Part 3D drawing intelligent price checking method and system

Abstract

The invention provides an intelligent price checking method and system for 3D drawings of parts. The method comprises the following steps: identifying a process type through a pre-trained process classification model, and matching a corresponding target equipment type according to the process type; analyzing the 3D drawing information of the part to obtain geometric machining characteristics of the part, matching a corresponding tool type from a preset tool library according to the process type and the geometric machining characteristics, and determining tool machining parameters according to equipment performance parameters corresponding to the target equipment type and tool performance parameters corresponding to the tool type; on the basis of the geometric machining features, the target equipment type and the tool machining parameters, the machining time of the part is predicted through a pre-trained time prediction model; and generating a target price checking result of the part based on the equipment use cost corresponding to the target equipment type, the material cost corresponding to the part raw material and the man-hour cost corresponding to the processing man-hour. According to the invention, the efficiency and accuracy of price checking are improved.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence machine learning technology, and particularly to an intelligent cost estimation method and system for three-dimensional (3D) drawings of parts. Background Art

[0002] With the rapid development of the manufacturing industry, especially the increasing complexity in the field of parts processing, enterprises have put forward higher requirements for the efficiency and accuracy of calculating the processing costs of parts. The traditional cost estimation methods are not only time-consuming and laborious, but also prone to errors due to human factors or calculation simplification, making it difficult to meet the needs of modern manufacturing for efficient and accurate cost accounting. Therefore, developing an intelligent cost estimation method to accurately predict processing man-hours and costs has become an urgent need in the industry.

[0003] Currently, there are mainly two methods for calculating the processing costs of parts in the market. The first calculation method is that the cost estimators open the 3D drawings through software, manually enter data relying on personal experience and relevant tables, and perform cost calculations using tools capable of addition, subtraction, multiplication, and division. The process of the second method is as follows: collect 3D design drawings and other relevant information through a software platform, perform simple calculations to obtain the weight and material type of the parts, and then estimate the material price by multiplying the weight by the unit price of the material. The cost estimators roughly estimate the production costs based on experience to form the final cost estimation result.

[0004] However, the first calculation method is time-consuming and inefficient, and due to the influence of human factors, it is difficult to guarantee the accuracy of the cost estimation results; although the second calculation method is simple, the results have a high error rate. Especially in the calculation of processing time, it often cannot accurately reflect the actual manufacturing costs, resulting in low accuracy of the final cost estimation results. Summary of the Invention

[0005] The embodiments of this application provide an intelligent cost estimation method and system for 3D drawings of parts to solve the problems of low cost estimation efficiency and low accuracy of cost estimation results existing in the prior art.

[0006] In a first aspect, the embodiments of this application provide an intelligent cost estimation method for 3D drawings of parts, including:

[0007] Input the 3D drawing information of the parts into a pre-trained process classification model, identify the process type, and match the corresponding target equipment type according to the process type;

[0008] Analyze and process the 3D drawing information of the component to obtain the geometric processing features of the component. According to the process type and the geometric processing features, match the corresponding tool type from the preset tool library, and determine the tool processing parameters based on the equipment performance parameters corresponding to the target equipment type and the tool performance parameters corresponding to the tool type;

[0009] Based on the geometric processing features, the target equipment type, and the tool processing parameters, predict the processing time of the component through a pre-trained processing time prediction model;

[0010] Generate the target pricing result of the component based on the equipment usage cost corresponding to the target equipment type, the material cost corresponding to the component raw material, and the labor cost corresponding to the processing time.

[0011] Optionally, the process classification model includes an analysis and extraction module, a multi-level processing module, and an identification module connected in sequence;

[0012] Inputting the 3D drawing information of the component into a pre-trained process classification model to identify the process type includes:

[0013] Based on the 3D drawing information of the component, use the analysis and extraction module to perform three-dimensional space topology analysis on the 3D drawing of the component to obtain a structure analysis result, and extract global geometric attributes from the structure analysis result. The global geometric attributes include surface curvature, hole and groove distribution, and contour boundary features;

[0014] Based on the global geometric attributes, use the first-level processing module in the multi-level processing module to divide the processing area in the 3D drawing of the component into multiple processing units, and perform local feature quantization analysis on each processing unit through the second-level processing module in the multi-level processing module to generate a process parameter set corresponding to each processing unit. The process parameter set includes processing depth, cutting angle, and tolerance range;

[0015] Use the identification module to identify the corresponding process type by determining whether the process parameter set meets the process rules corresponding to each process type. The process rules include constraint conditions such as equipment processing capacity thresholds, material hardness adaptability, and tool load limits.

[0016] Optionally, the performing local feature quantization analysis on each processing unit through the second-level processing module in the multi-level processing module to generate a process parameter set corresponding to each processing unit includes:

[0017] Through the second-level processing module in the multi-level processing module, the following processes are executed: For each processing unit, perform quantitative analysis on the geometry, surface roughness, and edge features of the processing unit to generate an initial local feature set, and enhance the local feature data in the initial local feature set to obtain an enhanced initial local feature set;

[0018] Based on the enhanced initial local feature sets corresponding to each processing unit, calculate the similarity between processing units to generate a similarity matrix;

[0019] According to the similarity matrix, optimize the enhanced initial local feature set corresponding to each processing unit to generate a process parameter set corresponding to each processing unit.

[0020] Optionally, the matching of the corresponding target device type according to the process type includes:

[0021] Based on the process type, screen a candidate device set that meets the machining accuracy, spindle speed range, and power requirements from a preset device database;

[0022] Rank the devices in the candidate device set to obtain a ranking result, and the ranking basis includes device idle rate, historical machining efficiency, and energy consumption coefficient;

[0023] According to the matching degree between the maximum cutting depth in the geometric machining feature and the rigidity parameters of each candidate device, select the candidate device with the largest matching degree from the ranking result as the target device type.

[0024] Optionally, the matching of the corresponding tool type from a preset tool library according to the process type and the geometric machining feature, and determining the tool machining parameters according to the device performance parameters corresponding to the target device type and the tool performance parameters corresponding to the tool type includes:

[0025] According to the process type and the geometric machining feature, construct tool selection constraint conditions, and the tool selection constraint conditions include the adaptation ranges of tool edge length, edge diameter, and coating material;

[0026] Select tools that meet the tool selection constraint conditions from a preset tool library to generate an initial tool set;

[0027] Perform secondary screening from the initial tool set according to multi-dimensional indicators to obtain the corresponding tool type, and the multi-dimensional indicators include the tensile strength of the part raw material, the bearing capacity of the target device, and the tool durability;

[0028] Based on the upper limit of the spindle speed, rigidity parameters, and equipment power limit in the equipment performance parameters, and in combination with the tensile strength and thermal conductivity of the raw materials of the components, calculate the initial feed rate and cutting depth;

[0029] According to the cutting resistance of the tool edge length, tool diameter, and coating material in the tool performance parameters, adjust the feed rate and cutting depth through a dynamic constraint optimization algorithm so that the load fluctuation range of the tool adapts to the bearing capacity of the target equipment;

[0030] According to the adjusted feed rate and cutting depth, calculate the tool overlap rate, tool cutting volume, cutting speed, and cutting path to generate tool processing parameters.

[0031] Optionally, predicting the processing time of the component through a pre-trained processing time prediction model based on the geometric processing features, the target equipment type, and the tool processing parameters includes:

[0032] Based on the geometric processing features, the target equipment type, and the tool processing parameters, construct a processing time prediction input feature set;

[0033] Input the processing time prediction input feature set into a pre-trained processing time prediction model. The processing time prediction model predicts the input features based on a deep learning algorithm to obtain the processing time of the component. The processing time is the sum of the processing times of each processing unit.

[0034] Optionally, generating the target pricing result of the component based on the equipment usage cost corresponding to the target equipment type, the material cost corresponding to the raw materials of the component, and the labor cost corresponding to the processing time includes:

[0035] Calculate the equipment usage cost according to the equipment usage rate corresponding to the target equipment type, calculate the material cost according to the type and quantity of the component materials, and calculate the labor cost according to the processing time and the preset labor rate;

[0036] Take the sum result of the equipment usage cost, the material cost, and the labor cost as the initial pricing result;

[0037] Combine the adaptive correction coefficient to adjust the initial pricing result to obtain the target pricing result.

[0038] In a second aspect, an intelligent pricing system for 3D drawings of components provided by an embodiment of the present application includes:

[0039] An identification and matching module, configured to input the 3D drawing information of the component into a pre-trained process classification model, identify the process type, and match the corresponding target equipment type according to the process type;

[0040] An analysis and determination module for analyzing and processing the 3D drawing information of the component to obtain the geometric processing features of the component, matching the corresponding tool type from a preset tool library according to the process type and the geometric processing features, and determining the tool processing parameters according to the equipment performance parameters corresponding to the target equipment type and the tool performance parameters corresponding to the tool type;

[0041] A prediction module for predicting the processing man-hours of the component through a pre-trained man-hour prediction model based on the geometric processing features, the target equipment type, and the tool processing parameters;

[0042] A generation module for generating the target pricing result of the component based on the equipment usage cost corresponding to the target equipment type, the material cost corresponding to the component raw material, and the man-hour cost corresponding to the processing man-hours.

[0043] In a third aspect, an embodiment of the present application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a 3D drawing intelligent pricing method for components as described in any item of the first aspect.

[0044] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing a computer program, and when the computer program is executed by a computer, it implements a 3D drawing intelligent pricing method for components as described in any item of the first aspect.

[0045] An embodiment of the present application provides a 3D drawing intelligent pricing method for components, including: inputting the 3D drawing information of the component into a pre-trained process classification model to identify the process type, and matching the corresponding target equipment type according to the process type; analyzing and processing the 3D drawing information of the component to obtain the geometric processing features of the component, matching the corresponding tool type from a preset tool library according to the process type and the geometric processing features, and determining the tool processing parameters according to the equipment performance parameters corresponding to the target equipment type and the tool performance parameters corresponding to the tool type; predicting the processing man-hours of the component through a pre-trained man-hour prediction model based on the geometric processing features, the target equipment type, and the tool processing parameters; generating the target pricing result of the component based on the equipment usage cost corresponding to the target equipment type, the material cost corresponding to the component raw material, and the man-hour cost corresponding to the processing man-hours.

[0046] Through the application of the process classification model, the embodiments of the present application have the capabilities of automatic analysis and calculation, can quickly output results within seconds, greatly improve the quotation efficiency, and meet the manufacturing requirements of rapid response; by combining various influencing factors such as equipment usage cost, material cost, and labor cost for price verification calculation, it ensures the accuracy of the quotation and reduces the cost risk caused by estimation errors; the target price verification result not only provides the final quotation, but also includes detailed breakdowns and calculation logics, enabling users to clearly understand each cost component and supporting a more scientific decision-making process. Further, by combining multi-level processing capabilities, refined processing of local feature quantization analysis, and intelligence of process rule matching, the intelligence level of process type recognition is improved.

[0047] These aspects or other aspects of the present application will be more clearly understood in the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application or in the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0049] Figure 1 It is a flowchart of an intelligent price verification method for 3D drawings of parts provided by an embodiment of the present application;

[0050] Figure 2 It is a schematic structural diagram of an intelligent price verification system for 3D drawings of parts provided by an embodiment of the present application;

[0051] Figure 3 It is a schematic structural diagram of a computing device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] In order to enable those skilled in the art to better understand the solutions of the present application, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application.

[0053] In some of the processes described in the specification, claims, and the above-mentioned drawings of this application, a number of operations appear in a specific order. However, it should be clearly understood that these operations may not be executed in the order in which they appear herein or may be executed in parallel. The serial numbers of the operations, such as 11, 12, etc., are only used to distinguish the different operations, and the serial numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions such as "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent a sequence, and do not limit that "first" and "second" are of different types.

[0054] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0055] In order to improve the efficiency and accuracy of the component cost estimation method, the embodiments of the present application can construct an intelligent cost estimation method by combining computer vision technology with geometric processing features in manufacturing. Specifically, the idea is as follows: Considering the characteristics of accurate classification of convolutional neural networks in the field of deep learning, especially in images, the embodiments of the present application establish a process classification model to automatically identify processing processes according to 3D drawing information. By accurately analyzing and identifying processing processes, this process classification model can intelligently match and recommend the most suitable processing equipment, improving the efficiency of equipment selection. Combining with the automatic prediction of processing man-hours, the cost estimation efficiency and the accuracy of cost estimation results are further improved.

[0056] Figure 1 The flowchart of an intelligent cost estimation method for component 3D drawings provided by the embodiments of the present application is as Figure 1 shown, and the method includes:

[0057] S11. Input the component 3D drawing information into a pre-trained process classification model to identify the process type, and match the corresponding target equipment type according to the process type.

[0058] It should be understood that component parts can refer to standardized component parts (such as bolts, nuts, etc.) or non-standard component parts. Non-standard component parts refer to component parts that are designed and manufactured according to specific requirements and do not conform to standardized specifications or dimensions. Non-standard component parts usually have unique geometric shapes, dimensions, materials, or functional requirements, and may include complex curved surfaces, holes, special-shaped structures, etc., and the requirements cannot be directly met by existing standard parts on the market. Non-standard component parts may require the use of special materials (such as high-strength alloys, high-temperature-resistant materials, etc.) or special processing techniques (such as precision machining, heat treatment, etc.). Non-standard component parts often have high requirements for machining accuracy and surface quality, and may involve complex machining processes.

[0059] It should also be understood that the process classification model can be obtained by constructing through a convolutional neural network in deep learning. Exemplarily, the process classification model can be a deep learning model composed of an analysis and extraction module, a multi-level processing module, and an identification module, and is used to extract geometric features from 3D drawings and identify machining processes. The process types include but are not limited to turning, milling, turning and milling compounding, etc., and are used to distinguish different machining processes. The target equipment type is the type of processing equipment applicable to the component parts corresponding to the 3D drawing information of the component parts. In other words, the target equipment type: the processing equipment screened based on the process type needs to meet requirements such as machining accuracy, spindle speed, power, etc., such as a CNC milling machine or a five-axis machining center.

[0060] Before the application of the process classification model, the embodiments of the present application can collect a series of 3D drawing sample data of component parts for training the process classification model. This model can identify the required machining processes in the 3D drawings of component parts, including but not limited to turning, milling, turning and milling compounding, etc. In the actual application process, the embodiments of the present application can identify the process type based on the topological features and dimensional parameters of the component parts through the process classification model. The application of the pre-trained process classification model and the matching of the corresponding target equipment type according to the process type both improve the efficiency of equipment selection and ensure the adaptability of the machining process and the machining quality during the manufacturing process.

[0061] Exemplarily, the non-standard component part contains complex curved surfaces and deep holes. After the process classification model analyzes its 3D drawing, it identifies that the "five-axis milling" process is required, and filters out a certain model of five-axis machining center with matching rigid parameters from the equipment library.

[0062] S12. Analyze and process the 3D drawing information of the component parts to obtain the geometric machining features of the component parts. According to the process type and geometric machining features, match the corresponding tool types from the preset tool library, and determine the tool machining parameters according to the equipment performance parameters corresponding to the target equipment type and the tool performance parameters corresponding to the tool types.

[0063] Among them, geometric machining features refer to the geometric characteristics of components related to machining, including but not limited to information such as dimensions, shapes, hole and groove distributions, etc. The geometric machining feature recognition algorithms in computer vision technology can be edge detection algorithms, Hough transform, scale-invariant feature transform, 3D reconstruction and analysis technology, deep learning models, etc. Among them, edge detection algorithms, such as Canny edge detection or Sobel operator, etc. Such algorithms are mainly used to identify the boundaries of components and analyze the contour boundary features of components. The Hough transform is a technique commonly used to identify simple shapes (such as straight lines, circles, etc.) in images and is mainly used to identify hole and groove distributions. Scale-invariant feature transform can identify and describe local features in drawings and can maintain high stability even under different perspectives, scales, and lighting conditions, and can identify complex surface characteristics and specific geometric shapes. 3D reconstruction and analysis technology, such as stereo matching, structured light scanning, etc., can extract detailed geometric information from 3D drawing information, including global geometric attributes such as surface curvature, hole sizes, and positions. Deep learning models can automatically learn and identify various types of geometric machining features without the need for manually designed feature extractors. As for the type, quantity, and specifications of features, they are determined according to specific application scenarios and requirements. For example, when analyzing a component, the following types of features may be concerned: geometric shapes (such as circles, rectangles, etc.), the quantity and distribution of holes and grooves, changes in surface curvature, the length and angle of edges, and the specific quantity and specifications of each feature will be determined according to the actual component design and its manufacturing requirements. By combining the above algorithms, efficient and accurate recognition of these features can be achieved, thereby supporting subsequent processes such as process classification, equipment selection, and cost estimation.

[0064] Optionally, in the case where there are multiple types of cutting tools and multiple types of target equipment, the embodiments of the present application can identify the optimal cutting tool type and equipment type for geometric machining features. This step ensures the efficiency and accuracy in the machining process.

[0065] In step S12, the equipment performance parameters include but are not limited to: upper limit of spindle speed, rigidity parameter, equipment power limit, positioning accuracy, repeat positioning accuracy, etc. Among them, the upper limit of spindle speed can refer to the maximum speed that the equipment spindle can reach, which affects the cutting speed and machining efficiency. The rigidity parameter reflects the ability of the equipment to resist deformation and is crucial for maintaining machining accuracy. The equipment power limit refers to the maximum power consumption during the operation of the equipment and is related to whether it can support high-intensity or high-speed machining tasks. The positioning accuracy refers to the position accuracy that the equipment can achieve when performing machining operations and has a direct impact on machining quality. The repeat positioning accuracy refers to the degree of consistency of the equipment's multiple positionings at the same position and is a key indicator for measuring the stability of the equipment.

[0066] The tool performance parameters include, but are not limited to, tool edge length, edge diameter, coating material, cutting resistance, durability, etc. Among them, the tool edge length can refer to the effective cutting length of the tool, which affects the amount of material that can be removed in a single feed. The edge diameter can refer to the diameter of the tool edge, which determines the cutting area and the suitable hole diameter for machining. The coating material can refer to the material covering the tool surface, which can improve the wear resistance of the tool, reduce the friction coefficient, etc., thereby extending the service life and improving the machining surface quality. The cutting resistance can refer to the magnitude of the resistance encountered by the tool during cutting, which is related to factors such as tool material, workpiece material, and cutting conditions. Durability can refer to the time that the tool can continuously be used or the amount of work completed under specific working conditions, and it is an important indicator for evaluating the economy of the tool. The equipment performance parameters and tool performance parameters provide accurate data support for subsequent tool selection and process planning, and ensure the efficiency and accuracy during the machining process. By comprehensively considering these parameters, the machining strategy can be optimized, costs can be reduced, and product quality can be improved.

[0067] Correspondingly, the parsing process can be implemented using computer vision technology. Specifically, in the embodiments of the present application, considering that computer vision technology has the characteristics of automation, high efficiency, high precision, and powerful feature extraction ability in image processing, the geometric machining feature recognition algorithm in computer vision technology is used to recognize the geometric machining features in the 3D drawing of the component. Combining with an adaptive artificial intelligence model, the embodiments of the present application can accurately recognize the geometric machining features in the 3D drawing of the component, not only deepening the understanding of the 3D drawing of the component, but also providing accurate basic data for subsequent tool selection and man-hour estimation. Optionally, the above artificial intelligence model can be a convolutional neural network, a voxel grid, a graph convolutional network, etc.

[0068] S13. Predict the machining man-hours of the component through a pre-trained man-hour prediction model based on the geometric machining features, the target equipment type, and the tool machining parameters.

[0069] It should be understood that the pre-trained man-hour prediction model can be a convolutional neural network, a multi-layer perceptron, a transformer model, etc. in the field of artificial intelligence.

[0070] Exemplarily, there are 20 hole grooves, the rotation speed of the five-axis machining center is 6000 rpm, the feed rate is 120 mm / min, the man-hour prediction model is trained through historical data, and after inputting the features, the time of each machining unit is output. For example, it takes 2.5 minutes for a single deep hole machining. By accumulating all the unit times (20 holes × 2.5 minutes + 15 minutes for surface machining), the total man-hours are 65 minutes.

[0071] Accordingly, based on geometric machining features, target equipment types, and cutting tool machining parameters, embodiments of the present application can also combine with the raw materials of parts and components, and accurately calculate the machining man-hours of parts and components through a pre-trained man-hour prediction model. This model can respond in real time to changes in different machining conditions, thereby providing an accurate time estimate for production scheduling and ensuring the efficient execution of production plans.

[0072] S14. Generate a target pricing result for the part based on the equipment usage cost corresponding to the target equipment type, the material cost corresponding to the raw material of the part, and the man-hour cost corresponding to the machining man-hours.

[0073] It should be understood that the custom pricing algorithm can be a genetic algorithm, a particle swarm optimization algorithm, fuzzy logic, a dynamic constraint optimization algorithm, etc. Among them, the dynamic constraint optimization algorithm is aimed at optimization problems with dynamically changing constraint conditions and can adjust the cutting tool parameters in real time to adapt to different machining requirements and condition changes.

[0074] Accordingly, based on equipment selection, material costs, and machining man-hours, embodiments of the present application can comprehensively calculate an accurate quotation through a custom pricing algorithm. This custom pricing algorithm takes into account various influencing factors to ensure that the target quotation result is not only accurate but also has market competitiveness.

[0075] Moreover, the target quotation result includes detailed cost components, such as material costs, equipment usage fees, and labor costs, etc. Each item is provided with a corresponding correction coefficient, and users can flexibly adjust the correction coefficients corresponding to each item according to actual needs. Exemplarily, the parameters of the costs are dynamically adjusted according to market fluctuations (such as a 10% increase in material prices) and process complexity (the surface machining difficulty coefficient is 1.2). Then, the custom pricing algorithm is used to automatically recalculate and update the quotation, providing users with convenient cost control and budget management functions. Therefore, embodiments of the present application have the advantages of transparent price details, supporting parameter adjustment and dynamic update.

[0076] Exemplarily, when calculating the initial pricing result, equipment usage cost (200 yuan / hour × 1.08 hours = 216 yuan) + material cost (titanium alloy unit price 500 yuan / kg × 0.5 kg = 250 yuan) + man-hour cost (50 yuan / hour × 1.08 hours = 54 yuan) = 520 yuan. Introduce the complexity coefficient 1.2 corresponding to the surface machining difficulty, and the final pricing = 520 yuan × 1.2 = 624 yuan. The detailed list indicates "surface machining surcharge 104 yuan".

[0077] By executing S11 to S14, the embodiments of the present application, through the application of process classification models, etc., have the ability of automatic analysis and calculation, can quickly output results within seconds, greatly improve the quotation efficiency, and meet the manufacturing requirements of rapid response; by combining various influencing factors such as equipment usage cost, material cost, and man-hour cost for price verification calculation, the accuracy of the quotation is ensured, and the cost risk caused by estimation errors is reduced; the target price verification result not only provides the final quotation, but also includes detailed details and calculation logics, enabling users to clearly understand each cost component and supporting a more scientific decision-making process. Users can set exclusive correction factors, thereby calculating a quotation that is more in line with the actual situation according to their own needs, enhancing the applicability and flexibility of the embodiments of the present application.

[0078] In some possible embodiments, the process classification model includes a parsing and extraction module, a multi-level processing module, and an identification module that are connected in sequence. Correspondingly, in S11, the 3D drawing information of the component is input into a pre-trained process classification model to identify the process type, including:

[0079] Step 111: Based on the 3D drawing information of the component, use the parsing and extraction module to perform three-dimensional spatial topology parsing on the 3D drawing of the component to obtain a structure parsing result, and extract global geometric attributes from the structure parsing result. The global geometric attributes include surface curvature, hole and groove distribution, and contour boundary features.

[0080] Among them, the global geometric attributes are extracted from the structure parsing result of the 3D drawing of the component, specifically including surface curvature, hole and groove distribution, and contour boundary features, etc. The global geometric attributes focus more on the overall description of the entire component rather than a specific processing area or unit. Therefore, it provides an overview of the overall geometric characteristics of the component. The surface curvature is used to describe the degree of curvature of the curved surface; the hole and groove distribution includes the number, position, and shape of the holes, and the contour boundary features can refer to the geometric parameters of the outer shape of the component.

[0081] In this step, the embodiments of the present application convert the 3D drawing into structured data through a three-dimensional reconstruction algorithm (such as point cloud processing or voxelization technology) to identify the spatial topology relationship of the component. Use differential geometry algorithms to calculate the curvature values of each point on the curved surface to generate a curvature distribution map. Use the Hough transform to detect circular / irregular holes and grooves, and record their positions, diameters, and depths. Extract the geometric parameters of the outer contour of the component, such as length, angle, etc., through an edge detection algorithm (such as the Canny operator). The global geometric attributes are stored in the form of structured data, such as a curvature matrix in a preset format, a hole and groove coordinate list, and a contour polygon vertex sequence.

[0082] Step 112: Based on the global geometric attributes, divide the machining areas in the 3D drawing of the component into multiple machining units through the first-level processing module in the multi-level processing module, and perform local feature quantization analysis on each machining unit through the second-level processing module in the multi-level processing module to generate a set of process parameters corresponding to each machining unit. The set of process parameters includes machining depth, cutting angle, and tolerance range.

[0083] Among them, the set of process parameters is generated based on the local feature quantization analysis of each machining unit, and contains more specific machining details and requirements. The set of process parameters includes parameters such as machining depth, cutting angle, and tolerance range. The machining depth can refer to the cutting thickness, the cutting angle refers to the tool feed direction, and the tolerance range refers to the allowable dimension error. These parameters are set to meet the machining requirements of specific process types and are of great significance for guiding the actual machining process. By judging whether these process parameters conform to the process rules of specific process types (such as equipment machining capacity threshold, material hardness adaptability, and tool load limit), the corresponding process type is identified.

[0084] Specifically, in the embodiment of the present application, based on the hole groove distribution and contour boundary in the global geometric attributes, the component is divided into independent machining units (for example, separating the deep hole area and the curved surface area). The region growing algorithm or clustering analysis algorithm is used to merge adjacent features according to geometric similarity. The shape features of the unit, such as aspect ratio and symmetry, are extracted by principal component analysis. The surface texture is quantified by the gray-level co-occurrence matrix. The edge sharpness and continuity are detected by wavelet transform. Combining with the machining process library, the machining depth (for example, the deep hole unit needs to be milled in layers, and the depth is set to 5 mm / layer), the cutting angle (adjust the tool inclination angle according to the surface curvature), and the tolerance range (the tolerance of the precision area is ±0.01 mm) are matched for each unit.

[0085] Step 113: Use the recognition module to identify the corresponding process type by judging whether the set of process parameters meets the process rules corresponding to each process type. The process rules include constraint conditions such as equipment machining capacity threshold, material hardness adaptability, and tool load limit.

[0086] Among them, the prepared processing capacity threshold can refer to the maximum cutting force, the material hardness adaptability can refer to the hardness matching between the tool and the workpiece, and the tool load limit can refer to the maximum feed rate. If the cutting force requirement (calculated according to the material hardness and cutting depth) of a certain processing unit exceeds the equipment threshold, it is marked as "multi-axis machining required". If the workpiece material is titanium alloy, the tool coating is screened as a high-temperature resistant coating. The feed rate is adjusted through the dynamic programming algorithm to ensure that the tool load fluctuation is within the equipment bearing range. If the processing unit contains deep holes (depth > 20 mm) and requires high precision (tolerance ±0.005 mm), it is determined as "precision milling"; if the unit is a complex surface (curvature change rate > 0.5 / mm), it is determined as "five-axis simultaneous machining".

[0087] Another exemplary example is that a certain engine turbine blade is a non-standard component, with special-shaped cooling holes with a depth of 25 mm and a complex surface with a drastic curvature change, and the processing process needs to be determined. The embodiments of the present application can determine the following information: the average curvature of the blade surface is 0.3 / mm, and the curvature of the blade tip is 0.8 / mm; there are 24 special-shaped cooling holes with a diameter of 3 mm and a depth of 25 mm; the total length of the blade is 150 mm, and there are 5 acute-angle turns at the edge. Then, at the first level, 24 cooling hole regions and the blade surface region are separated; at the second level, the processing depth (such as 25 mm layer milling) and cutting angle (such as vertical feed) of the cooling hole unit are quantified, and the tolerance of the surface unit is set to ±0.005 mm. For the cooling hole unit, since the depth exceeds 20 mm and a tolerance of ±0.005 mm is required, the "deep hole drilling + precision reaming" process is triggered; for the surface unit, since the curvature change rate exceeds the threshold, the "five-axis milling" process is triggered.

[0088] By executing steps 111 to 113, the embodiments of the present application achieve efficient three-dimensional topology analysis through the parsing and extraction module. The multi-level processing module decomposes complex components into multiple processing units and generates refined process parameters. The recognition module combines the process rule library to achieve accurate matching of process types. Taking the turbine blade as an example, the traditional manual price verification takes 4 hours and is prone to ignoring the precision requirements of deep holes. However, the embodiments of the present application can complete the process determination within 3 minutes, avoid the risk of tool overload, and improve the reliability of the processing plan.

[0089] As a possible implementation manner, in step 112, the second-level processing module in the multi-level processing module performs local feature quantization analysis on each processing unit to generate a set of process parameters corresponding to each processing unit, including:

[0090] Step a1: Through the second-level processing module in the multi-level processing module, perform the following process: For each processing unit, perform quantitative analysis on the geometric shape, surface roughness, and edge features of the processing unit to generate an initial local feature set, and enhance the local feature data in the initial local feature set to obtain an enhanced initial local feature set.

[0091] Exemplarily, in the embodiment of the present application, the geometric shape of the processing unit can be analyzed to extract geometric features, including curvature, angle, length, and width; the surface roughness of the processing unit can also be quantitatively analyzed to extract surface roughness features, including average roughness, peak value, and valley value; then, in the embodiment of the present application, the edge features of the processing unit can also be analyzed to extract edge features, including edge curvature, edge angle, and edge length, and the extracted geometric features, surface roughness features, and edge features are integrated into an initial local feature set.

[0092] In step a1, in the embodiment of the present application, the Gauss-Bonnet formula in differential geometry can be used to calculate the mean curvature and principal curvature of the surface of the processing unit. For example, the average curvature of a certain groove area is 0.2 diopters. The long axis and short axis directions of the unit are extracted through principal component analysis, and the aspect ratio (such as 3:1) and the key angle (such as a chamfer of 45°) are calculated. The gray-level co-occurrence matrix is used to analyze the surface texture, and the average roughness (Ra = 1.6 μm), peak value (Rz = 6.3 μm), and valley value (Rv = 5.8 μm) are extracted. The boundary points are located through Canny edge detection, and the edge curvature (such as the curvature of a sharp edge is 0.5 / mm) and the turning angle (such as a 90° right angle) are calculated by combining polynomial fitting. Then, in the embodiment of the present application, ±10% random perturbation is added to the surface roughness data to simulate measurement errors; the edge features are translated and rotated (±5°) to enhance the adaptability of the model to processing deviations. The enhanced feature set contains three types of features: geometry, roughness, and edge, and is stored in vector form (for example, [curvature 0.2, aspect ratio 3, Ra 1.6, Rz 6.3, edge curvature 0.5...]).

[0093] Step a2: Based on the enhanced initial local feature sets corresponding to each processing unit, calculate the similarity between the processing units to generate a similarity matrix.

[0094] Among them, the similarity matrix is a symmetric matrix that can reflect the feature similarity between processing units. The larger the matrix element value, the more similar the features. The weighted similarity of comprehensive geometry, roughness, and edge features is considered to avoid the influence of single-feature deviation on the global situation. Moreover, feature normalization standardizes features with different dimensions (such as curvature with the unit of / mm and roughness with the unit of μm) to the interval [0, 1]. The Euclidean distance is used to calculate the geometric feature difference (such as the geometric distance between unit A and B is 0.15); the cosine similarity is adopted to measure the surface texture matching degree (such as the similarity is 0.92); the dynamic time warping algorithm is used to match the edge curvature sequences (such as the matching degree is 85%). The comprehensive similarity is calculated according to the weight distribution (geometry 40%, roughness 30%, edge 30%) to generate an N×N matrix, where N is the number of processing units. Exemplarily, a certain component contains 10 processing units, generating a 10×10 matrix, and the similarity between unit 3 and 7 is 0.88 (high similarity, and the processing strategies may be combined).

[0095] Step a3: According to the similarity matrix, optimize the enhanced initial local feature set corresponding to each processing unit to generate the process parameter set corresponding to each processing unit.

[0096] Feature optimization can eliminate redundant data or merge similar units to improve the rationality of parameters. After optimization, the processing requirements corresponding to each processing unit, such as the processing depth stratification strategy and the optimized value of the cutting angle. The hierarchical clustering algorithm is used to group the units with a similarity > 0.8 into the same category (such as units 3, 7, and 9 are merged into the "deep hole group"). For the units in the same category, the embodiments of the present application inherit the optimal parameters within the group (such as the deepest hole in the group is 25 mm, and layer milling is uniformly adopted, 5 mm per layer); for the independent units, the embodiments of the present application independently generate parameters according to the features (such as the cutting angle of the high-curvature surface unit is adjusted to 30°). For example, if the edge length of a certain tool is 20 mm, a stratification alarm is triggered when the unit depth is 25 mm; for equipment with low rigidity, the feed rate is reduced by 20%. Among them, unit 3 (deep hole): processing depth 5 mm / layer × 5 layers, cutting angle 90°, tolerance ±0.01 mm; unit 8 (curved surface): cutting angle 30°, feed rate 100 mm / min, tolerance ±0.005 mm.

[0097] In another exemplary case, for the engine cylinder head of a certain automobile (including 20 machining units, including cooling water channel holes and combustion chamber surfaces), process parameters need to be generated. In the embodiment of the present application, the geometric features of the cooling water channel holes can be extracted, such as a diameter of 8 mm, a depth of 50 mm, a surface roughness Ra = 3.2 μm, and an edge curvature of 0.1 / mm. Then, a ±2 mm perturbation is added to the 50 mm depth to simulate the drill wear error. Next, the embodiment of the present application calculates the similarity of the 20 units and finds that the similarity of 5 water channel holes (depth 50 ± 2 mm) is >0.9, which are classified into the "deep hole group"; the similarity between the combustion chamber surface unit and other units is <0.3, which is marked as an independent unit. Unified parameters for the hole group: layer-by-layer machining (5 mm per layer, a total of 10 layers), a cutting angle of 90°, and a cemented carbide drill bit is used; for the combustion chamber surface: a cutting angle of 25°, a feed rate of 80 mm / min, and due to the large change in the surface curvature, the feed rate is reduced to prevent vibration of the tool.

[0098] By executing step a1 to step a3, the embodiment of the present application solves the problem of the singularity of machining unit features through local feature quantization and data enhancement; the similarity matrix realizes the strategy merging of similar units, reducing parameter redundancy; clustering optimization and constraint verification ensure the safety and feasibility of process parameters. Taking the engine cylinder head as an example, the traditional method needs to design parameters for 20 units one by one, which takes 3 hours, while this solution can be completed within a few minutes through an automated process, and reduces the tool load fluctuation by 40% and improves the machining efficiency by 25%.

[0099] In some possible embodiments, in step S11, matching the corresponding target device type according to the process type includes:

[0100] Step 113: Based on the process type, screen a set of candidate devices that meet the machining accuracy, spindle speed range, and power requirements from a preset device database.

[0101] Among them, the preset device database is a structured database storing device information, including parameters such as machining accuracy, spindle speed range, and power. Exemplarily, the machining accuracy is ±0.01 mm, the spindle speed range is 500 - 8000 rpm, and the power is 10 kW. The set of candidate devices refers to a list of devices initially screened that meet the basic requirements of the process type, such as devices that meet the "five-axis milling" process.

[0102] In step 113, the embodiment of the present application extracts associated parameters (machining accuracy ≤ ±0.005 mm, spindle speed ≥ 6000 rpm) from the database according to the process type (such as "precision milling"), excludes devices with an accuracy > ±0.005 mm, screens devices with a spindle speed range covering 6000 - 8000 rpm, and excludes devices with a power < 8 kW that cannot support high-intensity cutting, and finally generates a set of candidate devices (for example, 3 five-axis machining centers: device A, B, C).

[0103] Step 114: Perform priority sorting on the devices in the candidate device set to obtain a sorting result. The sorting basis includes device idle rate, historical processing efficiency, and energy consumption coefficient.

[0104] Among them, the device idle rate refers to the proportion of time that the device is not currently occupied (e.g., the idle rate of device A is 70%); the historical processing efficiency refers to the average working hours of the device for the same type of process in the past (e.g., the average time taken by device B to process the same type of parts is 50 minutes); the energy consumption coefficient refers to the energy consumption per unit time for processing (e.g., the energy consumption coefficient of device C is 1.2 kW·h / min).

[0105] Further, the weight of the device idle rate is 40%, the weight of the historical processing efficiency is 40%, and the weight of the energy consumption coefficient is 20%.

[0106] In this step 114, in the case where the idle rate of device A is 70%, the score for the idle rate is 70×0.4 = 28 points; in the case where the historical efficiency of device A is 50 minutes, the standardized efficiency score is (100 - 50)×0.4 = 20 points, where the shorter the efficiency, the better; the energy consumption of device A is 1.2 kW·h / min, and the standardized energy consumption score is (2.0 - 1.2) / 2.0×20 = 8 points. The total score of device A is 56 points. If the total score of device B is 70 points and the total score of device C is 40 points, then the sorting result is B, A, C.

[0107] Step 115: According to the matching degree between the maximum cutting depth in the geometric processing characteristics and the rigidity parameters of each candidate device, select the candidate device with the largest matching degree from the sorting result as the target device type.

[0108] Among them, the rigidity parameter is an index of the device's ability to resist deformation (e.g., the rigidity parameter of device B is 500 N / μm); the maximum cutting depth is the maximum thickness of the material to be removed in the processing unit (e.g., the cutting depth of a certain deep hole is 5 mm); the matching degree is the degree of adaptation between the device rigidity parameter and the cutting depth requirement.

[0109] Exemplarily, in the embodiment of the present application, according to the material hardness (such as titanium alloy) and the cutting depth (5 mm), the cutting force F = K×depth×width = 1200 N is calculated by the formula. If the rigidity of device B is 500 N / μm and the safety factor is taken as 1.5, then the matching degree = 500 / (1200×1.5) = 0.28; if the rigidity of device A is 400 N / μm, then the matching degree = 400 / (1200×1.5) = 0.22; the matching degree needs to be ≥0.2. Both device B and device A are qualified, but the matching degree of device B is higher. Select device B with the highest matching degree from the sorting result (B > A > C) as the target device type.

[0110] In other examples, a non-standard part (made of titanium alloy) needs to be precision machined with deep holes (cutting depth of 5 mm, tolerance of ±0.005 mm). The screening criteria are: machining accuracy ≤ ±0.005 mm, spindle speed ≥ 6000 rpm, power ≥ 8 kW; the candidate devices are: five-axis machining center B (accuracy of ±0.003 mm, speed of 8000 rpm, power of 10 kW), device A (±0.004 mm, 7000 rpm, 9 kW), and the machining accuracy of device C is ±0.006 mm, so device C is eliminated. If the idle rate of device B is 60%, the historical efficiency is 45 minutes, and the energy consumption is 1.1 kW·h / min, then the total score of device B = 60×0.4+(100 - 45)×0.4+(2.0 - 1.1) / 2.0×20 = 24 + 22 + 9 = 55 points; if the idle rate of device A is 50%, the historical efficiency is 55 minutes, and the energy consumption is 1.3 kW·h / min, then the total score of device A = 20 + 18 + 7 = 45 points; therefore, the sorting result is: B > A.

[0111] Calculate the cutting force F = 1200 N, the rigidity of device B is 500 N / μm, and the matching degree of device B is 0.28; the rigidity of device A is 400 N / μm, and the matching degree of device A is 0.22. Therefore, the embodiment of the present application selects device B as the target device to ensure sufficient rigidity during the machining process and avoid tool chatter.

[0112] By performing steps 113 to 115, the embodiment of the present application screens candidate devices driven by the process type, reducing the number of ineffective devices; combines multi-dimensional sorting of idle rate, efficiency, and energy consumption to optimize resource utilization; and ensures machining stability through rigid parameter matching.

[0113] In some possible embodiments, in step S12, according to the process type and geometric machining features, the corresponding tool type is matched from a preset tool library, and the tool machining parameters are determined according to the device performance parameters corresponding to the target device type and the tool performance parameters corresponding to the tool type, including:

[0114] Step 121, construct tool selection constraint conditions according to the process type and geometric machining features, and the tool selection constraint conditions include the adaptation ranges of tool edge length, edge diameter, and coating material.

[0115] In step 121, the tool selection constraint conditions are limit parameters based on the process type and geometric features. For example, the tool edge length needs to be ≥ the hole depth, the edge diameter needs to be ≤ 90% of the hole diameter, and the coating material needs to be adapted to the workpiece hardness.

[0116] In step 121, if the process is "deep hole drilling", the constraint conditions include the cutting edge length ≥ hole depth (e.g., 25 mm), and the cutting edge diameter ≤ hole diameter (e.g., an 8 mm tool is adapted to a 9 mm hole diameter). For the surface machining unit, the constraint is that the radius of the tool ball nose ≤ the minimum radius of curvature (e.g., a ball nose of R1 mm is required for a curvature radius of 2 mm). The constraint conditions are stored in a logical expression, for example, {cutting edge length ≥ 25 mm, cutting edge diameter ≤ 8 mm, coating = a certain material coating}.

[0117] Step 122: Select tools that meet the tool selection constraint conditions from the preset tool library to generate an initial tool set.

[0118] Among them, the preset tool library is a database containing parameters such as tool types (drills, milling cutters), cutting edge length, cutting edge diameter, and coating. The initial tool set is a list of tools that meet the constraint conditions of step 121. For example, 5 drills that meet the requirements of deep hole drilling are selected.

[0119] In step 122, tools with insufficient inventory or discontinued production can be excluded to generate an initial set (such as drills A, B, and C).

[0120] Step 123: Perform a secondary screening from the initial tool set according to multi-dimensional indicators to obtain the corresponding tool types. The multi-dimensional indicators include the tensile strength of the part raw material, the load-bearing capacity of the target equipment, and the tool durability.

[0121] Among them, the multi-dimensional indicators include the tensile strength of the raw material (e.g., the tensile strength of titanium alloy is 950 MPa), the load-bearing capacity of the equipment (e.g., the maximum feed force is 2000 N), and the tool durability (e.g., the life ≥ 200 minutes).

[0122] In step 123, if the workpiece material is titanium alloy (high tensile strength), the tool substrate material is screened as cemented carbide. Calculate the cutting force of the tool (F = K × feed rate × cutting depth). If the F of drill C = 1800 N exceeds the equipment load-bearing upper limit (2000 N × safety factor 0.8 = 1600 N), it is eliminated. According to historical data, the life of drill A is 250 minutes > the life of drill B is 200 minutes, and the final tool type is drill A.

[0123] An algorithm can be designed in the embodiment of the present application. The algorithm can be a rule-based expert system, a machine learning model, etc., to automatically execute steps 121 to 123 through the algorithm. From the descriptions of steps 121 to 123, it can be seen that the embodiment of the present application can select tools based on the process type, geometric machining features, tensile strength of the raw material, and load-bearing capacity of the target equipment.

[0124] Step 124: Calculate the initial feed rate and cutting depth based on the upper limit of the spindle speed, rigidity parameter, and equipment power limit in the equipment performance parameters, in combination with the tensile strength and thermal conductivity of the part raw material.

[0125] Among them, the initial feed rate Ra is the distance that the tool moves per minute. Exemplarily, Ra = spindle speed (rpm) × feed per tooth (mm / tooth) × number of teeth. The initial cutting depth ap is the thickness of the material cut in a single pass, which is limited by the rigidity of the equipment (such as 500 N / μm) and the thermal conductivity of the material (such as 16 W / m·K for titanium alloy).

[0126] In step 124, the embodiment of the present application can take a safety value of 6000 rpm according to the upper speed limit of 8000 rpm of equipment B. The cutting depth is 0.25 mm, and the feed per tooth is taken as 0.05 mm / tooth.

[0127] Step 125: Adjust the feed rate and cutting depth through a dynamic constraint optimization algorithm according to the cutting resistance of the tool edge length, tool diameter, and coating material in the tool performance parameters, so that the load fluctuation amplitude of the tool matches the bearing capacity of the target equipment.

[0128] Among them, the dynamic constraint optimization algorithm is used to adjust the parameters in real time so that the tool load fluctuation (such as cutting force change) matches the bearing capacity of the equipment. For example, the feed rate is reduced from 1200 mm / min to 1000 mm / min to reduce the load peak.

[0129] In step 125, the dynamic constraint optimization algorithm adjusts the feed rate and cutting depth in real time to ensure that the tool load is within the bearing range of the equipment.

[0130] Optionally, the embodiment of the present application simulates the cutting process through finite element analysis and detects that the cutting force peak is 1900 N when the feed rate is 1200 mm / min, exceeding the safety threshold of equipment B (such as 1600 N). The embodiment of the present application can also iteratively reduce the feed rate to 1000 mm / min through the particle swarm optimization algorithm, and at the same time increase the cutting depth to 0.3 mm to make the cutting force stable within 1500 N.

[0131] Step 126: Calculate the tool overlap rate, tool cutting volume, cutting speed, and cutting path according to the adjusted feed rate and cutting depth to generate tool processing parameters.

[0132] It should be understood that the tool processing parameters are also called key processing attributes. The tool overlap rate is the overlapping ratio of adjacent cutting paths (such as 30%), which affects the surface quality; the cutting path is the tool movement trajectory, such as spiral plunge cutting (suitable for deep holes) or contour milling (suitable for curved surfaces).

[0133] Correspondingly, the embodiment of the present application can automatically generate a tool path according to the geometric processing features, and calculate tool processing parameters such as the tool overlap rate, cutting volume, cutting speed, cutting path, and cutting length according to the adjusted feed rate and cutting depth.

[0134] Exemplarily, in the embodiments of the present application, edge detection is used to identify the contour, the Hough transform is used to locate the position of the hole and groove, and 3D reconstruction is used to extract the surface curvature. According to the process type (such as milling) and geometric features (hole depth of 10 mm), a coated carbide tool with a cutting edge length ≥ 12 mm is selected from the tool library. Considering the workpiece material (titanium alloy with a tensile strength of 50 MPa) and the equipment power (the upper limit of the spindle speed is 8000 rpm), tools with insufficient durability are excluded. Based on the equipment rigidity parameters and the cutting resistance of the tool, a dynamic algorithm adjusts the feed rate to 150 mm / min and the cutting depth to 0.5 mm.

[0135] Another exemplarily, in the embodiments of the present application, based on geometric features, a spiral path (descending 0.3 mm per turn) is adopted for deep holes, and a contour path (layer height of 0.2 mm) is adopted for curved surfaces. A machining parameter table is generated, including a feed rate of 1000 mm / min, a cutting depth of 0.3 mm, a helix angle of 30°, and an overlap ratio of 25%.

[0136] Yet another exemplarily, for the turbine disk of an engine, the material is nickel-based alloy with a tensile strength of 1200 MPa. It is required to machine 24 cooling holes with a hole diameter of 10 mm and a depth of 50 mm. The constraint conditions provided by the embodiments of the present application are that the cutting edge length ≥ 50 mm, the cutting edge diameter ≤ 9 mm, and the coating is a high-temperature resistant coating. Then, 3 carbide drills (cutting edge length of 55 mm, cutting edge diameter of 9 mm, high-temperature coating) are selected from the library. Drill C with an excessive cutting force is eliminated through secondary screening, and drill A with the highest durability is selected. Next, the initial feed rate of 800 mm / min and the cutting depth of 0.2 mm are calculated. After optimization, the feed rate is reduced to 700 mm / min and the cutting depth is increased to 0.25 mm, and the load is stabilized at 1400 N. Finally, the embodiments of the present application generate a spiral drilling path with an overlap ratio of 20%. The single-hole working hours = 50 mm / (0.25 mm / turn × 700 mm / min) = 28.6 minutes.

[0137] By performing Step 121 to Step 126, the embodiments of the present application construct and perform multi-dimensional screening through constraint conditions to accurately match high-durability tools; the dynamic optimization algorithm balances the machining efficiency and the equipment load to avoid the risk of overload; the path planning improves the machining quality.

[0138] Optionally, the embodiments of the present application can calculate the feed rate based on the spindle speed of the equipment (such as a machine tool), thereby improving the accurate calculation of the machining working hours. Exemplarily, Ra = sp × ci / 60, where Ra is the feed rate, representing the distance that the tool moves per minute. The feed rate affects the machining efficiency and the surface quality. sp is the spindle speed, representing the number of revolutions per minute of the machine tool spindle. The spindle speed determines the cutting speed, which in turn affects the machining efficiency and the tool life. ci is the chip load per revolution, representing the distance that the tool moves per revolution of the spindle. The chip load per revolution directly affects the cutting thickness and the machining surface quality.

[0139] Optionally, the embodiments of the present application can also simply calculate the cutting amount according to geometric machining features. Exemplarily, ar = (lp × wp - ma) / (2 × rate), where ar is the cutting amount, representing the volume of material to be removed. The cutting amount is an important parameter for calculating machining man-hours. lp is the height of the component, representing the dimension of the component in the machining direction and used to calculate the geometric volume of the component. wp is the width of the component, representing the dimension of the component perpendicular to the machining direction and used to calculate the geometric volume of the component. ma is the geometric machining feature, used to calculate the actual volume of material to be removed. rate is the conversion rate between the actual image and the component size, representing the proportional relationship between pixels in the image and the actual size, and used to convert the size of the component in the 3D drawing to the actual size.

[0140] In a possible embodiment, S13, predicting the machining man-hours of the component through a pre-trained man-hour prediction model based on geometric machining features, target equipment type, and tool machining parameters, includes:

[0141] Step 131, constructing a man-hour prediction input feature set based on geometric machining features, target equipment type, and tool machining parameters.

[0142] The man-hour prediction input feature set: a structured data set integrating geometric machining features (such as the number of holes and grooves, surface curvature), target equipment type (such as the five-axis machining center encoded as a categorical variable), and tool machining parameters (feed rate, cutting depth), and is used to input into the man-hour prediction model.

[0143] Feature normalization: performing standardization processing on features with different dimensions (such as the curvature unit is / mm, and the feed rate is mm / min) to make them fall into the [0, 1] interval and avoid model training deviation.

[0144] Identifying the number of holes (such as 24) and the average depth (such as 50 mm) through the Hough transform, and extracting the maximum curvature (such as 0.8 / mm) and surface area (such as 2000 mm 2 ) using the curvature calculation algorithm. Converting the target equipment type (such as "five-axis machining center") into a one-hot encoding ([1, 0, 0]). Extracting the feed rate (1000 mm / min), cutting depth (0.3 mm), and tool overlap rate (25%) from the tool machining parameters. Feature value = (original value - mean) / standard deviation; for example, the feed rate mean is 800 mm / min, and the standard deviation is 200 → the feed rate of 1000 mm / min is standardized to 1.0. The feature set is in vector form, such as [number of holes 24, maximum curvature 0.8, equipment encoding 1, 0, 0, feed rate 1.0, cutting depth 0.3].

[0145] Step 132: Input the set of input features for man-hour prediction into a pre-trained man-hour prediction model. The man-hour prediction model predicts the input features based on a deep learning algorithm to obtain the processing man-hours of the components. The processing man-hours are the sum of the processing times of each processing unit.

[0146] The man-hour prediction model refers to a deep learning model (such as a multi-layer perceptron) trained based on historical processing data. After inputting the set of input features, it outputs the predicted man-hours of each processing unit. The sum of the processing man-hours refers to the cumulative value of the man-hours of all processing units. For example, if the deep hole processing unit takes 30 minutes and the curved surface unit takes 20 minutes, the total man-hours are 50 minutes.

[0147] The model architecture of the deep learning model includes an input layer, hidden layers, and an output layer. Among them, the number of nodes in the input layer = the feature dimension, such as 6 dimensions; there are 3 hidden layers, with 128 nodes in each layer, and the activation function is the ReLU function; the output layer has 1 node, with a linear activation function, predicting the man-hours of a single processing unit. After the model training is completed, input the set of input features into the model, and output the man-hours of each unit (such as 28 minutes for the deep hole unit and 18 minutes for the curved surface unit); accumulate the man-hours of all units, and the total man-hours = 28 + 18 = 46 minutes.

[0148] Exemplarily, the maximum curvature of the special-shaped curved surface is 1.2 / mm, and the average depth of the deep hole is 40mm; a five-axis machining center (encoded [1, 0, 0]); the feed rate is 900mm / min, and the cutting depth is 0.25mm; the standardized features are [curved surface curvature 1.2, hole depth 40, equipment 1, 0, 0, feed rate 0.9, cutting depth 0.25]. After model prediction, the man-hours of a single curved surface unit = 22 minutes, and the deep hole unit = 18 minutes; the total man-hours = 10×22 + 15×18 = 220 + 270 = 490 minutes.

[0149] By performing Step 131 to Step 132, the embodiments of the present application achieve efficient and accurate estimation of processing man-hours through the construction of a structured feature set and the prediction of a deep learning model.

[0150] Optionally, to achieve simple calculation of processing man-hours, the embodiments of the present application can use the following formula: y = ar / (ap × Ra), where y is the processing man-hours, representing the time required to complete the processing, and the unit can be minutes, which is used to evaluate processing efficiency and cost. ar is the cutting volume, and the larger the cutting volume, the longer the processing man-hours. ap is the cutting depth, representing the thickness of each cut, and the cutting depth directly affects the material removal amount of each cut. Ra is the feed rate, and the higher the feed rate, the shorter the processing man-hours.

[0151] In a possible embodiment, S14: Generate a target pricing result for the component based on the equipment usage cost corresponding to the target equipment type, the material cost corresponding to the component raw material, and the man-hour cost corresponding to the processing man-hours, including:

[0152] Step 141: Calculate the equipment usage cost according to the equipment usage rate corresponding to the target equipment type, calculate the material cost according to the type and quantity of component materials, and calculate the labor cost according to the processing hours and the preset labor rate.

[0153] Among them, the equipment usage rate refers to the usage cost of the equipment per unit time (such as the rate of a five-axis machining center is 200 yuan / hour), which is obtained from the preset rate table according to the equipment type. The material cost refers to the total cost of the component raw materials, and the calculation formula is material quantity × material unit price, where the quantity is calculated by the volume of the 3D model (such as the volume of titanium alloy is 0.05m 3 × density 4.5g / cm 3 = 225 kg), and the unit price is obtained from the real-time market price database (such as titanium alloy is 500 yuan / kg). The labor rate refers to the processing labor cost per unit labor hour (such as 50 yuan / hour), which is determined by the preset standard of the enterprise.

[0154] Exemplarily, when the rate of equipment B (five-axis machining center) is 200 yuan / hour and the processing hours are 8.17 hours (490 minutes), the equipment usage cost = 200 × 8.17 = 1634 yuan. When the quantity of titanium alloy is 225 kg and the unit price is 500 yuan / kg, the material cost = 225 × 500 = 112,500 yuan. When the labor hours are 8.17 hours and the rate is 50 yuan / hour, the labor cost = 8.17 × 50 = 408.5 yuan.

[0155] Another exemplarily, in order to calculate the cost price of the processed product (i.e., the above-mentioned component), the embodiments of the present application can first calculate the quantities of various holes, corners, etc. of the processed product, or count the depths of the grooves in the three-dimensional coordinates. For example, for a drawing, the embodiments of the present application first count the numbers of through holes, blind holes and grooves in the drawing, and the prices of different through holes can also be customized according to the depth of the milling groove. For example, the price of 10 cm is different from that of 30 cm.

[0156] Step 142: Take the sum of the equipment usage cost, material cost and labor cost as the initial pricing result.

[0157] Among them, the initial pricing result can refer to the direct sum of the equipment, material and labor costs, reflecting the basic cost structure. Exemplarily, the initial pricing result = equipment cost 1634 yuan + material cost 112,500 yuan + labor cost 408.5 yuan = 114,542.5 yuan.

[0158] Step 143: Combine the adaptive correction coefficient to adjust the initial pricing result to obtain the target pricing result.

[0159] Optionally, the adaptive correction coefficient may refer to a comprehensive coefficient for dynamically adjusting the nuclear price, covering factors such as market fluctuations (e.g., a 10% increase in material prices), process complexity (a surface processing difficulty coefficient of 1.2), and volume discounts (e.g., a coefficient of 0.9 when the volume > 100 pieces). Exemplarily, when the market fluctuation is a 10% increase in the unit price of titanium alloy, the material coefficient = 1.1; when the process complexity represents 10 special-shaped surface units, the complexity coefficient = 1.2; when the volume discount is for an order quantity of 50 pieces (<100), the discount coefficient = 1.0; the comprehensive coefficient

[0160] = 1.1 × 1.2 × 1.0 = 1.32.

[0162] Alternatively, the adaptive correction coefficient may be a coefficient corresponding to a single indicator, with different indicators corresponding to different coefficients. For example, when the market fluctuation is a 10% increase in the unit price of titanium alloy, the adaptive correction coefficient corresponding to the material = 1.1; when the process complexity represents 10 special-shaped surface units, the adaptive correction coefficient corresponding to the complexity = 1.2; when the volume discount is for an order quantity of 50 pieces (<100), the adaptive correction coefficient corresponding to the discount = 1.0.

[0163] By performing Step 141 to Step 143, the embodiments of the present application ensure cost transparency and traceability through accurate accounting of equipment, materials, and man-hours; the adaptive correction coefficient dynamically responds to market and process changes, avoiding the risk of underestimated quotes. This solution controls the error within ±5% through dynamic correction and supports detailed breakdowns (such as material and process surcharges), which can help enterprises price scientifically and manage costs.

[0164] Furthermore, to facilitate unified management of the target nuclear price results corresponding to all the 3D drawing information of the components, the embodiments of the present application may set up a reference database including all the target nuclear price results, and each target nuclear price result includes equipment usage fees, material fees, labor fees, and other miscellaneous fees. Also, a flexible adjustment channel can be designed to allow users to adjust these cost parameters according to actual situations to achieve more accurate quotes and cost control. Through the above steps, the embodiments of the present application can automate the entire process from inputting the 3D drawing information of the components to outputting the processing quote and calculation details, providing users with an efficient and accurate quote and cost management tool.

[0165] In the above embodiments, to ensure the accurate extraction of geometric machining features, the embodiments of the present application can extract global geometric attributes from 3D drawings through a specified graphic feature extraction algorithm. And to improve the recognition ability of complex geometric machining features and ensure the accuracy of process planning, the embodiments of the present application can introduce an arbitrary shape cutting groove graphic recognition algorithm to recognize complex shapes and cutting groove features in parts. To provide sectional feature information of parts and support local feature quantification analysis, the embodiments of the present application can introduce a sectional extraction feature information algorithm to extract geometric features of parts. To improve the efficiency and accuracy of feature extraction and reduce redundant data, the embodiments of the present application can optimize the feature extraction process through the ant algorithm to identify key geometric features. In the process of calculating the similarity between different machining units, the embodiments of the present application can introduce the mean hash algorithm, the difference hash algorithm, and the perceptual hash algorithm, and use the fusion result of the three hash algorithms as the final feature similarity. To provide multi-dimensional similarity evaluation, the embodiments of the present application can introduce a three-channel histogram similarity difference algorithm to calculate the three-channel histogram of geometric features and generate a similarity matrix. To improve the robustness of local feature data, the embodiments of the present application can repair and enhance the extracted local feature data through an image inpainting enhancement algorithm.

[0166] Among them, the specified graphic feature extraction algorithm refers to Scale-Invariant Feature Transform (SIFT) or Speeded-Up Robust Features (SURF), which can identify and describe features in drawings. The arbitrary shape cutting groove graphic recognition algorithm refers to a convolutional neural network. The sectional extraction feature information algorithm refers to a technology based on computer tomography reconstruction or a sectional analysis algorithm. The three-channel histogram similarity difference algorithm is a method for evaluating the similarity of geometric features, which generates a similarity matrix by calculating the three-channel (for example, the three channels in the RGB color space, but may be different types of feature representations in geometric features) histogram of geometric features. The specific implementation can adopt color histogram matching technology extended to the field of geometric features, such as a histogram comparison method based on the distribution of geometric features, and measure the similarity by calculating the differences between different histograms (such as Euclidean distance, chi-square distance, etc.). The image inpainting enhancement algorithm refers to a repair network or morphological operations, including dilation, erosion and other technologies to improve the quality and integrity of feature extraction results.

[0167] Figure 2 FIG. is a schematic structural diagram of an intelligent pricing system for 3D drawings of parts provided by the embodiments of the present application, as Figure 2 shown, the system includes:

[0168] An identification and matching module 21, configured to input the 3D drawing information of the part into a pre-trained process classification model, identify the process type, and match the corresponding target device type according to the process type.

[0169] The parsing and determination module 22 is configured to parse and process the 3D drawing information of the component parts to obtain the geometric processing features of the component parts, match the corresponding tool types from a preset tool library according to the process type and geometric processing features, and determine the tool processing parameters based on the equipment performance parameters corresponding to the target equipment type and the tool performance parameters corresponding to the tool types.

[0170] The prediction module 23 is configured to predict the processing time of the component parts through a pre-trained man-hour prediction model based on the geometric processing features, target equipment type, and tool processing parameters.

[0171] The generation module 24 is configured to generate the target pricing result of the component parts based on the equipment usage cost corresponding to the target equipment type, the material cost corresponding to the raw materials of the component parts, and the man-hour cost corresponding to the processing time.

[0172] Figure 2 The intelligent pricing system for 3D drawings of the component parts described above can execute Figure 1 the intelligent pricing method for 3D drawings of the component parts described in the illustrated embodiment, and its implementation principle and technical effects will not be elaborated further.

[0173] For the intelligent pricing system for 3D drawings of the component parts in the above embodiment, the specific ways in which each module and unit perform operations have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0174] In a possible design, Figure 2 the intelligent pricing system for 3D drawings of the component parts in the illustrated embodiment can be implemented as a computing device, such as Figure 3 shown, and this computing device may include a storage component 31 and a processing component 32.

[0175] The storage component 31 stores one or more computer instructions, and among them, the one or more computer instructions are called and executed by the processing component 32.

[0176] The processing component 32 is configured to: input the 3D drawing information of the component parts into a pre-trained process classification model to identify the process type, and match the corresponding target equipment type according to the process type; parse and process the 3D drawing information of the component parts to obtain the geometric processing features of the component parts, match the corresponding tool types from a preset tool library according to the process type and geometric processing features, and determine the tool processing parameters based on the equipment performance parameters corresponding to the target equipment type and the tool performance parameters corresponding to the tool types; predict the processing time of the component parts through a pre-trained man-hour prediction model based on the geometric processing features, target equipment type, and tool processing parameters; generate the target pricing result of the component parts based on the equipment usage cost corresponding to the target equipment type, the material cost corresponding to the raw materials of the component parts, and the man-hour cost corresponding to the processing time.

[0177] Among them, the processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components for executing the above method.

[0178] The storage component 31 is configured to store various types of data to support the operation of the terminal. The storage component may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as random access memory (RAM), static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0179] Of course, the computing device may also necessarily include other components, such as input / output interfaces, display components, communication components, etc.

[0180] The input / output interface provides an interface between the processing component and the peripheral interface module, and the above peripheral interface module may be an output device, an input device, etc.

[0181] The communication component is configured to facilitate communication between the computing device and other devices in a wired or wireless manner, etc.

[0182] Among them, the computing device may be a physical device or an elastic computing host provided by a cloud computing platform, etc. At this time, the computing device may refer to a cloud server, and the above processing component, storage component, etc. may be basic server resources leased or purchased from the cloud computing platform.

[0183] The embodiments of the present application also provide a computer storage medium storing a computer program, and when the computer program is executed by a computer, it can implement the above-mentioned Figure 1 intelligent pricing method for 3D drawings of parts in the embodiments shown.

[0184] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0185] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0186] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0187] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of each embodiment of the present application.

Claims

1. An intelligent pricing method for 3D drawings of parts, characterized in that, Including: Input the 3D drawing information of the parts into a pre-trained process classification model, identify the process type, and match the corresponding target equipment type according to the process type; Perform parsing processing on the 3D drawing information of the parts to obtain the geometric processing features of the parts. According to the process type and the geometric processing features, match the corresponding tool type from a preset tool library, and determine the tool processing parameters according to the equipment performance parameters corresponding to the target equipment type and the tool performance parameters corresponding to the tool type; Based on the geometric processing features, the target equipment type, and the tool processing parameters, predict the processing man-hours of the parts through a pre-trained man-hour prediction model; Generate the target pricing result of the parts based on the equipment usage cost corresponding to the target equipment type, the material cost corresponding to the raw materials of the parts, and the man-hour cost corresponding to the processing man-hours.

2. The method according to claim 1, characterized in that, The process classification model includes a parsing and extraction module, a multi-level processing module, and an identification module connected in sequence; The inputting the 3D drawing information of the parts into a pre-trained process classification model to identify the process type includes: Based on the 3D drawing information of the parts, use the parsing and extraction module to perform three-dimensional space topology parsing on the 3D drawing of the parts to obtain a structure parsing result, and extract global geometric attributes from the structure parsing result. The global geometric attributes include surface curvature, hole and groove distribution, and contour boundary features; Based on the global geometric attributes, use the first-level processing module in the multi-level processing module to divide the processing area in the 3D drawing of the parts into multiple processing units, and perform local feature quantization analysis on each processing unit through the second-level processing module in the multi-level processing module to generate a process parameter set corresponding to each processing unit. The process parameter set includes processing depth, cutting angle, and tolerance range; Use the identification module to identify the corresponding process type by determining whether the process parameter set meets the process rules corresponding to each process type. The process rules include constraint conditions such as equipment processing capacity threshold, material hardness adaptability, and tool load limit.

3. The method according to claim 2, wherein The performing local feature quantization analysis on each processing unit through the second-level processing module in the multi-level processing module to generate a process parameter set corresponding to each processing unit includes: Through the second-level processing module in the multi-level processing module, perform the following process: for each processing unit, perform quantization analysis on the geometric shape, surface roughness, and edge features of the processing unit to generate an initial local feature set, and enhance the local feature data in the initial local feature set to obtain an enhanced initial local feature set; Based on the enhanced initial local feature sets corresponding to each processing unit, calculate the similarity between the processing units to generate a similarity matrix; According to the similarity matrix, optimize the enhanced initial local feature set corresponding to each processing unit to generate a process parameter set corresponding to each processing unit.

4. The method according to any one of claims 1 to 3, characterized in that, Matching the corresponding target equipment type according to the process type includes: Based on the process type, screening a set of candidate equipment that meets the machining accuracy, spindle speed range, and power requirements from a preset equipment database; Sorting the equipment in the set of candidate equipment by priority to obtain a sorting result, and the sorting basis includes equipment idle rate, historical processing efficiency, and energy consumption coefficient; According to the matching degree between the maximum cutting depth in the geometric machining feature and the rigidity parameters of each candidate equipment, selecting the candidate equipment with the largest matching degree from the sorting result as the target equipment type.

5. The method according to any one of claims 1 to 3, characterized in that, Matching the corresponding tool type from a preset tool library according to the process type and the geometric machining feature, and determining tool processing parameters according to the equipment performance parameters corresponding to the target equipment type and the tool performance parameters corresponding to the tool type, including: Constructing tool selection constraint conditions according to the process type and the geometric machining feature, where the tool selection constraint conditions include the adaptation ranges of tool edge length, edge diameter, and coating material; Selecting tools that meet the tool selection constraint conditions from a preset tool library to generate an initial tool set; Performing secondary screening from the initial tool set according to multi-dimensional indicators to obtain the corresponding tool type, and the multi-dimensional indicators include the tensile strength of the component raw material, the load-bearing capacity of the target equipment, and tool durability; Based on the upper limit of spindle speed, rigidity parameter, and equipment power limit in the equipment performance parameters, combined with the tensile strength and thermal conductivity coefficient of the component raw material, calculating the initial feed rate and cutting depth; According to the cutting resistance of the tool edge length, edge diameter, and coating material in the tool performance parameters, adjusting the feed rate and cutting depth through a dynamic constraint optimization algorithm so that the load fluctuation range of the tool adapts to the load-bearing capacity of the target equipment; Calculating the tool overlap rate, tool cutting amount, cutting speed, and cutting path according to the adjusted feed rate and cutting depth to generate tool processing parameters.

6. The method according to any one of claims 1 to 3, characterized in that Predicting the processing time of the component through a pre-trained processing time prediction model based on the geometric machining feature, the target equipment type, and the tool processing parameters, including: Constructing a processing time prediction input feature set based on the geometric machining feature, the target equipment type, and the tool processing parameters; Inputting the processing time prediction input feature set into a pre-trained processing time prediction model, and the processing time prediction model predicts the input features based on a deep learning algorithm to obtain the processing time of the component, where the processing time is the sum of the processing times of each processing unit.

7. The method according to any one of claims 1 to 3, characterized in that Generating the target pricing result of the component based on the equipment usage cost corresponding to the target equipment type, the material cost corresponding to the component raw material, and the processing time cost corresponding to the processing time, including: Calculating the equipment usage cost according to the equipment usage rate corresponding to the target equipment type, calculating the material cost according to the type and quantity of the component material, and calculating the processing time cost according to the processing time and the preset processing time rate; Take the sum of the equipment usage cost, the material cost, and the man-hour cost as the initial pricing result; Combine the adaptive correction coefficient to adjust the initial pricing result to obtain the target pricing result.

8. An intelligent pricing system for 3D drawings of parts, characterized in that, It includes: An identification and matching module, which is used to input the 3D drawing information of the parts into a pre-trained process classification model to identify the process type, and match the corresponding target equipment type according to the process type; An analysis and determination module, which is used to analyze and process the 3D drawing information of the parts to obtain the geometric processing features of the parts, match the corresponding tool type from the preset tool library according to the process type and the geometric processing features, and determine the tool processing parameters according to the equipment performance parameters corresponding to the target equipment type and the tool performance parameters corresponding to the tool type; A prediction module, which is used to predict the processing man-hours of the parts through a pre-trained man-hour prediction model based on the geometric processing features, the target equipment type, and the tool processing parameters; A generation module, which is used to generate the target pricing result of the parts based on the equipment usage cost corresponding to the target equipment type, the material cost corresponding to the raw materials of the parts, and the man-hour cost corresponding to the processing man-hours.

9. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a 3D drawing intelligent pricing method for parts as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that, A computer program is stored, and when the computer program is executed by the computer, it implements a 3D drawing intelligent pricing method for parts as described in any one of claims 1 to 7.

Images