Method and system for determining sheet metal working time and computer program product

Abstract

The invention provides a method and system for determining sheet metal working hours and a computer program product, and relates to the technical field of intelligent manufacturing. The method for determining the sheet metal working time comprises the steps that a three-dimensional model of a to-be-machined part is obtained, wherein the to-be-machined part is a sheet metal machining part; based on the three-dimensional model, matching a first historical model from a historical database, and taking related data of the first historical model as template data; metal plate to-be-machined features of the three-dimensional model are recognized through a part recognition model; based on the template data, processing requirements of to-be-processed characteristics of the metal plate are determined; identifying characteristic parameters of the to-be-processed characteristics of the metal plate by using a CAD module; and determining the sheet metal processing time of the to-be-processed part based on the processing requirements, the characteristic parameters and the sheet metal processing technology knowledge base of the to-be-processed characteristics of the sheet metal. According to the method, the defects of inconsistent man-hour estimation and the like caused by disjunction of a process and calculation and lack of intelligent decision in a traditional method are overcome.

Description

Technical Field

[0001] This application relates to the field of intelligent manufacturing technology, and in particular to a method, system and computer program product for determining sheet metal processing time. Background Technology

[0002] In the sheet metal manufacturing industry, accurately estimating machining time is a crucial step for companies in production scheduling, cost accounting, and pricing decisions. Traditional methods heavily rely on the personal experience of process engineers. Engineers must manually identify features such as bending lines and punching holes based on 2D drawings or 3D models, relying on experience manuals or memory to estimate the time for each bending, punching, and stretching process, and then manually summarize the results. This method is inefficient, time-consuming, and the accuracy of the estimation results fluctuates greatly. Different engineers may have significantly different time estimates for the same part, directly leading to the failure of production cost control and a loss of competitive pricing, becoming a prominent bottleneck restricting the company's responsiveness and profitability.

[0003] In recent years, although some computer-aided time estimation technologies have been applied, most have failed to fundamentally solve the problem of intelligent decision-making in sheet metal processes. Existing solutions are often limited to digitizing fixed calculation formulas or making simplistic analogies to historical data, lacking in-depth analysis and automatic planning of complex process constraints such as bending sequence interference, stamping die matching, and the influence of stretching direction in 3D models. The systems cannot optimize process sequences to minimize die changeover time like experienced engineers, nor can they dynamically adjust process routes based on real-time die inventory and equipment capacity. More importantly, these systems typically lack self-learning and continuous evolution capabilities; their knowledge bases cannot be automatically optimized with the accumulation of new processing cases, causing the estimation models to gradually lag behind actual production levels and technological advancements, failing to meet the high requirements of modern intelligent manufacturing for efficiency, accuracy, and adaptability.

[0004] Therefore, there is an urgent need in this field for a new solution that can make deep use of 3D model information, realize intelligent linkage between process and working time, and has self-evolution capabilities. Summary of the Invention

[0005] The embodiments in this specification aim to address the problems of existing sheet metal processing time determination methods, which rely excessively on manual experience, are inefficient, have poor accuracy, and fail to fully utilize 3D model information. Specifically, the embodiments in this specification provide an automated, intelligent, and continuously evolving sheet metal processing time determination method, system, and computer program product.

[0006] This specification provides one or more embodiments of a method for determining sheet metal processing time, comprising: acquiring a three-dimensional model of a part to be processed, wherein the part to be processed is a sheet metal processing part; matching a first historical model from a historical database based on the three-dimensional model, and using relevant data of the first historical model as template data; identifying sheet metal processing features of the three-dimensional model through a part recognition model; wherein the sheet metal processing features include at least one of bending features, stamping features, and stretching features; determining the processing requirements of the sheet metal processing features based on the template data; identifying the feature parameters of the sheet metal processing features using a CAD module; and determining the sheet metal processing time of the part to be processed based on the processing requirements, feature parameters, and a sheet metal processing technology knowledge base of the sheet metal processing features.

[0007] In some embodiments, matching a first historical model from a historical database based on the three-dimensional model includes: acquiring point cloud data and key attributes of the three-dimensional model; the key attributes include at least material type, shape, and overall outline size; comparing the point cloud data and key attributes with cases in the historical database in multiple dimensions, and determining the historical case model with the highest comprehensive similarity as the first historical model.

[0008] In some embodiments, identifying sheet metal features to be processed in the 3D model using a part recognition model includes: acquiring point cloud data of the 3D model; and identifying the sheet metal features to be processed using the part recognition model based on the point cloud data, wherein the part recognition model is a trained PointNet network model.

[0009] In some embodiments, determining the processing requirements of the sheet metal feature to be processed based on the template data includes: finding historical features that correspond to the current sheet metal feature to be processed in terms of type and geometry from the relevant data of the first historical model; and determining at least one of the processing accuracy, surface treatment requirements and process constraints associated with the historical features as the processing requirements of the current feature.

[0010] In some embodiments, determining the sheet metal processing time of the part to be processed based on the processing requirements, feature parameters, and sheet metal processing technology knowledge base of the sheet metal features to be processed includes: matching a set of feasible sheet metal processing technology routes from the technology knowledge base based on the processing requirements and feature parameters, and determining the processing time required for each process step in each technology route and the corresponding candidate mold; solving the set of technology routes using a preset optimization algorithm, with the shortest total processing time as the optimization objective, and determining the final optimized processing scheme from all feasible technology routes; and calculating the total processing time of the sheet metal part to be processed based on the technology routes, processing times required for each process step, and molds determined by the final optimized processing scheme.

[0011] In some embodiments, the process knowledge base is constructed by: structurally storing sheet metal processing features, processing parameters, actual process routes, process times and mold information from historical processing cases, and establishing mapping relationships and constraint rules between features and processes.

[0012] In some embodiments, the optimization algorithm is a genetic algorithm that encodes the process route as chromosomes, evaluates its total processing time using a fitness function, and searches for the optimal solution by iteratively performing selection, crossover, and mutation operations.

[0013] In some embodiments, the total processing time includes at least the processing time of each process, the time for mold preparation and replacement between processes, and the time for workpiece handling and repositioning between processes.

[0014] In some embodiments, the above method further includes: providing an interactive interface to display the sheet metal features to be processed, processing requirements, feature parameters, mold, processing plan, and calculated working hours; receiving adjustment instructions input by the user through the interactive interface, and updating the processing plan or working hour calculation results based on the instructions.

[0015] In some embodiments, the above method further includes adding the three-dimensional model, sheet metal features to be processed, processing requirements, feature parameters, molds, processing plans, and actual production feedback time data to the historical database and the sheet metal processing technology knowledge base.

[0016] In some embodiments, the above method further includes: when the part recognition model cannot recognize some or all of the features in the three-dimensional model, performing an overall time estimate based on the overall volume, surface area, and plate thickness of the three-dimensional model and referring to the template data.

[0017] This specification also provides a sheet metal processing time determination system in one or more embodiments, including a part acquisition module, a template matching module, a feature recognition module, a processing requirement determination module, a feature parameter recognition module, and a time determination module. The part acquisition module acquires a three-dimensional model of a part to be processed, wherein the part to be processed is a sheet metal processing part. The template matching module matches a first historical model from a historical database based on the three-dimensional model and uses the relevant data of the first historical model as template data. The feature recognition module identifies the sheet metal processing features of the three-dimensional model through the part recognition model; wherein the sheet metal processing features include at least one of bending features, stamping features, and stretching features. The processing requirement determination module determines the processing requirements of the sheet metal processing features based on the template data. The feature parameter recognition module identifies the feature parameters of the sheet metal processing features using a CAD module. The time determination module determines the sheet metal processing time of the part to be processed based on the processing requirements, feature parameters, and a sheet metal processing technology knowledge base of the sheet metal processing features.

[0018] One or more embodiments of this specification also provide a computer program product, including a computer program that, when at least a portion of the computer program is executed by a processor, enables the above-described method for determining sheet metal processing time. Attached Figure Description

[0019] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. The same numbers in the drawings denote the same structures or steps.

[0020] Figure 1 This is a schematic diagram illustrating an application scenario of a sheet metal processing time determination system according to some embodiments of this specification.

[0021] Figure 2 This is an exemplary flowchart illustrating a method for determining sheet metal processing time according to some embodiments of this specification.

[0022] Figure 3 This is a schematic diagram of the PointNet network model structure shown in some embodiments of this specification.

[0023] Figure 4 This is an exemplary flowchart illustrating the determination of sheet metal processing time for a part to be processed, according to some embodiments of this specification.

[0024] Figure 5 This is a schematic diagram of the functional modules of a sheet metal processing time determination system according to some embodiments of this specification. Detailed Implementation

[0025] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are merely some examples or embodiments of this application. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0026] It should be understood that the terms “system,” “device,” “unit,” and / or “module” used herein are one way to distinguish different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0027] As indicated in this application and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of explicitly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0028] Flowcharts are used in this application to illustrate the operations performed by the system according to embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0029] In the sheet metal manufacturing industry, accurately estimating machining time is a crucial step for companies in production scheduling, cost accounting, and pricing decisions. Traditional methods heavily rely on the personal experience of process engineers. Engineers must manually identify features such as bending lines and punching holes based on 2D drawings or 3D models, relying on experience manuals or memory to estimate the time for each bending, punching, and stretching process, and then manually summarize the results. This method is inefficient, time-consuming, and the accuracy of the estimation results fluctuates greatly. Different engineers may have significantly different time estimates for the same part, directly leading to the failure of production cost control and a loss of competitive pricing, becoming a prominent bottleneck restricting the company's responsiveness and profitability.

[0030] In recent years, although some computer-aided time estimation technologies have been applied, most have failed to fundamentally solve the problem of intelligent decision-making in sheet metal processes. Existing solutions are often limited to digitizing fixed calculation formulas or making simplistic comparisons with historical data, lacking in-depth analysis and automatic planning of complex process constraints such as bending sequence interference, stamping die matching, and the influence of stretching direction in 3D models. The systems cannot optimize process sequences to minimize die changeover time like experienced engineers, nor can they dynamically adjust process routes based on real-time die inventory and equipment capacity. More importantly, these systems typically lack self-learning and continuous evolution capabilities; their knowledge bases cannot be automatically optimized with the accumulation of new processing cases, causing the estimation models to gradually lag behind actual production levels and technological advancements, failing to meet the high demands of modern intelligent manufacturing for efficiency, accuracy, and adaptability. Furthermore, in practical applications, many parts to be processed only have 3D models without corresponding processing requirements, or the processing requirements are presented in paper or PDF formats, making them difficult for the system to read. Therefore, developing a method that can intelligently interpret 3D models, automatically plan processes, and accurately calculate time has become an urgent need to promote the digital transformation of the manufacturing industry and achieve cost reduction and efficiency improvement.

[0031] In view of this, some embodiments of this specification provide a method, system, and computer program product for determining sheet metal processing time.

[0032] Figure 1 This is a schematic diagram illustrating an application scenario of a sheet metal processing time determination system according to some embodiments of this specification. For example... Figure 1 As shown, the sheet metal processing time determination system 100 (hereinafter referred to as system 100) may include a server 110, a network 120, a client 130, a database 140, and a part to be processed 150.

[0033] Server 110 can be configured to process information and / or data related to sheet metal processing time calculation. In some embodiments, server 110 is the intelligent computing hub and business logic core of the system. It is typically one or more high-performance computers deployed in an enterprise data center or cloud platform, responsible for executing the most computationally intensive algorithm steps. For example, server 110 can load and run trained part recognition models such as PointNet, perform high-speed processing of uploaded 3D model point cloud data, complete part type confirmation and processing feature recognition; and perform intelligent matching of template data (or time calculation template), derivation of processing technology sequences, and dynamic time calculation based on physical formulas. It embeds the query and reasoning logic of the sheet metal processing technology knowledge base. In some embodiments, server 110 can be regional or remote. For example, server 110 can access information and / or data stored in client 130 and / or database 140 through network 120. As another example, server 110 can directly connect to client 130 and / or database 140 to access stored information and / or data. In some embodiments, server 110 can be implemented on a cloud platform. For example, a cloud platform may include private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, internal cloud, multi-cloud, or any combination of the above examples.

[0034] Network 120 facilitates the exchange of information and / or data, serving as the artery connecting system components and ensuring data flow. In some embodiments, one or more components of system 100 (e.g., server 110, client 130, database 140, and contract manufacturing part 150) can send information and / or data to other components of system 100 via network 120. For example, server 110 can exchange data with client 130 via network 120. In some embodiments, network 120 can be any of a wired or wireless network, or a combination thereof. In enterprise applications, it typically refers to a local area network (LAN), virtual private network (VPN), or secure wide area network (WAN). Network 120 is primarily responsible for transmitting 3D model data, reliably transferring large 3D model files uploaded by client 130 to server 110; it carries interactive commands and results, transmitting parameter adjustment commands, process plans, calculation results, and visualization data in real time between the interactive interface of client 130 and the computing engine of server 110; and it ensures data security and integrity: through encrypted transmission protocols (such as HTTPS / SSL) and enterprise firewall policies, it ensures the security of core process data and model assets during transmission, preventing leakage and tampering.

[0035] Client 130 is the core operating terminal and visual window for users (such as process engineers) to interact with the system. It typically manifests as a dedicated software application installed on a workstation computer or a web interface accessed via a browser. In some embodiments, client 130 may include a smart mobile device, smart tablet, computer, laptop, or any combination thereof. In some embodiments, client 130 can provide a user-friendly interface, allowing engineers to select a 3D model file of the part 150 to be processed from a local or integrated system and initiate a time analysis task. In some embodiments, client 130 can receive and clearly display the automatic identification results, recommended machining processes, tools, and parameters returned by server 110. Engineers can review, fine-tune, or confirm on this interface, achieving an organic combination of artificial intelligence and expert experience. In some embodiments, client 130 can display the final total time, cost analysis, and detailed process route in a graphical and structured manner (such as 3D highlighted features, time breakdown tree, and process cards), and supports one-click export of standard reports.

[0036] Database 140 can store data and / or instructions. In some embodiments, database 140 can be a high-performance, structured data storage system, typically an enterprise-level relational database or a time-series database. Database 140 persistently stores historical databases, sheet metal processing technology knowledge bases, and time calculation template libraries. In some embodiments, database 140 may include mass storage, removable storage, volatile read-write memory, read-only memory (ROM), or any combination thereof. In some embodiments, database 140 can be implemented on a cloud platform. By way of example only, a cloud platform may include private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, inter-cloud, multi-cloud, or any combination thereof. Database 140 not only provides real-time data support for system operation but also serves as the data foundation for achieving closed-loop optimization and self-learning of the system.

[0037] In some embodiments, database 140 may connect to network 120 and communicate with one or more components of system 100 (e.g., server 110, client 130, etc.). One or more components of system 100 may access data or instructions stored in database 140 via network 120. In some embodiments, database 140 may connect directly to or communicate with one or more components of system 100 (e.g., server 110, client 130, etc.). In some embodiments, database 140 may be part of server 110.

[0038] The part to be processed 150 is the starting point of the physical object processed by the system. In the digital workflow, it first participates in system interaction in the form of a three-dimensional digital model. In some embodiments, this model is a CAD file (such as STEP or PRT format) containing complete geometric information and material properties. As the data source of the entire process, its information enters the digital system through scanning, direct upload, etc., and is then parsed by the algorithm of server 110, ultimately transformed into an executable processing plan and quantified time data. In some embodiments, the part to be processed 150 can be a sheet metal part or other different types of parts.

[0039] Those skilled in the art will understand that when a component of system 100 performs, that component can perform by signals and / or electromagnetic signals. For example, when client 130 sends a service request to server 110, the processor of client 130 can generate a signal encoding the request. The processor of client 130 can then send the signal to an output port. If client 130 communicates with server 110 via a wired network, the output port can be physically connected to a cable that further sends the signal to the input port of server 110. If client 130 communicates with server 110 via a wireless network, the output port of client 130 can be one or more antennas that convert the signal into an electromagnetic signal. Within electronic devices such as client 130 and / or server 110, instructions and / or actions are performed by signals when the processor processes instructions, issues instructions, and / or performs actions. For example, when the processor retrieves or saves data from a storage medium, it can send signals to the read / write device of the storage medium, which can read or write structured data in the storage medium. The structured data can be transmitted to the processor in the form of signals via the bus of the electronic device. Here, a signal can refer to a single signal, a series of signals, and / or at least two discrete signals.

[0040] Figure 2 This is an exemplary flowchart illustrating a method for determining sheet metal processing time according to some embodiments of this specification. In some embodiments, Figure 2 The illustrated process 200 can be executed by server 110. In some embodiments, process 200 can be implemented by a sheet metal processing time determination system 500 deployed on server 110. In some embodiments, such as Figure 5 As shown, the sheet metal processing time determination system 500 may include a part acquisition module 510, a template matching module 520, a feature recognition module 530, a processing requirement determination module 540, a feature parameter recognition module 550, and a time determination module 560.

[0041] In some embodiments, such as Figure 2As shown, process 200 may include the following steps.

[0042] Step 210: Obtain a 3D model of the part to be processed. In some embodiments, step 210 can be achieved by... Figure 5 The parts acquisition module 510 is implemented in the system.

[0043] In some embodiments, in actual industrial environments, 3D model data may be stored in different systems and locations. Therefore, this embodiment supports flexible reading of 3D model files from multiple sources by constructing a unified data access interface. In some embodiments, the system can directly access a specified file server on the enterprise intranet or the local hard drive of an engineer's workstation, and read model files through standard file paths (such as \\NAS\Project X\CAD\Models or C:\Users\Workspace). This approach is suitable for enterprises whose data has not yet been centrally uploaded to the cloud or for small-batch, temporary analysis. In some embodiments, for enterprises that have adopted cloud storage, the system downloads the target model file from a specified cloud storage bucket by calling the API interface of a cloud service provider (such as Alibaba Cloud OSS or Tencent Cloud COS), using a secure access key and token for authentication. This approach facilitates centralized data management and remote collaboration. In some embodiments, the system can also initiate a data request to the PLM system by passing in a unique identifier such as the part's material code or drawing number through the standard RESTful API or SOAP Web Service interface provided by a PLM / PDM system (such as Teamcenter or Windchill). After authorization verification, the PLM system will return the corresponding 3D model file and metadata (such as materials and version number). This ensures that the model used is the latest approved and valid version, achieving seamless integration with the product data source.

[0044] In some embodiments, the parts to be processed are sheet metal parts. In some embodiments, sheet metal parts may include: structural components and supporting components (such as various mounting brackets, fixing plates, reinforcing ribs, cabinet frames, guide rails, trays, etc., mainly used for supporting, fixing or connecting other components), housings and covers (such as equipment chassis, controller housings, protective covers, covers, electrical boxes, etc., mainly used for housing, protection, sealing or aesthetic purposes), panels and door panels (such as equipment front panels, operation panels, cabinet doors, decorative panels, etc., usually with features such as display holes, button holes, ventilation holes, etc.), pipes and air ducts (such as ventilation ducts, exhaust hoods, air conditioning ducts, connecting elbows, etc., used to guide the flow of gas or liquid), connectors and standard parts (such as sheet metal flanges, connecting pieces, corner brackets, gaskets, etc., used for assembly connections between systems), and specific functional parts (such as heat sinks, shielding covers, filter cavities, antenna reflectors, springs, hinges, etc., with specific electrical, thermal or mechanical functions), etc.

[0045] Step 220: Based on the 3D model, match the first historical model from the historical database, and use the relevant data of the first historical model as template data. In some embodiments, step 220 can be performed by... Figure 5 The template matching module 520 is implemented in the middle.

[0046] In some embodiments, matching a first historical model from a historical database based on a 3D model includes: acquiring point cloud data and key attributes of the 3D model; the key attributes include at least material type, shape, and overall outline dimensions; comparing the point cloud data and key attributes with cases in the historical database in multiple dimensions, and determining the historical case model with the highest comprehensive similarity as the first historical model.

[0047] In some embodiments, the system first collects point cloud data from the uploaded 3D model. Through unified sampling, the geometric information of the model surface is transformed into a standard set of 3D spatial points, which are then normalized to eliminate the influence of absolute size and position. Subsequently, the system calculates the feature descriptor of the point cloud (e.g., a fast point feature histogram), and uses this as the core, combined with multi-dimensional key attributes such as material type, shape, and overall contour dimensions parsed from model metadata or associated process files, to construct a comprehensive retrieval vector. The system then performs a multi-dimensional similarity comparison between this vector and successfully processed case models in the historical database. Shape similarity is calculated by comparing the metric distance of the point cloud descriptors (e.g., chamfer distance); material type, shape, and overall contour dimensions are quantified and scored according to preset rules. Finally, the system employs a weighted fusion strategy to comprehensively evaluate the similarity across all dimensions, identifying the historical case model with the highest overall similarity score as the first historical model. Its related data (including processing features and requirements of the first historical model) is automatically extracted and used as template data for estimating the current part's machining time. In some embodiments, the relevant data of the first historical model also includes the historical time calculation scheme of the first historical model (including standard preparation time, mapping relationship of specific process parameters, and calculation coefficients, etc.). In some embodiments, the template data includes the historical processing requirements of the historical features of the first historical model. These historical processing requirements are used to determine the processing requirements of the sheet metal features to be processed in step 240.

[0048] Step 230: Identify the sheet metal features to be processed in the 3D model using the part recognition model. These sheet metal features include at least one of bending, stamping, and stretching features. In some embodiments, step 230 can be achieved through... Figure 5 The feature recognition module 530 is implemented in the system.

[0049] In some embodiments, identifying sheet metal processing features of a 3D model using a part recognition model includes: acquiring point cloud data of the 3D model; and identifying the sheet metal processing features based on the point cloud data using a part recognition model, where the part recognition model is a trained PointNet network model. In some embodiments, the system can collect point cloud data from the acquired 3D model. More details regarding acquiring point cloud data of the 3D model can be found in step 220 and its related description, and will not be repeated here. In some embodiments, the part recognition model can perform instance segmentation on the point cloud data of the 3D model to identify the sheet metal processing features of the 3D model.

[0050] Subsequently, the point cloud data can be input into a pre-trained part recognition model, which preferably adopts the PointNet network architecture because it can directly process unordered point cloud data and effectively learn global features. Figure 3This is a schematic diagram of the PointNet network model structure shown in some embodiments of this specification. PointNet network structure 300 is shown below. Figure 3 As shown, Figure 3 In this model, n represents the number of points in the input point cloud; C represents the concatenation of vectors; MLP stands for Multilayer Perceptron; and k represents the number of classifications. In some embodiments, the part recognition model mainly consists of a local feature extraction module and a non-local feature extraction module. The input point cloud data first passes through the local feature extraction module, which constructs local features point by point in the point cloud. Then, the extracted local features are vector-concatenated with the original point cloud data and attribute adjacency graph data, and input into a multi-head attention mechanism. By learning multiple attention weights, a more comprehensive and diverse feature representation can be learned, enabling the extraction of features from different subspaces, thereby obtaining higher-level features. After two rounds of feature extraction to enrich the semantics point by point, the feature results from the first three layers are concatenated and input into a shared MLP (256, 512, 1024) to increase the dimensionality of the features. Then, global features are extracted using max pooling, and finally input into a shared MLP (512, 256, k), outputting k classification results of the features.

[0051] In some embodiments, the construction and training of the part recognition model can be carried out according to the following steps.

[0052] Sample library construction and partitioning: A large number of 3D model samples of sheet metal parts, covering various processing feature types such as bending, stamping, and stretching features, were collected and labeled. Standardized point cloud data corresponding to these samples were generated using the aforementioned method to form a labeled dataset. This dataset was then randomly partitioned into training, validation, and test sets according to a preset ratio (e.g., 7:2:1).

[0053] Network training and feature learning.

[0054] Input: Input the point cloud data (N×3 matrix, where N is the number of points and 3 is the coordinates) from the training set into the PointNet network.

[0055] Feature extraction: The network first extracts features independently for each point using a shared multilayer perceptron (MLP). Then, it aggregates the features of all points using a symmetric function (such as max pooling) to generate a global feature vector that represents the shape of the entire part. This design makes the network invariant to the input order of points.

[0056] Classification output: The global feature vector is then classified through a fully connected layer, outputting the probability distribution of the part belonging to each processing feature category.

[0057] Loss calculation and optimization: The error between the predicted class and the true label is calculated using the cross-entropy loss function, and the network weight parameters are updated through a backpropagation algorithm (such as the Adam optimizer) to minimize the loss.

[0058] Model tuning and validation.

[0059] During training, the model performance is evaluated periodically using a validation set to monitor changes in accuracy and loss, in order to prevent overfitting.

[0060] We used methods such as grid search or Bayesian optimization to repeatedly tune and optimize hyperparameters of the network, such as learning rate, batch size, and number of MLP layers.

[0061] Once the model's performance on the validation set has stabilized and the accuracy has reached the expected level (e.g., exceeding 95%), a final evaluation is performed using a test set that was not used in training. The model is considered complete and ready for production use only when the accuracy of the test results exceeds a preset reliability threshold (e.g., 98%).

[0062] Step 240: Based on the template data, determine the processing requirements of the sheet metal features to be processed. In some embodiments, step 240 can be achieved by... Figure 5 The processing requirements determination module 540 is implemented in the process.

[0063] In some embodiments, determining the processing requirements of the sheet metal feature to be processed based on template data includes: finding historical features that correspond to the current sheet metal feature to be processed in terms of type and geometry from the relevant data of the first historical model; and determining at least one of the processing accuracy, surface treatment requirements and process constraints associated with the historical features as the processing requirements of the current feature.

[0064] In some embodiments, after completing part-level template matching, the system needs to perform precise mapping at the feature granularity. The core of this process is to determine the "target historical feature" corresponding to each "sheet metal feature to be processed" identified on the current 3D model in the selected "first historical model." This process is not a simple geometric dimension comparison, but a comprehensive judgment combining feature type, geometric attributes, and contextual relationships. In some embodiments, the system first filters feature sets of the same type in the first historical model based on the identification results, and then calculates the similarity between the two in key parameters. For complex features, the system can also analyze the topological relationships between features to match feature groups with similar processing sequences and clamping constraints. Through this multi-dimensional evaluation, the system determines the historical feature with the highest similarity as the target historical feature.

[0065] In some embodiments, after identifying the target historical feature, the system intelligently extracts and adapts the processing requirements for the current feature from its associated, proven "historical processing requirements." These parameters primarily and critically include processing accuracy, surface treatment requirements, and process constraints. In some embodiments, the system not only directly reads these parameter values ​​but also performs applicability verification in conjunction with the specific geometry and location of the current sheet metal feature to be processed. Simultaneously, the complete process context inherited from historical features, such as the machine tool accuracy class used, measurement methods, and even processing parameters, is also referenced as tacit knowledge to ensure the recommended accuracy and surface treatment requirements are process-achievable. The final generated processing requirements are clearly presented on the user interface for final confirmation and calibration by process engineers, ensuring the reliability of experience transfer.

[0066] Step 250: Identify the feature parameters of the sheet metal features to be processed using the CAD module. In some embodiments, step 250 can be achieved by... Figure 5 The feature parameter recognition module 550 is implemented in the system.

[0067] In some embodiments, the system invokes an integrated CAD module to perform precise feature parameter identification. This process can be achieved by silently launching a headless CAD session in the background and loading the original 3D model by calling the secondary development API (such as NX Open, SolidWorks API) of the target CAD software (such as Siemens NX, Dassault Systèmes SolidWorks). In some embodiments, for each identified sheet metal feature to be processed, the system extracts key geometric data according to its type. For example, for bending features, the system locates the flange edge forming the bend, and accurately calculates the bending angle, bending radius, and bending line length by querying the included angle and topological relationship of adjacent surfaces. It also analyzes the position of the neutral layer in the sheet metal to determine the bending allowance. For stamping features (such as holes, louvers, and protrusions), the system traverses the sketch outlines or stamping feature objects in the model to obtain their outline boundary geometry, positioning coordinates, outline perimeter, and area, and identifies whether they are standard features (such as round holes and square holes) to associate them with standard molds. For stretching features, the system analyzes the height of the sidewalls and the draft angle to extract the stretching depth, bottom fillet radius, and opening outline dimensions. All extracted parameters are verified by the geometry engine to ensure their accuracy and are ultimately organized into a structured feature parameter set, providing geometric data for subsequent knowledge-based process matching and time calculation.

[0068] Step 260: Based on the processing requirements, feature parameters, and sheet metal processing technology knowledge base of the sheet metal features to be processed, determine the sheet metal processing time for the part to be processed. In some embodiments, step 260 can be achieved by... Figure 5The time determination module 560 in the middle is used for implementation. In some embodiments, step 260 can be implemented by... Figure 4 The process shown in step 400 is implemented. Figure 4 This is an exemplary flowchart illustrating the determination of sheet metal processing time for a part to be processed, according to some embodiments of this specification. Figure 4 The illustrated process 400 can be executed by server 110. In some embodiments, process 400 can be implemented by a sheet metal processing time determination system 500 deployed on server 110. Further, process 400 can be... Figure 5 The working time determination module in the 560 is implemented.

[0069] In some embodiments, step 260 may further include the following steps.

[0070] Step 410: Based on processing requirements and feature parameters, match a set of feasible sheet metal processing routes from the process knowledge base, and determine the processing time required for each process step in each route and the corresponding candidate molds. In some embodiments, the process knowledge base is constructed in the following way: the sheet metal processing features, processing parameters, actual process routes, process times, and mold information in historical processing cases are stored in a structured manner, and a mapping relationship and constraint rules between features and processes are established.

[0071] In some embodiments, the system 500 first initiates a matching process with the process knowledge base based on the processing requirements (such as bending angle tolerance, punching burr standards, etc.) and precise feature parameters (such as bending length, punching contour dimensions, etc.) for each sheet metal feature to be processed obtained in the aforementioned steps. The knowledge base is built from the deep digitization of a large number of historical processing cases. The 3D unfolded model, the identified set of sheet metal processing features, the extracted processing parameters, the detailed process route actually used (including the process sequence, the specific mold number used in each process), the measured actual processing time of the process, and the equipment information in each completed case are cleaned, tagged, and structured and stored in a relational database. Then, the mapping relationship and constraint rules between features and processes are established through data mining technology. During the matching phase, the system takes the processing requirements and feature parameters of the current feature as input and performs multi-condition retrieval and rule-based reasoning in the knowledge base. For example, for a bending feature of "2mm thickness, 100mm length, 90-degree angle, and requirement of no indentation at the sharp angle," the system retrieves all historical cases that have processed similar features and met the quality requirements, extracting multiple feasible process routes. Each route is associated with a set of candidate molds used historically. Simultaneously, based on historical process time data and the parameters of the current feature, the system calculates the processing time for each route using a built-in standard time calculation formula. Finally, it outputs a set of feasible process routes, including several feasible routes, detailed process steps for each route, estimated processing time, and a list of optional molds, providing a decision space for subsequent optimization algorithms.

[0072] Step 420: Solve the set of process routes using a preset optimization algorithm, with the shortest total processing time as the optimization objective, and determine the final optimized processing scheme from all feasible process routes. In some embodiments, the optimization algorithm is a genetic algorithm, which encodes the process routes as chromosomes, evaluates their total processing time using a fitness function, and searches for the optimal solution by iteratively performing selection, crossover, and mutation operations.

[0073] In some embodiments, the implementation process of the genetic algorithm specifically includes the following steps.

[0074] Encoding and Initialization: A complete process route is encoded as a chromosome, where the gene positions sequentially represent the steps in the process sequence, the mold used in that step, and the connection logic between steps; an initial population is generated using heuristic rules based on the process knowledge base.

[0075] Fitness evaluation: The fitness function is defined as the reciprocal of the total processing time, i.e., Fitness = 1 / T total T total The total time is calculated based on the process route decoded from the chromosome; a higher fitness value indicates a better solution.

[0076] Evolutionary operations: In each generation of evolution, a roulette wheel selection strategy is used based on the fitness value; crossover is performed on the selected chromosomes with a preset probability to exchange some of their process sequences or mold selection gene segments; and mutation is performed with a lower probability to randomly change the value of a certain gene position.

[0077] Termination and Output: When the evolution reaches the preset maximum number of iterations, or when the optimal fitness value no longer increases for several consecutive generations, the evolution is terminated, and the chromosome with the highest fitness in the current population is decoded and output as the final optimized processing scheme.

[0078] Step 430: Based on the process route, processing time required for each step, and molds determined by the final optimized processing scheme, calculate the total processing time for the sheet metal part to be processed. The total processing time includes at least the processing time of each step, the mold preparation and change time between steps, and the time for handling and repositioning the workpiece between steps.

[0079] In some embodiments, the system 500 first parses the determined process route, the serialized process list, the estimated processing time for each process, and the designated molds from the final optimized processing scheme. The total processing time is dynamically accumulated from three core parts: the first part is the processing time of each process, which is directly derived from the estimated value of each process in the scheme. For example, the processing time of the stamping process is obtained by multiplying the number of stampings by the single stroke time, and the time of the bending process is calculated based on the bending length and the equipment speed; the second part is the mold preparation and replacement time between processes. Based on the mold usage sequence in the process route, the system queries the standardized installation, debugging, and disassembly time of the corresponding mold in the equipment parameter library and accumulates it each time the mold is switched. At the same time, it intelligently identifies processes that use the same mold consecutively to combine the preparation time; the third part is the workpiece handling and repositioning time between processes. Based on the digital model of the factory layout and the process sequence, the system simulates the logistics time required for the workpiece to be transferred from the previous process equipment (such as a laser cutting machine) to the next process equipment (such as a bending machine), and calculates the positioning, clamping, and calibration time necessary to complete the subsequent operations on the new processing equipment. The system seamlessly integrates and summarizes these three time components according to the process sequence, and finally outputs a total processing time accurate to the minute. This result not only includes the total value, but also breaks down and displays the details of each time component, providing reliable data that can be directly executed for production scheduling and cost analysis.

[0080] In some embodiments, process 200 further includes: providing an interactive interface to display one or more of the sheet metal features to be processed, processing requirements, feature parameters, molds, processing plans, and calculated working hours; receiving adjustment instructions input by the user through the interactive interface, and updating the processing plan or working hour calculation results based on the instructions.

[0081] In some embodiments, the system can achieve human-machine collaborative calibration through a graphical user interface, which automatically pops up or is presented as a standalone view after background calculations are completed. In some embodiments, the interface centrally displays all key information of the current sheet metal part in the form of a visual panel, including each sheet metal feature to be processed identified and highlighted in the 3D lightweight model, processing requirements determined from template data, precise feature parameters extracted using the CAD module, a recommended mold list matched from the knowledge base, and a detailed processing plan (including process route and process sequence) generated by the optimization algorithm, along with the calculated process and total processing time. Users can review and intervene in this interface, for example, by directly clicking on a bending feature and modifying its bending sequence in the parameter panel on the right to try to reduce mold changes, or by selecting another mold from the candidate mold library that is more suitable for the current process for a stamping process from the drop-down list, or by manually adjusting the estimated time of a process. In some embodiments, when a user issues an adjustment command, the system will capture the command in real time and immediately trigger the background recalculation logic. For example, when changing a mold, the system will re-query the standard parameters of the mold and recalculate the process time; when adjusting the bending sequence, the system will re-verify the geometric interference between processes and recalculate the handling and positioning time. In some embodiments, after all local updates are completed, the system will automatically and instantly update the display content of the processing plan and recalculate and refresh the total time result, while marking the modified parts with visual cues (such as color changes), thereby integrating intelligent planning with the experience judgment of human experts to form a flexible, reliable, and ultimately human-confirmed time determination system.

[0082] In some embodiments, process 200 further includes adding the 3D model, sheet metal features to be processed, processing requirements, feature parameters, mold, processing plan, and actual production feedback time data to the historical database and sheet metal processing technology knowledge base.

[0083] In some embodiments, after completing a sheet metal part time determination task and obtaining user confirmation or actual production verification, the system automatically performs a data recording and optimization feedback process. In some embodiments, the system first cleans, correlates, and encapsulates all key data generated throughout the task process, including the original 3D model file, all identified sheet metal features to be processed, corresponding processing requirements and feature parameters, the finally selected mold information, the confirmed processing plan (including complete process routes and process details), the calculated theoretical time, and the actual production feedback time data obtained from the manufacturing execution system, generating a structured data package with a unique task ID and version tag. This data package is then synchronously and persistently stored in a historical database and a sheet metal processing technology knowledge base through a secure data interface. In the historical database, it serves as a complete and traceable archive, with its multi-dimensional feature vectors (such as point cloud features, material thickness, and key dimensions) indexed to support rapid similarity retrieval and template matching for future new parts. Within the process knowledge base, this case package is deeply deconstructed and absorbed—its specific "feature-process-mold-parameter-result" combination serves as a high-value instance sample, used to expand the knowledge base's coverage and enhance its ability to handle complex or marginal features. In some embodiments, the system calculates the deviation rate by comparing theoretical and actual working hours, and uses statistical analysis models to assess and optimize the confidence of relevant process mapping rules in the knowledge base. For example, it may increase the recommendation weight of a certain type of efficient process route, or dynamically calibrate the parameters (such as machine tool idle travel speed and mold installation coefficient) in the standard working hour calculation formula based on actual processing efficiency. This drives the entire system's process recommendation and working hour calculation logic to achieve continuous and robust autonomous performance iteration and evolution without large-scale manual intervention.

[0084] In some embodiments, process 200 further includes: when the part recognition model cannot recognize some or all of the features in the three-dimensional model, the overall working time is estimated based on the overall volume, surface area and plate thickness of the three-dimensional model and with reference to the template data.

[0085] In some embodiments, when the system processes the point cloud data of a 3D model using a pre-trained PointNet part recognition model, if the confidence level of the model output is lower than a preset threshold or if a specific geometric region cannot be classified into any known processing feature category, the feature is determined to be unrecognizable. In this case, the system will automatically switch to a backup overall time estimation process to ensure functional robustness and timely delivery. This process first bypasses specific feature recognition and directly calculates the overall geometric macroscopic parameters of the part to be estimated through the CAD kernel interface, including its overall volume, total surface area, and sheet thickness obtained from model attributes or user input. Next, the system performs an analogy analysis with the template data (i.e., the relevant data of the first historical model) that has been successfully matched in the previous steps: the system calculates the scaling factor between the current part and the historical template in terms of volume and surface area, and considers the multiplicative effect of the difference in sheet thickness on processing difficulty. Based on these scaling factors and the known total time of the historical template, the system applies an experience-based scaling formula for estimation. Ultimately, the system generates a time estimate based on a holistic analogy, clearly indicating its source as a holistic estimate. It also provides the key parameters and proportions upon which the estimate is based for user review and decision-making. This mechanism ensures that even when intelligent identification is hindered, the system can still provide a well-founded and traceable preliminary time reference, maintaining the continuity and practicality of the method in real-world applications.

[0086] Some embodiments of this specification also provide a system for determining sheet metal processing time. Figure 5 This is a functional module diagram of a sheet metal processing time determination system according to some embodiments of this specification. In some embodiments, such as Figure 5 As shown, the sheet metal processing time determination system 500 may include a part acquisition module 510, a template matching module 520, a feature recognition module 530, a processing requirement determination module 540, a feature parameter recognition module 550, and a time determination module 560. In some embodiments, the various modules of the sheet metal processing time determination system 500 may be configured in the server 110.

[0087] In some embodiments, the part acquisition module 510 is used to acquire a three-dimensional model of the part to be processed, which is a sheet metal processing part.

[0088] In some embodiments, the template matching module 520 is used to match a first historical model from a historical database based on a 3D model, and use the relevant data of the first historical model as template data. In some embodiments, the template matching module 520 can also acquire point cloud data and key attributes of the 3D model; the key attributes include at least material type, shape, and overall outline dimensions; the point cloud data and key attributes are compared with cases in the historical database in multiple dimensions, and the historical case model with the highest comprehensive similarity is determined as the first historical model.

[0089] In some embodiments, the feature recognition module 530 is used to identify sheet metal features to be processed in a 3D model using a part recognition model. In some embodiments, the feature recognition module 530 can also acquire point cloud data of the 3D model; based on the point cloud data, the part recognition model is used to identify the sheet metal features to be processed, where the part recognition model is a trained PointNet network model.

[0090] In some embodiments, the processing requirement determination module 540 is used to determine the processing requirements of the sheet metal feature to be processed based on template data. In some embodiments, the processing requirement determination module 540 can also search for historical features that correspond to the current sheet metal feature to be processed in terms of type and geometry from the relevant data of the first historical model; and determine at least one of the processing accuracy, surface treatment requirements and process constraints associated with the historical features as the processing requirements of the current feature.

[0091] In some embodiments, the feature parameter recognition module 550 is used to recognize feature parameters of sheet metal features to be processed using a CAD module.

[0092] In some embodiments, the time determination module 560 is used to determine the sheet metal processing time of the part to be processed based on the processing requirements, feature parameters, and sheet metal processing technology knowledge base of the sheet metal features to be processed. In some embodiments, the time determination module 560 can also match a set of feasible sheet metal processing process routes from the process knowledge base based on the processing requirements and feature parameters, and determine the processing time required for each process step in each process route and the corresponding candidate mold; use a preset optimization algorithm to solve the set of process routes, take the shortest total processing time as the optimization objective, and determine the final optimized processing scheme from all feasible process routes; and calculate the total processing time of the sheet metal part to be processed based on the process route, the processing time required for each process step, and the mold determined by the final optimized processing scheme.

[0093] For more information on each module, please refer to [link / reference]. Figures 2 to 4 The relevant explanations will not be repeated here. It should be understood that... Figure 5The systems and modules shown can be implemented in various ways. For example, in some embodiments, the systems and modules can be implemented by hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the methods and systems described above can be implemented using computer-executable instructions and / or included in the control code of a processor, such as on a media such as a disk, CD, or DVD-ROM, or in the memory of a programmable device. The systems and modules of this specification can be implemented not only by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips and transistors, or programmable hardware devices such as field-programmable gate arrays and programmable logic devices, but also by software, for example, executed by various types of processors, or by a combination of the aforementioned hardware circuits and software (e.g., firmware).

[0094] This specification also provides a computer program product in some embodiments. This computer program product stores a computer program that, when at least a portion of the computer program is executed by a processor, enables the determination of sheet metal processing time as described in any embodiment of this specification. If the constituent modules of the above system are implemented as software functional units and sold or used as independent products, they can be stored in the computer program product.

[0095] The beneficial effects that the embodiments of this specification may bring include, but are not limited to: (1) Through the collaboration of intelligent recognition models based on point cloud data or graph neural networks and CAD modules, automated and high-precision recognition and parameter extraction of processing features such as bending, stamping and stretching on sheet metal three-dimensional unfolded models are realized, completely replacing the inefficient mode of relying on process engineers to manually read drawings and measure, greatly shortening the initial stage of determining working time from several hours to minutes, and fundamentally eliminating human subjective error, ensuring the consistency and reliability of input information; (2) A process knowledge base based on deep learning of historical cases is constructed and utilized, which can intelligently match the identified features and their parameters to feasible process routes and molds that have been verified in practice, and automatically solve the global optimal processing scheme with the goal of minimizing total working time through optimization algorithms such as genetic algorithms. The process simulates and surpasses the experience-based decision-making of senior process experts, solving the core pain points of the traditional method, such as the disconnect between process planning and time calculation, and the inability to automatically optimize multiple schemes. This makes the time estimation results highly feasible and economical. (3) The entire process is not a closed automation, but incorporates necessary manual review and fine-tuning capabilities through the human-computer interaction interface, ensuring the flexibility and acceptability of system decision-making. At the same time, all process and result data can form a closed loop to feed back to the knowledge base, driving the system to achieve continuous self-learning and performance evolution, so that its estimation accuracy will continue to improve over time. (4) A robust fault-tolerance mechanism is designed. Even when facing new and complex features that lead to recognition failure, it can make a reasonable overall estimation based on macroscopic geometric parameters and template data, ensuring the high availability and robustness of the method in practice. In summary, this invention not only significantly improves the efficiency, accuracy and intelligence level of sheet metal processing time determination, but also builds a continuously value-added process knowledge asset through data closed loop, providing strong core technical support for enterprises to achieve precise cost control, efficient production scheduling and rapid market quotation. It should be noted that different embodiments may produce different beneficial effects. In different embodiments, the beneficial effects may be any one or a combination of the above, or any other possible beneficial effects.

[0096] Unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this application are not intended to limit the order of the processes and methods of this application. Although some embodiments that are currently considered useful have been discussed through various examples in the foregoing disclosure, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments. Rather, the claims are intended to cover all modifications and equivalent combinations that conform to the substance and scope of the embodiments of this application. For example, while the system components described above can be implemented by hardware devices, they can also be implemented solely by software solutions, such as installing the described system on existing servers or mobile devices.

[0097] Similarly, it should be noted that, in order to simplify the description of the present application and thus aid in the understanding of one or more embodiments of the specification, the foregoing description of the embodiments of the present application sometimes combines multiple features into a single embodiment, drawing, or description thereof. However, this disclosure method does not imply that the subject matter of the present application requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of the single embodiments disclosed above.

[0098] For each patent, patent application, patent application publication, and other material such as articles, books, specifications, publications, and documents referenced in this application, the entire contents of that material are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this application, as well as documents that limit the broadest scope of the claims in this application (currently or subsequently appended to this application). It should be noted that if there is any inconsistency or conflict between the descriptions, definitions, and / or terminology used in the supplementary materials of this application and the content of this application, the descriptions, definitions, and / or terminology used in this application shall prevail.

[0099] Finally, it should be understood that the embodiments described in this application are merely illustrative of the principles of the embodiments of this application. Other modifications may also fall within the scope of this application. Therefore, alternative configurations of the embodiments of this application are considered as examples and not limitations, and are regarded as consistent with the teachings of this application. Accordingly, the embodiments of this application are not limited to the embodiments explicitly described and illustrated in this application.

Claims

1. A method for determining sheet metal processing time, characterized in that, include: Obtain a 3D model of the part to be processed, wherein the part to be processed is a sheet metal part; Based on the three-dimensional model, a first historical model is matched from the historical database, and the relevant data of the first historical model is used as template data. The sheet metal features to be processed in the three-dimensional model are identified by the part recognition model; wherein, the sheet metal features to be processed include at least one of bending features, stamping features and stretching features; Based on the template data, the processing requirements of the sheet metal features to be processed are determined; The CAD module is used to identify the feature parameters of the sheet metal to be processed. Based on the processing requirements, feature parameters, and sheet metal processing technology knowledge base of the sheet metal features to be processed, the sheet metal processing time of the parts to be processed is determined.

2. The method for determining sheet metal processing time according to claim 1, characterized in that, The step of matching the first historical model from the historical database based on the three-dimensional model includes: Obtain the point cloud data and key attributes of the 3D model; the key attributes include at least material type, shape, and overall outline dimensions; The point cloud data and key attributes are compared with cases in the historical database in multiple dimensions, and the historical case model with the highest comprehensive similarity is determined as the first historical model.

3. The method for determining sheet metal processing time according to claim 2, characterized in that, The process of identifying the sheet metal features to be processed in the 3D model through the part recognition model includes: Obtain the point cloud data of the three-dimensional model; Based on the point cloud data, the part recognition model is used to identify the sheet metal features to be processed. The part recognition model is a trained PointNet network model.

4. The method for determining sheet metal processing time according to claim 1, characterized in that, The step of determining the processing requirements of the sheet metal features to be processed based on the template data includes: From the relevant data of the first historical model, find the historical features that correspond to the current sheet metal features to be processed in terms of type and geometry; At least one of the processing precision, surface treatment requirements, and process constraints associated with the historical features is determined as the processing requirements for the current features.

5. The method for determining sheet metal processing time according to claim 1, characterized in that, The determination of sheet metal processing time for the part to be processed, based on the processing requirements, feature parameters, and sheet metal processing technology knowledge base of the sheet metal features to be processed, includes: Based on the processing requirements and characteristic parameters, a set of feasible sheet metal processing routes is matched from the process knowledge base, and the processing time required for each process in each route and the corresponding candidate mold are determined. The set of process routes is solved using a preset optimization algorithm, with the shortest total processing time as the optimization objective, and the final optimized processing scheme is determined from all feasible process routes. Based on the process route, processing time required for each step, and mold determined by the final optimized processing scheme, calculate the total processing time for the sheet metal part to be processed.

6. The method for determining sheet metal processing time according to claim 5, characterized in that, The process knowledge base is constructed in the following way: The sheet metal processing features, processing parameters, actual process routes, process times, and mold information from historical processing cases are stored in a structured manner, and a mapping relationship and constraint rules between features and processes are established.

7. The method for determining sheet metal processing time according to claim 5, characterized in that, The optimization algorithm is a genetic algorithm, which encodes the process route as chromosomes, evaluates its total processing time using a fitness function, and searches for the optimal solution by iteratively performing selection, crossover, and mutation operations.

8. The method for determining sheet metal processing time according to claim 5, characterized in that, The total processing time includes at least the processing time of each process, the time for mold preparation and replacement between processes, and the time for workpiece handling and repositioning between processes.

9. The method for determining sheet metal processing time according to claim 5, characterized in that, Also includes: Provide an interactive interface to display the sheet metal features to be processed, processing requirements, feature parameters, molds, processing plans, and calculated working hours; The system receives adjustment instructions input by the user through the interactive interface and updates the processing plan or time calculation results based on the instructions.

10. The method for determining sheet metal processing time according to claim 5, characterized in that, Also includes: The 3D model, sheet metal features to be processed, processing requirements, feature parameters, molds, processing plans, and actual production feedback time data are added to the historical database and the sheet metal processing technology knowledge base.

11. The method for determining sheet metal processing time according to claim 1, characterized in that, Also includes: When the part recognition model cannot recognize some or all of the features in the three-dimensional model, the overall working time is estimated based on the overall volume, surface area and plate thickness of the three-dimensional model and with reference to the template data.

12. A sheet metal processing time determination system, characterized in that it includes a part acquisition module, a template matching module, a feature recognition module, a processing requirement determination module, a feature parameter recognition module, and a time determination module; The part acquisition module is used to acquire a three-dimensional model of the part to be processed, which is a sheet metal part. The template matching module is used to match a first historical model from the historical database based on the three-dimensional model, and use the relevant data of the first historical model as template data; The feature recognition module is used to identify the sheet metal features to be processed in the 3D model through the part recognition model; wherein... The sheet metal processing features include at least one of bending features, stamping features, and stretching features; The processing requirement determination module is used to determine the processing requirements of the sheet metal features to be processed based on the template data; The feature parameter recognition module is used to identify the feature parameters of the sheet metal features to be processed using the CAD module; The time determination module is used to determine the sheet metal processing time of the part to be processed based on the processing requirements, feature parameters, and sheet metal processing technology knowledge base of the sheet metal features to be processed.

13. A computer program product, characterized in that, The method includes a computer program that, when at least a portion of the computer program is executed by a processor, enables the determination of sheet metal processing time as described in any one of claims 1 to 11.

Images