparing the K types of stamping features with the shape feature sets in the stamping feature process mapping. According to the matching results, the corresponding stamping process is called to obtain multiple target calling processes. The multiple target calling processes are the processing steps required to achieve specific stamping features. Step S440, rearrange the processes of the multiple target calling processes based on the standard stamping process sequence to generate the target process sequence. Specifically, based on the standard stamping process sequence, rearrange the multiple target calling processes to ensure that the sequence of the stamping processes meets the requirements of actual production and the processing logic of the stamping parts, and finally generate the target process sequence, which is a plurality of stamping processes arranged in the rearranged order. Step S450, reverse engineering analysis is performed on the target stamping part model according to the target process sequence to obtain the plurality of process stamping models. This implementation method can more accurately match stamping features with stamping processes and generate a more reasonable target process sequence by introducing knowledge graphs and process rearrangement based on the standard stamping process sequence, thereby achieving the technical effect of improving production efficiency and product quality.

[0071] In a possible implementation, step S450 further includes step S451, using the multiple target calling processes to divide the K types of stamping features to obtain multiple groups of stamping features. Specifically, the K types of stamping features are analyzed in terms of hierarchical relationship, sequence and mutual influence, and the K types of stamping features are allocated to each target calling process according to the logical relationship and dependency between the K types of stamping features. Through the division process, the K types of stamping features are divided into multiple groups, and each group of features corresponds to a target calling process. Step S452, according to the target process sequence, the multiple groups of stamping features are reversely sorted to obtain a stamping feature group sequence. Specifically, according to the target process sequence, that is, the stamping process sequence after optimization and rearrangement, the reverse processing order of the multiple groups of stamping features is determined, that is, starting from the state of the final product (target stamping part model), the product state before each process is reversely traced. After the reverse sorting process, a stamping feature group sequence is obtained, and the stamping feature group sequence arranges multiple groups of stamping features in an orderly manner according to the needs of reverse engineering analysis. Step S453, sequentially call the stamping feature group sequence to obtain the first group of sequential stamping features, and use the first group of sequential stamping features as constraints to perform feature elimination on the target stamping part model to obtain the first process stamping model. Specifically, according to the order of the stamping feature group sequence, first call the first group of sequential stamping features, then use the first group of sequential stamping features as constraints to perform feature elimination on the target stamping part model, that is, identify and remove the parts corresponding to the first group of sequential stamping features in the target stamping part model to show the intermediate state before the corresponding process is completed, and obtain the first process stamping model through feature elimination. The first process stamping model represents the product state before the last process is completed. Step S454, sequentially call the stamping feature group sequence to obtain the second group of sequential stamping features, and use the second group of sequential stamping features as constraints to perform feature elimination on the first process stamping model to obtain the second process stamping model. Specifically, continue to call the second group of sequential stamping features in the order of the stamping feature group sequence, and use the second group of sequential stamping features as a constraint to perform feature elimination on the first process stamping model obtained in step S453, that is, identify and remove the parts corresponding to the second group of sequential stamping features in the first process stamping model, and obtain the second process stamping model through this round of feature elimination, which represents the product state before the second-to-last process is completed. Step S455, and so on, use the stamping feature group sequence as a feature elimination constraint, perform multiple rounds of feature elimination and model retention on the target stamping part model, and obtain the multiple process stamping models.Specifically, the above steps are iteratively executed, the stamping feature group sequence is called in turn, and the process stamping model of the corresponding process is feature eliminated. Through multiple rounds of feature elimination and model retention operations, the intermediate state after each process in the entire production process is gradually restored. Finally, multiple process stamping models corresponding to the target process sequence are obtained. The multiple process stamping models show the gradual deformation process from raw materials to final products. This implementation method uses the feature elimination method to obtain multiple process stamping models. Feature elimination is based on the stamping features of the product, and these features are gradually removed from the target stamping part model, thereby restoring the product state after different processes are completed. This method can accurately simulate the impact of each process on the product, ensuring that the multiple process stamping models obtained are highly consistent with the actual production process, and achieving a high accuracy technical effect.

[0072] In a possible implementation, the target stamping part model is reverse-engineered and analyzed according to the target process sequence to obtain the multiple process stamping models. After that, step S400 further includes step S460, which groups the multiple process stamping models into adjacent groups according to the target process sequence to obtain multiple groups of process stamping models. Specifically, according to the target process sequence, the logical relationship between the various processes, the degree of dependence, and their impact on the final product are analyzed, and the multiple process stamping models are grouped according to the continuity of the processes, the similarity of the stamping features, the deformation stage of the material, and other characteristics to obtain multiple groups of process stamping models. The process stamping models in each group are logically adjacent or have similar stamping features. Step S470, obtain the first group of process stamping models from the multiple groups of process stamping models, perform stamping intersection analysis on the first group of process stamping models, and if there is a stamping intersection area, perform stamping deviation analysis on the stamping intersection area, and optimize the target process sequence based on the stamping deviation analysis results. Specifically, a first group is selected from the grouped multiple groups of process stamping models as the first group of process stamping models, and the geometric shape and stamping feature analysis is performed on the process stamping models in the first group of process stamping models, and the stamping intersection areas between different process stamping models, that is, the same or similar stamping feature areas, are identified, and the mutual influence and potential problems of these stamping intersection areas in the stamping process are analyzed. Deviation analysis is performed on the identified stamping intersection areas, including analyzing the causes of the deviation, such as mold design, stamping force control, material flow, etc., and quantifying the size and degree of the deviation, and evaluating the impact of the deviation on product quality and process stability. According to the results of the stamping deviation analysis, the process sequence involving the first group of process stamping models in the target process sequence is adjusted to optimize the process sequence and reduce the stamping deviation. Step S480, by analogy, by performing stamping intersection analysis on the multiple groups of process stamping models, and optimizing the target process sequence based on the analysis results, an optimized process sequence is generated. Specifically, repeat step S470, perform similar analysis and optimization on the process stamping models of the remaining groups, perform stamping intersection analysis, deviation analysis and reordering optimization on each group of process stamping models, and in each cycle, merge or update the optimized target process sequence with the previous results. The final optimization result includes the optimization adjustments of all groups, forming a complete optimized process sequence. Step S490, perform reordering optimization on the multiple process stamping models based on the optimized process sequence. Specifically, apply the final optimized target process sequence to all process stamping models, adjust or regenerate the process stamping model sequence according to the optimized process sequence, and ensure that the multiple process stamping models match the actual stamping process.This implementation method conducts stamping intersection analysis on the process stamping model, identifies the intersection areas that may exist in the stamping process, conducts deviation analysis on these intersection areas, finds out the causes of the deviations, and reorders the process sequence based on the analysis results. This optimizes the stamping process in these areas and reduces deviations, thereby achieving the technical effect of improving product quality and consistency.

[0073] Step S500, using the K kinds of stamping features to map the multiple process stamping models to perform stamping process parameter analysis, and obtain multiple sets of stamping process parameter sequences. Specifically, the K kinds of stamping features include punching, bending, stretching, shearing, etc. By analyzing the geometric shapes, material properties and stamping procedures in the multiple process stamping models, it is determined which stamping features correspond to which parts in the multiple process stamping models, and each stamping feature is mapped to the corresponding process stamping model. For each stamping feature mapped to the process stamping model, analyze its impact on the stamping process parameters, including analysis of changes and optimization of parameters such as stamping force, stamping speed, die gap, material flow, etc. According to the results of feature analysis, determine the key process parameters that affect the stamping process. According to the analysis results of the target process sequence and the stamping features, construct a framework of the stamping process parameter sequence. In the stamping process parameter sequence framework, fill in the corresponding process parameter values ​​according to the process requirements corresponding to each stamping feature. For each process stamping model, repeat the above steps to generate multiple groups of different stamping process parameter sequences. The multiple groups of stamping process parameter sequences reflect the optimal process parameter combinations of different stamping features on different process stamping models.

[0074] In a possible implementation, step S500 further includes step S510, performing a reverse call on the stamping feature group sequence to obtain a first group of reverse stamping features, and performing a reverse call on the multiple process stamping models to obtain a first reverse process stamping model. Specifically, performing a reverse call on the stamping feature group sequence determined in step S452, that is, calling these stamping features in an order opposite to the original order, and calling the first group of reverse stamping features, similarly, performing a reverse call on multiple process stamping models, that is, accessing and processing these process stamping models in an order opposite to the original order, and calling the first reverse process stamping model, wherein the first group of reverse stamping features corresponds to the first reverse process stamping model, that is, after the raw material of the stamping parts to be produced is processed by the first process, the first reverse process stamping model is obtained, and the first reverse process stamping model has the first group of reverse stamping features. Step S520, use the first group of reverse stamping features to traverse the first reverse process stamping model to perform parameter calls, and obtain a first stamping deformation parameter set. Specifically, use the first group of reverse stamping features to traverse the first reverse process stamping model, that is, apply the reverse stamping features to the reverse process stamping model, and analyze the interaction between them. During the traversal process, call the relevant deformation parameters, and collect the values ​​of these deformation parameters, and organize the collected deformation parameter values ​​into a set, namely, the first stamping deformation parameter set. Step S530, and so on, perform reverse calls on the stamping feature group sequence and the multiple process stamping models, and perform parameter calls based on the reverse call result mapping to obtain multiple stamping deformation parameter sets. Specifically, the process of reverse calling and parameter calling is repeated for the remaining stamping feature group sequences and process stamping models. For each reverse calling and parameter calling, a corresponding stamping deformation parameter set is formed. Finally, multiple stamping deformation parameter sets are obtained, each set corresponds to a specific reverse stamping feature and reverse process stamping model combination. Step S540, pre-constructing a stamping process matching network, wherein the stamping process matching network includes multiple stamping process matching branches mapped to the multiple stamping processes, and the multiple stamping process matching branches are connected in parallel. Specifically, according to the number and characteristics of standard stamping processes, a stamping process matching network is pre-constructed. This stamping process matching network consists of multiple stamping process matching branches, each stamping process matching branch corresponds to a stamping process, and these stamping process matching branches are connected in parallel to form a complete network structure. The stamping process matching network is a network structure specifically used to process stamping process-related data. It receives various types of input information, such as material properties, stamping part shapes, etc., and performs in-depth processing and analysis on this information through built-in algorithms. Based on these inputs, the stamping process matching network can identify the process parameter combination that matches specific stamping requirements.Step S550, mapping and activating the stamping process matching branches of the stamping process matching network according to the target process sequence, and obtaining a plurality of target matching branches. Specifically, according to the target process sequence, the corresponding stamping process matching branches are mapped and activated in the stamping process matching network, that is, only the branches related to the target process sequence are activated, and other irrelevant branches are ignored, and all activated branches together constitute the target matching branches. Step S560, mapping and synchronizing the multiple stamping deformation parameter sets to the multiple target matching branches according to the target process sequence, and obtaining the multiple groups of stamping process parameter sequences. Specifically, the multiple stamping deformation parameter sets obtained in step S530 are matched with the multiple target matching branches, and each stamping deformation parameter set is associated with a specific target matching branch. The multiple stamping deformation parameter sets are synchronized to the corresponding target matching branches, and each target matching branch outputs a set of process parameters that match it, and finally outputs multiple groups of stamping process parameter sequences, and the multiple groups of stamping process parameter sequences are used to guide the actual stamping production process. This implementation method uses a stamping process matching network to generate a stamping process parameter sequence. The stamping process matching network has undergone a lot of learning and training, and can accurately understand the various variables and factors in the stamping process, so as to quickly output the optimal process parameters, achieving the technical effect of improving the efficiency and accuracy of obtaining multiple sets of stamping process parameter sequences.

[0075] In a possible implementation, step S540 further includes step S541, interactively obtaining multiple sample stamping deformation dimensions, multiple sample stamping control parameters, and multiple sample stamping tool calls of the first stamping process among the multiple stamping processes. Specifically, sample data of the first stamping process is obtained through user input, database query, etc., including sample stamping deformation dimensions (recording stamping part deformation dimension data under different process conditions), sample stamping control parameters (including stamping speed, pressure, temperature and other control parameters), and sample stamping tool calls (recording the type, specification and order of use of the stamping tools used). Step S542, based on the knowledge graph, the associated storage of the multiple sample stamping deformation dimensions, multiple sample stamping control parameters and multiple sample stamping tool calls is performed to construct and obtain the first stamping process matching layer. Specifically, a graph database or a relational database is used to model the entities (stamping deformation size, stamping control parameters, stamping tool calls) in the sample data and the relationships between them, and the association and queryability between the data are ensured through associative storage, for example, the control parameters and tool calls corresponding to a specific deformation size can be queried. Step S543, interactively obtain the sample stamping sequences of the multiple sample stamping deformation sizes, and generate a first stamping process sorting layer based on the sample stamping sequence. Specifically, the sample stamping sequence data of the sample stamping deformation size obtained interactively are analyzed, the deformation law in the stamping process is analyzed, and the stamping process sorting is generated based on the analysis results of the sample stamping sequence, that is, the execution order of the stamping operation in the stamping process is determined, for example, by analyzing and determining that the large hole is punched first in the stamping process, and then the small hole is punched, to avoid the adverse effects on the shape and position of the small hole that may be caused by material deformation when punching the large hole, according to the determined stamping process sorting, the corresponding algorithm is designed and implemented as part of the first stamping process sorting layer, and the first stamping process sorting layer is generated. Step S544, connect the output end of the first stamping process matching layer with the input end of the first stamping process sorting layer to construct and generate the first stamping process matching branch; step S545, and so on, construct and obtain the multiple stamping process matching branches mapped to the multiple stamping processes, and connect the multiple stamping process matching branches in parallel to construct and generate the stamping process matching network. Specifically, for each stamping process, repeat steps S541 to S544 to construct the corresponding stamping process matching branch. In the network structure, each stamping process matching branch is connected in parallel to ensure that they can work simultaneously and share data. This implementation method determines the optimal punching sequence by analyzing the sample stamping sequence, ensuring that in the actual stamping process, punching is performed in a reasonable order, reducing possible material deformation, unnecessary pauses and waiting time, etc., thereby achieving the technical effect of optimizing production efficiency and improving product quality.

[0076] Step S600, fit the multiple groups of stamping process parameter sequences to the target process sequence, generate a stamping dynamic control sequence, and use the stamping dynamic control sequence to execute the batch production of the stamping parts to be produced. Specifically, according to the process sequence and stamping feature requirements of the target process sequence, determine the specific position where each stamping process parameter sequence should be fitted, and fit the multiple groups of stamping process parameter sequences to the target process sequence one by one according to the determined fitting points. According to the fitted target process sequence, construct a framework of the stamping dynamic control sequence, use the framework of the stamping dynamic control sequence as the basis for storing and organizing all control parameters in the stamping process, and fill in the corresponding control parameter values ​​in the framework of the stamping dynamic control sequence according to the fitted stamping process parameter sequence. These control parameter values ​​are used to guide the operation and operation of the stamping equipment. The generated stamping dynamic control sequence is imported into the control system of the stamping equipment. According to the requirements of the stamping dynamic control sequence, the stamping equipment is set and adjusted as necessary to ensure that the stamping equipment is in the best working state. The stamping equipment is started, and the stamping parts to be produced are mass-produced according to the stamping dynamic control sequence. The mass production process is monitored in real time, and production data and feedback information are collected. According to the monitoring results, the stamping dynamic control sequence is optimized and adjusted as necessary to improve production efficiency and product quality. The embodiment of the present application adopts technical means such as modeling and restoration of product design information, image acquisition, stamping feature recognition, process step sequence sorting, and stamping process parameter analysis to realize dynamic control of the stamping machine tool, thereby reducing dependence on manual experience and improving the technical effect of stamping accuracy and production efficiency.

[0077] In the above, refer to Figure 1 The dynamic control method for a stamping machine tool according to an embodiment of the present invention is described in detail. Figure 2 A dynamic control system for a stamping machine tool according to an embodiment of the present invention is described.

[0078] The dynamic control system for a stamping machine tool according to an embodiment of the present invention is used to solve the technical problem that the existing stamping machine tool control relies on manual experience, resulting in insufficient stamping precision and low production efficiency, so as to reduce the reliance on manual experience, thereby improving the stamping precision and production efficiency. The dynamic control system for a stamping machine tool includes: a target stamping part model acquisition module 10, a two-dimensional model image set acquisition module 20, a stamping feature recognition module 30, a plurality of process stamping model acquisition modules 40, a plurality of stamping process parameter sequence acquisition modules 50, and a stamping dynamic control sequence generation module 60.

[0079] The target stamping part model acquisition module 10 is used to interactively obtain product design information of the stamping part to be produced, and perform modeling and restoration according to the product design information to obtain the target stamping part model;

[0080] The two-dimensional model image set acquisition module 20 is used to preset a standard product acquisition angle, and configure the camera viewing angle based on the standard product acquisition angle to perform image acquisition and export on the target stamping part model to obtain a two-dimensional model image set;

[0081] The stamping feature recognition module 30 is used to pre-build a stamping feature recognition network, synchronize the two-dimensional model image set to the stamping feature recognition network for stamping feature recognition, and output K types of stamping features, where K is a positive integer;

[0082] The multiple process stamping model acquisition module 40 is used to pre-build a stamping feature process mapping, and to sort the process steps of the K types of stamping features according to the stamping feature process mapping to obtain a target process sequence, and to perform reverse engineering analysis on the target stamping part model according to the target process sequence to obtain multiple process stamping models;

[0083] The multiple sets of stamping process parameter sequence acquisition module 50 is used to perform stamping process parameter analysis on the multiple process stamping model mappings using the K types of stamping features to obtain multiple sets of stamping process parameter sequences;

[0084] The stamping dynamic control sequence generation module 60 is used to fit the multiple sets of stamping process parameter sequences to the target process sequence, generate a stamping dynamic control sequence, and use the stamping dynamic control sequence to execute the batch production of the stamping parts to be produced.

[0085] The specific configuration of the stamping feature recognition module 30 will be described in detail below. As described above, a stamping feature recognition network is pre-constructed, and the two-dimensional model image set is synchronized to the stamping feature recognition network for stamping feature recognition, and K types of stamping features are output. The stamping feature recognition module 30 may further include: a preset operator calling unit is used to pre-construct a random operator library, and according to the M product acquisition angles of the standard product acquisition angle, M preset operators are randomly called from the random operator library, where M is a positive integer; a plurality of stamping identification image set acquisition units are used to interactively obtain a standard stamping feature set, and based on the standard stamping feature set, stamping feature identification is performed on the multiple real-collected image sets of the multiple sample stamping parts obtained interactively, to obtain multiple stamping identification image sets; a stamping feature recognition sub-channel construction unit is used to obtain a plurality of stamping identification image sets based on the M preset operators. Construct and generate M stamping feature recognition sub-channels, and use the multiple stamping logo image sets to construct the M stamping feature recognition sub-channels; the stamping feature recognition network construction unit is used to set the M stamping feature recognition sub-channels in parallel, and configure a feature aggregation channel at the output end of the M stamping feature recognition sub-channels to complete the construction of the stamping feature recognition network; the stamping feature subset acquisition unit is used to synchronize the M model image mapping of the two-dimensional model image set to the M stamping feature recognition sub-channels of the stamping feature recognition network for stamping feature recognition, and obtain M stamping feature subsets; the K types of stamping feature acquisition units are used to synchronize the M stamping feature subsets to the feature aggregation channel for repeated feature aggregation, and output the K types of stamping features.

[0086] Next, the specific configuration of the multiple process stamping model acquisition modules 40 will be described in detail. As described above, a stamping feature process mapping is pre-constructed, and the process steps of the K types of stamping features are sorted in order according to the stamping feature process mapping to obtain a target process sequence, and the target stamping part model is reverse engineered according to the target process sequence to obtain multiple process stamping models. The multiple process stamping model acquisition module 40 may further include: a standard stamping process sequence acquisition unit is used to interactively obtain a standard stamping process sequence, wherein the multiple stamping processes constituting the standard stamping process sequence are mapped to multiple shape feature sets; a stamping feature process mapping generation unit is used to associate and store the multiple stamping processes and the multiple shape feature sets based on a knowledge graph to generate the stamping feature process mapping; multiple target calling process acquisition units are used to use the K types of stamping features to match the stamping feature process mapping to perform stamping process calls to obtain multiple target calling processes; a target process sequence generation unit is used to rearrange the processes of the multiple target calling processes based on the standard stamping process sequence to generate the target process sequence; a reverse engineering analysis unit is used to reverse engineer the target stamping part model according to the target process sequence to obtain the multiple process stamping models.

[0087] Among them, the target stamping part model is reverse-engineered and analyzed according to the target process sequence to obtain the multiple process stamping models. The reverse engineering analysis unit may further include: a plurality of stamping feature acquisition subunits for dividing the K types of stamping features using the multiple target calling procedures to obtain multiple groups of stamping features; a stamping feature group sequence acquisition subunit for reverse sorting the multiple groups of stamping features according to the target process sequence to obtain a stamping feature group sequence; a first process stamping model acquisition subunit for sequentially calling the stamping feature group sequence to obtain a first group of sequential stamping features, and using the The first group of sequential stamping features are used as constraints to perform feature elimination on the target stamping part model to obtain a first process stamping model; the second process stamping model acquisition subunit is used to execute sequential calls on the stamping feature group sequence to obtain a second group of sequential stamping features, and use the second group of sequential stamping features as constraints to perform feature elimination on the first process stamping model to obtain a second process stamping model; multiple process stamping model acquisition subunits are used to perform multiple rounds of feature elimination and model retention on the target stamping part model in this way, using the stamping feature group sequence as a feature elimination constraint, to obtain the multiple process stamping models.

[0088] The specific configuration of the multiple-group stamping process parameter sequence acquisition module 50 will be described in detail below. As described above, the K types of stamping features are used to perform stamping process parameter analysis on the multiple process stamping model mappings to obtain multiple groups of stamping process parameter sequences. The multiple-group stamping process parameter sequence acquisition module 50 may further include: a reverse order calling unit for performing a reverse order call on the stamping feature sequence to obtain a first group of reverse order stamping features, and performing a reverse order call on the multiple process stamping models to obtain a first reverse order process stamping model; a parameter calling unit for performing a parameter call on the first reverse order stamping model using the first group of reverse order stamping features to traverse the first reverse order process stamping model to obtain a first stamping deformation parameter set; and multiple stamping deformation parameter set acquisition units for performing a reverse order call on the stamping feature sequence and the multiple process stamping models by analogy. Call, and perform parameter calling based on the reverse call result mapping to obtain multiple stamping deformation parameter sets; the stamping process matching network construction unit is used to pre-build a stamping process matching network, wherein the stamping process matching network includes multiple stamping process matching branches mapped to the multiple stamping procedures, and the multiple stamping process matching branches are connected in parallel; multiple target matching branch acquisition units are used to activate the stamping process matching branch mapping of the stamping process matching network according to the target process sequence to obtain multiple target matching branches; multiple groups of stamping process parameter sequence acquisition units are used to synchronize the mapping of the multiple stamping deformation parameter sets to the multiple target matching branches according to the target process sequence to obtain the multiple groups of stamping process parameter sequences.

[0089] Among them, a stamping process matching network is pre-constructed, wherein the stamping process matching network includes multiple stamping process matching branches mapped to the multiple stamping processes, and the multiple stamping process matching branches are connected in parallel, and the stamping process matching network construction unit may further include: a sample stamping process acquisition subunit is used to interactively obtain multiple sample stamping deformation dimensions, multiple sample stamping control parameters and multiple sample stamping tool calls of the first stamping process among the multiple stamping processes; a first stamping process matching layer construction subunit is used to associate and store the multiple sample stamping deformation dimensions, multiple sample stamping control parameters and multiple sample stamping tool calls based on the knowledge graph, and construct the acquired The first stamping process matching layer is obtained; the first stamping process sorting layer generating subunit is used to interactively obtain the sample stamping sequence of the multiple sample stamping deformation sizes, and generate the first stamping process sorting layer based on the sample stamping sequence; the first stamping process matching branch construction subunit is used to connect the output end of the first stamping process matching layer with the input end of the first stamping process sorting layer to construct and generate the first stamping process matching branch; the stamping process matching network generating subunit is used to construct and obtain the multiple stamping process matching branches mapped to the multiple stamping procedures by analogy, and connect the multiple stamping process matching branches in parallel to construct and generate the stamping process matching network.

[0090] Among them, the target stamping part model is reverse-engineered and analyzed according to the target process sequence to obtain the multiple process stamping models. After that, the reverse engineering analysis unit may further include: an adjacent grouping subunit is used to adjacently group the multiple process stamping models according to the target process sequence to obtain multiple groups of process stamping models; a first group of process stamping model acquisition subunit is used to obtain a first group of process stamping models by calling from the multiple groups of process stamping models; a stamping intersection analysis subunit is used to perform stamping intersection analysis on the first group of process stamping models, if there is a stamping intersection area, perform stamping deviation analysis on the stamping intersection area, and perform reordering optimization on the target process sequence based on the stamping deviation analysis result; an optimized process sequence generation subunit is used to generate an optimized process sequence by analogy, by performing stamping intersection analysis on the multiple groups of process stamping models, and reordering and optimizing the target process sequence based on the analysis results; a reordering optimization subunit is used to reorder and optimize the multiple process stamping models based on the optimized process sequence.

[0091] The dynamic control system for a stamping machine provided by the embodiment of the present invention can execute the dynamic control method for a stamping machine provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0092] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0093] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A dynamic control method for a stamping machine tool, characterized in that: The method comprises: Interactively obtain product design information of the stamping part to be produced, and perform modeling and restoration based on the product design information to obtain a target stamping part model; Preset a standard product acquisition angle, and configure the camera viewing angle based on the standard product acquisition angle to perform image acquisition and export on the target stamping part model to obtain a two-dimensional model image set; Pre-constructing a stamping feature recognition network, and synchronizing the two-dimensional model image set to the stamping feature recognition network for stamping feature recognition, and outputting K types of stamping features, where K is a positive integer; Pre-constructing a stamping feature process mapping, and sorting the process steps of the K stamping features according to the stamping feature process mapping to obtain a target process sequence, and performing reverse engineering analysis on the target stamping part model according to the target process sequence to obtain multiple process stamping models, including: interactively obtaining a standard stamping process sequence, wherein a plurality of stamping processes constituting the standard stamping process sequence are mapped to a plurality of shape feature sets; Associatively storing the plurality of stamping processes and the plurality of shape feature sets based on the knowledge graph, and generating the stamping feature process mapping; The K types of stamping features are matched with the stamping feature process mapping to perform stamping process calling to obtain multiple target calling processes; Rearrange the processes of the plurality of target calling processes based on the standard stamping process sequence to generate the target process sequence; Using the K stamping features to map the multiple process stamping models to perform stamping process parameter analysis, and obtain multiple sets of stamping process parameter sequences; The multiple groups of stamping process parameter sequences are fitted to the target process sequence to generate a stamping dynamic control sequence, and the stamping dynamic control sequence is used to execute batch production of the stamping parts to be produced.

2. The dynamic control method for a stamping machine tool according to claim 1, characterized in that: Pre-constructing a stamping feature recognition network, synchronizing the two-dimensional model image set to the stamping feature recognition network for stamping feature recognition, and outputting K types of stamping features, the method further includes: Pre-constructing a random operator library, and randomly calling M preset operators from the random operator library according to the M product acquisition angles of the standard product acquisition angle, where M is a positive integer; Interactively obtain a standard stamping feature set, and perform stamping feature identification on multiple real-collected image sets of multiple sample stamping parts obtained interactively based on the standard stamping feature set to obtain multiple stamping identification image sets; Constructing and generating M stamping feature recognition sub-channels based on the M preset operators, and using the multiple stamping mark image sets to construct the M stamping feature recognition sub-channels; The M stamping feature recognition sub-channels are arranged in parallel, and a feature aggregation channel is configured at the output ends of the M stamping feature recognition sub-channels to complete the construction of the stamping feature recognition network; Synchronize the mapping of the M model images of the two-dimensional model image set to the M stamping feature recognition sub-channels of the stamping feature recognition network to perform stamping feature recognition, and obtain M stamping feature subsets; The M stamping feature subsets are synchronized to the feature aggregation channel for repeated feature aggregation, and the K stamping features are output.

3. The dynamic control method for a stamping machine tool according to claim 1, characterized in that: Performing reverse engineering analysis on the target stamping part model according to the target process sequence to obtain the multiple process stamping models, the method further comprising: The K types of stamping features are divided by using the multiple target calling processes to obtain multiple groups of stamping features; Reversely sorting the plurality of groups of stamping features according to the target process sequence to obtain a stamping feature group sequence; Execute sequential call on the stamping feature group sequence to obtain a first group of sequential stamping features, and perform feature elimination on the target stamping part model with the first group of sequential stamping features as constraints to obtain a first process stamping model; Execute sequential call on the stamping feature group sequence to obtain a second group of sequential stamping features, and perform feature elimination on the first process stamping model with the second group of sequential stamping features as constraints to obtain a second process stamping model; By analogy, the stamping feature group sequence is used as a feature elimination constraint, and multiple rounds of feature elimination and model retention are performed on the target stamping part model to obtain the multiple process stamping models.

4. The dynamic control method for a stamping machine tool according to claim 3, characterized in that: The K stamping features are used to map the multiple process stamping models to perform stamping process parameter analysis to obtain multiple sets of stamping process parameter sequences. The method further includes: Reversely call the stamping feature group sequence to obtain a first reverse stamping feature group, and reversely call the multiple process stamping models to obtain a first reverse process stamping model; The first set of reverse stamping features is used to traverse the first reverse process stamping model to perform parameter calling to obtain a first stamping deformation parameter set; Similarly, the stamping feature group sequence and the multiple process stamping models are reversely called, and parameter calls are performed based on the reverse call result mapping to obtain multiple stamping deformation parameter sets; Pre-constructing a stamping process matching network, wherein the stamping process matching network includes a plurality of stamping process matching branches mapped to the plurality of stamping processes, and the plurality of stamping process matching branches are connected in parallel; According to the target process sequence, the stamping process matching network is activated by stamping process matching branch mapping to obtain a plurality of target matching branches; The plurality of stamping deformation parameter sets are mapped and synchronized to the plurality of target matching branches according to the target process sequence to obtain the plurality of sets of stamping process parameter sequences.

5. The dynamic control method for a stamping machine tool according to claim 4, characterized in that: Pre-constructing a stamping process matching network, wherein the stamping process matching network includes a plurality of stamping process matching branches mapped to the plurality of stamping processes, and the plurality of stamping process matching branches are connected in parallel, and the method further includes: Interactively obtaining a plurality of sample stamping deformation dimensions, a plurality of sample stamping control parameters, and a plurality of sample stamping tool calls of a first stamping process among the plurality of stamping processes; Based on the knowledge graph, the plurality of sample stamping deformation dimensions, the plurality of sample stamping control parameters and the plurality of sample stamping tool calls are associated and stored to construct a first stamping process matching layer; Interactively obtaining a sample stamping sequence of the plurality of sample stamping deformation sizes, and generating a first stamping process sorting layer based on the sample stamping sequence; Connecting the output end of the first stamping process matching layer to the input end of the first stamping process sorting layer to construct and generate a first stamping process matching branch; By analogy, the multiple stamping process matching branches mapped to the multiple stamping procedures are constructed and obtained, and the multiple stamping process matching branches are connected in parallel to construct and generate the stamping process matching network.

6. The dynamic control method for a stamping machine tool according to claim 3, characterized in that: According to the target process sequence, reverse engineering analysis is performed on the target stamping part model to obtain the multiple process stamping models. After that, the method further includes: Grouping the multiple process stamping models adjacently according to the target process sequence to obtain multiple groups of process stamping models; Obtaining a first group of process stamping models from the multiple groups of process stamping models; Performing a stamping intersection analysis on the first group of process stamping models, and if a stamping intersection area exists, performing a stamping deviation analysis on the stamping intersection area, and optimizing the target process sequence based on the stamping deviation analysis result; Similarly, by performing stamping intersection analysis on the multiple groups of process stamping models, and optimizing the target process sequence based on the analysis results, an optimized process sequence is generated; The plurality of process stamping models are sequenced and optimized based on the optimized process sequence.

7. A dynamic control system for a stamping machine tool, characterized in that: The system is used to implement the dynamic control method for a stamping machine tool according to any one of claims 1 to 6, and the system comprises: A target stamping part model acquisition module is used to interactively obtain product design information of the stamping part to be produced, and perform modeling and restoration based on the product design information to obtain a target stamping part model; A two-dimensional model image set acquisition module, which is used to preset a standard product acquisition angle, and configure a camera viewing angle based on the standard product acquisition angle to perform image acquisition and export on the target stamping part model to obtain a two-dimensional model image set; A stamping feature recognition module is used to pre-build a stamping feature recognition network, synchronize the two-dimensional model image set to the stamping feature recognition network for stamping feature recognition, and output K types of stamping features, where K is a positive integer; Multiple process stamping model acquisition modules, used for pre-building stamping feature process mapping, and sorting the process steps of the K kinds of stamping features according to the stamping feature process mapping to obtain a target process sequence, and performing reverse engineering analysis on the target stamping part model according to the target process sequence to obtain multiple process stamping models; A multiple-group stamping process parameter sequence acquisition module is used to perform stamping process parameter analysis on the multiple process stamping model mappings using the K stamping features to obtain multiple groups of stamping process parameter sequences; The stamping dynamic control sequence generation module is used to fit the multiple sets of stamping process parameter sequences to the target process sequence, generate a stamping dynamic control sequence, and use the stamping dynamic control sequence to execute the batch production of the stamping parts to be produced.

Images