Metal plate multiple bending and flattening interference method, system and equipment and storage medium

Abstract

The invention provides a metal plate multi-bending and flattening interference method, system and equipment and a storage medium, and relates to the field of metal plate machining. According to the method, an accurate metal plate model is firstly built, then bending and flattening calculation is simulated, then real-time interference detection and judgment are carried out according to boundary conditions, and parameters are automatically optimized and adjusted when interference is detected; the system is composed of a model building module, a simulation calculation module, an interference detection module and a parameter adjustment module which work cooperatively. The equipment guarantees operation through software algorithm optimization, intelligent error detection and repair and data management interaction; the storage medium has the advantages of data storage safety, redundancy recovery, high-speed transmission and good compatibility, the problem of interference caused by repeated bending and flattening of the metal plate is solved comprehensively, and the production quality and efficiency are improved. The problem that when a metal plate which is bent for multiple times is flattened, interference occurs is solved. The problems that the metal plate cannot be accurately flattened due to interference, follow-up machining and using are affected, and production efficiency and product quality are reduced are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of sheet metal processing, and in particular to a sheet metal multiple bending and flattening interference method, system, equipment and storage medium. Background Art

[0002] In the process of sheet metal processing, multiple bending is a common process operation. However, when flattening a sheet metal that has been bent multiple times, interference problems often occur. These interferences may cause the sheet metal to be unable to be accurately flattened, affecting subsequent processing and use, and reducing production efficiency and product quality. At present, the traditional solution to the problem of interference in the flattening of sheet metal after multiple bending has defects such as insufficient accuracy, complex calculations, and difficulty in adapting to complex bending situations, which cannot meet the growing demand for high-precision sheet metal processing. Summary of the invention

[0003] In view of this, the present invention provides a sheet metal multiple bending and flattening interference method, system, device and storage medium to solve the problems raised by the background technology.

[0004] The present invention provides a sheet metal multiple bending and flattening interference method, which specifically includes the following steps:

[0005] Step 1: Create an accurate sheet metal model:

[0006] Using professional 3D modeling software, through the built-in coordinate positioning function of the software, based on the Cartesian coordinate system, the vertices of the sheet metal are positioned in the form of absolute coordinate values. The coordinate positioning function can be expressed as P (x, y, z), where x, y, and z represent the position values ​​in the directions of the three coordinate axes respectively; combined with the dimensioning function, according to the sheet metal design drawings, the initial length L, width W, and thickness t dimensional parameters of the sheet metal are accurately entered to the millimeter level; during the input process, the accuracy of the size is ensured through the precision control mechanism provided by the software;

[0007] For the bending angle θ, use the angle measurement tool provided by the software based on the inverse tangent function To determine, where (x 1 ,y 1 ) and (x 2 ,y 2 ) is the coordinate value of the two points on the two-dimensional plane to determine the angle; in order to achieve accurate input to 0.1°, the software uses a high-precision numerical calculation method to accurately convert the calculation results of the inverse tangent function; for the bending radius R, combined with the design requirements, the coordinates of the relevant points are determined based on the Pythagorean theorem and complex geometric relationships; when determining the bending radius, a right triangle is constructed and the Pythagorean theorem is used to calculate the bending radius. 2 +b 2 =c 2, combined with known geometric dimensions and angle relationships, calculate the coordinates of the relevant points and finally enter them accurately to two decimal places;

[0008] After completing the parameter entry, use the model verification tool that comes with the software, based on the geometric shape matching algorithm; this algorithm extracts the shape context features of the model and compares the shape context feature vector of the entered model with the standard shape context feature library; the standard shape context feature library is established by extracting and analyzing features of a large number of standard sheet metal models; in the comparison process, calculate the similarity between the two, set the similarity threshold, if it is lower than the threshold, it is determined that the model has geometric shape or dimensional accuracy problems, and further compare it with the actual sheet metal design document to ensure that the model is highly consistent with the actual sheet metal features;

[0009] Step 2: Simulate bending and flattening calculation:

[0010] Adopt an algorithm based on geometric transformation and material mechanics principles; in geometric transformation, use the rigid body transformation matrix To describe the spatial position change of the sheet metal, R is the rotation matrix, which is used to describe the rotation angle and direction of the sheet metal, and t is the translation vector, which is used to describe the translation distance and direction of the sheet metal in space; combined with the elastic modulus E and yield strength σ of the sheet metal material y As performance parameters, these parameters are obtained through material testing experiments, and the experimental process follows the relevant material testing standards;

[0011] According to the set bending sequence, starting from the first bending, the shape of the sheet metal after each bending is calculated step by step according to the algorithm rules; the elastic deformation and plastic deformation process of the material under the bending force is simulated according to the law σ=Eε, where σ is stress and ε is strain; in the calculation process, the nonlinear characteristics of the material are considered, and the constitutive relationship equation of the material is introduced to simulate the deformation of the material more accurately;

[0012] During the calculation process, the sheet metal is divided into multiple small units, and the force analysis of each unit is performed using the node force balance equation. Among them, F i is the force acting on node i; the deformation of the unit is calculated by analyzing and calculating the node force and combining the mechanical properties of the material; at the same time, the deformation coordination equation is used to ensure the continuity and coordination of the deformation between adjacent units; detailed bending process data is generated, including the displacement u, stress σ, strain ε and other information of each unit. The displacement calculation is based on the virtual work principle ∫ V σ ij δε ij dV=∫ S T i δu i dS, where V is the unit volume, S is the unit surface area, and T iis the surface force, u i is the displacement; in the calculation process, the numerical integration method is used to solve the above integral equation;

[0013] Step 3: Real-time interference detection:

[0014] In the simulation flattening process, based on the spatial geometric relationship, a three-dimensional spatial coordinate system is constructed, and the position and posture of each part of the sheet metal in the coordinate system are determined by spatial vector operations; To determine the plane normal vector, is the unit vector of the coordinate axis, a x 、a y 、a z and b x , b y , b x Respectively, vector and Components on the three coordinate axes;

[0015] Step 4: Determine the criteria based on the preset boundary conditions

[0016] Set the minimum safe distance d between the parts of the sheet metal min , d min The value of is determined based on the material properties of the sheet metal, processing requirements, and actual application scenarios; Apply step three;

[0017] The collision detection algorithm based on the bounding box, AABB bounding box, its judgment condition is A and B are two bounding boxes. During the detection process, the spatial position of each tiny unit in the flattening process is analyzed in real time, and the detection efficiency is improved through the kd-tree fast spatial search algorithm.

[0018] Once it is detected that the distance d between the units is less than the minimum safety distance d min , if it is determined that there is interference, the three-dimensional coordinates of the interference position and the interference type, such as point interference, line interference or surface interference, are immediately recorded; when recording the interference position, a high-precision coordinate recording method is used to ensure the accuracy of the position information; the interference type is determined by analyzing the geometric characteristics of the interference area and combining the preset interference type judgment rules;

[0019] Step 5: Automatic optimization and adjustment of parameters:

[0020] When interference is detected, the system automatically starts the parameter adjustment mechanism according to the preset adjustment strategy library; the adjustment strategy library is established through the analysis and summary of a large amount of sheet metal processing experimental data combined with theoretical research results;

[0021] If the interference is caused by the bending angle, the bending angle is adjusted according to a certain proportion based on the degree of interference. The degree of interference is expressed by the interference volume V inter and the total volume of the sheet metal V total The adjustment formula is Where α is the adjustment coefficient, and the value of α is determined according to the material properties of the sheet metal, the interference type, and the actual processing experience; if it is a bending radius problem, the bending radius is adjusted in steps of 0.1mm-0.5mm according to the material properties and interference conditions, and the adjustment is based on the bending rebound formula of the material This formula takes into account the elastic modulus E, yield strength σ y , thickness t and original bending radius R old Effect on bending springback;

[0022] For the bending sequence, a heuristic search algorithm is used to select the optimal sequence from multiple possible bending sequence combinations. The cost function f(n) = g(n) + h(n), where g(n) is the actual cost from the starting point to node n, and h(n) is the estimated cost from node n to the target point. When calculating h(n), a heuristic function based on geometric distance and interference risk assessment is used to improve the search efficiency. For the flattening path, the path planning Dijkstra algorithm is used to replan according to the interference position and the overall shape of the sheet metal. The Dijkstra algorithm searches for the optimal path by continuously updating the shortest distance from the node to the source point. In the implementation process, data structures such as priority queues are used to optimize the time complexity of the algorithm.

[0023] The adjusted parameters are input into the simulation calculation module, and the simulation calculation is performed again, and the cycle is iterated until there is no interference in the simulation flattening result; during the iteration process, the parameter adjustment information and simulation calculation results of each iteration are recorded for analysis and optimization.

[0024] Furthermore, in the step of establishing an accurate sheet metal model, the model is also lightweighted by removing redundant geometric elements in the model, using a feature recognition algorithm to automatically identify and mark these redundant elements, and then deleting or simplifying them.

[0025] Furthermore, in the simulation bending and flattening calculation steps, adaptive meshing technology is used for complex sheet metal shapes; the density of the mesh is dynamically adjusted according to the curvature changes and stress concentration conditions of different parts of the sheet metal; in areas with large curvature or stress concentration, the mesh is automatically encrypted to improve the calculation accuracy; in relatively flat areas, the mesh density is appropriately relaxed to reduce the amount of calculation.

[0026] Furthermore, in the real-time interference detection step, in addition to the collision detection algorithm based on the bounding box, a fast detection method based on the spatial hash table is introduced; the three-dimensional space is divided into multiple uniform grid units and the spatial hash table is constructed.

[0027] Furthermore, in the step of automatic parameter optimization and adjustment, when multiple parameter adjustments still fail to eliminate interference, the expert system is activated; based on a large number of actual cases and domain knowledge, the expert system uses a reasoning engine to analyze interference situations and provide targeted solutions.

[0028] The present invention provides a sheet metal multiple bending and flattening interference system, which specifically includes the following steps:

[0029] Step 1: Model building module:

[0030] Equipped with professional 3D modeling tools, it supports the import of drawings in various industrial standard formats; through the format conversion interface, it uses a specific format conversion function library, which is implemented based on the syntax and semantic analysis of files in various formats; it accurately converts drawings in different formats into model data that can be recognized by the system. During the conversion process, the data is cleaned and verified to ensure the integrity and accuracy of the data; it provides an intuitive parameter entry interface, using text boxes, drop-down menus, sliders and other interactive components to facilitate users to enter various parameters; after the parameters are entered, the digital range verification function checkRange (value, min, max) is used to check whether the input value is within a reasonable range; at the same time, it provides multiple verification methods such as data type verification and unit verification to ensure the correctness of the input parameters;

[0031] It has the function of model verification and correction, through the model comparison algorithm based on curvature features; the algorithm extracts the curvature features of the input model and the standard model, constructs the curvature feature vector, and calculates the difference between the two; if a difference is found, the user is automatically prompted to make corrections, and both manual and automatic corrections are supported; the automatic correction adopts a correction strategy based on the optimization algorithm, and adjusts the model parameters by minimizing the model difference function to achieve automatic correction of the model;

[0032] Step 2: Simulate the calculation module:

[0033] Built-in high-performance computing engine, using multi-threaded parallel computing technology, and utilizing the computing resources of multi-core processors; through the thread creation function provided by the operating system, thread creation, scheduling and management are realized; in the multi-threaded computing process, synchronization technologies such as lock mechanism and semaphore mechanism are used to ensure data consistency and calculation accuracy; improve computing speed, and give full play to the performance advantages of multi-core processors through reasonable task division and load balancing strategies;

[0034] Integrate advanced bending and flattening simulation algorithms and combine them with genetic algorithms; in genetic algorithms, the roulette wheel selection function is used to select individuals according to their fitness values, and individuals with high fitness values ​​have a greater probability of being selected into the next generation; the single-point crossover function exchanges genes between selected individuals to generate new individuals; the basic bit mutation function randomly mutates the genes of individuals to increase the diversity of the population;

[0035] Generate visual simulation results of bending and flattening process, use professional graphics library and graphics drawing function to display the simulation process in 3D animation; in the display process, use lighting model, texture mapping and other technologies to enhance the visualization effect, and realize playback control through time control function;

[0036] Step 3: Interference detection module:

[0037] Based on advanced spatial analysis algorithms, the octree structure is constructed and operated through octree node creation functions, insertion functions, query functions, etc. When constructing the octree, the depth and node size of the octree are reasonably determined according to the geometric size and spatial range of the sheet metal model; combined with efficient calculation logic, all-round interference detection is performed on the sheet metal model during the simulation flattening process;

[0038] It can quickly and accurately identify various types of interference through feature extraction and pattern matching algorithms. The feature extraction algorithm analyzes the scale space of the image to extract key points and their feature descriptors; the template matching algorithm matches the extracted features with the preset interference template and distinguishes different types of interference based on the matching results.

[0039] Provide a detailed interference report, including the interference location, type, severity assessment and possible solutions. The severity assessment is quantitatively evaluated through the interference index function, which comprehensively considers the interference volume, area, location and other factors, and obtains the severity value of the interference through certain weight distribution and calculation rules. According to the severity value, provide corresponding solutions.

[0040] Step 4: Parameter adjustment module:

[0041] It has an intelligent parameter adjustment strategy library, which stores a variety of adjustment strategies for different interference situations. The strategies are established based on a large amount of experimental data and theoretical analysis. Through the strategy generation function, the strategies are generated according to the experimental data and theoretical models. When generating strategies, the decision tree algorithm is used to analyze and mine the experimental data to establish the mapping relationship between the interference situation and the adjustment strategy.

[0042] When interference is detected, the decision tree algorithm automatically selects the appropriate adjustment strategy from the strategy library according to the type and degree of interference; the decision tree algorithm classifies and makes decisions on interference situations by building a decision tree model;

[0043] The relevant parameters can be adjusted accurately, and the adjusted parameters can be fed back to the simulation calculation module in real time through the data interface. The system supports manual intervention and automatic optimization of the parameter adjustment process. Manual intervention is achieved through the user interface interactive function, and users can adjust the parameters according to their own experience and judgment. Automatic optimization is achieved through the optimization algorithm, and the simulation flattening results are optimized by continuously optimizing the parameters.

[0044] Furthermore, the model building module also has the function of interacting with an external database, and can obtain standard sheet metal model library and material property library information from the database.

[0045] Furthermore, during the calculation process, the simulation calculation module records the intermediate data and calculation status of the calculation process in real time.

[0046] The present invention provides a sheet metal multiple bending and flattening device, which specifically includes:

[0047] Software algorithm optimization:

[0048] Intelligent task scheduling system: In this system, the agent is the task scheduler, and the environment is the computing resource status and task queue of the device; the state space includes the task priority, resource requirements, execution progress and real-time status of system resources; the action space covers the task start, pause, continue and resource allocation adjustment;

[0049] Reward function design: In the reward function R = α × task completion rate + β × resource utilization rate + γ × computing efficiency improvement rate, the task completion rate refers to the ratio of the number of tasks successfully completed within a certain period of time to the total number of tasks; the resource utilization rate is measured by calculating the ratio of used resources to total resources; the computing efficiency improvement rate is determined by comparing the current task execution time with the historical average execution time; α, β, and γ are weight coefficients that can be flexibly adjusted according to the characteristics and needs of the actual sheet metal processing tasks; for sheet metal processing tasks with tight time, the weight of α can be appropriately increased to prioritize the completion of tasks on time; for equipment with limited resources, the weight of β can be increased to optimize resource utilization;

[0050] Learning and optimization process: The intelligent task scheduling system continuously interacts with the environment. Based on the reward feedback obtained for each action, the agent will refer to the Q value table when selecting an action in each state. The Q value table records the expected cumulative reward for each action performed in each state. After each interaction, the agent selects the action based on the actual reward and the maximum Q value of the next state according to the formula Where s is the current state, a is the current action, r is the reward, s′ is the next state, α is the learning rate, γ is the discount factor, and the Q value is updated to gradually learn the optimal scheduling strategy to achieve dynamic and accurate allocation of computing resources, ensuring that the entire computing process is efficient and stable;

[0051] Adaptive numerical solution algorithm: The finite element method discretizes the continuous solution domain into a combination of finite units, and obtains an approximate solution to the entire solution domain through equations; the error estimate of unit e is Where Ω e is the area of ​​unit e, σ h represents the stress of the finite element solution;

[0052] Grid adaptive adjustment: According to the error estimate, the grid is adaptively adjusted; when η e When the error exceeds the preset threshold, it indicates that the calculation error of the unit is large, and the unit and its adjacent units need to be meshed and further subdivided using the dichotomy method. e When it is much smaller than the error threshold, it means that the calculation accuracy of the unit is high and the grid density can be appropriately relaxed;

[0053] Intelligent error detection and repair module:

[0054] In this module, the prior probability P(C i ) is an estimate of the probability of input data errors, improper algorithm parameter settings, and rounding errors in the calculation process based on historical data and domain knowledge; likelihood probability (P(E|C i ) indicates that in some error reason C i The probability of data anomaly or calculation error event E in the event of anomaly or calculation error; the posterior probability P(C i |E) is after observing event E, the error cause C i Probability update of ;

[0055] Model construction and parameter determination: Through learning and analysis of massive normal data, a normal state model of data and calculation results is constructed; for the input sheet metal model data, its reasonable value range, the correlation between data and other parameters are determined; for the calculation results, the normal stress and strain ranges are determined; these parameters are used as prior knowledge for subsequent error diagnosis;

[0056] Error diagnosis and repair: When data anomalies or calculation errors are detected, use the formula Calculate the posterior probabilities of different error causes; find the error cause with the highest posterior probability and automatically try to fix it; if the input data is wrong, correct it according to the normal range and correlation of the data; if the algorithm parameters are set improperly, readjust the parameters according to historical experience and optimization algorithms, and then recalculate to ensure the reliability of the calculation results;

[0057] Data management and interaction:

[0058] Efficient data storage and indexing system: When data needs to be stored, the key value of the data is first calculated to obtain a value, and then the key value is searched clockwise on the ring to find the first node that is greater than or equal to the value, and the data is stored in the node;

[0059] B+ tree index; in a B+ tree, the time complexity of insert, delete, and query operations is O(log n , m), where n is the branching factor of the node, that is, the maximum number of child nodes each node contains, and m is the number of data;

[0060] Data communication mechanism: Communicate with the back-end server with the help of AJAX technology; AJAX allows asynchronous data exchange with the server without reloading the entire page; by creating an XMLHttpRequest object, setting the request URL, request method, request header and request body, the user input data is sent to the back-end server; after the server processes the request, it returns the response data, and the front-end updates the page content by parsing the response data.

[0061] The present invention provides a sheet metal multiple bending and flattening storage medium, which specifically includes:

[0062] Data Storage and Security:

[0063] Advanced encryption storage technology:

[0064] When encrypting, first divide the plaintext into groups, each group is 128 bits long; the plaintext group is P i , the previous ciphertext is C i-1 , the encryption key is K, then the i-th group of ciphertext C i The calculation method is Where E K is the encryption function, The encryption function encrypts the input data through a series of byte replacement, row shift, column mixing and key addition operations; when the data is written to the storage medium, the data is automatically encrypted and stored;

[0065] Key management: To ensure the security of keys, a key function is used to derive multiple subkeys based on the master key input by the user, which are used for different data encryption and decryption operations. At the same time, a key encryption key mechanism is used to encrypt and store the derived subkeys;

[0066] Data redundancy and recovery mechanism:

[0067] The original data block is D 1 , D 2 , …, D k , generate nk redundant check blocks P by encoding 1 , P 2 ,…,P n-k , where n is the total number of data blocks, the original data blocks can be regarded as the coefficients of the polynomial, and redundant check blocks can be generated through a specific encoding algorithm; when some data blocks are lost or damaged; assuming that the index of the lost data block is i 1 ,i 2 ,…,i m , then the restored data block It can be expressed as Where D l It is the data that is not lost;

[0068] Data integrity check: Perform integrity check on stored data regularly, use hash algorithm to calculate the hash value of data blocks, and compare the calculated hash value with the hash value pre-saved in the storage medium. If the two are consistent, the data is complete; if they are inconsistent, it means that the data may be tampered with or damaged, and it is repaired or backed up in time;

[0069] Data transfer and compatibility:

[0070] High-speed data transmission protocol:

[0071] A custom transmission protocol based on UDP is used. On the basis of UDP, data verification, retransmission mechanism and flow control functions are added. The data verification adopts the CRC algorithm to generate a check code attached to the data frame. The CRC algorithm obtains a check code of fixed length by performing polynomial division operation on the data frame. After receiving the data frame, the receiver also performs CRC calculation and compares the calculated check code with the received check code. If they are inconsistent, it means that an error occurs during the data transmission process and a retransmission request is sent.

[0072] Retransmission mechanism and flow control: The retransmission mechanism uses a combination of timeout retransmission and fast retransmission. After sending a data frame, the sender starts a timer. If no confirmation response is received from the receiver before the timer times out, the data frame is retransmitted. If multiple repeated ACKs are received in a short period of time, it means that data frames may be lost. The fast retransmission mechanism is used to immediately retransmit the lost data frames. Flow control is achieved through a sliding window mechanism. The sender dynamically adjusts the sending window size according to the window size fed back by the receiver to ensure stable data transmission and avoid overflow of the receiver's buffer due to the sender sending data too quickly. By optimizing the encoding and decoding algorithms and the data transmission link, the actual data transmission speed is increased by 3-5 times compared with the traditional transmission protocol.

[0073] Good compatibility:

[0074] Cross-platform driver development: By writing cross-platform device drivers and interface libraries, seamless integration with various types of devices and operating systems can be achieved; for different operating systems, corresponding development frameworks are used to write drivers; for Windows systems, Windows Driver Kit is used to write device drivers, which communicate with devices through interfaces such as USB and Ethernet; WDK provides a series of tools and libraries to help developers write drivers that comply with Windows operating system specifications and implement functions such as device identification, initialization, and data transmission; for Linux systems, driver development frameworks provided by the Linux kernel are used to write drivers, such as the character device driver framework and the block device driver framework; the Linux kernel provides a wealth of functions and interfaces to facilitate developers to implement various functions of device drivers; for macOS systems, the I / OKit framework provided by the system is used to develop drivers; the I / OKit framework provides an object-oriented programming interface for developing various types of device drivers to achieve communication and control with devices; these measures make it convenient for users to use in different working environments.

[0075] Beneficial effects:

[0076] 1. Improve flattening accuracy and ensure product quality:

[0077] Accurate parameter acquisition: In the stage of establishing an accurate sheet metal model, professional 3D modeling software is used to locate the absolute coordinate values ​​of each vertex of the sheet metal based on the Cartesian coordinate system and the high-precision coordinate positioning function. The accuracy can reach ±0.01mm, and the initial length, width, and thickness dimension parameters of the sheet metal can be accurately entered to the millimeter level. For the bending angle, the software's built-in angle measurement tool is used, based on the inverse tangent function and combined with a high-precision numerical calculation method, to achieve an entry accurate to 0.1°; for the bending radius, a right triangle is constructed based on the Pythagorean theorem and complex geometric relationships for calculation, and finally the entry is accurate to two decimal places.

[0078] Advanced algorithm simulation: In the process of simulating bending and flattening calculations, an algorithm based on geometric transformation and material mechanics is used, combined with material performance parameters such as elastic modulus and yield strength, and considering the nonlinear characteristics of the material, and the material constitutive relationship equation is introduced to simulate the deformation of the material more accurately. By dividing the sheet metal into tiny units, using the node force balance equation and deformation coordination equation, it is ensured that the calculated deformation of each unit is accurate and the deformation of adjacent units is continuous and coordinated.

[0079] The actual effect is remarkable: after actual testing, the flattening accuracy of sheet metal using the method of the present invention is improved by 30% compared with the traditional method, so that the dimensional deviation of the final product can be stably controlled within a very small range of ±0.1mm. This greatly improves the qualified rate of products. In the actual application of a certain automobile manufacturing company, the qualified rate of products has been increased from the original 80% to 95%, which greatly enhances the stability of product quality, reduces the scrap rate caused by interference, and reduces production costs.

[0080] 2. Simplify the calculation process and improve calculation efficiency:

[0081] Algorithm optimization: The specific algorithms used in this invention, such as in the simulation of bending and flattening calculations, make full use of the computing resources of multi-core processors through reasonable task division and load balancing strategies combined with multi-threaded parallel computing technology. In the intelligent task scheduling system, by designing a reasonable reward function and a learning optimization process based on the Q value table, dynamic and accurate allocation of computing resources is achieved, effectively reducing unnecessary calculation steps and complex mathematical operations.

[0082] The calculation time is greatly shortened: compared with the traditional method, the calculation time is shortened by 50%. For example, when processing a complex sheet metal simulation calculation, the traditional method takes 8 hours, while the method of the present invention only takes 4 hours, which greatly shortens the preparation time before processing, enabling enterprises to respond to production needs more quickly, improve production efficiency, and enhance market competitiveness.

[0083] 3. Adapt to complex working conditions and expand the scope of application:

[0084] Strong adaptability: The method and system can handle various complex sheet metal multiple bending situations. Whether it is complex sheet metal parts with a variety of different bending angles (such as in the aerospace field, the bending angles of some sheet metal parts range from 5° to 175°), radii (such as small sheet metal parts in electronic equipment, the bending radius is as small as 0.5mm), or sheet metal processing with special shapes (such as special-shaped heat sinks) and process requirements (such as bending processing of ultra-high strength materials), the interference problem can be accurately simulated and effectively solved through precise model establishment, simulation calculation and interference detection.

[0085] Wide range of industry applications: It can be widely used in industries such as aerospace, automobile manufacturing, and electronic equipment that require extremely high precision in sheet metal processing. In the aerospace field, it can meet the complex sheet metal processing needs of aircraft fuselage structural parts to ensure the safety and reliability of the aircraft; in automobile manufacturing, it can be used to manufacture key components such as automobile engine hoods and doors to improve the overall performance and appearance quality of the car; in electronic equipment manufacturing, it can meet the requirements of miniaturization and high-precision sheet metal processing, such as internal structural parts of mobile phones, computers and other equipment. It provides reliable technical support for production in different fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the drawings of the embodiment are briefly introduced below.

[0087] The drawings described below are only related to some embodiments of the present invention, but are not intended to limit the present invention.

[0088] In the attached picture:

[0089] Figure 1 It is a schematic flow chart of a sheet metal multiple bending and flattening interference method according to an embodiment of the present invention.

[0090] Figure 2 It is a system diagram of a sheet metal multiple bending and flattening system according to an embodiment of the present invention.

[0091] Figure 3 This is a diagram of the use of a sheet metal multiple bending and flattening device according to an embodiment of the present invention.

[0092] Figure 4 It is a schematic diagram of a storage medium process for flattening sheet metal multiple times according to an embodiment of the present invention. DETAILED DESCRIPTION

[0093] The embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings and examples.

[0094] Example: Please refer to Figures 1 to 4 As shown:

[0095] A sheet metal multiple bending and flattening interference method specifically comprises the following steps:

[0096] Step 1: Create an accurate sheet metal model:

[0097] Using professional 3D modeling software, through the built-in coordinate positioning function of the software, based on the Cartesian coordinate system, the vertices of the sheet metal are positioned in the form of absolute coordinate values. The coordinate positioning function can be expressed as P (x, y, z), where x, y, and z represent the position values ​​in the directions of the three coordinate axes respectively; combined with the dimensioning function, according to the sheet metal design drawings, the initial length L, width W, and thickness t dimensional parameters of the sheet metal are accurately entered to the millimeter level; during the input process, the accuracy of the size is ensured through the precision control mechanism provided by the software;

[0098] For the bending angle θ, use the angle measurement tool provided by the software based on the inverse tangent function To determine, where (x 1 ,y 1 ) and (x 2 ,y 2 ) is the coordinate value of the two points on the two-dimensional plane to determine the angle; in order to achieve accurate input to 0.1°, the software uses a high-precision numerical calculation method to accurately convert the calculation results of the inverse tangent function; for the bending radius R, combined with the design requirements, the coordinates of the relevant points are determined based on the Pythagorean theorem and complex geometric relationships; when determining the bending radius, a right triangle is constructed and the Pythagorean theorem is used to calculate the bending radius. 2 +b 2 =c 2 , combined with known geometric dimensions and angle relationships, calculate the coordinates of the relevant points and finally enter them accurately to two decimal places;

[0099] After completing the parameter entry, use the model verification tool that comes with the software, based on the geometric shape matching algorithm; this algorithm extracts the shape context features of the model and compares the shape context feature vector of the entered model with the standard shape context feature library; the standard shape context feature library is established by extracting and analyzing features of a large number of standard sheet metal models; in the comparison process, calculate the similarity between the two, set the similarity threshold, if it is lower than the threshold, it is determined that the model has geometric shape or dimensional accuracy problems, and further compare it with the actual sheet metal design document to ensure that the model is highly consistent with the actual sheet metal features;

[0100] Step 2: Simulate bending and flattening calculation:

[0101] Adopt an algorithm based on geometric transformation and material mechanics principles; in geometric transformation, use the rigid body transformation matrix To describe the spatial position change of the sheet metal, R is the rotation matrix, which is used to describe the rotation angle and direction of the sheet metal, and t is the translation vector, which is used to describe the translation distance and direction of the sheet metal in space; combined with the elastic modulus E and yield strength σ of the sheet metal material y As performance parameters, these parameters are obtained through material testing experiments, and the experimental process follows the relevant material testing standards;

[0102] According to the set bending sequence, starting from the first bending, the shape of the sheet metal after each bending is calculated step by step according to the algorithm rules; the elastic deformation and plastic deformation process of the material under the bending force is simulated according to the law σ=Eε, where σ is stress and ε is strain; in the calculation process, the nonlinear characteristics of the material are considered, and the constitutive relationship equation of the material is introduced to simulate the deformation of the material more accurately;

[0103] During the calculation process, the sheet metal is divided into multiple small units, and the force analysis of each unit is performed using the node force balance equation. Among them, F i is the force acting on node i; the deformation of the unit is calculated by analyzing and calculating the node force and combining the mechanical properties of the material; at the same time, the deformation coordination equation is used to ensure the continuity and coordination of the deformation between adjacent units; detailed bending process data is generated, including the displacement u, stress σ, strain ε and other information of each unit. The displacement calculation is based on the virtual work principle ∫ V σ ij δε ij dV=∫ S T i δu i dS, where V is the unit volume, S is the unit surface area, and T i is the surface force, u i is the displacement; in the calculation process, the numerical integration method is used to solve the above integral equation;

[0104] Step 3: Real-time interference detection:

[0105] In the simulation flattening process, based on the spatial geometric relationship, a three-dimensional spatial coordinate system is constructed, and the position and posture of each part of the sheet metal in the coordinate system are determined by spatial vector operations; To determine the plane normal vector, is the unit vector of the coordinate axis, a x 、a y 、a ∑ and b x 、b y 、b z The vectors are and Components on the three coordinate axes;

[0106] Step 4: Determine the criteria based on the preset boundary conditions

[0107] Set the minimum safe distance d between the parts of the sheet metal min , d min The value of is determined based on the material properties of the sheet metal, processing requirements, and actual application scenarios; Apply step three;

[0108] The collision detection algorithm based on the bounding box, AABB bounding box, its judgment condition is A and B are two bounding boxes. During the detection process, the spatial position of each tiny unit in the flattening process is analyzed in real time, and the detection efficiency is improved through the kd-tree fast spatial search algorithm.

[0109] Once it is detected that the distance d between the units is less than the minimum safety distance d min , if it is determined that there is interference, the three-dimensional coordinates of the interference position and the interference type, such as point interference, line interference or surface interference, are immediately recorded; when recording the interference position, a high-precision coordinate recording method is used to ensure the accuracy of the position information; the interference type is determined by analyzing the geometric characteristics of the interference area and combining the preset interference type judgment rules;

[0110] Step 5: Automatic optimization and adjustment of parameters:

[0111] When interference is detected, the system automatically starts the parameter adjustment mechanism according to the preset adjustment strategy library; the adjustment strategy library is established through the analysis and summary of a large amount of sheet metal processing experimental data combined with theoretical research results;

[0112] If the interference is caused by the bending angle, the bending angle is adjusted according to a certain proportion based on the degree of interference. The degree of interference is expressed by the interference volume V inter and the total volume of the sheet metal V total The adjustment formula is Where α is the adjustment coefficient, and the value of α is determined according to the material properties of the sheet metal, the interference type, and the actual processing experience; if it is a bending radius problem, the bending radius is adjusted in steps of 0.1mm-0.5mm according to the material properties and interference conditions, and the adjustment is based on the bending rebound formula of the material This formula takes into account the elastic modulus E, yield strength σ y , thickness t and original bending radius R old Effect on bending springback;

[0113] For the bending sequence, a heuristic search algorithm is used to select the optimal sequence from multiple possible bending sequence combinations. The cost function f(n) = g(n) + h(n), where g(n) is the actual cost from the starting point to node n, and h(n) is the estimated cost from node n to the target point. When calculating h(n), a heuristic function based on geometric distance and interference risk assessment is used to improve the search efficiency. For the flattening path, the path planning Dijkstra algorithm is used to replan according to the interference position and the overall shape of the sheet metal. The Dijkstra algorithm searches for the optimal path by continuously updating the shortest distance from the node to the source point. In the implementation process, data structures such as priority queues are used to optimize the time complexity of the algorithm.

[0114] The adjusted parameters are input into the simulation calculation module, and the simulation calculation is performed again, and the cycle is iterated until there is no interference in the simulation flattening result; during the iteration process, the parameter adjustment information and simulation calculation results of each iteration are recorded for analysis and optimization.

[0115] Among them, the step of establishing a precise sheet metal model also includes lightweight processing of the model, by removing redundant geometric elements in the model, using feature recognition algorithms to automatically identify and mark these redundant elements, and then using deletion or simplification operations.

[0116] Algorithms derived from feature recognition algorithms are often used to more accurately and efficiently process sheet metal model-related tasks. The algorithms include the following from different stages and requirements of model processing:

[0117] Semantic segmentation algorithm based on deep learning: In sheet metal models, traditional feature recognition algorithms have limitations in judging tiny features of complex structures. Semantic segmentation algorithms, such as U-Net and DeepLab series algorithms, can classify three-dimensional images of sheet metal models pixel by pixel. Taking U-Net as an example, it uses an encoder-decoder structure to first downsample the input model image to extract features, then restore the image resolution through upsampling, and combines jump connections to fuse features of different scales, thereby accurately identifying various features in the model, including tiny transition fillets, small features, etc., providing a more accurate basis for subsequent lightweight processing.

[0118] Topology optimization algorithm: When lightweighting sheet metal models, it is not enough for feature recognition algorithms to only identify redundant elements. Topology optimization algorithms can further optimize the model structure. Taking the variable density method as an example, it introduces density variables to describe the distribution of model materials. Based on finite element analysis and optimization criteria, it continuously iterates and updates the material distribution. Under the premise of ensuring the mechanical properties of the model, it removes unnecessary material areas to achieve a better lightweight effect and ensure the structural stability and functionality of the model.

[0119] Hierarchical clustering algorithm based on feature association: In the process of establishing the standard shape context feature library and model verification, the features identified by the feature recognition algorithm need to be organized and compared more effectively. The hierarchical clustering algorithm can hierarchically cluster the features of the sheet metal model according to the similarity between the features. For example, using agglomerative hierarchical clustering, starting with each feature as a separate class, gradually merge classes with high similarity to form a clustering result with a tree structure. In this way, similar features can be quickly located and compared during model verification, improving verification efficiency and accuracy.

[0120] Multi-scale feature fusion algorithm: Sheet metal models have features of different scales. In order to fully and accurately identify these features, multi-scale feature fusion algorithms have emerged. Taking FPN (Feature Pyramid Network) as an example, it fuses convolutional feature maps of different scales through top-down and lateral connections. When identifying sheet metal model features, it can obtain high-resolution detail features and use low-resolution semantic information, thereby improving the recognition ability of features of various sizes, especially for the recognition of features of different scales in complex sheet metal models.

[0121] Among them, in the simulation bending and flattening calculation steps, adaptive meshing technology is used for complex sheet metal shapes; the density of the mesh is dynamically adjusted according to the curvature changes and stress concentration in different parts of the sheet metal; in areas with large curvature or stress concentration, the mesh is automatically encrypted to improve the calculation accuracy; in relatively flat areas, the mesh density is appropriately relaxed to reduce the amount of calculation.

[0122] Among them, in the real-time interference detection step, in addition to the collision detection algorithm based on the bounding box, a fast detection method based on the spatial hash table is also introduced; the three-dimensional space is divided into multiple uniform grid units to construct a spatial hash table.

[0123] Among them, in the parameter automatic optimization and adjustment step, when multiple adjustments to the parameters still cannot eliminate the interference, the expert system is started; based on a large number of actual cases and domain knowledge, the expert system uses the reasoning engine to analyze the interference situation and provide targeted solutions.

[0124] A sheet metal multiple bending and flattening interference system specifically comprises the following steps:

[0125] Step 1: Model building module:

[0126] Equipped with professional 3D modeling tools, it supports the import of drawings in various industrial standard formats; through the format conversion interface, it uses a specific format conversion function library, which is implemented based on the syntax and semantic analysis of files in various formats; it accurately converts drawings in different formats into model data that can be recognized by the system. During the conversion process, the data is cleaned and verified to ensure the integrity and accuracy of the data; it provides an intuitive parameter entry interface, using text boxes, drop-down menus, sliders and other interactive components to facilitate users to enter various parameters; after the parameters are entered, the digital range verification function checkRange (value, min, max) is used to check whether the input value is within a reasonable range; at the same time, it provides multiple verification methods such as data type verification and unit verification to ensure the correctness of the input parameters;

[0127] It has the function of model verification and correction, through the model comparison algorithm based on curvature features; the algorithm extracts the curvature features of the input model and the standard model, constructs the curvature feature vector, and calculates the difference between the two; if a difference is found, the user is automatically prompted to make corrections, and both manual and automatic corrections are supported; the automatic correction adopts a correction strategy based on the optimization algorithm, and adjusts the model parameters by minimizing the model difference function to achieve automatic correction of the model;

[0128] Step 2: Simulate the calculation module:

[0129] Built-in high-performance computing engine, using multi-threaded parallel computing technology, and utilizing the computing resources of multi-core processors; through the thread creation function provided by the operating system, thread creation, scheduling and management are realized; in the multi-threaded computing process, synchronization technologies such as lock mechanism and semaphore mechanism are used to ensure data consistency and calculation accuracy; improve computing speed, and give full play to the performance advantages of multi-core processors through reasonable task division and load balancing strategies;

[0130] Integrate advanced bending and flattening simulation algorithms and combine them with genetic algorithms; in genetic algorithms, the roulette wheel selection function is used to select individuals according to their fitness values, and individuals with high fitness values ​​have a greater probability of being selected into the next generation; the single-point crossover function exchanges genes between selected individuals to generate new individuals; the basic bit mutation function randomly mutates the genes of individuals to increase the diversity of the population;

[0131] Generate visual simulation results of bending and flattening process, use professional graphics library and graphics drawing function to display the simulation process in 3D animation; in the display process, use lighting model, texture mapping and other technologies to enhance the visualization effect, and realize playback control through time control function;

[0132] Step 3: Interference detection module:

[0133] Based on advanced spatial analysis algorithms, the octree structure is constructed and operated through octree node creation functions, insertion functions, query functions, etc. When constructing the octree, the depth and node size of the octree are reasonably determined according to the geometric size and spatial range of the sheet metal model; combined with efficient calculation logic, all-round interference detection is performed on the sheet metal model during the simulation flattening process;

[0134] It can quickly and accurately identify various types of interference through feature extraction and pattern matching algorithms. The feature extraction algorithm analyzes the scale space of the image to extract key points and their feature descriptors; the template matching algorithm matches the extracted features with the preset interference template and distinguishes different types of interference based on the matching results.

[0135] Provide a detailed interference report, including the interference location, type, severity assessment and possible solutions. The severity assessment is quantitatively evaluated through the interference index function, which comprehensively considers the interference volume, area, location and other factors, and obtains the severity value of the interference through certain weight distribution and calculation rules. According to the severity value, provide corresponding solutions.

[0136] Step 4: Parameter adjustment module:

[0137] It has an intelligent parameter adjustment strategy library, which stores a variety of adjustment strategies for different interference situations. The strategies are established based on a large amount of experimental data and theoretical analysis. Through the strategy generation function, the strategies are generated according to the experimental data and theoretical models. When generating strategies, the decision tree algorithm is used to analyze and mine the experimental data to establish the mapping relationship between the interference situation and the adjustment strategy.

[0138] When interference is detected, the decision tree algorithm automatically selects the appropriate adjustment strategy from the strategy library according to the type and degree of interference; the decision tree algorithm classifies and makes decisions on interference situations by building a decision tree model;

[0139] The relevant parameters can be adjusted accurately, and the adjusted parameters can be fed back to the simulation calculation module in real time through the data interface. The system supports manual intervention and automatic optimization of the parameter adjustment process. Manual intervention is achieved through the user interface interactive function, and users can adjust the parameters according to their own experience and judgment. Automatic optimization is achieved through the optimization algorithm, and the simulation flattening results are optimized by continuously optimizing the parameters.

[0140] Among them, the model building module also has the function of interacting with external databases, and can obtain standard sheet metal model library and material property library information from the database.

[0141] Among them, the simulation calculation module records the intermediate data and calculation status of the calculation process in real time during the calculation process.

[0142] A sheet metal multiple bending and flattening device, specifically comprising:

[0143] Software algorithm optimization:

[0144] Intelligent task scheduling system: In this system, the agent is the task scheduler, and the environment is the computing resource status and task queue of the device; the state space includes the task priority, resource requirements, execution progress and real-time status of system resources; the action space covers the task start, pause, continue and resource allocation adjustment;

[0145] Reward function design: In the reward function R = α × task completion rate + β × resource utilization rate + γ × computing efficiency improvement rate, the task completion rate refers to the ratio of the number of tasks successfully completed within a certain period of time to the total number of tasks; the resource utilization rate is measured by calculating the ratio of used resources to total resources; the computing efficiency improvement rate is determined by comparing the current task execution time with the historical average execution time; α, β, and γ are weight coefficients that can be flexibly adjusted according to the characteristics and needs of the actual sheet metal processing tasks; for sheet metal processing tasks with tight time, the weight of α can be appropriately increased to prioritize the completion of tasks on time; for equipment with limited resources, the weight of β can be increased to optimize resource utilization;

[0146] Learning and optimization process: The intelligent task scheduling system continuously interacts with the environment. Based on the reward feedback obtained for each action, the agent will refer to the Q value table when selecting an action in each state. The Q value table records the expected cumulative reward for each action performed in each state. After each interaction, the agent selects the action based on the actual reward and the maximum Q value of the next state according to the formula Where s is the current state, a is the current action, r is the reward, s′ is the next state, α is the learning rate, γ is the discount factor, and the Q value is updated to gradually learn the optimal scheduling strategy to achieve dynamic and accurate allocation of computing resources, ensuring that the entire computing process is efficient and stable;

[0147] Adaptive numerical solution algorithm: The finite element method discretizes the continuous solution domain into a combination of finite units, and obtains an approximate solution to the entire solution domain through equations; the error estimate of unit e is Where Ω e is the area of ​​unit e, σ h represents the stress of the finite element solution;

[0148] Grid adaptive adjustment: According to the error estimate, the grid is adaptively adjusted; when η e When the error exceeds the preset threshold, it indicates that the calculation error of the unit is large, and the unit and its adjacent units need to be meshed and further subdivided using the dichotomy method. eWhen it is much smaller than the error threshold, it means that the calculation accuracy of the unit is high and the grid density can be appropriately relaxed;

[0149] Intelligent error detection and repair module:

[0150] In this module, the prior probability P(C i ) is an estimate of the probability of input data errors, improper algorithm parameter settings, and rounding errors in the calculation process based on historical data and domain knowledge; likelihood probability (P(E|C i ) indicates that in some error reason C i The probability of data anomaly or calculation error event E in the event of anomaly or calculation error; the posterior probability P(C i |E) is after observing event E, the error cause C i Probability update of ;

[0151] Model construction and parameter determination: Through learning and analysis of massive normal data, a normal state model of data and calculation results is constructed; for the input sheet metal model data, its reasonable value range, the correlation between data and other parameters are determined; for the calculation results, the normal stress and strain ranges are determined; these parameters are used as prior knowledge for subsequent error diagnosis;

[0152] Error diagnosis and repair: When data anomalies or calculation errors are detected, use the formula Calculate the posterior probabilities of different error causes; find the error cause with the highest posterior probability and automatically try to fix it; if the input data is wrong, correct it according to the normal range and correlation of the data; if the algorithm parameters are set improperly, readjust the parameters according to historical experience and optimization algorithms, and then recalculate to ensure the reliability of the calculation results;

[0153] Data management and interaction:

[0154] Efficient data storage and indexing system:

[0155] When data needs to be stored, the key value of the data is calculated first to get a value, then search clockwise on the ring to find the first node that is greater than or equal to the value, and store the data in the node;

[0156] B+ tree index; in a B+ tree, the time complexity of insert, delete, and query operations is O(log n m), where n is the branching factor of the node, that is, the maximum number of child nodes each node contains, and m is the number of data;

[0157] Data communication mechanism: Communicate with the back-end server with the help of AJAX technology; AJAX allows asynchronous data exchange with the server without reloading the entire page; by creating an XMLHttpRequest object, setting the request URL, request method, request header and request body, the user input data is sent to the back-end server; after the server processes the request, it returns the response data, and the front-end updates the page content by parsing the response data.

[0158] A sheet metal multiple bending and flattening storage medium, specifically comprising:

[0159] Data Storage and Security:

[0160] Advanced encryption storage technology:

[0161] When encrypting, first divide the plaintext into groups, each group is 128 bits long; the plaintext group is P i , the previous ciphertext is C i-1 , the encryption key is K, then the i-th group of ciphertext C i The calculation method is Where E K is the encryption function, The encryption function encrypts the input data through a series of byte replacement, row shift, column mixing and key addition operations; when the data is written to the storage medium, the data is automatically encrypted and stored;

[0162] Key management: To ensure the security of keys, a key function is used to derive multiple subkeys based on the master key input by the user, which are used for different data encryption and decryption operations. At the same time, a key encryption key mechanism is used to encrypt and store the derived subkeys;

[0163] Data redundancy and recovery mechanism:

[0164] The original data block is D 1 , D 2 , …, D k , generate nk redundant check blocks P by encoding 1 , P 2 ,…,P n-k , where n is the total number of data blocks, the original data blocks can be regarded as the coefficients of the polynomial, and redundant check blocks can be generated through a specific encoding algorithm; when some data blocks are lost or damaged; assuming that the index of the lost data block is i 1 ,i 2 ,…,i m , then the restored data block It can be expressed as Where D l It is the data that is not lost;

[0165] Data integrity check: Perform integrity check on stored data regularly, use hash algorithm to calculate the hash value of data blocks, and compare the calculated hash value with the hash value pre-saved in the storage medium. If the two are consistent, the data is complete; if they are inconsistent, it means that the data may be tampered with or damaged, and it is repaired or backed up in time;

[0166] Data transfer and compatibility:

[0167] High-speed data transmission protocol:

[0168] A custom transmission protocol based on UDP is used. On the basis of UDP, data verification, retransmission mechanism and flow control functions are added. The data verification adopts the CRC algorithm to generate a check code attached to the data frame. The CRC algorithm obtains a check code of fixed length by performing polynomial division operation on the data frame. After receiving the data frame, the receiver also performs CRC calculation and compares the calculated check code with the received check code. If they are inconsistent, it means that an error occurs during the data transmission process and a retransmission request is sent.

[0169] Retransmission mechanism and flow control: The retransmission mechanism uses a combination of timeout retransmission and fast retransmission. After sending a data frame, the sender starts a timer. If no confirmation response is received from the receiver before the timer times out, the data frame is retransmitted. If multiple repeated ACKs are received in a short period of time, it means that data frames may be lost. The fast retransmission mechanism is used to immediately retransmit the lost data frames. Flow control is achieved through a sliding window mechanism. The sender dynamically adjusts the sending window size according to the window size fed back by the receiver to ensure stable data transmission and avoid overflow of the receiver's buffer due to the sender sending data too quickly. By optimizing the encoding and decoding algorithms and the data transmission link, the actual data transmission speed is increased by 3-5 times compared with the traditional transmission protocol.

[0170] Good compatibility:

[0171] Cross-platform driver development: By writing cross-platform device drivers and interface libraries, seamless integration with various types of devices and operating systems can be achieved; for different operating systems, corresponding development frameworks are used to write drivers; for Windows systems, Windows Driver Kit is used to write device drivers, which communicate with devices through interfaces such as USB and Ethernet; WDK provides a series of tools and libraries to help developers write drivers that comply with Windows operating system specifications and implement functions such as device identification, initialization, and data transmission; for Linux systems, driver development frameworks provided by the Linux kernel are used to write drivers, such as the character device driver framework and the block device driver framework; the Linux kernel provides a wealth of functions and interfaces to facilitate developers to implement various functions of device drivers; for macOS systems, the I / OKit framework provided by the system is used to develop drivers; the I / OKit framework provides an object-oriented programming interface for developing various types of device drivers to achieve communication and control with devices; these measures make it convenient for users to use in different working environments.

[0172] 1. Establish precise sheet metal model experiment

[0173] Experimental preparation: Collect 10 representative sheet metal drawings from the fields of automobile manufacturing, aerospace, electronic equipment manufacturing, etc., covering simple regular shapes (such as rectangular plates, cylindrical tubes), medium complexity (such as automobile door inner panels, local sheet metal parts of aircraft engine air intakes), and complex special-shaped structures (such as complex bending parts of aircraft wing skins, multi-curved shapes of automobile engine hoods). Prepare professional 3D modeling software, such as SolidWorks, CATIA, etc., as well as traditional modeling tools.

[0174] Experimental steps:

[0175] New modeling method: Using professional 3D modeling software, based on the Cartesian coordinate system, the software's built-in coordinate positioning function is used to accurately locate each vertex of the sheet metal with absolute coordinate values. Combined with the dimensioning function, according to the high-precision sheet metal design drawings, the length, width, thickness and other dimensional parameters are accurately entered to the millimeter level. For the bending angle, with the help of the software's built-in angle measurement tool, it is determined based on the inverse tangent function, and the software uses a high-precision numerical calculation method to achieve accurate entry to 0.1°. For the bending radius, the coordinates of the relevant points are determined based on the Pythagorean theorem and complex geometric relationships, and are entered accurately to two decimal places. After completing the parameter entry, use the model verification tool based on the geometric shape matching algorithm (such as the shape context matching algorithm) to compare with the standard shape context feature library. At the same time, the feature recognition algorithm is used to automatically identify and mark redundant geometric elements in the model, and lightweight processing is achieved by deleting or simplifying operations.

[0176] Traditional modeling: Use traditional modeling tools to manually draw sketches, build models through basic operations such as stretching and rotation, and rely on manual experience for dimensioning and parameter setting. There is a lack of precise verification and automated lightweight processing steps.

[0177] Experimental results:

[0178] Compare Projects New approach Traditional methods Average modeling time for simple models (minutes) 3.5 6.0 Average modeling time for a moderately complex model (minutes) 5.0 8.5 Average modeling time for complex models (minutes) 7.0 12.0 Simple model accuracy (deviation from actual size, mm) ±0.08 ±0.3 Moderately complex model accuracy (deviation from actual size, mm) ±0.1 ±0.4 Complex model accuracy (deviation from actual size, mm) ±0.12 ±0.5 Simple model data size (MB) 8.0 15.0 Medium-complexity model data size (MB) 12.0 20.0 Complex model data volume (MB) 18.0 30.0 Model verification pass rate (%) 95 70

[0179] 2. Simulation of bending and flattening calculation experiment

[0180] Experimental preparation: Based on the 10 sheet metal models that have been built, for different material types (such as aluminum alloy, stainless steel, and ordinary carbon steel), obtain the material properties of each model, such as elastic modulus, yield strength, etc., through professional material testing equipment and material testing experiments that follow relevant standards. Prepare high-performance computers equipped with computing software related to the new method and traditional computing software.

[0181] Experimental steps:

[0182] New method calculation: An improved algorithm based on geometric transformation and material mechanics is used. In geometric transformation, the rigid body transformation matrix is ​​used to describe the spatial position change of the sheet metal, in which the rotation matrix accurately describes the rotation angle and direction, and the translation vector accurately describes the translation distance and direction. Combined with the material performance parameters, according to the set bending sequence, starting from the first bending, the sheet metal shape after each bending is calculated step by step according to the algorithm rules. The elastic deformation and plastic deformation process of the material under the bending force is simulated according to the law, and the nonlinear characteristics of the material are considered at the same time, and the elastic-plastic constitutive relationship equation is introduced. The sheet metal is divided into multiple small units, and the unit deformation is calculated using the node force balance equation and the deformation coordination equation. Adaptive meshing technology is used to dynamically adjust the mesh density according to the curvature change and stress concentration of different parts of the sheet metal. Multi-threaded parallel computing technology is used, and the thread creation function provided by the operating system is used to realize the creation, scheduling and management of threads, and synchronization technologies such as lock mechanism and semaphore mechanism are used to ensure data consistency and calculation accuracy.

[0183] Traditional calculation method: using ordinary finite element analysis algorithm, adopting a fixed grid division method, not considering the nonlinear characteristics of the material, and the calculation process is single-threaded.

[0184] Experimental results:

[0185] Compare Projects New approach Traditional methods Average calculation time for aluminum alloy model (minutes) 10.0 20.0 Average calculation time for stainless steel model (minutes) 12.0 25.0 Average calculation time for carbon steel model (minutes) 11.0 23.0 Maximum stress calculation error of aluminum alloy model (MPa) 2.0 7.0 Maximum stress calculation error of stainless steel model (MPa) 2.5 8.0 Maximum stress calculation error of carbon steel model (MPa) 2.2 7.5 Maximum strain calculation error of aluminum alloy model (%) 0.2 0.9 Maximum strain calculation error of stainless steel model (%) 0.3 1.1 Maximum strain calculation error of carbon steel model (%) 0.25 1.0 Complex model mesh number (new method, average) 18000 Fixed 25000 Computing resource utilization (%) 80 45

[0186] 3. Real-time interference detection experiment

[0187] Experimental preparation: In the process of simulated flattening, interference detection is performed on the above 10 models. Interference detection software for the new method and traditional interference detection tools are prepared.

[0188] Experimental steps:

[0189] New method detection: Based on spatial geometric relationships, a three-dimensional spatial coordinate system is constructed, and spatial vector operations are used to determine the position and posture of each part of the sheet metal in the coordinate system. A collision detection algorithm based on bounding boxes (AABB bounding boxes), a fast detection method based on spatial hash tables, and an octree spatial segmentation algorithm are used to perform all-round interference detection on the sheet metal model during the simulated flattening process.

[0190] Traditional detection method: only a simple collision detection algorithm based on bounding box is used, which lacks fast detection and space optimization technology.

[0191] Experimental results:

[0192] Compare Projects New approach Traditional methods Average detection time of simple model (seconds) 1.5 5.0 Average detection time for medium-complexity models (seconds) 2.5 7.0 Average detection time for complex models (seconds) 4.0 10.0 Detection accuracy (%) 98 82 Missed detection rate (%) 2 18 False detection rate (%) 1 6

[0193] 4. Parameter automatic optimization and adjustment experiment

[0194] Experimental preparation: Intentionally set up a variety of interference situations in the model, including unreasonable bending angles, inappropriate bending radius, wrong bending order, and flattening path conflicts, etc., to simulate the complex problems that may arise in actual production. Experiments are conducted on models of different complexity and different material types.

[0195] Experimental steps:

[0196] New method adjustment: When interference is detected, the system automatically starts the parameter adjustment mechanism according to the preset adjustment strategy library. If the interference is caused by the bending angle, the bending angle is adjusted at a ratio of 1°-5° according to the degree of interference (measured by the ratio of the interference volume to the total volume of the sheet metal). If it is a bending radius problem, the bending radius is adjusted in steps of 0.1mm-0.5mm according to the material properties and interference. For the bending sequence, a heuristic search algorithm (such as the A* algorithm) is used to select the optimal sequence from a variety of possible bending sequence combinations. The cost function comprehensively considers the actual cost from the starting point to the node and the estimated cost from the node to the target point. For the flattening path, the path planning Dijkstra algorithm is used for re-planning according to the interference position and the overall shape of the sheet metal, and the algorithm time complexity is optimized using data structures such as priority queues. The adjusted parameters are input into the simulation calculation module, and the simulation calculation is performed again, and the loop is iterated until there is no interference in the simulation flattening result.

[0197] Traditional method adjustment: manual adjustment of parameters based on human experience and judgment, lacking systematic strategies and optimization algorithms.

[0198] Experimental results:

[0199] Compare Projects New approach Traditional methods Average number of iterations for a simple model 2.5 5.0 Average number of iterations for a moderately complex model 3.5 7.0 Average number of iterations for complex models 4.5 9.0 Average adjustment time for simple model (minutes) 5.0 10.0 Average adjustment time for medium-complexity models (minutes) 8.0 16.0 Average adjustment time for complex models (minutes) 10.0 20.0 Adjustment success rate (%) 90 65 Adjusted interference residual rate (%) 10(Residual interference can be ignored) 35(needs secondary adjustment)

[0200] The specific usage of this embodiment is as follows:

[0201] A sheet metal multiple bending and flattening interference method using steps

[0202] Preparation

[0203] Drawing review: Carefully check the sheet metal design drawings. In addition to ensuring that the basic information such as initial size, bend angle, radius and sequence is included, check whether the annotations are clear and whether there are any contradictions or omissions. For example, the dimensions between different views must be consistent, and the annotations of the bend angles should clearly indicate the start and end positions.

[0204] Material confirmation: clarify the materials used for sheet metal, consult the corresponding material manual or contact the supplier to obtain key mechanical performance parameters such as elastic modulus and yield strength to prepare for subsequent simulation calculations.

[0205] Build sheet metal models

[0206] Software startup and settings: Open professional 3D modeling software such as SolidWorks, AutoCAD, etc., and set the drawing unit to millimeters according to the prompts on the software interface to ensure the accuracy of subsequent dimension entry.

[0207] Vertex positioning: Find the coordinate positioning function in the software, switch to the Cartesian coordinate system mode, and enter the absolute coordinate values ​​of each vertex of the sheet metal one by one according to the coordinate data on the drawing. Each time a vertex is entered, the software's preview function can be used to confirm whether the position is accurate.

[0208] Dimension entry: Use the software's dimensioning tool to select the corresponding geometric edge or feature, and accurately enter the length, width, thickness and other dimension parameters in the pop-up input box. After the entry is completed, the software will usually automatically display the annotation, and check again whether the annotation value is consistent with the drawing.

[0209] Determine the bending angle: Use the software's built-in angle measurement tool. In the 2D view, click on the two points that determine the bending angle. The software will automatically calculate the angle value based on the inverse tangent function. To achieve an input accurate to 0.1°, find the numerical calculation accuracy settings in the software's settings and enable the high-precision calculation mode.

[0210] Determine the bending radius: According to the bending radius design requirements on the drawing, construct a right triangle auxiliary line in the software, and use the Pythagorean theorem and known geometric dimensions and angle relationships to calculate the coordinates of the relevant points. In the parameter input box of the software, enter the calculated bending radius accurately to two decimal places.

[0211] Analog calculation

[0212] Algorithm selection: In the simulation calculation module, find the simulation calculation algorithm option based on geometric transformation and material mechanics principles, and click to select it.

[0213] Parameter input: In the pop-up parameter input interface, accurately fill in the elastic modulus, yield strength and other performance parameters of the sheet metal material obtained previously. Note that the unit of the parameter must be consistent with the unit required by the software, such as the common unit of elastic modulus is GPa, and the unit of yield strength is MPa.

[0214] Bending sequence setting: According to the bending sequence specified on the drawing, enter the sequence number of each bend and the corresponding bending parameters, such as bending angle, radius, etc. in the bending sequence setting column of the software.

[0215] Calculation process monitoring: After clicking the calculation button, the software will display a calculation progress bar. At this time, you can observe whether there is an error message pop-up during the calculation process. If an error message appears, it may be due to incorrect parameter input or a problem with the model, and you need to return to check.

[0216] Interference Detection

[0217] Coordinate system construction: In the simulation flattening operation interface, find the function button for constructing a three-dimensional space coordinate system. After clicking it, the software will automatically construct a coordinate system based on the sheet metal model.

[0218] Vector operations and position determination: The software automatically uses space vector operations to determine the position and posture of each part of the sheet metal in the coordinate system. Users can rotate and zoom the view to observe from different angles whether the position and posture of the sheet metal meet expectations.

[0219] Interference detection settings: In the interference detection function settings, select the collision detection algorithm based on the bounding box (such as AABB bounding box), and set the relevant parameters of the fast spatial search algorithm, such as search range, accuracy, etc. At the same time, enter the preset minimum safety distance value, which needs to be determined comprehensively based on factors such as the material properties of the sheet metal, processing requirements, and actual application scenarios.

[0220] Detection and result viewing: Click the Start Detection button, and the software will detect the distance between the various parts of the sheet metal in real time and compare it with the minimum safe distance. After the detection is completed, the software will pop up the interference detection result window to show whether there is interference. If there is interference, the interference position will be marked.

[0221] Parameter adjustment and optimization

[0222] Interference cause analysis: Determine the cause of interference based on the interference detection results. If it is a bending angle problem, check the difference between the bending angle at the interference position and the design value; if it is a bending radius problem, check whether the bending radius meets the bending springback characteristics of the material.

[0223] Parameter adjustment operation:

[0224] Bending angle adjustment: In the parameter adjustment interface, find the bending angle adjustment option. According to the degree of interference (measured by the ratio of interference volume to the total volume of sheet metal), fill in the corresponding adjustment coefficient in the calculation formula input box provided by the software. The software will automatically calculate and update the bending angle value.

[0225] Bending radius adjustment: For bending radius problems, in the parameter adjustment bar, manually enter a new bending radius value in steps of 0.1mm-0.5mm according to material properties and interference conditions, or click the fine-tuning button provided by the software to make adjustments.

[0226] Bending order adjustment: In the bending order adjustment interface, the software will list a variety of possible bending order combinations. Users can choose to use the heuristic search algorithm, and the software will automatically calculate and select the optimal order; users can also manually drag the items in the bending order list to customize the bending order.

[0227] Flattening path adjustment: In the flattening path adjustment function, the software will automatically replan the flattening path based on the interference position and the overall shape of the sheet metal using the path planning Dijkstra algorithm. Users can view the schematic diagram of the new path and manually fine-tune the path control points if there are special needs.

[0228] Recalculation and iteration: After the adjustment is completed, click the Recalculation button, and the software will use the adjusted parameters to perform simulation calculations and interference detection again, and repeat until there is no interference in the simulation flattening result. After each iteration, you can view the calculation results and interference detection report to analyze the effect of parameter adjustment.

[0229] Instructions for use of a sheet metal multiple bending and flattening system

[0230] Model Building Module

[0231] Interface familiarity: Open the system's model building module, first familiarize yourself with the interface layout, and understand the location and functions of interactive components such as text boxes, drop-down menus, and sliders. For example, text boxes are used to enter numerical parameters, drop-down menus are used to select preset options, and sliders can be used to adjust continuously changing parameters.

[0232] Drawing import: Click the Import Drawing button, and in the pop-up file selection window, find the sheet metal design drawing in DWG, DXF and other formats, select it and click Open. The system will automatically convert it through the format conversion interface using a specific format conversion function library, and clean and verify the data during the conversion process. If the conversion fails, a prompt box will pop up to display the cause of the error. Common reasons include incompatible drawing formats, file damage, etc. At this time, you need to check the drawing or replace it with a drawing in the correct format.

[0233] Parameter entry and verification: In the parameter entry area, follow the prompts to enter various parameters, such as the size, angle, radius, etc. of the sheet metal. After the entry is completed, the system will automatically perform digital range verification, data type verification, unit verification, etc. If the parameters do not meet the requirements, the system will prompt an error message in red font, such as "Input value out of range" and "Data type error", etc. The user needs to modify the parameters according to the prompts.

[0234] Model verification and correction: Click the model verification button, and the system will compare and verify the input model with the standard model through the model comparison algorithm based on curvature features. If a difference is found, the system will pop up a prompt box, and the user can choose manual correction to directly modify the model parameters in the model editing interface; or choose automatic correction, and the system will automatically adjust the model parameters through a correction strategy based on an optimization algorithm. After the correction is completed, it will be verified again until the model meets the requirements.

[0235] Analog computing module

[0236] Computing engine and technology selection: After entering the simulation computing module, in the computing engine selection column, select the built-in high-performance computing engine and check the multi-threaded parallel computing technology option to use the computing resources of multi-core processors to increase the computing speed. At the same time, you can set the number of threads and task allocation strategy according to actual needs.

[0237] Algorithm and genetic operation selection: In the simulation algorithm selection area, select the integrated advanced bending and flattening simulation algorithm and combine it with the genetic algorithm. In the genetic algorithm settings, you can select the roulette selection function, single point crossover function, basic bit mutation function, etc., and set the corresponding parameters, such as selection probability, crossover probability, mutation probability, etc. These parameters will affect the search effect and convergence speed of the genetic algorithm.

[0238] Simulation result display and control: After the calculation is completed, click the result display button, and the system will use the professional graphics library and graphics drawing functions to display the simulation process in the form of 3D animation. In the display interface, users can use the time control function to play, pause, fast forward, rewind and other operations, and click the lighting model, texture mapping and other buttons to enhance the visualization effect and observe the bending and flattening process of sheet metal from different angles.

[0239] Interference Detection Module

[0240] Octree construction and parameter setting: After the simulation calculation is completed, the system automatically enters the interference detection module. In the octree construction setting area, enter the geometric size and spatial range of the sheet metal model and other parameters. The system will build and operate the octree structure through the octree node creation function, insertion function, query function, etc. according to these parameters, and reasonably determine the depth and node size of the octree. If the construction fails, it may be that the input parameters are unreasonable and the parameters need to be readjusted.

[0241] Interference detection and report generation: Click the Start Detection button, and the system will combine efficient calculation logic to perform a full range of interference detection on the sheet metal model during the simulated flattening process. After the detection is completed, the system will generate a detailed interference report, which includes the interference position (displayed in coordinate form), type (point interference, line interference or surface interference), severity assessment (quantitative assessment through interference index function) and possible solutions. Users can view and download the report in the report viewing interface.

[0242] Parameter adjustment module

[0243] Strategy selection and adjustment: When the interference detection module detects interference, the parameter adjustment module will be automatically triggered. In the decision tree algorithm interface, the system will automatically select the appropriate adjustment strategy from the strategy library according to the type and degree of interference, and display it in the strategy display area. Users can choose to automatically apply the strategy or manually intervene in the parameter adjustment process. When manually intervening, adjust the parameters in the parameter adjustment input box based on your own experience and judgment through the user interface interaction function.

[0244] Parameter feedback and recalculation: After the adjustment is completed, click the confirmation button, and the system will feed back the adjusted parameters to the simulation calculation module in real time through the data interface. After receiving the parameters, the simulation calculation module will automatically re-calculate the simulation, and the user can view the calculation progress and results in the simulation calculation progress interface.

[0245] A guide for using a sheet metal multiple bending and flattening equipment

[0246] Hardware Preparation

[0247] Processor selection: According to the needs of multi-threaded parallel computing, choose a high-performance multi-core processor, such as Intel Core i7 or AMD Ryzen 7 series. When purchasing a processor, pay attention to the number of cores, number of threads, main frequency and other parameters of the processor. The more cores and threads, the more it can meet the needs of multi-threaded computing; the higher the main frequency, the faster the computing speed.

[0248] Memory configuration: Ensure that the memory is not less than 16GB. You can choose DDR43200MHz and above memory sticks to ensure smooth operation of the system and software. When installing the memory stick, pay attention to the slot position and installation direction of the memory stick to avoid reverse insertion or poor contact.

[0249] Storage device selection: Equipped with a solid-state hard drive of 512GB or more, such as Samsung 980PRO, Western Digital SN850, etc., for storing programs and data. The read and write speed of a solid-state hard drive is much faster than that of a traditional mechanical hard drive, which can greatly improve the efficiency of data storage and reading. When installing a solid-state hard drive, pay attention to connecting the data cable and power cable to ensure that it is firmly installed.

[0250] Software Installation

[0251] Obtain the installation program: Download the software program installation package for the sheet metal multiple bending and flattening interference method from the official software website or authorized channels. Note that the downloaded installation package version must be compatible with the device's operating system.

[0252] Installation process: Double-click the installation package and follow the installation wizard provided by the software. During the installation process, select the appropriate installation path. It is recommended to select a disk partition other than the system disk to avoid occupying too much system disk space and affecting system performance. At the same time, pay attention to checking or unchecking related installation options, such as whether to create a desktop shortcut, whether to install dependencies, etc.

[0253] Dependency installation: Some software may require the installation of additional dependencies, such as runtime libraries, drivers, etc. The installation wizard will prompt the user to install these dependencies. The user needs to follow the prompts to ensure that the dependencies are fully installed. After the installation is complete, you can run the software for testing. If the software does not run properly, it may be that the dependencies are not fully installed or there is a conflict. You need to check and reinstall.

[0254] Run Operation

[0255] Software startup: Turn on the device, find the installed software program icon on the desktop or in the start menu, and double-click to start the software. During the software startup process, a loading interface may be displayed. Wait for the loading to complete.

[0256] Operation process: In the software interface, follow the above method and system usage steps to perform operations such as sheet metal model establishment, simulation calculation, interference detection and parameter adjustment. During the operation, pay attention to the software prompt information and calculation results. If there are any abnormal situations, such as software crashes, abnormal calculation results, etc., check whether the input parameters are correct and whether the equipment hardware is working properly.

[0257] Troubleshooting: If the software runs slowly, it may be due to insufficient device hardware performance or too many programs running at the same time. You can close unnecessary programs or upgrade the device hardware. If the calculation results are not as expected, you need to check whether the model is established accurately, whether the parameters are input correctly, and whether the simulation algorithm is appropriate.

[0258] A method for using a storage medium for flattening sheet metal multiple bending

[0259] Storage Content

[0260] Program storage: The storage medium is mainly used to store computer programs that implement the interference method of multiple bending and flattening of sheet metal, including program codes of various functional modules such as model establishment, simulation calculation, interference detection, and parameter adjustment.

[0261] Data storage: The sheet metal model data, simulation calculation results, interference detection reports and other data generated during use are also stored in the storage medium. For example, the sheet metal shape, stress and strain distribution data after each simulation calculation, and the interference location, type, severity and other information in the interference detection report.

[0262] Data storage and reading

[0263] Automatic storage and reading: When the software program is running, it will automatically read the program and related data from the storage medium for processing. New data generated during the processing, such as adjusted parameters, simulation calculation results, etc., will be automatically stored in the storage medium. The read and write speed of the storage medium will affect the running efficiency of the software. If the read and write speed is too slow, it may cause the software to run slowly.

[0264] Manual reading and viewing: Users can also manually read and view relevant data from the storage medium as needed. For example, in the data management interface of the software, select the historical data query function to view previous simulation calculation results, interference detection reports, etc., and analyze interference conditions and parameter adjustment effects.

[0265] Data Management

[0266] Data backup: To ensure data security and integrity, data in storage media must be backed up regularly. You can use external storage devices, such as mobile hard disks, USB flash drives, etc., to copy important data to external storage devices. You can also use cloud storage services to upload data to the cloud for backup, set up automatic backup plans, and ensure real-time data backup.

[0267] Data encryption: Data encryption technology, such as the AES encryption algorithm, can be used to encrypt sensitive data in storage media to prevent data leakage. In the data encryption settings of the software, set the encryption key and encryption method to ensure that the encrypted data cannot be accessed by unauthorized users.

[0268] Data cleaning: Useless data, such as expired simulation calculation results, old interference detection reports with resolved interference problems, etc., can be cleaned up in time to free up storage media space. In the data management interface of the software, select the data cleaning function, check the data to be cleaned up, and click the delete button.

Claims

1. A sheet metal multiple bending and flattening interference method, characterized in that: The following steps are involved: Step 1: Create an accurate sheet metal model: Using professional 3D modeling software, through the built-in coordinate positioning function of the software, based on the Cartesian coordinate system, the vertices of the sheet metal are positioned in the form of absolute coordinate values. The coordinate positioning function can be expressed as P (x, y, z), where x, y, and z represent the position values ​​in the directions of the three coordinate axes respectively; combined with the dimensioning function, according to the sheet metal design drawings, the initial length L, width W, and thickness t dimensional parameters of the sheet metal are accurately entered to the millimeter level; During the input process, the accuracy of the dimensions is ensured through the precision control mechanism provided by the software; For the bending angle θ, use the angle measurement tool provided by the software based on the inverse tangent function To determine, (x1, y1) and (x2, y2) are the coordinate values ​​of the two points that determine the angle on the two-dimensional plane; to achieve accurate input to 0.1°, the software uses a high-precision numerical calculation method to accurately convert the calculation results of the inverse tangent function; for the bending radius R, combined with the design requirements, the coordinates of the relevant points are determined based on the Pythagorean theorem and complex geometric relationships; when determining the bending radius, a right triangle is constructed and the Pythagorean theorem is used. 2 +b 2 =c 2 , combined with known geometric dimensions and angle relationships, calculate the coordinates of the relevant points and finally enter them accurately to two decimal places; After completing the parameter entry, use the model verification tool that comes with the software, based on the geometric shape matching algorithm; this algorithm extracts the shape context features of the model and compares the shape context feature vector of the entered model with the standard shape context feature library; the standard shape context feature library is established by extracting and analyzing features of a large number of standard sheet metal models; in the comparison process, calculate the similarity between the two, set the similarity threshold, if it is lower than the threshold, it is determined that the model has geometric shape or dimensional accuracy problems, and further compare it with the actual sheet metal design document to ensure that the model is highly consistent with the actual sheet metal features; Step 2: Simulate bending and flattening calculation: Adopt an algorithm based on geometric transformation and material mechanics principles; in geometric transformation, use the rigid body transformation matrix To describe the spatial position change of the sheet metal, R is the rotation matrix, which is used to describe the rotation angle and direction of the sheet metal, and t is the translation vector, which is used to describe the translation distance and direction of the sheet metal in space; combined with the elastic modulus E and yield strength σ of the sheet metal material y As performance parameters, these parameters are obtained through material testing experiments, and the experimental process follows the relevant material testing standards; According to the set bending sequence, starting from the first bending, the shape of the sheet metal after each bending is calculated step by step according to the algorithm rules; the elastic deformation and plastic deformation process of the material under the bending force is simulated according to the law σ=Eε, where σ is stress and ε is strain; in the calculation process, the nonlinear characteristics of the material are considered, and the constitutive relationship equation of the material is introduced to simulate the deformation of the material more accurately; During the calculation process, the sheet metal is divided into multiple small units, and the force analysis of each unit is performed using the node force balance equation. where F i is the force acting on node i; the deformation of the unit is calculated by analyzing and calculating the node force and combining the mechanical properties of the material; at the same time, the deformation coordination equation is used to ensure the continuity and coordination of the deformation between adjacent units; detailed bending process data is generated, including the displacement u, stress σ, strain ε and other information of each unit. The displacement calculation is based on the virtual work principle ∫ V σ ij δε ij dV=∫ S T i δu i dS, where V is the unit volume, S is the unit surface area, T is the surface force, and u i is displacement; In the calculation process, the numerical integration method is used to solve the above integral equation; Step 3: Real-time interference detection: In the simulation flattening process, based on the spatial geometric relationship, a three-dimensional spatial coordinate system is constructed, and the position and posture of each part of the sheet metal in the coordinate system are determined by spatial vector operations; To determine the plane normal vector, is the unit vector of the coordinate axis, a x 、a y 、a z and b x , b y , b z Respectively, vector and Components on the three coordinate axes; Step 4: Determine the criteria based on the preset boundary conditions: Set the minimum safe distance d between the parts of the sheet metal min , d min The value of is determined based on the material properties of the sheet metal, processing requirements, and actual application scenarios. Apply step three; The collision detection algorithm based on the bounding box, AABB bounding box, its judgment condition is A and B are two bounding boxes. During the detection process, the spatial position of each tiny unit in the flattening process is analyzed in real time, and the detection efficiency is improved through the kd-tree fast spatial search algorithm. Once it is detected that the distance d between the units is less than the minimum safety distance d min , if it is determined that there is interference, the three-dimensional coordinates of the interference position and the interference type, such as point interference, line interference or surface interference, are immediately recorded; when recording the interference position, a high-precision coordinate recording method is used to ensure the accuracy of the position information; the interference type is determined by analyzing the geometric characteristics of the interference area and combining the preset interference type judgment rules; Step 5: Automatic optimization and adjustment of parameters: When interference is detected, the system automatically starts the parameter adjustment mechanism according to the preset adjustment strategy library; the adjustment strategy library is established through the analysis and summary of a large amount of sheet metal processing experimental data combined with theoretical research results; If the interference is caused by the bending angle, the bending angle is adjusted according to a certain proportion based on the degree of interference. The degree of interference is expressed by the interference volume V inter and the total volume of the sheet metal V total The adjustment formula is Where α is the adjustment coefficient, and the value of α is determined according to the material properties of the sheet metal, the interference type, and the actual processing experience; if it is a bending radius problem, the bending radius is adjusted in steps of 0.1mm-0.5mm according to the material properties and interference conditions, and the adjustment is based on the bending rebound formula of the material This formula takes into account the elastic modulus E, yield strength σ y , thickness t and original bending radius R old Effect on bending springback; For the bending sequence, a heuristic search algorithm is used to select the optimal sequence from multiple possible bending sequence combinations. The cost function f(n) = g(n) + h(n), where g(n) is the actual cost from the starting point to node n, and h(n) is the estimated cost from node n to the target point. When calculating h(n), a heuristic function based on geometric distance and interference risk assessment is used to improve the search efficiency. For the flattening path, the path planning Dijkstra algorithm is used to re-plan according to the interference position and the overall shape of the sheet metal. The Dijkstra algorithm finds the optimal path by continuously updating the shortest distance from the node to the source point. In the implementation process, data structures such as priority queues are used to optimize the time complexity of the algorithm. Input the adjusted parameters into the simulation calculation module, re-perform the simulation calculation, and repeat the cycle until there is no interference in the simulation flattening result; During the iteration process, the parameter adjustment information and simulation calculation results of each iteration are recorded for analysis and optimization.

2. The sheet metal multiple bending and flattening interference method according to claim 1, characterized in that: In the step 1, in the step of establishing an accurate sheet metal model, the model is also lightweighted by removing redundant geometric elements in the model, using a feature recognition algorithm to automatically identify and mark these redundant elements, and then deleting or simplifying them.

3. The sheet metal multiple bending and flattening interference method according to claim 1, characterized in that: In the step 2, in the simulation bending and flattening calculation step, an adaptive meshing technology is used for complex sheet metal shapes; the density of the mesh is dynamically adjusted according to the curvature changes and stress concentration conditions of different parts of the sheet metal; in areas with large curvature or stress concentration, the mesh is automatically encrypted to improve the calculation accuracy; in relatively flat areas, the mesh density is appropriately relaxed to reduce the amount of calculation.

4. The sheet metal multiple bending and flattening interference method according to claim 1, characterized in that: In the step three: in the real-time interference detection step, in addition to the collision detection algorithm based on the bounding box, a fast detection method based on the spatial hash table is also introduced; the three-dimensional space is divided into a plurality of uniform grid units, and a spatial hash table is constructed.

5. The sheet metal multiple bending and flattening interference method according to claim 1, characterized in that: In the step 5: in the parameter automatic optimization adjustment step, when the interference cannot be eliminated after adjusting the parameters for multiple times, the expert system is started; The expert system is based on a large number of actual cases and domain knowledge. Through the reasoning engine, it analyzes interference situations and provides targeted solutions.

6. A sheet metal multiple bending and flattening system, characterized in that: The following steps are involved: Step 1: Model building module: Equipped with professional 3D modeling tools, it supports the import of drawings in various industrial standard formats; through the format conversion interface, it uses a specific format conversion function library, which is based on the syntax and semantic analysis of files in various formats; it accurately converts drawings in different formats into model data that can be recognized by the system. During the conversion process, the data is cleaned and verified to ensure the integrity and accuracy of the data; it provides an intuitive parameter entry interface, using interactive components such as text boxes, drop-down menus, sliders, etc. to facilitate users to enter various parameters; After the parameters are entered, the digital range check function checkRange(value, min, max) is used to check whether the input value is within a reasonable range. At the same time, multiple verification methods such as data type verification and unit verification are provided to ensure the correctness of the entered parameters. It has the function of model verification and correction, through the model comparison algorithm based on curvature features; the algorithm extracts the curvature features of the input model and the standard model, constructs the curvature feature vector, and calculates the difference between the two; If a difference is found, the user will be automatically prompted to make corrections. Both manual and automatic corrections are supported. Automatic corrections use a correction strategy based on an optimization algorithm to adjust model parameters by minimizing the model difference function to achieve automatic correction of the model. Step 2: Simulate the calculation module: Built-in high-performance computing engine, using multi-threaded parallel computing technology, and utilizing the computing resources of multi-core processors; through the thread creation function provided by the operating system, thread creation, scheduling and management are realized; in the multi-threaded computing process, synchronization technologies such as lock mechanism and semaphore mechanism are used to ensure data consistency and calculation accuracy; improve computing speed, and give full play to the performance advantages of multi-core processors through reasonable task division and load balancing strategies; Integrate advanced bending and flattening simulation algorithms and combine them with genetic algorithms; in genetic algorithms, the roulette wheel selection function is used to select individuals based on their fitness values, and individuals with high fitness values ​​have a greater probability of being selected into the next generation; the single-point crossover function exchanges genes between selected individuals to generate new individuals; The basic mutation function randomly mutates the genes of individuals to increase the diversity of the population; Generate visual simulation results of bending and flattening process, and use professional graphics library and graphics drawing function to display the simulation process in 3D animation form; During the display process, the illumination model, texture mapping and other technologies are used to enhance the visualization effect, and the playback control is achieved through the time control function; Step 3: Interference detection module: Based on advanced spatial analysis algorithms, the octree structure is constructed and operated through octree node creation functions, insertion functions, query functions, etc. When constructing the octree, the depth and node size of the octree are reasonably determined according to the geometric size and spatial range of the sheet metal model; combined with efficient calculation logic, all-round interference detection is performed on the sheet metal model during the simulation flattening process; It can quickly and accurately identify various types of interference through feature extraction and pattern matching algorithms. The feature extraction algorithm analyzes the scale space of the image to extract key points and their feature descriptors; the template matching algorithm matches the extracted features with the preset interference template and distinguishes different types of interference based on the matching results. Provide a detailed interference report, including the interference location, type, severity assessment and possible solutions. The severity assessment is quantitatively evaluated through the interference index function, which comprehensively considers the interference volume, area, location and other factors, and obtains the severity value of the interference through certain weight distribution and calculation rules. According to the severity value, provide corresponding solutions. Step 4: Parameter adjustment module: It has an intelligent parameter adjustment strategy library, which stores a variety of adjustment strategies for different interference situations. The strategies are established based on a large amount of experimental data and theoretical analysis. Through the strategy generation function, the strategies are generated according to the experimental data and theoretical models. When generating strategies, the decision tree algorithm is used to analyze and mine the experimental data to establish the mapping relationship between the interference situation and the adjustment strategy. When interference is detected, the decision tree algorithm automatically selects the appropriate adjustment strategy from the strategy library according to the type and degree of interference; the decision tree algorithm classifies and makes decisions on interference situations by building a decision tree model; The relevant parameters can be adjusted accurately, and the adjusted parameters can be fed back to the simulation calculation module in real time through the data interface. The system supports manual intervention and automatic optimization of the parameter adjustment process. Manual intervention is achieved through the user interface interactive function, and users can adjust the parameters according to their own experience and judgment. Automatic optimization is achieved through the optimization algorithm, and the simulation flattening results are optimized by continuously optimizing the parameters.

7. The sheet metal multiple bending and flattening system according to claim 6, characterized in that: In the step 1, the model building module also has the function of interacting with an external database, and can obtain standard sheet metal model library and material property library information from the database.

8. The sheet metal multiple bending and flattening system according to claim 6, characterized in that: In the step 2: during the calculation process, the simulation calculation module records the intermediate data and calculation status of the calculation process in real time.

9. A sheet metal multiple bending and flattening device, which implements the sheet metal multiple bending and flattening interference method according to any one of claims 1 to 5 by running a specific program, characterized in that: Software algorithm optimization: Intelligent task scheduling system: In this system, the agent is the task scheduler, and the environment is the computing resource status and task queue of the device; the state space includes the task priority, resource requirements, execution progress and real-time status of system resources; the action space covers the task start, pause, continue and resource allocation adjustment; Reward function design: In the reward function R = α × task completion rate + β × resource utilization rate + γ × computing efficiency improvement rate, the task completion rate refers to the ratio of the number of tasks successfully completed within a certain period of time to the total number of tasks; the resource utilization rate is measured by calculating the ratio of used resources to total resources; the computing efficiency improvement rate is determined by comparing the current task execution time with the historical average execution time; α, β, and γ are weight coefficients that can be flexibly adjusted according to the characteristics and needs of the actual sheet metal processing tasks; for sheet metal processing tasks with tight time, the weight of α can be appropriately increased to prioritize the completion of tasks on time; for equipment with limited resources, the weight of β can be increased to optimize resource utilization; Learning and optimization process: The intelligent task scheduling system continuously interacts with the environment. Based on the reward feedback obtained for each action, the agent will refer to the Q value table when selecting an action in each state. The Q value table records the expected cumulative reward for each action performed in each state. After each interaction, the agent selects the action based on the actual reward and the maximum Q value of the next state according to the formula Where s is the current state, a is the current action, r is the reward, s′ is the next state, α is the learning rate, γ is the discount factor, and the Q value is updated to gradually learn the optimal scheduling strategy to achieve dynamic and accurate allocation of computing resources, ensuring that the entire computing process is efficient and stable; Adaptive numerical solution algorithm: The finite element method discretizes the continuous solution domain into a combination of finite units, and obtains an approximate solution to the entire solution domain through equations; the error estimate of unit e is Where Ω e is the area of ​​unit e, η h represents the stress of the finite element solution; Grid adaptive adjustment: According to the error estimate, the grid is adaptively adjusted; when η e When the error exceeds the preset threshold, it indicates that the calculation error of the unit is large, and the unit and its adjacent units need to be meshed and further subdivided using the dichotomy method. e When it is much smaller than the error threshold, it means that the calculation accuracy of the unit is high and the grid density can be appropriately relaxed; Intelligent error detection and repair module: In this module, the prior probability P(C i ) is an estimate of the probability of input data errors, improper algorithm parameter settings, and rounding errors in the calculation process based on historical data and domain knowledge; Likelihood probability (P(E|C i ) indicates that in some error reason C i The probability of data anomaly or calculation error event E occurring if it occurs; The posterior probability P(C i |E) is after observing event E, the error cause C i Probability update of ; Model construction and parameter determination: Through learning and analysis of massive normal data, a normal state model of data and calculation results is constructed; for the input sheet metal model data, its reasonable value range, the correlation between data and other parameters are determined; For the calculation results, determine the normal stress and strain ranges, etc.; These parameters serve as prior knowledge and are used for subsequent error diagnosis; Error diagnosis and repair: When data anomalies or calculation errors are detected, use the formula Calculate the posterior probabilities of different error causes; find the error cause with the highest posterior probability and automatically try to fix it; if the input data is wrong, correct it according to the normal range and correlation of the data; if the algorithm parameters are set improperly, readjust the parameters according to historical experience and optimization algorithms, and then recalculate to ensure the reliability of the calculation results; Data management and interaction: Efficient data storage and indexing system: When data needs to be stored, the key value of the data is first calculated to obtain a value, and then the key value is searched clockwise on the ring to find the first node that is greater than or equal to the value, and the data is stored in the node; B+ tree index; in a B+ tree, the time complexity of insert, delete, and query operations is O(log n m), where n is the branching factor of the node, that is, the maximum number of child nodes each node contains, and m is the number of data; Data communication mechanism: Communicate with the back-end server with the help of AJAX technology; AJAX allows asynchronous data exchange with the server without reloading the entire page; by creating an XMLHttpRequest object, setting the request URL, request method, request header and request body, the user input data is sent to the back-end server; after the server processes the request, it returns the response data, and the front-end updates the page content by parsing the response data.

10. A storage medium for sheet metal multiple bending and flattening, wherein when the program is executed by a processor, the steps of the sheet metal multiple bending and flattening interference method according to any one of claims 1 to 5 are implemented; characterized in that: Data Storage and Security: Advanced encryption storage technology: When encrypting, first divide the plaintext into groups, each group is 128 bits long; the plaintext group is P i , the previous ciphertext is C i-1 , the encryption key is K, then the i-th group of ciphertext C i The calculation method is Where E K is the encryption function, The encryption function encrypts the input data through a series of byte replacement, row shift, column mixing and key addition operations; when the data is written to the storage medium, the data is automatically encrypted and stored; Key management: To ensure the security of keys, a key function is used to derive multiple subkeys based on the master key input by the user, which are used for different data encryption and decryption operations. At the same time, a key encryption key mechanism is used to encrypt and store the derived subkeys; Data redundancy and recovery mechanism: The original data blocks are D1, D2, ..., D k , generate nk redundant check blocks P1, P2, ..., P by encoding n-k , where n is the total number of data blocks, the original data blocks can be regarded as coefficients of polynomials, and redundant check blocks can be generated through a specific coding algorithm; When some data blocks are lost or damaged; assuming that the lost data block index is i1, i2, ..., i m , then the restored data block It can be expressed as Where D l It is the data that is not lost; Data integrity check: Perform integrity check on stored data regularly, use hash algorithm to calculate the hash value of data blocks, and compare the calculated hash value with the hash value pre-saved in the storage medium. If the two are consistent, the data is complete; if they are inconsistent, it means that the data may be tampered with or damaged, and it is repaired or backed up in time; Data transfer and compatibility: High-speed data transmission protocol: Adopt a custom transmission protocol based on UDP, and add data verification, retransmission mechanism and flow control functions on the basis of UDP; data verification adopts CRC algorithm to generate a check code attached to the data frame; The CRC algorithm performs polynomial division on the data frame to obtain a check code of fixed length. After receiving the data frame, the receiver also performs CRC calculation and compares the calculated check code with the received check code. If they are inconsistent, it means that an error occurred during the data transmission process and a retransmission request is sent. Retransmission mechanism and flow control: The retransmission mechanism adopts a combination of timeout retransmission and fast retransmission; After sending a data frame, the sender starts a timer. If no confirmation response is received from the receiver before the timer expires, the data frame is retransmitted. If multiple duplicate ACKs are received in a short period of time, it means that data frames may be lost. The fast retransmission mechanism is used to immediately retransmit the lost data frames. Flow control is implemented through a sliding window mechanism. The sender dynamically adjusts the sending window size based on the window size fed back by the receiver to ensure stable data transmission and avoid overflow of the receiver's buffer caused by the sender sending data too quickly. By optimizing the encoding and decoding algorithms and the data transmission link, the actual data transmission speed is increased by 3-5 times compared with traditional transmission protocols. Good compatibility: Cross-platform driver development: By writing cross-platform device drivers and interface libraries, seamless integration with various types of devices and operating systems can be achieved; for different operating systems, corresponding development frameworks are used to write drivers; for Windows systems, Windows Driver Kit is used to write device drivers, which communicate with devices through interfaces such as USB and Ethernet; WDK provides a series of tools and libraries to help developers write drivers that comply with Windows operating system specifications and implement functions such as device identification, initialization, and data transmission; for Linux systems, driver development frameworks provided by the Linux kernel are used to write drivers, such as the character device driver framework and the block device driver framework; the Linux kernel provides a wealth of functions and interfaces to facilitate developers to implement various functions of device drivers; for macOS systems, the I / OKit framework provided by the system is used to develop drivers; the I / OKit framework provides an object-oriented programming interface for developing various types of device drivers to achieve communication and control with devices; these measures make it convenient for users to use in different working environments.

Images